ORIGINAL ARTICLE

Evaluation of facial hard and soft tissue

asymmetry using cone-beam computed
tomography
€ lin Arunc
Rahime Burcu Nur,a Derya Germeç Çakan,b and Tu
Istanbul, Turkey

Introduction: The purposes of this study were to (1) evaluate facial asymmetry 3 dimensionally using cone-
beam computed tomography (CBCT) and (2) compare the right and left facial hard and soft tissues
volumetrically and their interferences on each other. Methods: The CBCT data of 49 asymmetric (soft tissue
menton deviation, $4 mm; distance from the facial midline) (mean age, 19.9 6 5.6 years) and 39 symmetric pa-
tients (soft tissue menton deviation, \4 mm) (mean age, 17.8 6 5.5 years) were exported to the MIMICS soft-
ware program (version 13.0; Materialise, Leuven, Belgium). Linear, surface distance, angular, volumetric, and
surface area measurements were performed 3 dimensionally to assess and compare intragroup and intergroup
differences. Results: In the asymmetry group, linear measurements such as ramus height, mandibular effective
and corpus length, and absolute mandibular volumetric measurements were significantly decreased (P\0.001),
whereas facial mandibular, and soft and hard tissue volumetric measurements made on 3-dimensional images,
and linear measurements on 2-dimensional images were increased (P \0.001) on the deviation side.
Conclusions: Facial hard and soft tissue asymmetries can be precisely quantified using CBCT. However,
especially in the gonial region where the surface topography shows alterations caused by asymmetry, many
anatomic landmarks should be chosen for the assessment of asymmetry. At the gonial level, the compensation
of the soft tissues for the hard tissues was found on 2-dimensional images; nevertheless, 3-dimensional right
and left volumetric soft tissue evaluations provide evidence for asymmetry. (Am J Orthod Dentofacial Orthop
2016;149:225-37)

I
n positive interpersonal communications, the first
important domain is the appearance or beauty of
the face. Symmetric faces are perceived as more
attractive and healthy, serving as a rapid indicator of
higher genetic quality.1 Consequently, people critically
evaluate their own faces and wish to correct facial dis-
crepancies that they consider unacceptable. Facial asym-
metry has been determined to be the reason for patients
to seek orthodontic evaluation in 25% of cases.2
The human face is not always symmetric over the
facial midline. Asymmetries and deviations within limits
From the Department of Orthodontics, Faculty of Dentistry, Yeditepe University,
Istanbul, Turkey.
a
Teaching assistant.
b
Associate professor.
c
Professor; private practice, Istanbul, Turkey.
All authors have completed and submitted the ICMJE Form for Disclosure of Po-
tential Conflicts of Interest, and none were reported.
Address correspondence to: Rahime Burcu Nur, Department of Orthodontics,
Faculty of Dentistry, Yeditepe University, Bagdat Cad. No: 238, G€
oztepe, Istanbul
34728, Turkey; e-mail, drburcunur@gmail.com.
Submitted, August 2014; revised and accepted, July 2015.
0889-5406/$36.00
Copyright Ó 2016 by the American Association of Orthodontists.
http://dx.doi.org/10.1016/j.ajodo.2015.07.038
(fundamental asymmetries) are recognized as normal.3
However, severe or pathologic asymmetry of the cranio-
facial complex affecting teeth, soft tissues, and skeletal
units is not acceptable and needs combined orthodontic
and surgical treatment.4
Facial asymmetry is more frequently identified in the
lower third of the face. Severt and Proffit5 demonstrated
that facial asymmetry affects the upper face in only 5%,
the midface in 36%, and the lower face in 74% of the
examined subjects. The fact that the mandible is a mobile
bone that grows over a longer period than the maxilla has
been reported as the reason for the increased frequency of
lower facial asymmetry.6,7 Mandibular asymmetry may
occur due to right and left condylar or ramal vertical
dimensional discrepancies, differences between the
corpus lengths of the 2 sides or a rotational and
deviated position of the mandible. Although asymmetry
is more significant in the mandible, there can also be
different levels of asymmetry in the maxilla.
Facial appearance is basically composed of both the
hard and soft tissues (muscles, skin, and fat) that make
up this skeletal unit. Consequently, not only the skeleton
but also the soft tissues of the craniofacial region must
225
226
be evaluated for adequate diagnosis of asymmetry.
McCance et al8 showed that facial skeletal asymmetries
are apparent at clinical observations, whereas Burstone9
and Shah and Joshi10 demonstrated that the overlying
soft tissue components can compensate for underlying
skeletal asymmetry. Additionally, Masuoka et al11 re-
ported that patients who are clinically classified as sym-
metric or mildly asymmetric might have severe skeletal
asymmetries when assessed on posteroanterior radio-
graphs.
Determination of the tissues and regions with cranio-
facial asymmetry undoubtedly leads to better treatment
plans. For many years 2-dimensional (2D) diagnostic
tools have been used to diagnose asymmetry, but
magnification, distortion, and superimposition problems
could not be overcome.11,12 Moreover, 2D radiographs
provide only limited information about soft tissues.
Three-dimensional (3D) evaluation is needed to un-
derstand the complex nature of craniofacial asymmetry
and to differentiate the fundamental phenomena from
pathologic cases. Cone-beam computed tomography
(CBCT) allows evaluation of asymmetry via volumetric,
surface area, and distance measurements, which can
show asymmetry and morphology more realistically.
Interest in facial asymmetry in orthodontics has
increased in recent years. In many studies, asymmetry
was identified using 2D diagnostic tools that caused un-
derestimation of the 3D nature of the pathology.13
Three-dimensional evaluation studies highlight the
hard tissues but do not show the soft tissue components
of the facial asymmetry.14 Studies investigating the rela-
tionship between soft and hard tissues have been based
on a few measurements of limited regions of the face.15
Therefore, the purposes of this study were (1) to evaluate
facial asymmetry 3 dimensionally with linear, surface
distance, angular, volumetric, and surface area measure-
ments with CBCT and (2) to compare the right and left
facial hard and soft tissue volumetric measurements to
assess whether there were compensations.
MATERIAL AND METHODS
The chin was found to be strongly related to facial
asymmetry; therefore, asymmetry was defined by the
extent of soft-tissue menton (Me0 ) deviation from the
midsagittal reference line, crossing from glabella and
perpendicular to the pupillary line, by clinical evaluation
in this study. Patients with less than 4 mm of Me0 devi-
ation from the midsagittal reference line were catego-
rized as symmetric and comprised the control group,
whereas patients with 4 mm or more of Me0 deviation
comprised the asymmetry group, similar to the study
of Haraguchi et al.16 Additionally, the asymmetry group
was subdivided into 2 groups: (1) the ramus-originated
February 2016  Vol 149  Issue 2 American Journal of Orthodontics and Dentofacial Orthopedics
Nur, Çakan, and Arun
asymmetry group in which bilateral ramus height mea-
surement differences were greater than bilateral corpus
length measurement differences and (2) the corpus-
originated asymmetry group in which bilateral corpus
length measurement differences were greater than bilat-
eral ramus height measurement differences.
Forty-nine patients with asymmetry were enrolled in
this study, and CBCT scans were taken for diagnostic
purposes at the onset of orthodontic treatment. Data
from 39 control patients were taken from the archive
of the Department of Orthodontics at Yeditepe Univer-
sity in Istanbul, Turkey; they had undergone a 3D
CBCT scan for other purposes, such as identification of
impacted tooth locations or mini-implant placement.
Patient characteristics (sex, age, malocclusion classifica-
tion, and Me0 deviation) of both groups are given in
Table I. Patients with syndromes or significant pathol-
ogies affecting craniofacial appearance, such as trauma
or infection, or degenerative disease of the temporo-
mandibular joint, or who had undergone functional or-
thopedic treatment were excluded from the study. The
mean ages of the symmetry and asymmetry groups
were 17.8 6 5.5 and 19.9 6 5.6 years, respectively. All
subjects (minimum age, 12 years; maximum age,
30 years), evaluated by hand and wrist radiographs,
were at the end of the pubertal growth spurt or adults.
This study was approved by the ethical committee of
Yeditepe University.
All 3D scans were obtained using a CBCT scanner
(Iluma; IMTEC Imaging, Diegem, Belgium) at 120 kV,
1.0 mA, voxel size of 0.4 mm, and field of view of
21.1 3 14.2 cm. The patients were scanned for 40 sec-
onds while sitting upright with the head in a natural po-
sition in centric occlusion. After the scanning, the slices
were reconstructed after the reorientation of head posi-
tion. By paralleling the right and left orbitales to the
floor on the coronal section and by perpendicularly
setting the sella and nasion line to the screen on the axial
section, tilting and right-left orientation of the head
were prevented.
The data were stored in the DICOM (digital images
and communication in medicine) format and exported
to the MIMICS software program (version 13.0; Materi-
alise, Leuven, Belgium) for analysis. The plane crossing
the bilateral orbitale and right porion points was defined
as the horizontal plane. The midsagittal reference plane,
crossing the anatomic landmarks sella and nasion
perpendicular to the horizontal plane, was defined as
the facial midline (FML), similar to the definition used
in the study of Ryckman et al17 (Fig 1). With the auto-
matic function of gray-scale thresholding of the soft-
ware program, soft tissues (–700-225 HU) were
removed from the hard tissues (226-3071 HU), and
Nur, Çakan, and Arun
Table I. Patient characteristics in the control and asymmetry groups
Sex
Male (n) Female (n) Total (n)
Control group 18 21 39
Asymmetry group 27 22 49
Fig 1. The construction of facial midline, reference line, crossing sella and nasion points perpendicular
to the horizontal plane: a 3D view.
data from each were stored in separate files. Hyoid bone,
vertebrae, and structures posterior to the meatus acusti-
cus externus were cropped from the field of view. After
thresholding, 2 files were created: soft tissues and hard
tissues. The mandible was erased using the negative
construction function from the hard tissue file, and the
new file was saved as a maxillary complex file. Using
the Boolean operation function, the maxillary complex
file was extracted from the original hard tissue file,
and a new file including the mandible was constructed.
The duplicated hard and soft tissue files were then sepa-
rated into right and left segments through the FML.
Additionally, the mandible was also divided into 2 volu-
metric segments via the absolute mandibular reference
line passing through the anatomic hard tissue menton
and mental spine points perpendicular to the horizontal
plane, which were defined as the absolute mandibular
segments (Fig 2).18 The side toward which the Me0 was
shifted was categorized as the deviation side (A), and
the opposite side as the nondeviation side (B).
Anatomic landmarks were located on the 3D images
bilaterally, and their locations were verified in all 3
planes (Table II). Linear, surface distance, and angular
American Journal of Orthodontics and Dentofacial Orthopedics
227
Age (y) Malocclusion class Me0 deviation (mm)
Mean 6 SD I II III Mean 6 SD
17.8 6 5.5 22 10 7 2.17 6 1.11
19.9 6 5.6 18 18 3 5.41 6 1.37
measurements were made on hard and soft tissues
(Table III; Fig 3). Linear and surface distance measure-
ments were defined as the shortest linear distance and
the shortest distance in regard to the topography of
the anatomic surface between 2 points, respectively.
These are 3D measurements; thus, they measured the
distance of 3D anatomic points on 3D images. Skeletal
and corresponding 2D soft tissue linear measurements
were also made as described by Hong et al15 using the
translucency function of the software program, which
shows both the soft and hard tissue components at the
same frontal image (Table III; Fig 4). The distance be-
tween the midpoint of the maxillary central incisors
and the most lateral border of the mandible was defined
as the mandibular occlusal plane (MOP), whereas the
extension of the same measurement to the most lateral
soft tissue border at the same level was called the lateral
facial occlusal plane (LFOP). The measurements of
Golat-FML and the extension to the soft tissue border
were recorded as the mandibular basal plane (MBP)
and the lateral facial basal plane (LFBP), respectively.
MOP, LFOP, MBP, and LFBP are the only 2D linear mea-
surements projected on the frontal view and were
February 2016  Vol 149  Issue 2
228 Nur, Çakan, and Arun

Fig 2. Separation of the mandible into right and left segments. A, According to the absolute midline of
the mandible, a reference line passing through the anatomic hard tissue menton and mental spine
points perpendicular to the horizontal plane: frontal view (right); basilar view (left). B, According to
the facial midline, a reference line, crossing the anatomic landmarks sella and nasion perpendicular
to the horizontal plane: frontal view (left); basilar view (right). The basilar view enabled the comparison
of the differences in volumetric and surface area measurements.

preferred because of the poor reliability of 3D soft tissue
points. See Tables II and III for definitions of the
abbreviations.
Volumetric and surface area measurements were per-
formed on the right and left segments of the maxillary
complex, the facial mandible, the absolute mandible,
and the total facial hard and soft tissues. All measure-
ments were redone in 20 CBCT data sets and reassessed
after 30 days by the same investigator (R.B.N.).
Statistical analysis
All statistical analyses were performed using the SPSS
software package for Windows (version 15.0; SPSS, Chi-
cago, Ill). In addition to standard descriptive statistical
calculations (means and standard deviations), the Wil-
coxon test was used to compare the measurements be-
tween the nondeviated and deviated sides in each
group. Measurement differences between the nondevi-
ated and deviated sides were compared using the
Mann-Whitney U test between the 2 groups. The
Spearman correlation test was performed to determine
correlations between hard and soft tissue measurements.
To evaluate intraobserver reliability and method errors,
the Wilcoxon test and the Dahlberg formula19 were
used. The statistical significance level was set at P \0.05.
RESULTS
The Wilcoxon test showed no significant differences
between the first and second assessments of all measure-
ments (P .0.05). The random errors for the linear and
volumetric measurements were between 0.01 and
0.9 mm and 0.21 and 0.64 cm3, respectively.
The mean linear and angular measurements for hard
and soft tissue menton deviations from the midline were
significantly longer and higher in the asymmetry group
than in the control group (Table IV).
We compared the measurements between the devi-
ated and nondeviated sides in the asymmetry and

February 2016  Vol 149  Issue 2 American Journal of Orthodontics and Dentofacial Orthopedics
Nur, Çakan, and Arun 229

Table II. Three-dimensional landmark definitions

Landmark Definition
Sella (S) Center of the hypophyseal fossa
Nasion (N) Midpoint of the frontonasal suture
Zygomaticofrontal (Z) Most medial and anterior point of the zygomaticofrontal suture at the level of the lateral orbital rim
Zygomatic arch (ZA) Most anterior, lateral, and midpoint of the zygomaticomaxillary suture on the zygomatic arch
Condylar (Co) Most superior midpoint of the condylar head
Gonion lateralis (Golat) Most lateral point between the mandibular corpus and the ramus junction
Gonion inferius (Goinf) Most inferior point between the mandibular corpus and the ramus junction
Gonion posterius (Gopost) Most posterior point between the mandibular corpus and the ramus junction
Menton (Me) Most inferior midpoint in the symphysis
Soft tissue menton (Me0 ) Most inferior midpoint of the chin
Antegonion (Ag) Deepest point of the concavity between the mandibular corpus and the ramus junction
Jugulare (J) Most inferior midpoint of the concavity at the zygomaticomaxillary process
Premolar (P) Most superior point of the buccal cusp of the maxillary first premolar

control groups. Significant differences in the measure-
ments of ZA-Goinf, J-FML, J-N, Golat-FML, Co-Goinf,
Golat-Me, Gopost-Me, Co-Me, P-N, MOP, and MBP
were found between the deviated (A) and nondeviated
(B) sides in the asymmetry group (Table V; Fig 3, A
and C). In the control group, ZA-Goinf, J-N, Co-Goinf,
Golat-Me, Gopost-Me, Co-Me, Golat-N, and P-N
showed statistically significant differences between the
2 sides (Table V).
Golat-N measurements were significantly different
between the asymmetry subgroups. The results of side
A compared with side B were shorter in the ramus-
originated group and longer in the corpus-originated
asymmetry group (Table VI; Fig 3, B). Golat-Me surface
distance measurements between the 2 sides were signif-
icantly different in the asymmetry group but not in the
symmetry group. In both groups, there was no signifi-
cant difference between the gonial angle of the deviated
(A) and nondeviated (B) sides.
In the asymmetry group, the facial mandibular vol-
ume and surface area measurements showed significant
differences between the 2 sides (Table VII). However,
neither the side A or B maxillary measurements in either
group nor the side A or B mandibular measurements in
the control group showed any significant differences.
In the asymmetry group, the mandibular absolute vol-
ume and the surface area measurements on the nonde-
viated side (B) were significantly greater than on the
deviated side. Significant differences were also found
in the hard and soft tissue volumes and surface area
measurements between the 2 sides (Table VII).
Comparisons of the differences between the mea-
surements of sides A and B of both groups showed
that ZA-Goinf, J-FML, Golat-FML, Co-Goinf, Gopost-
Me, Co-Me, and P-N were significantly greater in the
asymmetry group (Table V). The difference between
the MOP measurements of the asymmetry group was
greater than that of the control group, whereas the dif-
ferences in the MBP, LFOP, and LFBP measurements
were not significant between the groups (Table V; Fig 4).
The absolute Golat-N difference of the control group
was lower than that of the ramus-originated and the
corpus-originated asymmetry subgroups. Golat-N was
longer on the deviated side in both the control group
and the ramus-originated asymmetry subgroup, but
shorter in the corpus-originated asymmetry subgroup
(Table VI).
Neither the surface distance nor the angular mea-
surement differences between the groups were statisti-
cally significant.
Mandibular volume and surface area measurements
related to facial and absolute mandibular midlines
were significantly different between the 2 groups
(Table VII). In the asymmetry group, the volumetric
and surface area measurements related to the facial
midline were smaller on the nondeviated side; neverthe-
less, the same measurements related to the absolute
mandibular midline were greater on the same side. The
differences in total hard and soft tissue volumetric mea-
surements were statistically significant between the
groups.
The angular and linear measurements of the hard and
soft tissue menton points and the facial midline were
significantly correlated with each other (r 5 0.84,
r 5 0.89, P 5 0.000).
DISCUSSION
Craniofacial asymmetry has previously been inves-
tigated using several different diagnostic methods. In
the past, posteroanterior radiographs were suggested
for evaluation of facial asymmetry. However, studies
showed that these radiographs have limited reliability
and provide incomplete assessment of soft tissues.
Furthermore, they are subject to superimposition

American Journal of Orthodontics and Dentofacial Orthopedics February 2016  Vol 149  Issue 2
230 Nur, Çakan, and Arun

Misdiagnosis of facial asymmetry can result in inac-

Table III. Linear, surface distance, angular, volu-
curate orthodontic treatment plans. Precise evaluation
metric, and surface area measurements
of facial asymmetry is a key step in orthodontic diag-
Linear (3D)
Z-Ag distance
nosis.28 Consequently, in recent years, the use of
Z-FML distance CBCT for evaluation of facial asymmetry has become
ZA-N distance more common. CBCT not only overcomes the disadvan-
ZA-Goinf distance tages of 2D radiographs without exposing the patient to
J-FML distance high levels of radiation and great expense,29,30 but also
J-N distance
Go-FML distance
improves the ability to understand the 3D nature of
Co-Golat distance facial asymmetry, enabling the simultaneous
Co-Goinf distance (ramus height) evaluation of both hard and soft tissues, and
Golat-Me distance enhances treatment outcomes.7,24,31 Recent studies
Gopost-Me distance (corpus length) have recommended the use of CBCT for diagnosing
Co-Me distance
Golat-N distance
facial asymmetry and classified 3D imaging as the
P-N distance best method for understanding a patient's
Me-FML distance (linear hard tissue menton deviation) morphology.32,33
Me0 -FML distance (linear soft tissue menton deviation) In the literature, there are different definitions of
Linear (2D) FML for the selection of the landmarks.7,14,17,24,33
MOP distance (mandibular occlusal plane)
LFOP distance (lateral facial occlusal plane)
However, to construct an optimal FML, planes that
MBP distance (mandibular basal plane) do not include landmarks compensating for the
LFBP distance (lateral facial basal plane) asymmetry and that are minimally affected by facial
Co-Golat distance asymmetry should be selected. Kwon et al7 mentioned
Surface distance (3D) that the morphology of the cranial base was similar
Golat-Me distance
CoGolat-MeGolat (gonial angle)
between symmetric and asymmetric patients. Therefore,
Angular (3D) in this study, as in the study of Ryckman et al,17 the
NMe-FML (angular hard tissue menton deviation) reference line, crossing sella and nasion perpendicular
NMe0 -FML (angular soft tissue menton deviation) to the horizontal plane, was preferred as the FML.
ANB The quantitative measurement of chin deviation for
Maxillary complex volume
Volumetric (3D)
the determination of facial asymmetry is a useful,
Mandibular volume repeatable, and objective method.6,11,16 Haraguchi
Absolute mandibular volume et al6 reported that patients with Me0 deviation from
Hard tissue volume the FML of 4 mm or more were classified as asymmetric
Soft tissue volume by orthodontists. Similarly, other CBCT studies on facial
Maxillary complex surface area
Surface area (3D)
asymmetry used the same amount of chin deviation as
Mandibular surface area the defining line between the symmetry and asymmetry
Absolute mandibular surface area groups.7,34,35 Hence, in our study, the criterion for
Hard tissue surface area selection of asymmetric patients was 4 mm or more of
Soft tissue surface area soft tissue chin deviation from the FML.
In the literature, in addition to the linear measure-
ments, the severity of the asymmetry was also classified
problems for deeper bony structures and rotation according to the angular soft tissue chin deviation from
effects related to the patient's head position.10,20,21 the midline. In our study, the mean angular soft tissue
Hwang et al22 reported that this technique is not suit- chin deviation of the symmetric group was 1.1 , corre-
able for determining chin morphology, which is the re- sponding to symmetric morphology, whereas that of
gion of the craniofacial complex most affected by the asymmetric group was 2.79 , corresponding to mod-
asymmetry. To overcome the limitations of 2D tech- erate asymmetric morphology according to the classifi-
niques and to obtain a 3D image of the pathology, cation of Lee et al.28
combined radiographs were used but achieved The hard tissues are covered with a thin layer of soft
little success because of changes in head position be- tissues at menton and in the chin region. McCance et al8
tween the shots.22-24 Meanwhile, soft tissue and Wermker et al36 reported that the soft tissues fol-
assessment techniques such as clinical evaluation or lowed the skeletal movements in the chin region in the
photography also provided no information about the same direction and distance after orthognathic surgery.
hard tissues.25-27 Similar to these results, in both groups of our study, the

February 2016  Vol 149  Issue 2 American Journal of Orthodontics and Dentofacial Orthopedics
Nur, Çakan, and Arun 231

Fig 3. Examples for linear, surface distance, angular measurements: A, Z-FML, ZA-N, Z-Ag, and J-
FML linear measurements; B, Golat-N linear measurments to categorize the asymmetry group into
subgroups; C, mandibular effective length (Co-Me), ramus height (Co-Goinf), and corpus length (Go-
post-Me) linear and ramus height and corpus length surface distance measurements.

Fig 4. Presentation of translucency function of the software program for: A, mandibular occlusal plane
(MOP) and mandibular basal plane (MBP); B, lateral facial occlusal plane (LFOP) and lateral facial
basal plane (LFBP) measurements to compare the asymmetry of hard and soft tissue components
at the same level.

In mandibular asymmetry, buccal pressure increases

Table IV. Comparison of linear and angular hard and
on the deviated side and decreases on the opposite
soft tissue menton deviation measurements between
side, causing morphologic alterations of the dentoalveo-
groups
lar segment in 3 dimensions.37,38 The distance between
Asymmetry group Control group the jugular point, as a reference for the dentoalveolar
(n 5 49) (n 5 39) segment of the maxilla, and the FML differed
Measurement Mean SD Mean SD P value significantly between the 2 sides in the asymmetry
Me-FML (mm) 4.83 1.32 1.86 1.03 0.000* group as well as between the 2 groups (Table V). It ap-
NMe-FML ( ) 2.57 0.81 0.97 0.51 0.000* pears that the jugular point did not compensate for
Me0 -FML (mm) 5.41 1.37 2.17 1.11 0.000* the asymmetry in the transverse dimension; instead, it
NMe0 -FML ( ) 2.79 0.86 1.11 0.60 0.000*
showed dentoalveolar adaptation. Additionally, in asym-
Mann-Whitney U test, *P \0.001. metric patients, not only the transverse but also the ver-
tical dimensions between the deviated and nondeviated
angular and linear measurements from the hard and soft sides differed because of functional adaptation.37
tissue menton to the FML showed positive correlations Several previous studies mentioned that the facial and
with each other (r 5 0.84, r 5 0.89, P 5 0.000). the dentoalveolar vertical height measurements were

American Journal of Orthodontics and Dentofacial Orthopedics February 2016  Vol 149  Issue 2
February 2016  Vol 149  Issue 2

232
Table V. Comparison of linear, surface distance, and angular measurements between sides (deviated and nondeviated) and comparison of deviated and non-
deviated sides linear, surface distance, and angular measurement differences between groups (asymmetry and control)
Comparison between sides

Asymmetry group (n 5 49) Control group (n 5 39) Comparison between groups

Deviated side (A) Nondeviated side (B) Deviated side (A) Nondeviated side (B) Asymmetry group (n 5 49) Control group (n 5 39)

Measurement Mean SD Mean SD P Mean SD Mean SD P Mean SD Mean SD P

Z-Ag 95.73 6.99 96.75 6.64 0.073 93.22 5.91 93.58 6.12 0.267 1.02 4.91 0.36 2.29 0.502
Z-FML 78.10 5.35 77.73 5.48 0.138 76.68 5.80 76.71 5.71 0.717 0.49 2.65 1.38 10.01 0.740
ZA-N 52.51 3.37 52.02 3.64 0.140 51.97 3.52 50.59 9.03 0.722 0.38 1.75 0.03 1.51 0.264
ZA-Goinf 65.43 5.91 67.84 5.51 0.000z 63.55 5.60 64.24 5.80 0.000z 2.41 4.40 0.68 0.81 0.000z
American Journal of Orthodontics and Dentofacial Orthopedics

J-FML 43.11 3.09 40.82 2.87 0.000z 41.62 3.41 41.36 3.03 0.346 2.29 2.86 0.25 1.98 0.000z
J-N 70.65 13.41 71.58 13.15 0.000z 66.91 8.11 67.35 8.07 0.003y 0.93 1.30 0.44 0.85 0.095
Golat-FML 84.22 4.86 81.15 5.47 0.000z 79.74 5.79 78.95 5.40 0.120 3.07 3.43 0.79 2.71 0.000z
Co-Golat 59.82 4.97 59.69 5.77 0.877 57.05 5.49 56.79 5.84 0.516 0.14 4.49 0.26 2.59 0.568
Co-Goinf 63.80 6.49 66.82 6.17 0.000z 63.44 6.22 64.54 5.99 0.001y 3.01 3.78 1.09 1.76 0.002y
Golat-Me 81.40 5.19 82.76 5.01 0.000z 78.51 5.70 79.89 6.75 0.023* 2.38 1.70 1.38 3.58 0.059
Gopost-Me 86.74 5.91 89.69 5.87 0.000z 85.35 6.61 86.53 6.66 0.000z 2.95 1.37 1.18 0.98 0.000z
Co-Me 119.69 7.33 122.63 8.03 0.000z 115.20 8.94 116.41 8.57 0.000z 2.94 2.74 1.21 1.56 0.002y
P-N 83.99 7.14 85.00 7.39 0.000z 82.76 12.01 83.19 11.85 0.019* 1.01 1.00 0.43 1.04 0.025*
MOP 56.95 5.19 54.13 5.72 0.001y 53.76 4.98 53.60 5.14 0.978 2.82 6.30 0.16 5.23 0.011*
LFOP 95.47 8.15 92.82 9.86 0.095 92.21 7.52 93.26 8.02 0.315 2.65 8.66 1.05 5.75 0.059
MBP 57.54 5.95 54.72 6.14 0.007y 54.25 5.36 53.40 6.76 0.410 2.82 6.85 0.86 5.62 0.171
LFBP 86.98 7.53 86.14 9.32 0.933 82.12 6.37 82.96 8.61 0.406 0.84 7.99 0.84 6.79 0.629
Surface distance (Co-Golat) 70.80 6.19 71.20 6.92 0.274 67.23 6.24 67.23 6.90 0.558 0.40 4.86 0.00 3.37 0.303
Surface distance (Golat-Me) 93.78 6.63 95.65 6.61 0.000z 89.99 7.44 91.26 7.17 0.078 2.57 3.30 1.28 3.88 0.297
CoGolat-MeGolat (gonial angle) 119.32 5.96 119.11 7.02 0.435 116.57 5.05 116.09 4.68 0.331 0.04 4.09 0.48 2.94 0.346

Nur, Çakan, and Arun

Wilcoxon and Mann-Whitney U tests were used to compare nondeviation and deviation side within groups and the symmetry and asymmetry groups, respectively.
*P \0.05; yP \0.01; zP \0.001.
Nur, Çakan, and Arun 233

shorter on the deviated side.34,37 The linear

1.22 0.000y
Table VI. Comparison of Golat-N linear measurement between sides (deviated and nondeviated) of the subgroups and groups (ramus-originated and corpus-
measurements between nasion and the premolar point

P
differed significantly between the 2 groups: they were

group (n 5 39)
approximately 1 mm shorter on the deviated side in

SD
Control
the asymmetry group (Table V). Similarly, ZA-Goinf

0.72
measurements were 2 mm longer on the nondeviated

Mean
Comparison between groups

side than on the deviated side in the asymmetry group

(Table V).
asymmetry subgroup asymmetry subgroup
Corpus-originated

2.52
SD
Craniofacial asymmetry is more frequently identified

Wilcoxon and Mann-Whitney U tests were used to compare nondeviation and deviation sides within groups and the symmetry and asymmetry groups, respectively.
in the lower third of the face.5 In a study that described
(n 5 31)

an asymmetry index for every cephalometric point, it was

Mean

stated that the index values increased from the top to the
1.83

bottom of the face.38 Similar to these findings, in our

study, linear measurements of the upper face (Z-Ag,
Ramus-originated

Z-FML, and ZA-N) and volumetric and surface area mea-

3.97
SD

surements of the maxillary complex showed no signifi-

(n 5 18)

cant differences between the groups or between the

3.22

sides, whereas almost all measurements of the lower

Mean

face showed significant differences (Tables V and VII).

Mandibular asymmetry was classified according to
Corpus-originated asymmetry subgroup (n 5 31)

the affected sides, as absolute corpus, ramus, and a com-

0.000y
P

bination of both, or as the inclination between these

anatomic parts.14,34 Furthermore, in some symmetric
Deviated side (A) Nondeviated side (B)

patients, a negative correlation could be seen between

5.13
SD

corpus length and ramus height on the same side to

compensate for a fundamental asymmetry.7 In our
120.29

study, no differences were found in the measurements

Mean

of gonial angles between the 2 groups. Nevertheless,

ramus height (Co-Goinf), corpus (Gopost-Me), and
mandibular effective length (Co-Me) measurements
5.22
SD

were smaller on the deviation side and greater on the

Comparison between sides

opposite side in both the asymmetry and the control

122.12
Mean

groups (Table V). In this manner, the asymmetry group

in our study can be categorized as having combined
asymmetry of both mandibular ramus and corpus.
Ramus-originated asymmetry subgroup (n 5 18)

0.002*
orginated asymmetry subgroups and control group)

Gonion is a point located at the mandibular angle. In

P

traditional evaluations, it is determined (1) by drawing

Deviated side (A) Nondeviated side (B)

tangent lines to both the posterior margin of the

8.04
SD

mandibular ramus and the inferior margin of the

mandibular body and (2) by dropping a perpendicular
from the intersection of these lines to the mandibular
123.01
Mean

angulus region.39 However, in asymmetric subjects, the

gonial morphology may show 3D surface alterations.
Therefore, the gonial region was evaluated with 3 points
7.08
SD

on the 3D images by Baek et al34: the most posterior

(Gopost), most inferior (Goinf), and most lateral (Golat)
119.79

points. Baek et al34 reported that gonion moved to a

Mean

*P \0.01; yP \0.001.

more superior and posterior location in asymmetric sub-

jects. Similarly, in our study, Golat-FML measurements
Measurements

were 3 mm longer on the deviation side in the asymme-

try group. The measurements of ramus height on the de-
Golat-N

viation side decreased; this probably led to the lateral

ramus inclination and the lateral displacement of gonion

American Journal of Orthodontics and Dentofacial Orthopedics February 2016  Vol 149  Issue 2
234 Nur, Çakan, and Arun

lateralis. Additionally, the ramus height measurements

P value
Asymmetry group (n 5 49) Control group (n 5 39)

0.000z
0.000z
0.002y

0.000z
0.047*

0.010*

0.013*
0.371

0.576

0.152
between Golat-Co showed no differences, whereas the
Goinf-Co measurement demonstrated significant differ-
ences (3.01 6 3.78 mm in the asymmetry group, and

15.57 79.81
1.74 31.09

0.21 18.18
3.31 32.81
9.45 110.62
4.23
1.75
0.49
6.44

9.75
SD
Comparison between groups

1.09 6 1.76 mm in the control group) in our study

(Table V). Moreover, measurement differences of
0.33
0.49
0.64
0.71

1.92
Mean

Gopost-Me (2.95 6 1.37 mm in the asymmetry group,

and 1.18 6 0.98 mm in the control group) were greater
than between Golat-Me (2.95 6 1.37 mm in the asym-

112.14
5.49
3.07
1.85
6.84

3.07
94.96
30.27

10.79
33.58
SD

metry group, and 1.18 6 0.98 mm in the control group),

Wilcoxon and Mann-Whitney U tests were used to compare nondeviation and deviation side within groups and the symmetry and asymmetry groups, respectively.
even though both represent corpus length. These alter-
ations in gonial morphology may be correlated with
asymmetry. The relationship of gonial modifications
0.81
2.55
1.96
3.36

5.09

6.64
33.99

25.54

14.78
57.88
Mean

and asymmetry types should be evaluated in future

studies. However, according to our results, we concluded
that to quantify the differences between the 2 sides in
P value

6.85 0.000z
14.93 0.241
6.85 0.204

1369.12 0.118
122.32 0.185
76.52 0.922
23.08 0.089
23.50 0.755
160.11 0.675
181.89 0.802

asymmetry subjects, the gonial region should be

analyzed using 3 points (Golat, Goinf, and Gopost).
Deviated side (A) Nondeviated side (B)

As previously mentioned, the study sample consisted

SD
Control group (n 5 39)

of patients with corpus and ramus asymmetry, but 1

anatomic part was always more dominant. Therefore,
Table VII. Comparison of volumetric and surface area measurements between sides and groups

376.27
587.83
499.57
136.59
137.52
656.62
1358.04
77.53
31.19
31.69

the asymmetry group was further categorized into 2 sub-

Mean

groups based on differences of the Golat-N measure-

ments into ramus-originated and corpus-originated
77.20 17.09

0.000z 123.35 96.96

501.32 89.40
0.000z 138.51 24.10
0,000z 135.41 28.01
0.001y 603.40 101.30

0.004y 659.93 163.05

0.000z 1367.49 222.80
6.97
6.79

groups. Our results showed statistically significant differ-

SD
Comparison between sides

ences in the Golat-N measurements between the sub-

groups and also from the control group (Table VI; Fig
31.68
31.05
P value Mean

3, B). There was a difference of approximately 5 mm be-

tween the ramus-originated and the corpus-originated
0.000z
0.000z

asymmetry subgroups in this measurement. Golat-N dis-

0.626

0.335

tances were longer in the corpus-originated subgroup,

and shorter in the ramus-originated asymmetry subgroup
Deviated side (A) Nondeviated side (B)

15.69

19.82

91.12
20.88
21.71
112.07

101.42
172.04
6.56
6.44
Asymmetry group (n 5 49)

SD

on the deviated side. Studies showed that muscle volumes

and intrinsic contraction strength decrease in high-angle
patients and increase in low-angle patients.40,41 The
81.15
32.89
34.72
114.04
622.63
508.01
142.60
150.32
650.28
1391.49
Mean

differences in volume and function of the muscles can

modify the skeletal units of the craniofacial
complex.14,42-44 In light of this knowledge, gonion
665.06 103.58
1449.37 166.37
6.68
6.26
81.95 16.68

117.40 20.72
656.62 68.05
Maxillary complex (surface area-cm2) 513.11 95.40
152.29 21.95
Absolute mandible (surface area-cm2) 143.67 21.69
SD

morphology might be modified in the vertical axis in

the ramus-originated asymmetry subgroup because of
35.45
32.76
Mean

the increase in vertical dimensions and the decrease in

muscular activity on the undeviated side. Similarly, the
gonial region might be elongated in the horizontal axis
Absolute mandible (volumetric-cm3)
Maxillary complex (volumetric-cm3)

in the corpus-originated asymmetry subgroup because

*P \0.05; yP \0.01; zP \0.001.

of increased muscular activity on the deviation side.

Hard tissue (surface area-cm2)
Soft tissue (surface area-cm2)
Hard tissue (volumetric-cm3)
Soft tissue (volumetric-cm3)

Mandible(surface area-cm2)

The soft tissue components may compensate for the

Mandible (volumetric-cm3)

skeletal asymmetries.9-11 A thin soft tissue layer such as

in the menton region is related to the underlying
skeletal units in a 1:1 ratio; nevertheless, in regions
Measurement

with thicker layers, the potential for compensation

increases. In the asymmetry group of our study,
statistically significant differences were found in MOP
and MBP measurements, but not in LFOP or LFBP

February 2016  Vol 149  Issue 2 American Journal of Orthodontics and Dentofacial Orthopedics
Nur, Çakan, and Arun
measurements on the 2D images (Table V; Fig 3). Skeletal
transverse asymmetries were apparent at the gonial (MBP)
and occlusal (MOP) levels, but probably because of
compensation by soft tissue thickness, no soft tissue
transverse discrepancies (LFOP, LFBP) were identified at
the same level. On the other hand, in asymmetric patients,
3D right and left volumetric soft tissue evaluations pro-
vide evidence for asymmetry on the 3D images
(P \0.001).
We emphasize that these measurements (MOP, LFOP,
MBP, and LFBP) are 2D linear measurements projected
on the frontal view, using the translucency function of
the software program. These measurements were
preferred only because of the lack of precise and repeat-
able soft tissue landmarks.45,46 Our aim was to get an
idea about the relationships between soft and hard
tissues at different levels.
Symmetric subjects may also have mild asymmetric
components, defined as fundamental asymmetries. In
our study, some linear measurements showed signifi-
cant differences between the deviated and contralateral
sides in accordance with the literature.47 Golat-Me
linear and surface distance measurements for the
same landmarks did not differ between sides. A possible
explanation for this finding may be the topographic
alteration of the surface. On the other hand, in the
asymmetry group, because the asymmetry was moder-
ate, the topographic changes could not mimic the skel-
etal asymmetry; therefore, significant differences were
found in both Golat-Me linear and surface distance
measurements.
The selection of the reference planes, dividing the
anatomic units into right and left parts, is of crucial
importance in asymmetry evaluation. The absolute volu-
metric asymmetry or the displacement because of the
asymmetry of the evaluated unit can be diagnosed ac-
cording to the selected reference plane.48 In the same
way, the right and left volumetric and surface area mea-
surements of the mandibular segments evaluated ac-
cording to the distance from the FML and the absolute
mandibular line demonstrated the displacement and
the absolute bilateral asymmetry of the mandible,
respectively. Wong et al48 reported a loss of 75.6% of
symmetry if a mandibular asymmetry analysis was per-
formed according to the maxillary midline. Moreover,
they suggested that this was induced because of the
divergence of 3.125 between the midline axes of the
maxilla and the mandible. Similarly, in our asymmetry
group, when divided through the FML, the mandibular
volumetric and surface area measurements of the seg-
ments were larger on the deviated side, but smaller
when divided through the absolute mandibular midline
(Table VII; Fig 3).
American Journal of Orthodontics and Dentofacial Orthopedics
235
Asymmetry was rarely identified in the maxilla. The
mandible, as a moveable joint, grows for a longer time
than does the maxilla. Therefore, the mandible must
compensate for asymmetries in the craniofacial
skeleton and modify its volume and surface
areas.6,7 As expected, in the asymmetry group, the
volumetric and surface area measurements of the
mandible, but not of the maxilla, showed significant
differences.
CONCLUSIONS
1. Morphologic changes at the gonial region were
observed; therefore, 3D evaluation of this region us-
ing multiple landmarks (Goinf, Gopost, Golat)
should be considered for a detailed diagnosis of
asymmetry.
2. In asymmetric patients, volumetric and surface area
measurement differences of the deviated and non-
deviated sides were more frequently identified in
the lower third and less often in the middle third
of the face.
3. Subjects classified as symmetric, who seem to have
no hard or soft tissue deviations, also show mild
asymmetries that are compensated for by surface
modifications of the skeletal units or soft tissue ad-
aptations.
4. In asymmetric patients, the soft tissues follow the
hard tissues to a large extent in the chin region.
5. Although in asymmetric patients the right and left-
segments of the mandible divided through the facial
midline demonstrate greater volumetric and surface
area measurements at the deviated side, the seg-
ments separated through the absolute mandibular
midline showed the contrary.
6. At the occlusal and gonial levels, asymmetry was
apparent in the skeleton but not in the soft tissue
on the 2D images. Probably the soft tissues decrease
the severity of skeletal asymmetry and compensate
for it at different levels.
7. The comparison of right and left volumetric and
surface area measurements of the soft tissue seg-
ments provides evidence of soft tissue asymmetry
in the craniofacial region. To create an ideal treat-
ment plan in asymmetric patients, not only the
hard tissues but also the soft tissues should be
evaluated.
ACKNOWLEDGMENTS
We thank Kutsal Tuaç and the 4C Medical Group in
Istanbul, Turkey, for technical support.
February 2016  Vol 149  Issue 2
236
REFERENCES
1. Fink B, Neave N, Manning JT, Grammer K. Facial symmetry and
judgements of attractiveness, health and personality. Pers Individ
Dif 2006;41:491-9.
2. Proffit WR, Philips C, Dann C. Who seeks surgical-orthodontic
treatment? Int J Adult Orthod Orthognath Surg 1990;5:153-60.
3. Lundstr€ om A. Some asymmetries of dental arches, jaws and skull
and their etiological significance. Am J Orthod 1961;47:81-106.
4. Bishara SE, Burkey PS, Kharouf JG. Dental and facial asymmetries:
a review. Angle Orthod 1994;64:89-98.
5. Severt TR, Proffit WR. The prevalence of facial asymmetry in den-
tofacial deformities population at the University of North Carolina.
Int J Adult Orthod Orthognath Surg 1997;12:171-6.
6. Haraguchi S, Takada K, Yasuda Y. Facial asymmetry in subjects
with skeletal Class III deformity. Angle Orthod 2002;72:28-35.
7. Kwon TG, Park HS, Ryoo HM, Lee SH. A comparison of craniofacial
morphology in patients with and without facial asymmetry—a
three dimensional analysis with computed tomography. Int J
Oral Maxillofac Surg 2006;35:43-8.
8. McCance AM, Moss JP, Fright WR, Linney AD, James DR. Three
dimensional analysis techniques—part 1: three dimensional soft-
tissue analysis of 24 adult cleft palate patients following Le Fort
I maxillary advancement: a preliminary report. Cleft Palate Cranio-
fac J 1997;34:36-45.
9. Burstone CJ. Diagnosis and treatment planning of patients with
asymmetries. Semin Orthod 1998;4:153-64.
10. Shah SM, Joshi MR. An assessment of asymmetry in the normal
craniofacial complex. Angle Orthod 1978;48:141-8.
11. Masuoka N, Momoi Y, Arji Y, Nawa H, Muratmatsu A, Goto S, et al.
Can cephalometric indices and subjective evaluation be consistent
for facial asymmetry? Angle Orthod 2005;75:651-5.
12. Pirttiniemi P, Miettinen J, Kantomaa T. Combined effects of errors
in frontal-view asymmetry diagnosis. Eur J Orthod 1996;18:
629-36.
13. Hayashi K, Muguruma T, Hamaya M, Mizoguchi I. Morphologic
characteristics of the dentition and palate in cases of skeletal
asymmetry. Angle Orthod 2004;74:26-30.
14. Maeda M, Katsumata A, Ariji Y, Muramatsu A, Yoshida K, Goto S,
et al. 3D-CT evaluation of facial asymmetry in patients with maxil-
lofacial deformities. Oral Surg Oral Med Oral Pathol Oral Radiol En-
dod 2006;102:382-90.
15. Hong HJ, Hong JW, Koh SH, Kim YO, Park BY. A three-dimensional
analysis of the relationship among lower facial width, bony width,
and masseter muscle volume in subjects with prominent mandible
angles. J Craniofac Surg 2009;20:1114-9.
16. Haraguchi S, Iguchi Y, Takada K. Asymmetry of the face in ortho-
dontic patients. Angle Orthod 2008;78:421-6.
17. Ryckman MS, Harrison S, Oliver D, Sander C, Boryor AA,
Hohmann AA, et al. Soft tissue changes after maxillomandibular
advancement surgery assessed with cone-beam computed tomog-
raphy. Am J Orthod Dentofacial Orthop 2010;137(Suppl):S86-93.
18. Lo LJ, Wong FH, Chen YR, Wong HF. Mandibular dysmorphology
in patients with unilateral cleft lip and cleft palate. Chang Gung
Med J 2002;25:502-8.
19. Dahlberg G. Statistical methods for medical and biological stu-
dents. New York: Interscience Publications; 1940.
20. Adams GL, Gansky SA, Miller AJ, Harrell WE Jr, Hatcher DC. Com-
parison between traditional 2-dimensional cephalometry and a
3-dimensional approach on human dry skulls. Am J Orthod Den-
tofacial Orthop 2004;126:397-409.
21. Trpkova B, Prasad NG, Lam EW, Raboud D, Glover KE, Major PW.
Assessment of facial asymmetries from posteroanterior cephalo-
February 2016  Vol 149  Issue 2 American Journal of Orthodontics and Dentofacial Orthopedics
Nur, Çakan, and Arun
grams: validity of reference lines. Am J Orthod Dentofacial Orthop
2003;123:512-20.
22. Hwang HS, Hwang CH, Lee KH, Kang BC. Maxillofacial 3-
dimensional image analysis for the diagnosis of facial asymmetry.
Am J Orthod Dentofacial Orthop 2006;130:779-85.
23. Janson G, Cruz KS, Woodside DG, Metaas A, Freitas MR,
Henriques JF. Dentoskeletal treatment changes in Class II sub-
division malocclusions in submentovertex and posteroanterior
radiographs. Am J Orthod Dentofacial Orthop 2004;126:
450-62.
24. Park SH, Yu HS, Kim KD, Lee KJ, Baik HS. A proposal for a new
analysis of craniofacial morphology by 3-dimensional computed
tomography. Am J Orthod Dentofacial Orthop 2006;129:
600.e23-34.
25. Edler R, Wertheim D, Greenhill D. Clinical and computerized
assessment of mandibular asymmetry. Eur J Orthod 2001;23:
485-94.
26. Edler R, Wertheim D, Greenhill D. Comparison of radiographic and
photographic measurement of mandibular asymmetry. Am J Or-
thod Dentofacial Orthop 2003;123:167-74.
27. Hechler SL. Cone beam CT: application in orthodontics. Dent Clin
North Am 2008;52:809-23.
28. Lee K, Hwang H, Curry S, Boyd RL, Norris K, Baumrind S. Effect of
cephalometer misalignment on calculations of facial asymmetry.
Am J Orthod Dentofacial Orthop 2007;132:15-7.
29. Brooks LS. CBCT dosimetry: orthodontic considerations. Semin Or-
thod 2009;15:14-8.
30. Mah JK, Danforth RA, Bumann A, Hatcher D. Radiation absorbed
in maxillofacial imaging with a new dental computed tomography
device. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2003;96:
508-13.
31. Cakirer B, Dean D, Palomo JM, Hans MG. Orthognathic surgery
outcome analysis: 3-dimensional landmark geometric morpho-
metrics. Int J Adult Orthod Orthognath Surg 2002;17:116-32.
32. White SC, Pae EK. Patient image selection criteria for cone
beam computed tomography imaging. Semin Orthod 2009;15:
19-28.
33. Moro A, Correra P, Boniello R, Gasparini G, Pelo S. Three-dimen-
sional analysis in facial asymmetry: comparison with model anal-
ysis and conventional two-dimensional analysis. J Craniofac Surg
2009;20:417-22.
34. Baek SH, Cho IS, Chang YI, Kim MJ. Skeletodental factors affecting
chin point deviation in female patients with class III malocclusion
and facial asymmetry: a three-dimensional analysis using
computed tomography. Oral Surg Oral Med Oral Pathol Oral Radiol
Endod 2007;104:628-39.
35. Jung YJ, Kim MJ, Baek SH. Hard and soft tissue changes after
correction of mandibular prognathism and facial asymmetry by
mandibular setback surgery: three- dimensional analysis using
computerized tomography. Oral Surg Oral Med Oral Pathol Oral
Radiol Endod 2009;107:763-71.
36. Wermker K, Kleinheinz J, Jung S, Dirksen D. Soft tissue response
and facial symmetry after orthognathic surgery. J Craniomaxillofac
Surg 2014;42:e339-45.
37. Ishizaki K, Suzuki K, Mito T, Tanaka EM, Sato S. Morphologic,
functional, and occlusal characterization of mandibular lateral
displacement malocclusion. Am J Orthod Dentofacial Orthop
2010;137:454.e1-9.
38. Kusayama M, Motohashi N, Kuroda T. Relationship between trans-
verse dental anomalies and skeletal asymmetry. Am J Orthod Den-
tofacial Orthop 2003;123:329-37.
39. Swennen GRJ, Schutyser F, Hausamen JE. Three dimensional
cephalometry. Berlin, Germany: Springer; 2005. p. 116-85.
Nur, Çakan, and Arun
40. van Spronsen PH. Long-face craniofacial morphology: cause or ef-
fect of weak masticatory musculature? Semin Orthod 2010;16:
88-117.
41. Satıroglu F, Arun T, Işık F. Comparative data on facial morphology
and muscle thickness using ultrasonography. Eur J Orthod 2005;
27:562-7.
42. Rodrigues L, Traina AA, Nakamai LF, Luz JG. Effects of the unilat-
eral removal and dissection of the masseter muscle on the facial
growth of young rats. Braz Oral Res 2009;23:89-95.
43. Kitai N, Fujii Y, Murakami S, Furukawa S, Kreiborg S, Takada K.
Human masticatory muscle volume and zygomatico-mandibular
form in adults with mandibular prognathism. J Dent Res 2002;
81:752-6.
American Journal of Orthodontics and Dentofacial Orthopedics
237
44. Koizumi Y, Ishii T, Nishii Y, Nojima K, Sueishi K. Influence of exper-
imental hemi-occlusion on mandibular morphology and internal
structure in growing rabbit. Orthod Waves 2010;69:58-65.
45. Duffy S, Noar JH, Evans RD, Sanders R. Three-dimensional analysis
of the child cleft face. Cleft Palate Craniofac J 2000;37:137-44.
46. Germec-Cakan D, Canter HI, Nur RB, Arun T. Comparison of facial
soft tissue measurements on 3D images and models obtained with
different methods. J Craniofac Surg 2010;21:1393-9.
47. Erickson KL, Bell WH, Goldsmith DH. Analysis model surgery. In:
Bell WH, editor. Modern practice in orthognathic and reconstruc-
tive surgery. Philadelphia: Saunders; 1992.
48. Wong TY, Fang JJ, Wu TC. A novel method of quantifying facial
asymmetry. Int Congr Ser 2005;1281:1223-6.
February 2016  Vol 149  Issue 2