ORIGINAL ARTICLE

Three-dimensional evaluation of
dentofacial transverse widths in adults
with different sagittal facial patterns
Soonshin Hwang, Jueon Song, Joongoo Lee, Yoon Jeong Choi, Chooryung J. Chung, and Kyung-Ho Kim
Seoul, Korea

Introduction: The aim of this study was to evaluate the dentofacial transverse dimensions of subjects with
different sagittal facial patterns using 3-dimensional cone-beam computed tomography images. Methods:
Cone-beam computed tomography images of 63 men and 80 women were divided into skeletal Class I, Class
II, and Class III groups. Skeletal and dental evaluations were made on frontal views and coronal cross-sections
of the images. Independent 2-sample t tests and 1-way analysis of variance followed by post hoc Tukey tests
were used for sex and group differences. Pearson correlation analysis was used to identify factors related to
changes in ANB angle. Results: The Class II subjects did not show differences in maxillomandibular width
and maxillary width compared with Class I subjects; however, their maxillary molars were more lingually tipped.
The Class III subjects showed greater maxillomandibular width differences and smaller maxillary widths and
maxillary buccolingual alveolar widths at midroot level compared with Class I subjects. The maxillary molars
were buccally inclined, and the mandibular molars were lingually compensated in Class III subjects. The ANB
angle showed positive correlations with jugal process width, maxillary width, and maxillary buccolingual alveolar
width at midroot level as well as mandibular molar buccal inclination; negative correlations were found in max-
illomandibular width difference, mandibular width at midroot level, and maxillary molar buccal inclination.
Conclusions: A relative comparison of Class I, Class II, and Class III subjects showed that dental compensation
had occurred to overcome the transverse skeletal discrepancies in the maxillary posterior segments of Class II
and Class III subjects. This could escalate unidentified periodontal and functional problems in the long term.
Future studies of transverse dentofacial dimensions, including periodontal evaluations and occlusal forces,
would be useful for delivering proper orthodontic treatment for skeletal Class II and Class III subjects. (Am J
Orthod Dentofacial Orthop 2018;154:365-74)

T
ransverse discrepancy is commonly assessed upon
intraoral examination by a posterior crossbite,
with prevalence rates from 8% to 22%.1-4 The
etiology of this malocclusion is complex and is
associated with hereditary or environmental factors
such as skeletal or dental abnormalities, congenital
From the Department of Orthodontics, Gangnam Severance Dental Hospital,
Institute of Craniofacial Deformity, College of Dentistry, Yonsei University, Seoul,
Korea.
All authors have completed and submitted the ICMJE Form for Disclosure of Po-
tential Conflicts of Interest, and none were reported.
Supported by the National Research Foundation of Korea grant
(2017R1D1A1B03030851) funded by the Korean government (Ministry of Sci-
ence, ICT, and Future Planning; Ministry of Science, ICT, and Future Planning).
Address correspondence to: Kyung-Ho Kim, Department of Orthodontics, Gang-
nam Severance Dental Hospital, Institute of Craniofacial Deformity, College of
Dentistry, 211 Eonjuro, Gangnam-gu, Seoul 135-720, Korea; e-mail, khkim@
yuhs.ac.
Submitted, July 2017; revised and accepted, November 2017.
0889-5406/$36.00
Ó 2018 by the American Association of Orthodontists. All rights reserved.
https://doi.org/10.1016/j.ajodo.2017.11.041
anomalies including cleft palate, nonnutritive sucking
habits, mouth breathing, lowered tongue posture,
and so on.5-9 The early presence of a transverse
discrepancy should not be overlooked since it could
affect dentofacial growth in adverse directions because
sagittal and vertical growth continues even after
transverse dimensions are established.10 Moreover,
transverse coordination of the dental arches plays an
important role in maintaining stable periodontal condi-
tions and functional occlusion in the long term.11
In patients without apparent posterior crossbite, or-
thodontic diagnosis and treatment is generally focused
on sagittal discrepancies, which are assessed with lateral
cephalographs and study casts to categorize the subject
into Class I, Class II, or Class III malocclusion on the basis
of the anteroposterior position of the maxilla and
mandible as well as first molar relationships. However,
in complex malocclusion cases, additional posteroante-
rior radiographs or 3-dimensional (3D) cone-beam
computed tomography (CBCT) images would aid in
365
366
detecting unidentified transverse discrepancies that may
be clinically masked by the severity of the sagittal
discrepancy and dental misalignment.
Previous studies have used 2-dimensional posteroan-
terior radiographs or study casts to compare the dental
and skeletal components of transverse dimensions
in subjects with different sagittal facial patterns.12-14
These studies reported transverse discrepancy of
deciduous molars in skeletal Class II subjects due to a
narrow maxillary arch even with seemingly normal
maxillomandibular width relationships that continued
throughout the mixed dentition stage.13-15 In skeletal
Class III subjects, transverse discrepancies have been
indirectly indicated by clinical studies that included rapid
palatal expansion combined with facemask usage.12,14
To our knowledge, there have been no studies
directly comparing dentofacial transverse dimensions
using 3D images in adults with different sagittal
facial patterns. Evaluations and comparisons of den-
tofacial widths of skeletal Class I, Class II, and Class
III adults without apparent posterior crossbites would
aid in identifying characteristics of the “hidden”
transverse discrepancies that are difficult to diagnose
clinically. Therefore, the aim of this study was to
evaluate whether adults with different sagittal facial
patterns show differences in dentofacial transverse
dimensions and first molar inclinations using 3D
CBCT images.
MATERIAL AND METHODS
In this study, we evaluated adults who visited Gang-
nam Severance Dental Hospital in Seoul, Korea, from
January 2011 to February 2017. They had CBCT scans
(Pax-Zenith 3D; Vatech, Gyeonggi-Do, Korea) taken in
intercuspal occlusion for diagnostic purposes of
impacted third molars. The images were taken with a
scan time of 24 seconds, tube voltage of 105 kV(p),
and 0.3-mm voxel size. Pretreatment cephalograms
were generated from the CBCT scans to analyze the
subjects based on maxillomandibular sagittal and verti-
cal discrepancies. The subjects were included if the ver-
tical facial patterns were normodivergent, within the
range of 28
to 37
measured by the angle between
the sella-nasion line and the mandibular plane. This
value represented about 1 SD from the mean mandib-
ular plane angle of adults reported by Riedel.16 Sub-
jects were excluded from the study if they had
posterior crossbite, facial asymmetry greater than
2 mm, dental crowding or spacing greater than
5 mm, prosthetic treatment of the first molars, missing
or extracted permanent teeth excluding the third mo-
lars, prior orthodontic treatment, and significant med-
ical or dental history such as cleft lip and palate,
September 2018  Vol 154  Issue 3 American Journal of Orthodontics and Dentofacial Orthopedics
Hwang et al
craniofacial syndromes, or trauma. The final sample
included 143 adults (63 men, 80 women).
The sample was further divided into 3 groups depend-
ing on the maxillary and mandibular sagittal differences
by evaluating the ANB angle followed by checking the
first molar relationships: (1) skeletal Class I group, ANB
angle between 0
and 4
and an Angle Class I molar rela-
tionship17; (2) skeletal Class II group, ANB angle greater
than 4
and an Angle Class II molar relationship17; and
(3) skeletal Class III group, ANB angle less than 0
and
an Angle Class III molar relationship.17 In the Class I
group, the mesiobuccal cusp tip of the maxillary first
molar was located at the buccal groove of the mandibular
first molar. The mean ANB angles and ages of the subjects
in the 3 groups are shown in Table I. This study was
approved by the institution of research review board of
Gangnam Severance Hospital (3-2017-0034).
Skeletal and dental evaluations were performed on
the coronal cross-sections of the CBCT scans with OnDe-
mand3D imaging software (CyberMed, Seoul, Korea).
Cross-sections of 5-mm thickness were used to visualize
both the mesiobuccal and palatal roots of the maxillary
first molar on the same section. This was because a
thinner section might show a portion of the maxillary
root that could mislead the investigator as to the loca-
tion of the furcation.18 The following reference planes
were used to ensure consistent orientation of the 2-
dimensional coronal slices: (1) the axial plane was
defined as the Frankfort horizontal, which was the plane
passing through the bilateral orbitales and the right po-
rion; (2) the “first coronal plane” was perpendicular to
the axial plane, passing the buccal groove of the maxil-
lary right first molar; (3) the “second coronal plane” was
perpendicular to the axial plane, passing the mesiobuc-
cal groove of the mandibular right first molar; and (4) the
sagittal plane was perpendicular to the axial and coronal
planes passing the midpoint of the medial rims of the or-
bits. The first coronal plane was used to evaluate the
dentoskeletal transverse measurements of all 3 groups,
and the second coronal plane was used for additional
evaluation of the mandibular skeletal parameters of
the Class II and Class III groups (Fig 1).
From the frontal view of the CBCT scan, the trans-
verse distances between the right and left jugal processes
and the bilateral antegonial notches were measured.
Maxillomandibular transverse dimensions were obtained
on the first coronal plane using the following landmarks:
(1) buccal alveolar crest point, (2) 7-mm apical point
from the buccal alveolar crest, (3) lingual alveolar crest
point, (4) 7-mm apical point from the lingual alveolar
crest, and (5) most convex point on the buccal side of
the first molar. Additionally, the mandibular skeletal
transverse width was measured using the same
Hwang et al 367

Table I. Demographic data of the subjects

Class I Class II Class III

Men Women Men Women Men Women

n 20 30 17 28 26 22
Age (y) 22.63 6 3.99 25.95 6 8.54 20.76 6 3.84 26.95 6 8.99 20.44 6 2.81 23.52 6 6.98
ANB (
) 2.33 6 0.98 2.83 6 0.62 5.65 6 1.07 5.28 6 0.92 2.04 6 1.71 2.09 6 1.73
SN-MP (
) 35.48 6 1.73 33.83 6 1.65 34.4 6 2.12 33.96 6 1.88 33.93 6 1.40 33.93 6 1.85
ANB, Angle formed by A-point–nasion–B-point; SN-MP, angle between sella-nasion line and mandibular plane.

Fig 1. Transverse measurements in the frontal view and location of the first and second coronal planes
in the midsagittal plane view in Class I, Class II, and Class III groups: J-J width, tranverse width at bilat-
eral jugal process point; AG-AG width, transverse width at bilateral antegonial notch point; MxMn dif-
ference, difference between AG-AG width and J-J width.

landmarks on the second coronal plane in the skeletal
Class II and Class III groups. The molar widths and incli-
nations were measured on the first coronal plane of all 3
groups. The inclinations of the maxillary first molars
were obtained by measuring the angles formed by the
long axis of the molars and the Frankfort horizontal
plane, and for the mandibular first molars, angles
formed by the long axes of the molars and the line con-
necting the mandibular inferior borders were used.19,20
Palatal height was assessed by measuring the distance
between the functional occlusal plane and the most
superior point on the palate.21
For further evaluations, the following width differ-
ences were calculated from the obtained measure-
ments: (1) the maxillomandibular width difference
(bilateral antegonial notch width minus right and
left jugal process width), (2) the maxillary buccolin-
gual alveolar widths at the alveolar crest and 7 mm
apical from the alveolar crest, and (3) the mandibular
buccolingual alveolar widths at the alveolar crest and
7 mm apical from the alveolar crest in the first and
second coronal planes. All landmarks and measure-
ments used in this study are shown in Table II and
Figure 2.

American Journal of Orthodontics and Dentofacial Orthopedics September 2018  Vol 154  Issue 3
368 Hwang et al

Table II. Sex differences of transverse measurements and molar inclinations according to sagittal facial pattern
Class I Class II Class III

Variable Men Women P Men Women P Men Women P

Interjugular W (mm) 69.69 6 2.44 66.14 6 2.99 z 68.50 6 2.37 65.25 6 2.29 z 65.93 6 3.36 64.13 6 3.63
Antegonial W (mm) 90.38 6 3.99 86.50 6 2.70 z 91.89 6 4.26 85.42 6 3.22 z 91.42 6 3.98 89.06 6 4.24
MxMn difference (mm) 20.69 6 3.15 20.36 6 3.26 23.39 6 4.62 20.17 6 3.21 y 25.49 6 4.30 24.93 6 3.66
Mx alveolar crest W (mm) 59.95 6 3.07 58.15 6 2.52 * 60.48 6 3.23 57.79 6 3.00 y 59.31 6 3.06 56.51 6 2.93 *
Mx 7 mm W (mm) 66.72 6 2.58 63.39 6 3.19 z 66.32 6 3.19 63.06 6 2.56 z 62.77 6 4.17 58.79 6 4.12 *
1Mn alveolar crest W (mm) 58.34 6 4.98 57.06 6 2.42 54.93 6 2.73 53.36 6 2.34 * 60.49 6 3.33 57.35 6 2.74 *
1Mn 7 mm W (mm) 65.69 6 4.38 63.05 6 3.94 * 57.45 6 3.65 55.10 6 4.41 68.42 6 5.73 65.62 6 5.59
2Mn alveolar crest W (mm) 58.34 6 4.98 57.06 6 2.42 57.63 6 2.94 55.88 6 2.28 * 58.70 6 2.33 56.30 6 2.46 *
2Mn 7 mm W (mm) 65.69 6 4.38 63.05 6 3.94 * 63.96 6 5.38 61.50 6 5.03 64.50 6 4.86 62.29 6 5.49
Mx intermolar W (mm) 57.39 6 2.87 55.60 6 2.36 * 56.47 6 2.24 53.66 6 1.64 z 57.78 6 2.90 54.72 6 2.63 y
Mn intermolar W (mm) 48.75 6 2.97 48.76 6 2.62 50.56 6 3.43 49.37 6 2.44 y 53.66 6 3.69 50.50 6 2.83 y
Mx Rt inclination (
) 93.79 6 2.37 93.57 6 3.73 89.29 6 3.17 91.03 6 3.49 97.60 6 5.23 97.08 6 4.26
Mx Lt inclination (
) 94.66 6 2.62 94.40 6 3.81 92.61 6 3.95 93.49 6 3.69 98.57 6 5.15 100.13 6 3.15
Mn Rt inclination (
) 78.99 6 3.64 79.93 6 2.49 82.46 6 3.68 83.06 6 5.87 73.26 6 7.81 75.64 6 4.68
Mn Lt inclination (
) 79.85 6 3.04 79.32 6 2.65 81.77 6 4.84 81.39 6 5.12 73.63 6 7.54 74.3 6 4.22
Palatal height (mm) 20.03 6 2.09 19.28 6 2.02 17.70 6 1.48 18.10 6 2.22 19.86 6 2.04 19.25 6 1.66
BL Mx alveolar crest W (mm) 11.09 6 1.01 10.26 6 0.85 y 11.71 6 1.69 10.78 6 1.85 10.00 6 1.83 9.30 6 1.91
BL Mx 7 mm W (mm) 16.93 6 1.05 15.32 6 1.40 z 18.24 6 1.39 16.39 6 1.58 z 15.37 6 2.60 14.42 6 2.14
1BL Mn alveolar crest W (mm) 9.89 6 1.12 9.55 6 0.89 9.57 6 0.77 8.78 6 0.79 y 9.15 6 0.59 8.81 6 0.78
1BL Mn 7 mm W (mm) 17.77 6 2.43 15.84 6 2.40 y 16.44 6 2.29 15.40 6 2.11 15.19 6 3.44 14.76 6 2.70
2BL Mn alveolar crest W (mm) 9.89 6 1.12 9.55 6 0.89 9.77 6 0.87 9.13 6 0.91 * 9.19 6 0.71 8.96 6 1.12
2BL Mn 7 mm W (mm) 17.77 6 2.43 15.84 6 2.40 y 17.02 6 2.20 15.93 6 1.92 16.37 6 2.86 15.24 6 2.35
W, Width; Mx, maxillary; Mn, mandibular; MxMn difference, difference of antegonial and interjugular widths; 1, measurement from first coronal
plane; 2, measurement from second coronal plane; Rt, right side; Lt, left side; BL, buccolingual.
*P \0.05; yP \0.01; zP \0.001.

Table III. Difference of transverse and angular parameters according to sagittal facial pattern in men (ANOVA)
Variable (men) Class I Class II Class III P value
Interjugular W (mm) 69.9 6 2.44a 68.50 6 2.37a 65.93 6 3.36b 0.0001z
Antegonial W (mm) 90.38 6 3.99 91.89 6 4.26 91.42 6 3.98 0.5080
MxMn difference (mm) 20.69 6 3.15a 23.39 6 4.62ab 25.49 6 4.30b 0.0009z
Mx alveolar crest W (mm) 59.95 6 3.07 60.48 6 3.23 59.31 6 3.06 0.4775
Mx 7 mm W (mm) 66.72 6 2.58a 66.32 6 3.19a 62.57 6 4.17b 0.0002z
1Mn alveolar crest W (mm) 58.34 6 4.98a 54.93 6 2.73b 60.49 6 3.33a \0.0001z
1Mn 7 mm W (mm) 65.69 6 4.38a 57.45 6 3.65b 68.42 6 5.73a \0.0001z
2Mn alveolar crest W (mm) 58.34 6 4.98 57.63 6 2.94 58.70 6 2.33 0.6189
2Mn 7 mm W (mm) 65.69 6 4.38 63.96 6 5.38 64.50 6 4.86 0.5331
Mx intermolar W (mm) 57.39 6 2.87 56.47 6 2.24 57.78 6 2.90 0.3364
Mn intermolar W (mm) 48.75 6 2.97a 50.56 6 3.43a 53.66 6 3.69b \0.0001z
Mx Rt inclination (
) 93.76 6 2.37a 89.29 6 3.17b 97.60 6 5.23c \0.0001z
Mx Lt inclination (
) 94.66 6 2.62a 92.61 6 3.95a 98.75 6 5.15b \0.0001z
Mn Rt inclination (
) 78.99 6 3.64a 82.46 6 3.68a 73.26 6 7.81b \0.0001z
Mn Lt inclination (
) 79.85 6 3.04a 81.77 6 4.84a 73.63 6 7.54b \0.0001z
Palatal height (mm) 20.03 6 2.09a 17.70 6 1.48b 19.86 6 2.04a 0.0009z
BL Mx alveolar crest W (mm) 11.09 6 1.01ab 11.71 6 1.69a 10.00 6 1.83b 0.0042y
BL Mx 7 mm W (mm) 16.93 6 1.05a 18.24 6 1.39a 15.37 6 2.60b 0.0001z
1BL Mn alveolar crest W (mm) 9.89 6 1.12a 9.57 6 0.77ab 9.15 6 0.59b 0.0186*
1BL Mn 7 mm W (mm) 17.77 6 2.43 16.44 6 2.29 15.19 6 3.44 0.3513
2BL Mn alvoelar crest W (mm) 9.89 6 1.12a 9.77 6 0.87ab 9.19 6 0.71b 0.0352*
2BL Mn 7 mm W (mm) 17.77 6 2.43 17.02 6 2.20 16.37 6 2.86 0.1945
W, Width; Mx, maxillary; Mn, mandibular; MxMn difference, difference of antegonial and interjugular widths; 1, measurement from first coronal
plane; 2, measurement from second coronal plane; Rt, right side; Lt, left side; BL, buccolingual.
*P \0.05; yP \0.01; zP \0.001.
Superscript letters, values with the same letters in the superscript do not differ significantly (p.0.05).

September 2018  Vol 154  Issue 3 American Journal of Orthodontics and Dentofacial Orthopedics
Hwang et al 369

Fig 2. Transverse and angular measurements in the coronal plane: A1-B1 width, maxillary transverse
width at alveolar crest; C1-D1 width, maxillary transverse width at 7 mm apical from alveolar crest; E1-
F1 width, mandibular transverse width at alveolar crest; G1-H1 width, mandibular transverse width at
7 mm apical from alveolar crest; M1-M2 width, maxillary intermolar width; M3-M4 width, mandibular in-
termolar width; angle I, maxillary right first molar inclination; angle J, maxillary left first molar inclination;
angle K, mandibular right first molar inclination; angle L, mandibular left first molar inclination; palatal
height, distance from functional occlusal plane to most superior point on palate; buccolingual width
at maxillary alveolar crest, difference between A1-B1 width and A2-B2 width divided by 2; buccolingual
width 7 mm from maxillary alveolar crest, difference between C1-D1 width and C2-D2 width divided by
2; buccolingual width at mandibular alveolar crest, difference between E1-F1 width and E2-F2 width
divided by 2; buccolingual width 7 mm from mandibular alveolar crest, difference between G1-H1 width
and G2-H2 width divided by 2.

Statistical analysis 2-sample t tests. Differences between the Class I, Class

All transverse measurements were performed by 2
examiners (J.S., J.L.), and 20 randomly selected samples
were assessed for a second time at an interval of
2 weeks. The intraexaminer and interexaminer correla-
tion coefficients showed high reliability (r 5 0.90-
0.94). Based on a pilot study, a minimum sample size
of 16 subjects per group was determined (G* Power
3; Dusseldorf, Germany) at a 5 0.05, power of 80%,
and effect size of 0.47 to detect differences in the
transverse width at the 7-mm apical level from the
alveolar crest among the Class I, Class II, and Class III
groups by using analysis of variance (ANOVA). The
normality of the variables was tested by the
Kolmogorov-Smirnov test. Descriptive statistics were
calculated for all variables as means and standard de-
viations. Sex differences were analyzed by independent
II, and Class III groups were compared by 1-way AN-
OVA followed by post hoc Tukey HSD tests. Factors
affected by a change in the ANB angle were further
checked by Pearson correlation analysis. A P value
less than 0.05 was considered statistically significant.
All statistical analyses was performed using the SPSS
software (version 20.0; IBM, Armonk, NY).
RESULTS
Transverse widths measured at the maxillary alveolar
crest and 7 mm apical from the alveolar crest, and maxil-
lary intermolar width were significantly greater in the
men for all 3 groups.
In the Class I group, men had greater width values
at the jugal process and antegonial notch, 7 mm apical
from the mandibular alveolar crest measured on both

American Journal of Orthodontics and Dentofacial Orthopedics September 2018  Vol 154  Issue 3
370
coronal planes, buccolingual maxillary widths at the
alveolar crest and 7 mm apical from the alveolar crest,
and the buccolingual mandibular width 7 mm apical
from the alveolar crest measured on both coronal
planes.
In the Class II group, men had greater values for the
jugal process and antegonial notch width, maxilloman-
dibular width difference value (bilateral antegonial
notches minus right and left jugal processes), mandib-
ular alveolar crest measured on both coronal planes,
mandibular intermolar width, buccolingual maxillary
width 7 mm apical from the alveolar crest, and buccolin-
gual mandibular width at the alveolar crest measured on
both coronal planes. In the Class III group, men had
greater widths at the mandibular alveolar crest measured
on both coronal planes and greater mandibular intermo-
lar width (Table II).
For the sagittal facial pattern difference in men, the
interjugular width was smaller in the Class III group
compared with the other 2 groups; this resulted in a
greater maxillomandibular width difference for the Class
III group compared with the Class I group. The maxillary
width 7 mm apical from the alveolar crest was signifi-
cantly smaller in the Class III group compared with the
other 2 groups. The mandibular widths at the alveolar
crest and 7 mm apical from the alveolar crest were
smaller in the Class II group compared with the other 2
groups when measured on the first coronal plane. The
mandibular intermolar width was significantly greater
in the Class III group compared with the Class I group.
The maxillary right and left first molar inclinations
were significantly greater, and the mandibular right
and left first molar inclinations were significantly smaller
in the Class III group compared with the Class I and Class
II groups. The palatal height was smaller in the Class II
group compared with the other groups. The buccolin-
gual maxillary alveolar width at the alveolar crest was
thinner in the Class III group compared with the Class
II group. The buccolingual maxillary alveolar width
7 mm apical from the alveolar crest was also thinner in
the Class III group compared with the Class I and Class
II groups. The buccolingual mandibular width at the
alveolar crest was significantly greater in the Class I
group compared with the Class III group measured on
both coronal planes (Table III).
In the women, the antegonial width was signifi-
cantly greater in the Class III group compared with
the Class I group; this resulted in a greater maxilloman-
dibular width difference in the Class III group
compared with the Class I and Class II groups. The
maxillary width 7 mm apical from the alveolar crest
was smaller in the Class III group compared with the
September 2018  Vol 154  Issue 3 American Journal of Orthodontics and Dentofacial Orthopedics
Hwang et al
other 2 groups. The mandibular widths at the alveolar
crest and 7 mm apical from the alveolar crest were
smaller in the Class II group compared with the other
2 groups when measured from the first coronal plane.
The maxillary intermolar width was smaller in the Class
II group compared with the other 2 groups, and the
mandibular intermolar width was greater in the Class
III group compared with the Class I group. The maxil-
lary right and left first molar inclinations were signifi-
cantly greatest in the Class III group, whereas the
Class II group had the smallest inclination. The
mandibular right and left first molar inclinations were
significantly smaller in the Class III group compared
with the Class I and Class II groups. The palatal height
was lower in the Class II group compared with the other
2 groups. The buccolingual maxillary widths at the
alveolar crest and 7 mm apical from the alveolar crest
were greater in the Class II group compared with the
Class III group. The buccolingual mandibular width at
the alveolar crest was greater in the Class I group
compared wtih the other 2 groups measured from the
first coronal plane (Table IV).
Differences between mandibular width measure-
ments on the first and second coronal planes were
analyzed in the Class II and Class III groups. The mandib-
ular widths at the alveolar crest and 7 mm apical to the
alveolar crest were significantly greater on the second
coronal plane compared with the first coronal plane in
the Class II group for both sexes. In the Class III group,
the mandibular width 7 mm apical from the alveolar
crest was significantly greater on the first coronal plane
compared with the second coronal plane for both sexes.
The mandibular width at the alveolar crest was also
significantly greater on the first coronal plane in the
Class III group for the men (Table V).
As the ANB angle increased, the jugal process width,
maxillary width 7 mm apical from the alveolar crest,
mandibular right and left first molar inclinations, maxil-
lary buccolingual widths at the alveolar crest and 7 mm
apical from the alveolar crest increased whereas the
maxillomandibular width difference, mandibular widths
at the alveolar crest and 7 mm apical from the alveolar
crest measured from the first coronal plane, and the
maxillary right and left first molar inclinations decreased
in both sexes. In the men, palatal height decreased and
buccolingual mandibular width at the alveolar crest
measured on the first and second coronal planes
increased, and the buccolingual mandibular width
7 mm apical from the alveolar crest measured on the first
coronal plane increased as the ANB angle increased. In
the women, the antegonial width decreased as the
ANB angle increased. Most of the variables that were
Hwang et al 371

Table IV. Difference of transverse and angular parameters according to sagittal facial pattern in women (ANOVA)
Variable (women) Class I Class II Class III P value
Interjugular W (mm) 66.14 6 2.99 65.25 6 2.29 64.13 6 3.63 0.061
Antegonial W (mm) 86.50 6 2.70a 85.42 6 3.22ab 89.06 6 4.24b 0.0011*
MxMn difference (mm) 20.36 6 3.26a 20.17 6 3.21a 24.93 6 3.66b \0.0001y
Mx alveolar crest W (mm) 58.15 6 2.52 57.79 6 3.00 56.51 6 2.93 0.1086
Mx 7 mm W (mm) 63.39 6 3.19a 63.06 6 2.56a 58.79 6 4.12b \0.0001y
1Mn alveolar crest W (mm) 57.06 6 2.42a 53.36 6 2.34b 57.35 6 2.74a \0.0001y
1Mn 7 mm W (mm) 63.05 6 3.94a 55.10 6 4.41b 65.62 6 5.59a \0.0001y
2Mn alveolar crest W (mm) 57.06 6 2.42 55.88 6 2.28 56.30 6 2.46 0.1676
2Mn 7 mm W (mm) 63.05 6 3.94 61.50 6 5.03 62.29 6 5.49 0.4718
Mx intermolar W (mm) 55.60 6 2.36a 53.66 6 1.64b 54.72 6 2.63a 0.0072*
Mn intermolar W (mm) 48.76 6 2.62a 49.37 6 2.44ab 50.08 6 2.70b 0.0007y
Mx Rt inclination (
) 93.57 6 3.73a 91.03 6 3.49b 97.08 6 4.26c \0.0001y
Mx Lt inclination (
) 94.40 6 3.81a 93.49 6 3.69a 100.13 6 3.15b \0.0001y
Mn Rt inclination (
) 79.93 6 2.49a 82.55 6 5.88a 75.64 6 4.68b \0.0001y
Mn Lt inclination (
) 79.32 6 2.65a 81.39 6 5.12a 74.30 6 4.22b \0.0001y
Palatal height (mm) 19.28 6 2.02 18.10 6 2.22 19.25 6 1.66 0.054
BL Mx alveolar crest W (mm) 10.26 6 0.85ab 10.78 6 1.85a 9.30 6 1.91b 0.006*
BL Mx 7 mm W (mm) 15.32 6 1.40a 16.39 6 1.58b 14.42 6 2.14a 0.0005y
1BL Mn alveolar crest W (mm) 9.55 6 0.89a 8.78 6 0.79b 8.81 6 0.78b 0.0008y
1BL Mn 7 mm W (mm) 15.84 6 2.40 15.40 6 2.11 14.76 6 2.70 0.6701
2BL Mn alvoelar crest W (mm) 9.55 6 0.89 9.13 6 0.91 8.96 6 1.12 0.0748
2BL Mn 7 mm W (mm) 15.84 6 2.40 15.93 6 1.92 15.24 6 2.35 0.5147
W, Width; Mx, maxillary; Mn, mandibular; MxMn difference, difference of antegonial and interjugular widths; 1, measurement from first coronal
plane; 2, measurement from second coronal plane; Rt, right side; Lt, left side; BL, buccolingual.
*P \0.01; yP \0.001.
Superscript letters, values with the same letters in the superscript do not differ significantly (p.0.05).

correlated showed moderate correlation values to
changes in the ANB angle (Table VI).
DISCUSSION
Clinicians are focused on improving sagittal discrep-
ancies and lateral facial profiles unless there are evident
posterior crossbites in adult orthodontic patients. It is
easy to underestimate transverse discrepancies that are
masked by dental compensation. Therefore, in this
study, we evaluated and compared transverse parame-
ters of skeletal Class I, Class II, and Class III subjects
without clinically diagnosed posterior crossbites that
may have been hidden by dental compensations.
Men had significantly greater transverse widths at
the maxillary alveolar crest and midroot level as well
as maxillary intermolar widths compared with women
in all 3 groups. Although there were some variations
between the groups, most skeletal and dental trans-
verse measurements had greater values in men; this
agreed with previous studies that mentioned that cra-
nial and facial widths are significantly affected by
sex.22-24 Accordingly, dentofacial widths in different
sagittal facial patterns were evaluated in both sexes
(Table II).
When comparing Class II with Class I subjects, there
were no significant differences in jugal process,
antegonial width, or maxillomandibular width differ-
ence between the 2 groups. The Class II group did not
show significant differences in maxillary width at the
alveolar crest or midroot level compared with the Class
I group. However, mandibular transverse measurements
in the Class II group at the alveolar crest and midroot
level measured on the first coronal plane were signifi-
cantly smaller than those of the Class I group. This was
expected because of the relatively posterior position of
the mandible compared with the maxilla in the Class II
group and the fact that alveolar widths in the premolar
areas are narrower compared with molars.25 The maxil-
lary first molar inclinations in the Class II group were
significantly less than those of the Class I group in
women, with a similar trend in men, but the mandibular
molar inclinations and mandibular intermolar widths did
not show significant differences between both groups
for both sexes. This implies that lingual tipping had
occurred in the maxillary first molars, and there was no
significant change in the mandibular first molars to
overcome the transverse discrepancy caused by the
sagittal discrepancy. The results of this study agree
with previous reports in that transverse discrepancies
in Class II subjects originate from the maxillary posterior
dentition and not from the maxillary alveolar base.26,27
In addition, there were no significant differences

American Journal of Orthodontics and Dentofacial Orthopedics September 2018  Vol 154  Issue 3
372 Hwang et al

between Class I and Class II subjects in the buccolingual

P value
maxillary width at the alveolar crest for both sexes and at
the midroot level for men. Although the Class II women

*
showed significantly greater buccolingual maxillary

56.30 6 2.46
61.89 6 5.30
8.96 6 1.12
15.24 6 2.35
Plane 2
widths at the midroot level, it is reasonable to assume
Class III that this would not have clinical significance in terms
of periodontal health because lingual tipping of the
57.35 6 2.74 maxillary first molars had occurred in thicker maxillary
65.62 6 5.59
8.81 6 0.78
14.76 6 2.70
buccolingual alveolar bone to achieve proper
Plane 1

transverse occlusion. However, occlusal function may

be affected because masticatory forces of lingually
Women

tipped maxillary molars are not vertically directed

P value

parallel to the long axis of the tooth.28

Table V. Differences in mandibular transverse measurements in coronal planes 1 and 2 for Class II and Class III subjects

In the comparison of Class III with Class I subjects, the

z
z

Plane 1, Measurement from first coronal plane; Plane 2, measuremen from second coronal plane; W, width; Mn, mandibular; BL, buccolingual.

jugal process width was significantly smaller in the Class

55.88 6 2.28
61.50 6 5.03
9.13 6 0.91
15.93 6 1.92

III group for men, with a similar tendency in women. This

Plane 2

resulted in greater maxillomandibular width differences

Class II

of 25.49 mm in men and 24.93 mm in women; these

values are about 5 mm greater than the normal value
53.36 6 2.34
55.10 6 4.41
8.78 6 0.79
15.40 6 2.11

of 19.6 mm used in the maxillomandibular transverse

Plane 1

differential index at growth completion reported by

Ricketts et al.29 The maxillary width at midroot level
was also significantly smaller in the Class III group;
this agreed with previous studies that showed narrower
P value

maxillary arches in Class III patients.30,31 However,

z
y

there was no significant difference of mandibular

58.70 6 2.33
64.26 6 4.80
9.19 6 0.71
16.37 6 2.86

transverse width at the alveolar crest or midroot level

Plane 2

between the Class III and Class I groups measured

Class III

from the first and second coronal planes that were

different from the results of the Class II group.
Nonetheless, the skeletal width difference between the
60.49 6 3.33
68.42 6 5.73
9.15 6 0.59
15.19 6 3.44

maxilla and the mandible in Class III patients was

Plane 1

compensated by significant buccally flared maxillary

and lingually tipped mandibular molars. The results
Men

from this study suggest that Class III adults who

P value

undergo camouflage treatment would need transverse

dental compensation to overcome the narrow maxilla
z
z

and the anteroposterior overjet correction. However,

57.63 6 2.94
63.96 6 5.38
9.77 6 0.87
17.02 6 2.20

the maxillary buccolingual alveolar width at the

Plane 2
Class II

midroot level was significantly smaller in the Class III

group compared with that of the Class I group in men,
with a similar trend in women. This necessitates
54.93 6 2.73
57.45 6 3.67
9.57 6 0.77
16.44 6 2.29

preliminary evaluation of the maxillary alveolar width

Plane 1

in Class III patients even without posterior crossbites

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

before orthodontic treatment to maintain periodontal

health as the maxillary molars are buccally tipped to
achieve a proper posterior occlusion. In some subjects
BL Mn alveolar crest W
Mn alveolar crest W

with thin maxillary buccolingual alveolar widths,

maxillary expansion using miniscrew-assisted rapid
BL Mn 7 mm W
Mn 7 mm W

palatal expansion could be used to increase functional

occlusion, periodontal health, and long-term stability.32
Most parameters with statistical significance in the
Pearson correlation analysis had moderate correlation

September 2018  Vol 154  Issue 3 American Journal of Orthodontics and Dentofacial Orthopedics
Hwang et al 373

Table VI. Transverse or angular measurements related to increased ANB angle (Pearson correlation)
ANB angle

Variable r (men) P value (men) r (women) P value (women)

Interjugular W (mm) 0.4067 0.0009z 0.2438 0.0293*
Antegonial W (mm) 0.0117 0.9276 0.3352 0.0024*
MxMn difference (mm) 0.3049 0.0151* 0.4988 \0.0001z
Mx alveolar crest W (mm) 0.1029 0.4221 0.1946 0.0837
Mx 7 mm W (mm) 0.4681 0.0001z 0.5192 \0.0001z
1Mn alveolar crest W (mm) 0.5137 \0.0001z 0.4732 \0.0001z
1Mn 7 mm W (mm) 0.6410 \0.0001z 0.6431 \0.0001z
2 Mn alveolar crest W (mm) 0.1653 0.1993 0.0230 0.8398
2 Mn 7 mm W (mm) 0.1194 0.3515 0.0420 0.7118
Mx intermolar W (mm) 0.2405 0.0666 0.1357 0.2392
Mn intermolar W (mm) 0.1977 0.1404 0.1032 0.3625
Mx Rt inclination (
) 0.6595 \0.0001z 0.5598 \0.0001z
Mx Lt inclination (
) 0.5248 \0.0001z 0.5815 \0.0001z
Mn Rt inclination (
) 0.5721 \0.0001z 0.5940 \0.0001z
Mn Lt inclination (
) 0.4799 0.0002z 0.6411 \0.0001z
Palatal height (mm) 0.3356 0.0094y 0.1873 0.0984
BL Mx alveolar crest W (mm) 0.4303 0.0007z 0.2781 0.0131*
BL Mx 7 mm W (mm) 0.5979 \0.0001z 0.4533 \0.0001z
1 BL Mn alveolar crest W (mm) 0.3029 0.0197* 0.0261 0.8186
1 BL Mn 7 mm W (mm) 0.3031 0.0166* 0.0113 0.9217
2 BL Mn alveolar crest W (mm) 0.3261 0.0142* 0.0917 0.4183
2 BL Mn 7 mm W (mm) 0.1532 0.2384 0.1518 0.1845
W, Width; Mx, maxillary; Mn, mandibular; MxMn difference, difference of antegonial and interjugular width; 1, measurement from first coronal
plane; 2, measurement from second coronal plane; Rt, right side; Lt, left side; BL, buccolingual.
*P \0.05; yP \0.01; zP \0.001.

values. As the subjects showed a tendency toward skel-
etal Class II by an increase in the ANB angle, the jugal
process width, maxillary alveolar width at the midroot
level, maxillary buccolingual alveolar widths at the alve-
olar crest and midroot level increased for both sexes. On
the contrary, the maxillomandibular width difference
and mandibular widths at the alveolar crest and midroot
level measured on the first coronal plane decreased. The
skeletal transverse discrepancies were compensated by a
tendency of lingual tipping of the maxillary molars and
buccal flaring of the mandibular molars as the ANB
angle increased.
In this retrospective study, we compared the trans-
verse skeletal and dental parameters of Class I, Class II,
and Class III subjects without obvious clinical posterior
crossbites who had completed their growth. Therefore,
extreme cases of sagittal and transverse discrepancies
were not included; this was a limitation of this study.
In addition, although the subjects had normodivergent
profiles, individual variations in occlusal and mandibular
plane angles should have been considered. Despite the
possibility of dental compensation in Class I subjects
with good clinical posterior occlusion,18 a relative com-
parison of Class I, Class II, and Class III subjects showed
that dental compensation had occurred to overcome the
transverse skeletal discrepancy in the maxillary posterior
segments of both Class II and Class III subjects; this may
escalate unidentified periodontal and functional prob-
lems in the long term. Future studies of transverse den-
tofacial dimensions, including periodontal evaluations
and occlusal forces, would be useful for delivering
proper orthodontic treatment for skeletal Class II and
Class III subjects.
CONCLUSIONS
1. Men had greater transverse widths measured at the
maxillary alveolar crest and midroot level as well as
maxillary intermolar widths compared with women
in the Class I, Class II, and Class III groups.
2. Class II subjects did not show a significant differ-
ence in maxillomandibular width and maxillary
width at the alveolar crest or midroot level
compared with the Class I group.
3. In the Class II subjects, the maxillary first molars
were lingually tipped, but there was no significant
difference in the mandibular first molars to over-
come the skeletal transverse discrepancy.
4. The Class III subjects showed significantly greater
maxillomandibular width differences, and smaller

American Journal of Orthodontics and Dentofacial Orthopedics September 2018  Vol 154  Issue 3
374
maxillary and buccolingual alveolar widths at the
midroot level compared with Class I subjects.
5. In Class III subjects, skeletal transverse discrepancy
was compensated by significant buccally flared
maxillary and lingually tipped mandibular first mo-
lars.
6. The ANB angle showed positive correlations with ju-
gal process width, maxillary width, and maxillary
buccolingual alveolar width at midroot level as
well as mandibular molar buccal inclination; nega-
tive correlations were found in maxillomandibular
width difference, mandibular width at midroot level,
and maxillary molar buccal inclination.
A relative comparison of Class I, Class II, and Class III
subjects showed that dental compensation had occurred
to overcome the transverse skeletal discrepancy in the
maxillary posterior segments of both Class II and Class
III subjects. This could escalate unidentified periodontal
and functional problems in the long term. Future studies
of transverse dentofacial dimensions, including peri-
odontal evaluations and occlusal forces, would be useful
for delivering proper orthodontic treatment for skeletal
Class II and Class III subjects.
REFERENCES
1. Kisling E. Occlusal interferences in the primary dentition. ASDC J
Dent Child 1981;48:181-91.
2. Almeida RR, Almeida MR, Oltramari-Navarro PV, Conti AC,
Navarro Rde L, Marques HV. Posterior crossbite—treatment and
stability. J Appl Oral Sci 2012;20:286-94.
3. Foster TD, Hamilton MC. Occlusion in the primary dentition. Study
of children at 2 and one-half to 3 years of age. Br Dent J 1969;126:
76-9.
4. Day AJ, Foster TD. An investigation into the prevalence of molar
crossbite and some associated aetiological conditions. Dent Pract
Dent Rec 1971;21:402-10.
5. Kantomaa T. Correction of unilateral crossbite in the deciduous
dentition. Eur J Orthod 1986;8:80-3.
6. Kutin G, Hawes RR. Posterior cross-bites in the deciduous and
mixed dentitions. Am J Orthod 1969;56:491-504.
7. Bishara SE, Warren JJ, Broffitt B, Levy SM. Changes in the preva-
lence of nonnutritive sucking patterns in the first 8 years of life. Am
J Orthod Dentofacial Orthop 2006;130:31-6.
8. Lofstrand-Tidestrom B, Thilander B, Ahlqvist-Rastad J,
Jakobsson O, Hultcrantz E. Breathing obstruction in relation to
craniofacial and dental arch morphology in 4-year-old children.
Eur J Orthod 1999;21:323-32.
9. Luzzi V, Ierardo G, Viscogliosi A, Fabbrizi M, Consoli G, Vozza I,
et al. Allergic rhinitis as a possible risk factor for malocclusion: a
case-control study in children. Int J Paediatr Dent 2013;23:
274-8.
10. Cortella S, Shofer FS, Ghafari J. Transverse development of the
jaws: norms for the posteroanterior cephalometric analysis. Am J
Orthod Dentofacial Orthop 1997;112:519-22.
11. Vanarsdall RL, White RP Jr. Three-dimensional analysis for skeletal
problems. Am J Orthod Dentofacial Orthop 1995;107(2):22a-23a.
September 2018  Vol 154  Issue 3 American Journal of Orthodontics and Dentofacial Orthopedics
Hwang et al
12. Baik HS. Clinical results of the maxillary protraction in Korean chil-
dren. Am J Orthod Dentofacial Orthop 1995;108:583-92.
13. Tollaro I, Baccetti T, Franchi L, Tanasescu CD. Role of posterior
transverse interarch discrepancy in Class II, Division 1 malocclusion
during the mixed dentition phase. Am J Orthod Dentofacial Orthop
1996;110:417-22.
14. Franchi L, Baccetti T. Transverse maxillary deficiency in Class II
and Class III malocclusions: a cephalometric and morphometric
study on postero-anterior films. Orthod Craniofac Res 2005;8:
21-8.
15. Baccetti T, Franchi L, McNamara JA Jr, Tollaro I. Early dentofacial
features of Class II malocclusion: a longitudinal study from the de-
ciduous through the mixed dentition. Am J Orthod Dentofacial Or-
thop 1997;111:502-9.
16. Riedel R. The relation of maxillary structures to cranium in
malocclusion and normal occlusion. Angle Orthod 1952;22:
142-55.
17. Angle EH. Treatment of malocclusion of the teeth: Angle's system.
Philadelphia: S. S. White Dental Manufacturing Company; 1907.
18. Miner RM, Al Qabandi S, Rigali PH, Will LA. Cone-beam computed
tomography transverse analysis. Part I: normative data. Am J Or-
thod Dentofacial Orthop 2012;142:300-7.
19. Shewinvanakitkul W, Hans M, Narendran S, Palomo JM.
Measuring buccolingual inclination of mandibular canines
and first molars using CBCT. Orthod Craniofac Res 2011;14:
168-74.
20. Ahn J, Kim SJ, Lee JY, Chung CJ, Kim KH. Transverse dental
compensation in relation to sagittal and transverse skeletal dis-
crepancies in skeletal Class III patients. Am J Orthod Dentofacial
Orthop 2017;151:148-56.
21. Jacobson A. Update on the Wits appraisal. Angle Orthod 1988;58:
205-19.
22. Forster CM, Sunga E, Chung CH. Relationship between dental arch
width and vertical facial morphology in untreated adults. Eur J Or-
thod 2008;30:288-94.
23. Meredith HV. Growth in bizygomatic face breadth during child-
hood. Growth 1954;18:111-34.
24. Yavuz I, Ikbal A, Baydas B, Ceylan I. Longitudinal posteroanterior
changes in transverse and vertical craniofacial structures between
10 and 14 years of age. Angle Orthod 2004;74:624-9.
25. Uysal T, Memili B, Usumez S, Sari Z. Dental and alveolar arch
widths in normal occlusion, class II division 1 and class II division
2. Angle Orthod 2005;75:941-7.
26. Staley RN, Stuntz WR, Peterson LC. A comparison of arch widths in
adults with normal occlusion and adults with Class II, Division 1
malocclusion. Am J Orthod 1985;88:163-9.
27. Sayin MO, Turkkahraman H. Comparison of dental arch and alve-
olar widths of patients with Class II, division 1 malocclusion and
subjects with Class I ideal occlusion. Angle Orthod 2004;74:
356-60.
28. Dawson E. Evaluation, diagnosis and treatment of occlusal prob-
lems. St Louis: C. V. Mosby; 1989.
29. Ricketts RM, Bench RW, Hilgers JJ, Schulhof R. An overview of
computerized cephalometrics. Am J Orthod 1972;61:1-28.
30. Slaj M, Spalj S, Pavlin D, Illes D, Slaj M. Dental archforms in den-
toalveolar Class I, II and III. Angle Orthod 2010;80:919-24.
31. Braun S, Hnat WP, Fender DE, Legan HL. The form of the human
dental arch. Angle Orthod 1998;68:29-36.
32. Park JJ, Park YC, Lee KJ, Cha JY, Tahk JH, Choi YJ. Skeletal and
dentoalveolar changes after miniscrew-assisted rapid palatal
expansion in young adults: a cone-beam computed tomography
study. Korean J Orthod 2017;47:77-86.