ORIGINAL ARTICLE

Comparison of different midsagittal

plane configurations for evaluating
craniofacial asymmetry by expert
preference
SangIn An,a Ji-Yeon Lee,b Chooryung J. Chung,c and Kyung-Ho Kimc
Seoul and Gyeonggi-do, Korea

Introduction: In this study, we aimed to compare 8 candidate midsagittal planes (MSPs) constructed from
different median landmarks to determine the most appropriate one for evaluating craniofacial asymmetry.
Methods: We included 30 patients (18 men, 12 women; mean age, 25.7 6 6.03 years) who visited the National
Health Insurance Service Ilsan Hospital in Gyeonggi-do, Korea, with a complaint of facial asymmetry. Four
MSPs passing through 2 median landmarks perpendicular to the Frankfort horizontal plane and 4 other MSPs
passing through 3 median landmarks were constructed. Menton, anterior nasal spine, and anterior nasal
spine-to-posterior nasal spine line deviations were evaluated using these 8 MSPs. Eight MSPs from 30
subjects were shown to 6 experts, who selected the planes that they considered the most appropriate.
Results: Experts most frequently selected the plane passing through nasion and basion perpendicular to the
Frankfort horizontal plane (66 of 180 times; P\0.05). In evaluating craniofacial asymmetry, using MSPs passing
through 3 median landmarks in the cranial base can lead to underestimation of the asymmetry of the menton,
anterior nasal spine, and anterior nasal spine-to-posterior nasal spine line. Conclusions: We suggest using
MSPs perpendicular to the Frankfort horizontal plane or a plane passing through anterior nasal spine in clinical
practice. (Am J Orthod Dentofacial Orthop 2017;152:788-97)

P
atients’ clinical photographs and cephalometric
radiographic images have been used to diagnose
craniofacial asymmetry. Until the advent of comput-
erized tomography (CT), 2-dimensional cephalometric
analysis had been important in diagnosing patients with
craniofacial malformation and malocclusion.1 However,
this approach suffers from the limitations of 2-
dimensional imaging: distortion or enlargement of images
and difficulty in superimposition of anatomic structures.2-5
a
Department of Orthodontics, School of Dentistry, Yonsei University, Seoul, Korea.
b
Department of Orthodontics, National Health Insurance Service Ilsan Hospital,
Goyang, Gyeonggi-do, Korea.
c
Department of Orthodontics, Gangnam Severance Hospital, Institute of Craniofacial
Deformity, College of Dentistry, Yonsei University, Seoul, Korea.
SangIn An and Ji-Yeon Lee are joint first authors and contributed equally to this
work.
All authors have completed and submitted the ICMJE Form for Disclosure of Po-
tential Conflicts of Interest, and none were reported.
Address correspondence to: Kyung-Ho Kim, Department of Orthodontics, Gang-
nam Severance Hospital, Institute of Craniofacial Deformity, College of Dentistry,
Yonsei University, 211 Eonjuro, Gangnam-gu, Seoul 135-720, Korea; e-mail,
khkim@yuhs.ac.
Submitted, October 2016; revised and accepted, April 2017.
0889-5406/$36.00
Ó 2017 by the American Association of Orthodontists. All rights reserved.
http://dx.doi.org/10.1016/j.ajodo.2017.04.024
788
Analysis through 3-dimensional (3D) CT plays a major role
in overcoming the limitations of 2-dimensional analysis.6-9
When establishing an orthognathic surgery plan for pa-
tients with craniofacial asymmetry using 3D cephalometric
analysis, construction of a midsagittal plane (MSP) is
crucial, and the degree of asymmetry could be increased
or reduced, depending on the way of constructing the
MSP.10,11 Many studies have investigated methods for
constructing a MSP; median landmarks and bilateral
symmetric landmarks of the craniofacial area have
commonly been used.10,12,13 However, in patients with
craniofacial asymmetry, the craniofacial structures are
likely to lack symmetry.14,15 Thus, constructing the MSP
in patients with craniofacial asymmetry is more difficult
than in the patients with facial symmetry, because the
reference for determining the landmark with ideal
symmetry or that passes correctly through the middle of
the craniofacial area among the landmarks used for
determining the MSP is acquired after constructing it.
However, to date, only a few studies of MSP config-
urations in patients with craniofacial asymmetry have
been conducted.10,16 Thus, in this study, 8 MSPs
passing through median landmarks or planes
An et al
perpendicular to the horizontal reference plane were
selected as candidate MSPs for evaluating craniofacial
asymmetry on 3D CT images of patients with facial
asymmetry. We planned to investigate which plane
would be the most appropriate from among the 8
MSPs by having them assessed by experts.
MATERIAL AND METHODS
This study was approved by the institutional review
board of National Health Insurance Service Ilsan Hos-
pital (NHIMC 2016-06-006-001) in Gyeonggi-do, Ko-
rea. Six orthodontists and oral and maxillofacial
surgeons who had diagnosed more than 100 orthog-
nathic surgery patients selected the MSP that they
judged to be the most appropriate among the 8 candi-
date MSPs, based on images of 30 patients with facial
asymmetry.
Among the 82 patients who visited National Health
Insurance Service Ilsan Hospital from 2008 to 2014
with a complaint of facial asymmetry and who had CT im-
aging, 30 subjects with facial asymmetry were selected
using the following criteria: (1) adults over 19 years old;
(2) no systemic diseases; (3) no congenital deformities,
including cleft lip and palate; (4) no temporomandibular
joint disease; (5) no facial injury or fracture; and (6) no
history of orthognathic surgery. The mean age of these
patients (18 men, 12 women) was 25.7 6 6.03 years
(range, 19-43 years).
To obtain 3D CT images, the patients were scanned
by multislice CT (SOMATOM Sensation 64-slice;
Siemens, Malvern, Pa) in the supine position under the
following conditions: 120 kV, 200 mA, scanning time
1 second, 512 3 512 pixels with 1-mm slice thickness.
The images were saved as DICOM files and reconstructed
with a 3D stereoscopic medical image diagnosis program
(Simplant version 14.0; Materialise Dental, Leuven,
Belgium).
Eleven reference points were set in the 3D images
(Fig 1); these landmarks and their definitions are
presented in Table I. The Frankfort horizontal (FH)
plane passing through the left and right orbitales and
the median point on the left and right porions was
set as the horizontal reference plane, and the line pass-
ing through anterior nasal spine (ANS) and posterior
nasal spine (PNS) (A-P line), was set as the reference
line for evaluating the yaw of the maxilla (Fig 1;
Table I).
Eight MSPs for this study were constructed (Fig 2;
Table II). The horizontal distances from the 8 MSPs to
menton and ANS (menton and ANS deviations) were
measured (Fig 3). Angles between the 8 MSPs and the
A-P line (A-P line deviation) were measured (Fig 3).
American Journal of Orthodontics and Dentofacial Orthopedics
789
The directions in which deviation occurred from each
MSP were sorted into positive values and the directions
of the nondeviation side into negative values.
Three-dimensional images of the craniofacial struc-
tures and 8 MSPs of the 30 subjects were shown to 6
experts. They were not given the definition of each
plane, and the 8 MSPs were shown to them in a random
sequence. They selected 180 MSPs that they considered
the most appropriate planes. The horizontal distances
from all MSPs to menton and ANS, and the angles to
the A-P line were measured. The values and absolute
values of differences in deviation measurements using
the candidate MSPs were determined, and their mean
values were calculated. The values are shown in Table
III. Table IV and Figure 4 show the process involved in
finding the values and absolute values of differences in
the menton deviation measurements.
In terms of absolute values of differences in the
measurement of the menton deviation, absolute values
of differences in the measurement of the ANS deviation,
and absolute values of differences in the measurement
of the A-P line deviation, the closer these values were to
0, the more similar the measurement values to the
selected MSP would be to the diagnosis of asymmetry.
For value differences in the menton deviation measure-
ment, value differences in ANS deviation measurement,
and value differences in A-P line deviation measurement,
the positive and negative values of distances and angles
were sorted by setting the direction of the asymmetry
deviation side as the positive value in each patient.
The greater these values, the greater the chance that the
asymmetry could be overestimated compared with the
selected MSP, and the smaller these values, the greater
the chance of underestimating the asymmetry.
Statistical analysis
The power analysis of this study for sample size was
conducted as follows. The clinically significant differ-
ence between MSPs 1 through 8 was 1, the common
standard deviation was 2, type 1 error was 0.00179,
and power was 0.8. Given these parameters, the total
number of samples required for an independent t test
was 258. Even considering a dropout rate of 10%, a
sample size of 1440, collected by 6 experts using 8
MSPs in 30 subjects, was more than adequate.
The coronal reference plane, horizontal reference
plane, and sagittal reference plane were set in 3D
images. The vertical distances from these planes to the
measurement points were measured. All measurements
were repeated after a 4-week interval. The intraclass cor-
relation coefficient was used to test the intraexaminer
reliability and to assess the reproducibility of the
December 2017
Vol 152
Issue 6
790 An et al

Fig 1. Establishment of craniofacial reference points and planes: A, landmarks; B, FH plane. (Land-
marks on 3D image surfaces have red points, and landmarks projected through 3D image surfaces
have yellow points.)

Table I. Landmarks, reference plane, and line

Landmark Definition
Na (nasion) Most posterior point of curvature between frontal bone and nasal bone in the MSP
Cg (crista galli) Most superior point of crista galli of the ethmoid bone
S (sella) Center of sella turcica
Ba (basion) Middorsal point of the anterior margin of the foramen magnum
Or Rt (orbitale right) Lowest point of lower margin of the right orbit
Or Lt (orbitale left) Lowest point of lower margin of the left orbit
Po Rt (porion right) Most superior point of right external auditory meatus
Po Lt (porion left) Most superior point of left external auditory meatus
Anterior nasal spine (ANS) Most anterior point of nasal floor
Posterior nasal spine (PNS) Most posterior point of nasal floor
Me (menton) Most inferior part of the bony chin in the median plane
Reference plane
Frankfort horizontal (FH) plane Plane passing through right orbitale, left orbitale and midpoint of left and right porions
Reference line
A-P line Line passing through ANS and PNS

measurements. To investigate the plane most preferred
by the experts, the frequency with which the 6 experts
chose a MSP from among the data obtained from the
30 subjects was analyzed with a generalized linear mixed
model, and we evaluated the level of agreement among
the experts to select the most appropriate MSP for each
subject with kappa statistics. Additionally, the differ-
ences that would arise from choosing each candidate
MSP when evaluating asymmetry were measured by
analyzing the differences between the measurement
values based on the candidate MSP and the most
commonly selected MSP. The difference values of each
MSP and the means of their absolute values were
analyzed with a linear mixed model; the Bonferroni
test was used to correct for multiple comparisons. All
statistical analyses were performed using SPSS software
(version 22; IBM, Armonk, NY), and P \0.05 was re-
garded as statistically significant.

December 2017
Vol 152
Issue 6 American Journal of Orthodontics and Dentofacial Orthopedics
An et al 791

Fig 2. Candidate MSPs: A, MSPs passing through 2 median landmarks, perpendicular to the FH plane
(MSP 1, FH-Na-Ba; MSP 2, FH-Na-S; MSP 3, FH-Cg-Ba; MSP 4, FH-Cg-S); B, MSPs passing through
3 median landmarks (MSP 5, Ba-Na-S; MSP 6, Ba-Cg-S; MSP 7, Ba-Na-ANS; MSP 8, Ba-Cg-ANS).

kappa values were 0.499 between experts 1 and 4;

Table II. Definitions of candidate MSPs 0.436 between experts 1 and 5; 0.303 between experts
MSP Definition 1 and 6; and 0.341 between experts 4 and 5.
MSP 1 (FH-Na-Ba) Passing through Na and Ba while The absolute values of differences in the measurement
perpendicular to the FH plane of the menton deviation of MSP 5 (5.30 mm) and MSP 6
MSP 2 (FH-Na-S) Passing through Na and S while (5.96 mm) had significantly larger values than those of the
perpendicular to the FH plane
other MSPs (0.81-2.17 mm) (P\0.05; Table VI). The value
MSP 3 (FH-Cg-Ba) Passing through Cg and Ba while
perpendicular to the FH plane differences in the menton deviation measurement of MSP
MSP 4 (FH-Cg-S) Passing through Cg and S while 5 (2.53 mm) and MSP 6 (2.97 mm) showed statistically
perpendicular to the FH plane significant larger negative values than those of the other
MSP 5 (Ba-Na-S) Passing through Ba, Na, and S MSPs (0.76-0.14 mm) (P\0.05; Table VI). The absolute
MSP 6 (Ba-Cg-S) Passing through Ba, Cg, and S
values of differences in the measurement of the menton
MSP 7 (Ba-Na-ANS) Passing through Ba, Na, and ANS
MSP 8 (Ba-Cg-ANS) Passing through Ba, Cg, and ANS deviation of MSP 7 (1.85 mm) and MSP 8 (2.17 mm)
had statistically significant larger values than those of
MSPs 1 and 2 (0.81 mm) (P \0.05; Table VI). But in
RESULTS some studies, a critical menton deviation that clinically
The intraclass correlation coefficient for the intraexa- differentiates symmetry from asymmetry is approximately
miner reliability showed the high reliability of the 4 mm.17,18 The absolute value of differences in the
repeated measurements, ranging from 0.91 to 0.95 measurement of ANS deviation and value differences in
(P \0.001). The most frequently selected plane was ANS deviation showed a similar pattern to the absolute
MSP 1, which passed through nasion and basion and values of differences in the measurement of the menton
was perpendicular to the FH plane. MSP 1 was selected deviation and value differences in menton deviation
66 times from among the 180 planes, at a frequency measurement. The absolute values of differences in
of 36.67%; it was statistically significantly selected measurement of the A-P line deviation of MSPs 4, 5,
more often than the other 7 MSPs (P \0.05; Table V). and 6 showed statistically significant larger values than
In the evaluation of agreement among the experts, the those of the other MSPs (P \0.05; Table VI). The value

American Journal of Orthodontics and Dentofacial Orthopedics December 2017
Vol 152
Issue 6
792 An et al

Fig 3. A, Measurement of horizontal distance between MSP and points; B, measurement of angle be-
tween MSP and A-P line.

differences in the A-P line deviation measurement of MSP DISCUSSION

6 had statistically significant greater negative values than
those of MSPs 1, 2, 3, 7, and 8 (P \0.05; Table VI).
Therefore, although MSP 1 was the most frequently
chosen, it had a high degree of similarity with the other
planes, excluding MSPs 5 and 6.
Based on the findings of this study, we suggest that
MSPs constructed from median landmarks in the cranial
base are not recommended in clinical practice.
In many studies on the configurations of MSPs for
evaluating craniofacial asymmetry, median landmarks

December 2017
Vol 152
Issue 6 American Journal of Orthodontics and Dentofacial Orthopedics
An et al 793

Table III. Definitions of the values and absolute values of differences in menton, ANS, A-P line deviation measure-
ments
Absolute value of differences in
menton deviation measurements
Absolute value of differences in
ANS deviation measurements
Absolute value of differences in
A-P line deviation measurements
Value of differences in
menton deviation measurements
Value of differences in
ANS deviation measurements
Value of differences in
A-P line deviation measurements
Definition Abbreviation
P
jMenton deviation measured by candidate MSP -menton deviation AVDMe
Pmeasured by selected MSPj / 180
jANS deviation measured by candidate MSP -ANS deviation measured by AVDANS
selected MSPj / 180
P
jA-P deviation measured by candidate MSP – A-P deviation measured by AVDAP
Pselected MSPj / 180
(Menton deviation measured by candidate MSP -menton deviation VDMe
measured by selected MSP) / 180
P
(ANS deviation measured by candidate MSP -ANS deviation measured by VDANS
Pselected MSP) / 180
(A-P deviation measured by candidate MSP – A-P deviation measured by VDAP
selected MSP) / 180

or bilateral symmetric landmarks of the craniofacial area
have been used.19 However, the current knowledge on
whether cranial asymmetry and facial asymmetry can
be shown in the same way is insufficient.20 It has even
been reported that the MSPs of the face and the cranium
are not correlated.10 Therefore, a MSP constructed
according to the landmarks of the cranial base may be
appropriate for evaluating the asymmetry of the cranium
but could be inappropriate for evaluating facial asym-
metry. To evaluate facial asymmetry more appropriately,
a MSP passing through ANS would be preferable,21,22
but ANS could be affected by asymmetry of the
cranium.23,24 Thus, Ras et al25 recommended use of a
plane perpendicular to the horizontal reference plane
composed of the angulus oculi lateralis. Since growth
in the orbitale area is completed by 9 years of age, it
could be a more accurate reference for patients with
facial asymmetry.26 In our study, 4 planes (MSPs 1-4)
passing through 2 median landmarks perpendicular to
the FH plane passing through the right and left orbitales,
and 2 planes (MSPs 5 and 6) passing through 3 median
landmarks of the cranial base, and 2 planes (MSPs 7 and
8) passing through median landmarks of the cranial base
and ANS were constructed as candidate MSPs.
The plane that was the most frequently selected by the
experts as the appropriate MSP for use in patients with
craniofacial asymmetry was MSP 1, passing through
nasion and basion and perpendicular to the FH plane
(Table V). In the evaluation of agreement among the
experts, statistically significant reliability was shown
between experts 1, 4, 5, and 6. This study was
designed for the experts to select the most appropriate
MSP intuitively, not to examine whether all 6 experts
selected a specific plane, which resulted in a low kappa
value. Because MSP 1 showed the smallest absolute value
of differences in the measurement of the menton devia-
tion, absolute value of differences in the measurement
of the ANS deviation, and absolute value of differnces
in measurement of the A-P line deviation, the use of
MSP 1 would be appropriate for evaluating craniofacial
asymmetry (Table VI). MSPs 2, 3, and 4, which are perpen-
dicular to the FH plane, also showed relatively small abso-
lute value of differences in the measurement of the
menton deviation, absolute value of differences in the
measurement of the ANS deviation, and absolute value
of differences in measurement of the A-P line deviation,
indicating that they could be used to evaluate craniofacial
asymmetry appropriately. Particularly, MSP 4, which
passes through crista galli and sella, was selected only 6
of 180 times, but showed no statistically significant
difference in absolute value of differences in the measure-
ment of the menton deviation, absolute value of differ-
ences in the measurement of the ANS deviation
compared with MSP 1. Thus, MSPs 1, 2, 3, and 4, which
are perpendicular to the FH plane, were formed similarly
in the facial area. MSPs 7 and 8, passing through 2 refer-
ence points (cranial base and ANS), showed statistically
significantly larger absolute values of differences in the
measurement of the menton deviation value than those
in MSPs 1 and 2 (Table VI). However, these differences
were not clinically significant for evaluating craniofacial
asymmetry. Additionally, MSPs 7 and 8 showed no statis-
tically significant differences in absolute value of differ-
ences in the measurement of the menton deviation and
absolute value of differences in the measurement of the
ANS deviation compared with MSPs 3 and 4. Jeon
et al12 reported that MSPs passing through ANS and crista
galli could divide the face more symmetrically than MSPs
perpendicular to the FH plane passing through the crista
galli in patients without asymmetry. This suggests that
ANS deviation is uncommon. However, MSPs 7 and 8
may have limited use when evaluating patients with
asymmetry, because ANS can be distorted according to
the appearance of asymmetry.

American Journal of Orthodontics and Dentofacial Orthopedics December 2017
Vol 152
Issue 6
December 2017
Vol 152
Issue 6

794
Table IV. Finding the value and absolute value of differences in menton deviation measurements
Expert 1 Expert 2 . Expert 6 n 5 180

Patient 1 Patient 2 . Patient 30 Patient 1 Patient 2 . Patient 30 Patient 1 Patient 2 . Patient 30 Mean SD
Menton deviation
MSP 1 14.07 7.95 9.06 14.07 7.95 9.06 14.07 7.95 9.06 5.76 3.23
MSP 2 14.06 7.94 9.05 14.06 7.94 9.05 14.06 7.94 9.05 5.82 3.22
MSP 3 15.41 8 9.44 15.41 8 9.44 15.41 8 9.44 5.72 3.75
.
MSP 8 11.65 3.55 5.24 11.65 3.55 5.24 11.65 3.55 5.24 5.07 3.6
American Journal of Orthodontics and Dentofacial Orthopedics

MSP 1 MSP 3 MSP 1 MSP 3 MSP 2 MSP 1 MSP 2 MSP 6 MSP 1

Selected MSP (FH-Na-Ba) (FH-Cg-Ba) (FH-Na-Ba) (FH-Na-Ba) (FH-Na-S) (FH-Na-Ba) (FH-Na-S) (Ba-Cg-S) (FH-Na-Ba)
VDMe
MSP 1 0 0.05 0 1.34 0.01 0 0.01 1.25 0 0.08 1.56
MSP 2 0.01 0.06 0.01 1.35 0 0.01 0 1.24 0.01 0.14 1.54
MSP 3 1.34 0 0.38 0 0.06 0.38 1.35 1.3 0.38 0.04 1.94
.
MSP 8 2.42 4.45 3.82 3.76 4.39 3.82 2.41 3.15 3.82 0.61 2.65
AVDMe
MSP 1 0 0.05 0 1.34 0.01 0 0.01 1.25 0 0.81 1.33
MSP 2 0.01 0.06 0.01 1.35 0 0.01 0 1.24 0.01 0.81 1.31
MSP 3 1.34 0 0.38 0 0.06 0.38 1.35 1.3 0.38 1.37 1.37
.
MSP 8 2.42 4.45 3.82 3.76 4.39 3.82 2.41 3.15 3.82 2.17 1.63
VDMe, Value of differences in menton deviation measurements; AVDMe, absolute value of differences in menton deviation measurements.

An et al
An et al 795

Table V. Number of times MSPs were selected by the 6

experts
Expert
Percentage Multiple
MSP 1 2 3 4 5 6 Total (%) comparisons
MSP 1 (FH- 9 9 12 11 14 11 66 36.67 A
Na-Ba)
MSP 2 7 8 2 3 3 10 33 18.33 B
(FH-Na-S)
MSP 3 (FH- 5 1 5 2 6 1 20 11.11 B, C
Cg-Ba)
MSP 4 (FH- 2 2 0 1 1 0 6 3.33 C
Cg-S)
MSP 5 0 1 3 2 0 0 6 3.33 C
(Ba-Na-S)
MSP 6 0 0 1 1 1 1 4 2.22 C
(Ba-Cg-S)
MSP 7 (Ba- 6 8 3 8 5 6 36 20 B
Na-ANS)
MSP 8 (Ba- 1 1 4 2 0 1 9 5 C
Cg-ANS)
Total 30 30 30 30 30 30 180 100

Data were analyzed by generalized linear mixed model and multiple

comparison correction with Bonferroni tests, at a significance level
of P \0.05.

Fig 4. Example of menton deviation measured by candi-
date MSP or selected MSP.
However, the absolute value of differences were
significantly different in the A-P line deviation measure-
ment of MSP 4 and MSP 1, 2 or 3. It is assumed that the
amounts of lateral deviation of sella and crista galli
would strongly affect the yaw of the plane because of
the relatively short anteroposterior distance from sella
to crista galli.
On the other hand, MSPs 5 and 6, constructed based
on 3 median landmarks in the cranial base, were found
to be inappropriate for evaluating craniofacial asymmetry
(Table VI). In the report by Kim et al,13 10 MSPs con-
structed from median landmarks in the cranial base
were suggested as appropriate, compared with other
MSPs for evaluating craniofacial asymmetry, since the
distances to the median landmarks and middle points of
bilateral symmetric landmarks in the cranial base were
small. Nevertheless, according to our study, constructing
MSPs by using only landmarks in the cranial base made a
marked difference. It is assumed that the distortion in
facial asymmetry would be increased with little change
in the reference point positions in the left or right direc-
tion, since the vertical height difference between the
reference points in the cranial base was not larger than
that of the facial reference points. In addition, MSPs 5
and 6 would result in distortion of the patient's facial
asymmetry, with a tendency for underestimating the
menton and ANS deviations (Table VI). In the study of
Kim et al,10 the MSP constructed with only a reference
point in the cranial base could magnify the maxilloman-
dibular asymmetry compared with a MSP that included
the reference points in a facial area such as ANS, because
the subjects were not grouped by whether they had
craniofacial asymmetry.
In this study, MSPs 1, 2, 3, and 4, which are perpendic-
ular to the FH plane, showed fewer differences than did
MSPs passing through the 3 cranial base points when
evaluating the menton and ANS deviations of patients
with craniofacial asymmetry. Among MSPs 1, 2, 3, and
4, perpendicular to the FH plane, MSPs passing through
nasion in the anterior area showed fewer differences in
evaluating the menton deviation. In the posterior area,
there were no differences when evaluating the menton
deviation, regardless of whether we selected basion or
sella. Because menton is a landmark located in the ante-
rior area of the face, asymmetry evaluation results in the
facial area not markedly affected irrespective of the point
selected as the posterior reference point when construct-
ing the MSP. On the other hand, the anterior reference
point plays an important role when evaluating menton.
Jeon et al12 reported that using crista galli as the anterior
reference point makes it easier to set the center between
the left and right sides, because of the sharp form of crista
galli. However, according to our study, using nasion,

American Journal of Orthodontics and Dentofacial Orthopedics December 2017
Vol 152
Issue 6
796 An et al

which is a more anterior landmark, resulted in fewer dif-

AVDMe, Absolute value of differences in menton deviation measurements; AVDANS, absolute value of differences in ANS deviation measurements; AVDAP, absolute value of differences in A-P line
(Mean 6 SD)

deviation measurements; VDMe, value of differences in Menton deviation measurements; VDANS, value of differences in ANS deviation measurements; VDAP, value of differences in A-P line de-

*Data analyzed by linear mixed model and multiple comparison correction with Bonferroni test at significance level of P \0.05. Identical letters (A,B,C) indicate the lack of a significant difference
(Ba-Cg-ANS)

2.17 6 1.63

0.82 6 0.78

0.70 6 0.66

–0.61 6 2.65

–0.31 6 1.09

–0.15 6 0.95
ferences rather than when using crista galli as the anterior
MSP 8

reference point.
B

B
On the other hand, posterior landmarks as well as
anterior landmarks are important when evaluating
yaw. In this study, MSPs passing through nasion in the
(Ba-Na-ANS)

anterior region and basion in the posterior region could

(Mean 6 SD)
1.85 6 1.64

0.82 6 0.79

0.67 6 0.67

–0.76 6 2.36

–0.32 6 1.09

–0.23 6 0.92
MSP 7

result in fewer differences when evaluating yaw. There-

fore, the wider the measurement points located in the
B

B
anteroposterior area, the less the deviation. In the future,
studies should compare a MSP that includes opisthion,
the hindmost point of the foramen magnum, or other
(Ba-Cg-S) (Mean 6 SD)

posterior measurement points. Further studies should

be designed to evaluate the agreement of examiners in
5.96 6 6.42

3.11 6 3.47

1.95 6 2.52

–2.97 6 8.25

–1.81 6 4.30

–0.84 6 3.08

choosing the most appropriate MSP to diagnose the

MSP 6
Table VI. Values and absolute values of differences in menton, ANS, A-P line deviation measurements (n 5 180)

same asymmetry condition because this methodology

B

B
C

may reduce bias.

In craniofacial asymmetry, evaluations of soft and
hard tissues should be made concurrently. Further
studies will be needed to compare different MSP config-
urations on soft tissues. Additionally, a study involving
(Mean 6 SD)
5.30 6 5.41

2.45 6 2.52

1.53 6 1.91

–2.53 6 7.15

–1.10 6 3.34

–0.59 6 2.37
(Ba-Na-S)

the ratio of hard tissue to soft tissue changes would be

MSP 5

B, C

useful for diagnosing and establishing a treatment

B

B
C

plan for patients with craniofacial asymmetry.

CONCLUSIONS
(Mean 6 SD)
1.43 6 1.46

1.13 6 1.13

1.22 6 1.32

–0.12 6 2.04

–0.28 6 1.57

–0.44 6 1.74
(FH-Cg-S)

For menton deviation evaluation, the MSP passing

MSP 4

through nasion and basion and perpendicular to the FH

A, B

B, C
A

plane was preferred by the experts. However, the use of

MSPs perpendicular to the FH plane is not the most appro-
priate method to apply clinically in evaluating craniofacial
asymmetry. Using MSPs passing through 3 median
(Mean 6 SD)
1.37 6 1.37

1.04 6 0.96

0.77 6 0.71

0.04 6 1.94

–0.02 6 1.41

0.19 6 1.03
(FH-Cg-Ba)

landmarks in the cranial base can lead to underestimation

MSP 3

A, B

A, B

of the asymmetry of menton, ANS, and the A-P line.

A

Therefore, we suggest using MSPs perpendicular to the

FH plane or passing through ANS in clinical practice.
(Mean 6 SD)
0.81 6 1.31

0.44 6 0.66

0.82 6 1.05

0.14 6 1.54

0.08 6 0.79

–0.28 6 1.30

REFERENCES
(FH-Na-S)
MSP 2

1. Broadbent BH. A new x-ray technique and its application to

A

orthodontia. Angle Orthod 1931;1:45-66.

2. Ahlqvist J, Eliasson S, Welander U. The cephalometric projection.
Part II. Principles of image distortion in cephalography.
Dentomaxillofac Radiol 1983;12:101-8.
(Mean 6 SD)

3. Gravely JF, Benzies PM. The clinical significance of tracing error in

(FH-Na-Ba)

0.81 6 1.33

0.44 6 0.66

0.43 6 0.59

0.08 6 1.56

0.05 6 0.79

0.10 6 0.72

cephalometry. Br J Orthod 1974;1:95-101.

MSP 1

A, B

4. Houston WJ. The analysis of errors in orthodontic measurements.

A

Am J Orthod 1983;83:382-90.
5. Terajima M, Nakasima A, Aoki Y, Goto TK, Tokumori K, Mori N,
viation measurements.

et al. A 3-dimensional method for analyzing the morphology of

patients with maxillofacial deformities. Am J Orthod Dentofacial
between mean.
AVDMe (mm)*

VDANS (mm)*

Orthop 2009;136:857-67.
VDMe (mm)*
AVDAP ( )*

6. Kragskov J, Bosch C, Gyldensted C, Sindet-Pedersen S. Comparison

VDAP ( )*
AVDANS

of the reliability of craniofacial anatomic landmarks based on

(mm)*

cephalometric radiographs and three-dimensional CT scans. Cleft

Palate Craniofac J 1997;34:111-6.

December 2017
Vol 152
Issue 6 American Journal of Orthodontics and Dentofacial Orthopedics
An et al
7. Matteson SR, Bechtold W, Phillips C, Staab EV. A method for three-
dimensional image reformation for quantitative cephalometric
analysis. J Oral Maxillofac Surg 1989;47:1053-61.
8. De Momi E, Chapuis J, Pappas I, Ferrigno G, Hallermann W,
Schramm A, et al. Automatic extraction of the mid-facial plane
for cranio-maxillofacial surgery planning. Int J Oral Maxillofac
Surg 2006;35:636-42.
9. Hartmann J, Meyer-Marcotty P, Benz M, Hausler G, Stellzig-
Eisenhauer A. Reliability of a method for computing facial
symmetry plane and degree of asymmetry nased on 3D-data. J
Orofac Orthop 2007;68:477-90.
10. Kim TY, Baik JS, Park JY, Chae HS, Huh KH, Choi SC. Determina-
tion of midsagittal plane for evaluation of facial asymmetry using
three-dimensional computed tomography. Imaging Sci Dent
2011;41:79-84.
11. Yoon KW, Yoon SJ, Kang BC, Kim YH, Kook MS, Lee JS, et al. De-
viation of landmarks in accordance with methods of establishing
reference planes in three-dimensional facial CT evaluation.
Imaging Sci Dent 2014;44:207-12.
12. Jeon YN, Lee KH, Hwang HS. Validity of midsagittal reference planes
constructed in 3D CT images. Korean J Orthod 2007;37:182-91.
13. Kim HJ, Kim BC, Kim JG, Zhengguo P, Kang SH, Lee SH. Construc-
tion and validation of the midsagittal reference plane based on the
skull base symmetry for three-dimensional cephalometric
craniofacial analysis. J Craniofac Surg 2014;25:338-42.
14. Klingenberg CP, Barluenga M, Meyer A. Shape analysis of symmet-
ric structures: quantifying variation among individuals and
asymmetry. Evolution 2002;56:1909-20.
15. Gawlikowska A, Szczurowski J, Czerwinski F, Miklaszewska D,
Adamiec E, Dzieciolowska E. The fluctuating asymmetry of
medieval and modern human skulls. Homo 2007;58:159-72.
16. Damstra J, Fourie Z, De Wit M, Ren Y. A three-dimensional com-
parison of a morphometric and conventional cephalometric
American Journal of Orthodontics and Dentofacial Orthopedics
797
midsagittal planes for craniofacial asymmetry. Clin Oral Investig
2012;16:285-94.
17. Haraguchi S, Takada K, Yasuda Y. Facial asymmetry in subjects
with skeletal Class III deformity. Angle Orthod 2002;72:28-35.
18. Masuoka N, Muramatsu A, Ariji Y, Nawa H, Goto S, Ariji E. Discrim-
inative thresholds of cephalometric indexes in the subjective
evaluation of facial asymmetry. Am J Orthod Dentofacial Orthop
2007;131:609-13.
19. Moss ML, Salentijn L. Differences between the functional matrices
in anterior open-bite and in deep overbite. Am J Orthod 1971;60:
264-80.
20. 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.
21. Grummons DC, Kappeyne van de Coppello MA. A frontal asymme-
try analysis. J Clin Orthod 1987;21:448-65.
22. Tuncer BB, Atac MS, Yuksel S. A case report comparing 3-D
evaluation in the diagnosis and treatment planning of hemiman-
dibular hyperplasia with conventional radiography. J Craniomaxil-
lofac Surg 2009;37:312-9.
23. Pirttiniemi P, Miettinen J, Kantomaa T. Combined effects of errors in
frontal-view asymmetry diagnosis. Eur J Orthod 1996;18:629-36.
24. Trpkova B, Prasad NG, Lam EW, Raboud D, Glover KE, Major PW.
Assessment of facial asymmetries from posteroanterior
cephalograms: validity of reference lines. Am J Orthod Dentofacial
Orthop 2003;123:512-20.
25. Ras F, Habets LL, van Ginkel FC, Prahl-Andersen B. Method for
quantifying facial asymmetry in three dimensions using stereo-
photogrammetry. Angle Orthod 1995;65:233-9.
26. Vig PS, Hewitt AB. Asymmetry of the human facial skeleton. Angle
Orthod 1975;45:125-9.
December 2017
Vol 152
Issue 6