See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/12510043

Three-dimensional morphology of the masseter muscle in patients with

mandibular prognathism

Article  in  Dentomaxillofacial Radiology · March 2000

DOI: 10.1038/sj/dmfr/4600515 · Source: PubMed

CITATIONS READS

32 166

7 authors, including:

Yoshiko Ariji Akitoshi Katsumata

Aichi Gakuin University Asahi University
125 PUBLICATIONS   2,096 CITATIONS    144 PUBLICATIONS   2,317 CITATIONS   

SEE PROFILE SEE PROFILE

Kenji Yoshida Eiichiro Ariji

Aichi Gakuin University Aichi Gakuin University
165 PUBLICATIONS   860 CITATIONS    225 PUBLICATIONS   4,192 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

photon counting imaging View project

Application development by use our developed energy resolved photon counting X ray detector. View project

All content following this page was uploaded by Akitoshi Katsumata on 02 June 2014.

The user has requested enhancement of the downloaded file.

 Dentomaxillofacial Radiology (2000) 29, 113 ± 118
ã 2000 Macmillan Publishers Ltd. All rights reserved 0250 ± 832X/00 $15.00
www.nature.com/dmfr

Three-dimensional morphology of the masseter muscle in patients

with mandibular prognathism
Y Ariji*,1, A Kawamata2, K Yoshida3, S Sakuma4, H Nawa5, M Fujishita2 and E Ariji1
1
Department of Oral and Maxillofacial Radiology, School of Dentistry, Aichi-Gakuin University, Nagoya, Japan; 2Department of
Oral Radiology, School of Dentistry, Asahi University, Gifu, Japan; 3First Department of Oral and Maxillofacial Surgery, School of
Dentistry, Aichi-Gakuin University; Nagoya, Japan; 4Department of Fixed Prosthodontics, School of Dentistry, Aichi-Gakuin
University, Nagoya, Japan; 5Department of Orthodontics, School of Dentistry, Aichi-Gakuin University, Nagoya, Japan

Objective: To compare the morphology of the masseter muscle in patients with mandibular
prognathism with that of normal subjects.
Methods: Three-dimensional X-ray computed tomography (CT) was performed on 69 patients
with mandibular prognathism and compared with 91 normal subjects. The angle of the muscle
direction in relation to the Frankfurt horizontal plane and the area and the ratio of length of
the short to long axes (s/l ratio) on the section perpendicular to the muscle direction were
measured.
Results: The mean angle, area and s/l ratio in patients with mandibular prognathism was 76.68
(s.d. 4.48), 318.3 mm2 (s.d. 77.2 mm2) and 0.312 (s.d. 0.049), respectively. Those of the normal
subjects were 65.18 (s.d. 4.48), 368.3 mm2 (s.d. 97.2 mm2) and 0.393 (s.d. 0.054), respectively.
The angle was signi®cantly larger, and the area and s/l ratio were signi®cantly smaller than
those of normal subjects (P50.001).
Conclusion: The morphology of the masseter muscle in mandibular prognathism is
signi®cantly di€erent from that of normal subjects. Our results may be helpful in evaluating
the results of orthognathic surgery.

Keywords: prognathism; masseter muscle; tomography, X-ray computed

Introduction

The function of the jaw musculature is considered to
be a determinant of craniofacial growth and develop-
ment1 and a signi®cant relationship has been found
between the masticatory muscles and facial morphol-
ogy in normal subjects.1 ± 4 Functional activity and bite
force have been shown to di€er signi®cantly from
normal subjects in patients with mandibular prognath-
ism.5 ± 12 These factors are also reported to alter after
orthognathic surgery.13 ± 16 Muscle morphology could
also be a signi®cant parameter to consider in planning
surgery and predicting the outcome.
Various methods have been described for measuring
muscle morphology. The direction and cross-sectional
area of the masseter muscle have frequently been
*Correspondence to: Y Ariji, Department of Oral and Maxillofacial Radiology,
School of Dentistry, Aichi-Gakuin University, 2 ± 11 Suemori-dori, Chikusa-ku,
Nagoya 464 ± 8651, Japan
Received 21 September 1999, accepted 13 December 1999
measured from cross-sectional images obtained by CT
and MRI.2,17,18 Cross-sectional images are usually
obtained by scanning parallel to the horizontal or
occlusal plane. Wejis and coworkers have asserted the
importance of obtaining images perpendicular to the
muscle ®bres and proposed scanning at 308 to the
Frankfurt horizontal (FH) plane for the masseter and
medial pterygoid muscles.17 However, this technique
cannot be applied to patients with mandibular
prognathism, since the direction of the masseter
muscle is considerably di€erent from normal.4,9,10,19
So-called centroid methods have been proposed for
determination of the direction of muscle contraction.20
While it is possible to calculate the direction and cross-
sectional area for individual subjects with this method,
it is time-consuming and inappropriate for clinical use
since each image has to be individually traced.
In the present study, we describe a method for
determining the direction and cross-sectional area of
the masseter muscle from three-dimensional CT
 Masseter muscle morphology
Y Ariji et al
114
reconstruction, and de®ne its three-dimensional mor-
phological characteristics in patients with mandibular a
prognathism.

Methods

Subjects
Sixty-nine patients with mandibular prognathism with-
out asymmetry were studied (31 men and 38 women,
aged between 17 and 28 years (mean, 22.5 s.d. 3.8). The
patients underwent CT examination 12 months before
orthognathic surgery in the Department of Oral and
Maxillofacial Radiology, Aichi-Gakuin University
Dental Hospital and Department of Oral Radiology,
Asahi University Dental Hospital between January 1993
and February 1998. Ninety-one age- and sex- matched
normal subjects were retrospectively selected as a control
group from patients who had undergone CT for other
maxillofacial diseases, such as a tumor or infection b
during the same period. Subjects were selected on the
basis that both masseter muscles were entirely scanned
and the muscles were asymptomatic and free from any
pathology. The ®nal control group consisted of 49 men
and 42 women ranging in age from 17 to 29 years (mean
24.2 s.d. 3.5).

CT examination and three-dimensional reconstruction

The CT machines used were a Somatom Art (Siemens
AG, Erlangen, Germany) and an X-Vision (Toshiba,
Tokyo, Japan). The slice thickness of the reconstructed
images was 1 or 2 mm. The occlusal or Frankfurt
horizontal (FH) planes were used as the radiographic
base line and the region extending from the orbit to the
chin was scanned in all examinations. The resultant data
were stored on a magneto-optical disk and then
converted to three-dimensional CT images with one or
other of two personal computers (Power Macintosh c
9500/200 and 8600/250, Apple Computer, Cupertino,
CA, USA). The software used was Voxel view (Vital
Images, Inc., Tokyo, Japan) for 3D reconstruction and
measurement of the direction and Med Vision (Innet
Systems, Castine, ME, USA) for window control,
®ltration and measurement of the cross-sectional images.

Measurement of muscle morphology

The skin surface and subcutaneous fat were deleted on
each axial image and the masseter muscle exposed
through the adjustment of the window width and
smoothing ®lter (Figure 1). A lateral view of the 3D-
image was then reconstructed by superimposing the
external auditory meati. The FH plane was determined
with reference to the facial skeleton and the anterior
border of the masseter muscle de®ned (Figure 2). The
angle between the anterior border of the masseter and Figure 1 Image processing sequence used for measurements of the
masseter muscle. (a) Original axial CT image transferred to the Apple
the FH plane was then measured as the average of computer system. (b) Image after adjustment of the window width.
three repeated measurements to give the muscle (c) Image after processing with the smoothing ®lter. The skin was
direction. then removed and the masseter muscle brought to the surface

Dentomaxillofacial Radiology

The area and ratio of the length of short to long axis
(s/l ratio) of the masseter muscle were calculated
according to the following formulae [at right angles
to the muscle direction].
area ˆ a0
cos 
s=1 ratio ˆ s=…l0
cos†
where: a0, s, and l0 are the area, length of the short
axis, and length of the long axis measured on the
original axial image at a level approximately 1 cm
above the occlusal plane and y is the angle between the
axial image and the section perpendicular to the muscle
a
Figure 2 Lateral 3D-CT image of the masseter muscle. The angle
between the anterior border of the masseter muscle and the Frankfurt
horizontal plane was measured as the muscle direction. (a) Muscle
direction in a normal subject. (b) Muscle direction in a patient with
prognathism
Masseter muscle morphology
Y Ariji et al
115
direction (Figure 3). Each value was obtained as the
mean of three measurements. The scanning level for
the measurements was chosen so that metal artifacts
were avoided and corresponded to the roots of the
maxillary teeth. Furthermore, previous reports and our
preliminary tests showed that the measured values were
relatively stable and correlated well with the volume of
the masseter muscle at the level of the scan.18
Values of prognathic and normal subjects were
compared using Student's t-test with P50.05 signifi-
cant.
Intra-observer precision was calculated from ®ve
repeated measurements of one image by the radiologist
(YA.).
Results
The errors of intra-observer precision exposed as
coecient of variance were: 0.5% coecient of variance
for (angle), 3.76% for area and 4.99% for s/l ratio.
There was no signi®cant di€erence between the right
and left values for all three parameters in either the
prognathic patients or the normal subjects. In
prognathic patients, the right and left directions were
77.48 and 76.68, the right and left areas 320.1 mm2 and
318.3 mm2, and the s/l ratios 0.320 and 0.312. The
correlation coecients between right and left were
0.771, 0.940 and 0.700 for direction, area and s/l ratio,
respectively. In normal subjects, the right and left
values were 65.18 and 65.18, 384.7 mm2 and 368.3 mm2,
and 0.391 and 0.393 for direction, area and s/l ratio,
Figure 3 Methods for calculating the area (a) and ratio of short to
long dimensions (s/l ratio) on a section perpendicular to the direction
of the masseter muscle. The area (a) was obtained by multiplying a0
by cosy where: a0=the cross-sectional area measured on an axial
image, y=the angle between the axial image and the section
perpendicular to the muscle direction. The ratio of short to long
axis (s/l ratio) was calculated as follows : s/l ratio=s/(l0 cosy) where:
.
s=the short axis on an axial image, l0=the long axis on an axial
image, y=the angle between the axial image and the section
perpendicular to the muscle direction
Dentomaxillofacial Radiology
 Masseter muscle morphology
Y Ariji et al
116
Table 1 The mean (s.d.) values of angle, area and s/l ratio of the masseter muscle in patients with mandibular prognathism and normal subjects
Angle (8) Area (mm2) s/l ratio
n Mean s.d. Mean s.d. Mean s.d.
Prognathism
All 66 76.6 4.4 318.3 77.2 0.312 0.049
Male 29 77.4 5.6 362.0 71.0 0.329 0.049
Female 37 76.0 3.3 284.1 64.0 0.298 0.046
Normal
All 90 65.1 4.4 368.3 97.2 0.393 0.054
Male 49 65.2 4.0 406.7 96.6 0.396 0.059
Female 41 64.9 4.9 322.4 76.5 0.389 0.047
s, short axis of the masseter muscle; l, long axis of the masseter muscle

Figure 4 Correlation between the areas of the right and left masseter Figure 5 Comparison of the two-dimensional distribution of the
muscles. The dotted line shows the 95% con®dence interval of the angle and s/l ratio of the masseter muscle in patients with mandibular
regression curve in the normal group prognathism and normal subjects

Table 2 Discriminant analysis of values for measurements of angle, and one normal subject with one or more of the three
area and s/l ratio of the masseter muscle in patients with mandibular parameters outside this range were excluded from the
prognathism and normal subjects
subsequent calculations since this study focused on
Sensitivity Specificity Accuracy symmetrical subjects. Values of the left side were used
(%) (%) (%)
for comparisons. The mean angle in patients with
Variables for discriminant analysis mandibular prognathism was signi®cantly larger than
Angle 95.7 93.1 94.4
Area 64.1 56.3 60.3
normal subjects (P50.01), while the mean area and s/l
s/l ratio 87.0 73.6 80.5 ratio was smaller (P50.01).
Angle and area 97.8 89.7 93.9 Table 2 shows the sensitivity, speci®city and
Angle and s/l ratio 96.7 94.3 95.5 accuracy for discriminating the masseter muscle in
Area and s/l ratio 87.0 73.6 80.5 the patients with mandibular prognathism from that
Angle and area and s/l ratio 98.9 90.8 95.0
of the control subjects. The angle was the most
s, short axis of the masseter muscle; l, long axis of the masseter
muscle
e€ective parameter for discriminating between the
two groups. Figure 5 shows the diagnostic power
when the measured values were combined two-
dimensionally. When the angle and s/l ratio were
combined, the sensitivity, speci®city and accuracy
respectively and the respective correlation coecients increased to a maximum of 96.7%, 94.3% and
were 0.744, 0.963 and 0.765 (Table 1). Almost all 95.5%, respectively. Further, when the three para-
values were within the 95% con®dence interval of the meters were combined, the sensitivity increased to
regression line (Figure 4). Three prognathic patients 98.9% (Table 2).
Dentomaxillofacial Radiology

Discussion
There have been di€erences of opinion on whether the
area of the masseter muscle di€ers between the two
sides, although it is recognized that there is a close
correlation.2,18,21 In contrast, there is no di€erence
between the right and left sides in thickness of the
muscle,22 ± 25 corresponding to the short axis in the
present study. Koolstra found no asymmetry in the
muscle direction,26 with a high correlation between the
two sides.1 In this study, we found no signi®cant
di€erence, and a close correlation, between both sides
for all three parameters. Previous authors have used
various methods to determine the representative
parameters.3,8 ± 10,18,24 Weiji et al. and Bakke et al. used
the mean of the right and left sides,3,8 while Hannam et
al. and Sasaki et al. used one side only.9,10 We chose to
analyse the left side since Hannam et al. found no
consistent trend in the data from either side.10
The anterior border of the masseter muscle was
clearly de®ned on the lateral 3D-CT images, and was
therefore chosen as indicating the direction of the
masseter muscle and the angle of its border to the FH
plane was measured. We found that by measuring the
angle between it and the FH plane we were able to
discriminate well between prognathic patients and
normal subjects. Furthermore, the direction was the
most e€ective single value in de®ning these patients in
the discriminant analysis, and the diagnostic power was
highest when combined with the s/l ratio. Therefore,
we consider it an appropriate parameter to indicate the
direction of the masseter muscle, although it will be
necessary to con®rm this conclusion by comparison
with other authors' methods.
There have been a few reports on the three-
dimensional orientation of the masseter mus-
cle.4,9,10,19,27 Hannam reported that the angle between
the muscle and the functional occlusal plane in the
sagittal (lateral) view was 748.10 van Spronsen analysed
the di€erences between normal and long-face subjects
with normal occlusion, and found that the latter had a
slightly larger angle to the FH plane.19 However, the
di€erence was not signi®cant in discriminant analysis.
In the present study, the angle was signi®cantly larger
in prognathism. This result suggests that muscle
References
1. van Spronsen PH, Koolstra JH, van Ginkel FC, Weijs WA, Valk
J, Prahl-Andersen B. Relationships between the orientation and
moment arms of the human jaw muscles and normal craniofacial
morphology. Eur J Orthod 1997; 19: 313 ± 328.
2. van Spronsen PH, Weijs WA, Valk J, Prahl-Andersen B, van
Ginkel FC. Relationships between jaw muscle cross-sections and
craniofacial morphology in normal adults, studied with
magnetic resonance imaging. Eur J Orthod 1991; 13: 351 ± 361.
3. Weijs WA, Hillen B. Relationships between masticatory muscle
cross-section and skull shape. J Dent Res 1984; 63: 1154 ± 1157.
4. Weijs WA. The functional signi®cance of morphological
variation of the human mandible and masticatory muscles.
Acta Morphol Neerl Scand 1989; 27: 149 ± 162.
Masseter muscle morphology
Y Ariji et al
117
direction correlates closely with the skeletal morphol-
ogy and could be an important parameter for
evaluating any changes in muscle morphology.
The crosssectional area has been used frequently as
a parameter of muscle size because it has a high
correlation with the volume.18,24 Di€erent levels have
been proposed for measurement of the cross-sectional
area of the masseter muscle, such as 8 mm above the
mandibular foramen,18 20 mm below the FH plane28
or 30 mm ventrocranially to the angle of the
mandible.11 It has also been reported that the cross-
section area in the middle of the masseter is relatively
constant over a range of 12 mm.2,11 This region
corresponds the level of the maxillary roots which
was used in this study.
There have been many reports on the cross-sectional
area of the masseter muscle using CT or MRI in
subjects with normal craniofacial morphology with
values ranging from 447 mm2 to 690 mm2 2,9 ± 11,21,24,29 all
of which are larger than our results. This may be due
to di€erences in race or the method of measurement.
There have been two studies of Japanese subjects using
CT. Xu et al. reported that the area was 570 mm2 in
males and 487 mm2 in females18 while Ando found that
it was 381 to 399 mm2 in males and 288 to 293 mm2 in
females.28 Our result is similar to the latter, 407 mm2 in
males and 322 mm2 in females, with a signi®cant sex
di€erence. The area on cross-sectional images is
reported to correlate well with skeletal morphology.
van Spronsen reported that the area of the masseter in
long-faced subjects was smaller than that in normal
people.11,21 Prognathic patients have a long-face with
the mandibles positioned more anteriorly than those
with a normal occlusion. The present study, therefore,
is consistent with van Spronsen's results.
In conclusion, the masseter muscle is thin and small
in prognathic patients in comparison with normal
subjects with its long axis closer to a right angle to the
FH plane.
Acknowledgements
We thank Dr H Yuasa of the Department of the Oral and
Maxillofacial Surgery, Nagoya City Jyohoku Municipal
Hospital for his advice on the statistical analysis.
5. Bakke M, Michler L. Temporalis and masseter muscle activity in
patients with anterior open bite and craniomandibular disorders.
Scand J Dent Res 1991; 99: 219 ± 228.
6. Kayukawa H. Malocclusion and masticatory muscle activity: a
comparison of four types of malocclusion. J Clin Pediatr Dent
1992; 16: 162 ± 177.
7. Ueda HM, Ishizuka Y, Miyamoto K, Morimoto N, Tanne K.
Relationship between masticatory muscle activity and vertical
craniofacial morphology. Angle Orthod 1998; 68: 233 ± 238.
Dentomaxillofacial Radiology
 Masseter muscle morphology
Y Ariji et al
118
8. Bakke M, Tuxen A, Vilmann P, Jensen BR, Vilmann A, Toft M.
Ultrasound image of human masseter muscle related to bite
force, electromyography, facial morphology, and occlusal
factors. Scand J Dent Res 1992; 100: 164 ± 171.
9. Sasaki K, Hannam AG, Wood WW. Relationship between the
size, position, and angulation of human jaw muscles and
unilateral ®rst molar bite force. J Dent Res 1989; 68: 499 ± 503.
10. Hannam AG, Wood WW. Relationships between the size and
spatial morphology of human masseter and medial pterygoid
muscles, the craniofacial skeleton, and jaw biomechanics. Am J
Phys Anthropol 1989; 80: 429 ± 445.
11. van Spronsen PH, Weijs WA, Valk J, Prahl-Andersen B, van
Ginkel FC. Comparison of jaw-muscle bite-force cross-sections
obtained by means of magnetic resonance imaging and high-
resolution CT scanning. J Dent Res 1989; 68: 1765 ± 1770.
12. Ingervall B, Minder C. Correlation between maximum bite force
and facial morphology in children. Angle Orthod 1997; 67: 415 ±
422.
13. Hirose K. The study of the relationships between the
masticatory muscles activity and the craniofacial morphology
in mandibular prognathism. Shigaku 1990; 78: 49 ± 62.
14. Zarrinkelk HM, Throckmorton GS, Ellis III E, Sinn DP.
Functional and morphologic alterations secondary to superior
repositioning of the maxilla. J Oral Maxillofac Surg 1995; 53:
1258 ± 1267.
15. Harper RP, de Bruin H, Burcea I. Muscle activity during
mandibular movements in normal and mandibular retrognathic
subjects. J Oral Maxillofac Surg 1997; 55: 225 ± 233.
16. Iwase M, Sugimori M, Kurachi Y, Nagumo M. Changes in bite
force and occlusal contacts in patients treated for mandibular
prognathism by orthognathic surgery. J Oral Maxillofac Surg
1998; 56: 850 ± 855.
17. Weijs WA, Hillen B. Relationships between the physiological
cross-section of the human jaw muscles and their cross-sectional
area in computer tomograms. Acta Anat 1984; 118: 129 ± 138.
18. Xu JA, Yuasa K, Yoshiura K, Kanda S. Quantitative analysis of
masticatory muscles using computed tomography. Dentomax-
illofac Radiol 1994; 23: 154 ± 158.
Dentomaxillofacial Radiology
View publication stats
19. van Spronsen PH, Weijs WA, van Ginkel FC, Prahl-Andersen B.
Jaw muscle orientation and moment arms of long-face and
normal adults. J Dent Res 1996; 75: 1372 ± 1380.
20. Koolstar JH, van Eijden TMGJ, Weijs WA. An iterative
procedure to estimate muscle lines of action in vivo. J
Biomechanics 1989; 22: 911 ± 920.
21. van Spronsen PH, Weijs WA, Valk J, Prahl-Andersen B, van
Ginkel FC. A comparison of jaw muscle cross-sections of long-
face and normal adults. J Dent Res 1992; 71: 1279 ± 1285.
22. Raadsheer MC, Kiliaridis S, van Eijden TM, van Ginkel FC,
Prahl-Andersen B. Masseter muscle thickness in growing
individuals and its relation to facial morphology. Arch Oral
Biol 1996; 41: 323 ± 332.
23. Kiliaridis S, Kalebo P. Masseter muscle thickness measured by
ultrasonography and its relation to facial morphology. J Dent
Res 1991; 70: 1262 ± 1265.
24. Gionhaku N, Lowe AA. Relationship between jaw muscle
volume and craniofacial form. J Dent Res 1989; 68: 805 ± 809.
25. Prabhu NT, Munshi AK. Measurement of masseter and
temporalis muscle thickness using ultrasonographic technique.
J Clin Pediatr Dent 1994; 19: 41 ± 44.
26. Koolstra JH, van Eijden TM, van Spronsen PH, Weijs WA, Valk
J. Computer-assisted estimation of lines of action of human
masticatory muscles reconstructed in vivo by means of magnetic
resonance imaging of parallel sections. Arch Oral Biol 1990; 35:
549 ± 556.
27. Kasai K, Richards LC, Kanazawa E, Ozaki T, Iwasawa T.
Relationship between attachment of the super®cial masseter
muscle and craniofacial morphology in dentate and edentulous
humans. J Dent Res 1994; 73: 1142 ± 1149.
28. Ando T. Observation of the masticatory muscles and buccal fat
pad on the CT images. J Natl Med Coll 1991; 16: 30 ± 41.
29. Weijs WA, Hillen B. Physiologic cross-section of the human jaw
muscles. Acta Anat 1985; 121: 31 ± 35.