ORIGINAL CONTRIBUTIONS
Age- and sex-related differences in
masseter size and its role in oral
functions
Chia-Shu Lin, DDS, DPhil; Ching-Yi Wu, DDS, PhD; Shih-Yun
Wu, DDS; Kai-Hsiang Chuang, DDS; Hsiao-Han Lin, MSc;
Dong-Hui Cheng, DDS; Wen-Liang Lo, DDS, PhD
T
he masseter muscle plays a key functional and
structural role in the stomatognathic system.
Masseter muscle activity is associated with
chewing behavior1 and swallowing,2 and disrup-
tion in its activity may be linked to temporomandibular
disorders.3 The masseter is the largest jaw elevator
muscle and the major contributor of the strength of jaw
closure,4 and its size is associated closely with bite
force.5,6 Cumulative evidence suggests that variation in
masseter muscle size may be a critical factor related to
individual differences in oral functions.4-6
Researchers have used masticatory performance (MP)
and salivary flow rate (SFR) as the clinical metrics for
evaluating chewing and swallowing abilities. Both a
person’s age and sex contribute in a critical manner to
the individual differences found in MP and SFR.7,8
However, researchers have not yet investigated system-
atically the effect of age and sex on masseter muscle size
and the association of masseter muscle size with other
clinical metrics. Investigators of previously published
studies have reported that magnetic resonance imaging
(MRI) data can be used for assessing the morphology of
orofacial muscles9,10 with good validity.11-13 In our study,
we reasoned that T1-weighted MRI data of the head,
which can be acquired commonly during a brain scan,
can be a suitable and convenient source for assessing
masseter muscle volume (MMV).
We had 3 major goals for our study. First, we aimed to
develop a voxel-based approach for assessing MMV,
using T1-weighted MRI data. Second, because the
investigators of previously published studies focused
Copyright ª 2017 American Dental Association. All rights reserved.
ABSTRACT
Background. The masseter muscle plays a key structural
and functional role in the stomatognathic system. Re-
searchers’ cumulative evidence has suggested that the
variation in the size of a person’s masseter muscle may be a
critical factor related to individual differences in oral
functions. However, researchers have not yet investigated
systematically the effect of a person’s age and sex on
masseter muscle size and the association of masseter
muscle size with other clinical metrics, including mastica-
tory performance (MP) and salivary flow rate (SFR). Using
T1-weighted magnetic resonance imaging (MRI) data
provides a noninvasive method for assessing masseter
muscle volume (MMV).
Methods. Using T1-weighted MRI data, the authors
developed a voxel-based method to assess MMV and
investigated the associations among MMV, MP, and
SFR.
Results. The authors acquired T1-weighted MRI data
from scans of the heads of 62 healthy adults and assessed
MMV by means of using a voxel-based approach. The
authors’ assessment results had acceptable rates of inter-
rater and intrarater reliability. MMV was significantly
lower in the older subgroup and in the female subgroup. In
addition, the correlation for MMV was significantly posi-
tive with MP and stimulated SFR.
Conclusions. The study results revealed evidence that
the authors’ voxel-based approach, which they designed on
the basis of T1-weighted MRI data, would be a reliable
method for quantifying MMV.
Practical Implications. The findings suggest that the
variation in masseter muscle size may be a critical factor to
assess individual differences in oral functions.
Key Words. Aging; magnetic resonance imaging;
mastication; masseter muscle.
JADA 2017:-(-):---
http://dx.doi.org/10.1016/j.adaj.2017.03.001
JADA ( )
- - http://jada.ada.org - 2017 1
 ORIGINAL CONTRIBUTIONS
mainly on younger adults and used smaller sample
sizes,14-17 we determined that, by means of collecting data
from a larger sample, we could investigate the age-
and sex-related effects on MMV. Because previous
investigators’ findings revealed age-related differences in
the masseter cross-sectional area18 and sex-related dif-
ferences in the masseter muscle thickness,19 we hypoth-
esized that a reduced MMV would be associated with
older age (hypothesis 1) and female sex (hypothesis 2).
Third, we investigated the association between MMV
and the major clinical metrics of oral functions: MP,
unstimulated SFR (uSFR), and stimulated SFR (sSFR).
We hypothesized that MMV would correlate positively
with MP (hypothesis 3) and that an increased MMV
would be associated with increased sSFR7 (hypothesis 4).
METHODS
Study group. We recruited 62 healthy study participants
(39 women and 23 men; age range, 23-74 years) to receive
a T1-weighted MRI scan and assessments of oral func-
tions, including MP, uSFR, and sSFR. We recruited
volunteers by advertising in local community centers and
the Taipei Veterans General Hospital (Taiwan). We
excluded from participation volunteers who had a his-
tory of major physical or psychiatric disorders, brain
injury, or brain surgery, as well as those who were unable
to undergo MRI owing to physical or psychological
contraindications. We received written informed consent
from all participants before the experiment initiated. The
institutional review boards of National Yang-Ming
University (Taipei, Taiwan) and Taipei Veterans General
Hospital approved the study.
Acquisition of head T1-weighted MRI. Using a 3
Tesla Siemens MRI scanner (Siemens Magnetom Trio,
A Tim System, Siemens), we performed all of the
T1-weighted MRIs at the 3T MRI Laboratory of National
Yang-Ming University. We acquired the images by
means of using a 32-channel head coil, using the
magnetization prepared rapid gradient-echo sequence
(repetition time, 2,530 milliseconds; echo time, 3.03 ms;
flip angle, 7
; matrix size, 256  256  192; voxel size, 1 
1  1 cubic millimeters), which we based on the protocol
that we reported in our previous studies.20
Protocol for the MMV assessment. We assessed
MMV using a voxel-based approach, which we based on
the following protocol:
- Originally, we stored the acquired images according
to the Digital Imaging and Communications in Medicine
standard and then transformed the images to the
Neuroimaging Informatics Technology Initiative data
format. We performed all investigations by means of
using the imaging tool FSLView 3.2.0 (Oxford Centre for
Functional MRI of the Brain).
- One author (C.-S.L.) manually defined the anatomic
region of the masseter muscle without additional
between-image registration or intensity normalization.
2 JADA ( )
- - http://jada.ada.org - 2017
We defined the right and left masseter muscle
separately.
- Primarily, we based the delineation of the region of
interest (ROI) of the masseter muscle on the anatomic
landmarks that we could clearly and repeatedly identify
in a T1-weighted image (Figure 1A). First, we delineated
the border of the ROI on the sagittal view from the
lateral to the medial position (Figure 1A).
- We examined the resulting masseter ROI from the
axial view and the coronal view.
- Using another software tool (fslstats, Oxford Centre
for Functional MRI of the Brain), we quantified the
muscle volume for each side according to the number of
voxels within the defined masseter ROI. We determined
the MMV value by means of calculating the average of
the muscle volume from both sides.
Reliability of the MMV assessment. To evaluate the
intrarater reliability of the method, 1 of the authors
(K.-H.C.) independently assessed the MRI data of 20
study participants, by means of using the same protocol.
To evaluate the intrarater test-retest reliability, 1 of the
authors (C.-S.L.) performed a second assessment of the
MRI data of 20 study participants 1 month after the first
assessment. We quantitatively assessed both of the reli-
ability tests, by means of using the intraclass correlation
coefficient (ICC), and we qualitatively assessed the tests,
by means of using the Bland-Altman plot.
Assessment of MP. Using the colorimetric method,
we assessed MP.21 After receiving the scan, each partic-
ipant chewed a piece of color-changeable chewing gum
(Lotte) for 3 minutes.22,23 We digitized the chewed gum,24
and we assessed the color change by means of using the
L*a*b* chromatic system.22
Assessment of SFR. We collected saliva by means of
modifying procedures that investigators had described
previously.25 The study participants refrained from
eating, drinking, or toothbrushing for an hour before
we collected the samples. Among all the participants
(N ¼ 62), we excluded the results or did not perform an
assessment in 26 participants, because those participants
had eaten a meal or had a drink right before the
assessment. To collect unstimulated saliva, we asked
participants to drool saliva passively into a tube, the
weight of which we measured, for 10 minutes after
swallowing the remaining saliva. We limited study par-
ticipants’ movements of the head and neck regions, and
we asked them to limit their speech. To obtain samples of
participants’ stimulated saliva, we asked the participants
ABBREVIATION KEY. MMV: Masseter muscle volume.
MP: Masticatory performance. MRI: Magnetic resonance
imaging. MW: Mann-Whitney U test. Q1: Lower quartile. Q3:
Upper quartile. ROI: Region of interest. SFR: Salivary flow rate.
sSFR: Stimulated salivary flow rate. SW: Shapiro-Wilk. uSFR:
Unstimulated salivary flow rate.
 ORIGINAL CONTRIBUTIONS

Interrater assessment 6
RATER 2 – RATER 1, cm3

Mean + 2 SD
of MMV Mean – 2 SD
(K.-H. CHUANG), cm3

50 4
Mean of difference
40 2
RATER 2

30 0
20 0 10 20 30 40 50
–2
10
–4
0
0 10 20 30 40 50 –6
RATER 1 (C.-S. LIN), cm3 (RATER 1 + RATER 2)/2, cm3
TIME POINT 2 – TIME POINT 1, cm3

6
Interrater assessment Mean + 2 SD
of MMV 4 Mean – 2 SD
TIME POINT 2, cm3

50 Mean of difference
40 2
30 0
20 0 10 20 30 40 50
–2
10
–4
0
0 10 20 30 40 50 –6

B TIME POINT 1, cm3 (TIME POINT 1 + TIME POINT 2)/2, cm3

Figure 1. Manual delineation and assessment of the masseter muscle from the head T1-weighted magnetic resonance imaging scan. A. The
upper panels show the surrounding structures identified at a lower position (near the mandibular ramus, the upper left panel) and a higher position
(near the mandibular condyle, the upper right panel). The lower panel shows the sagittal view of the delineated masseter region (red) from 1
participant, from the lateral position (upper left) to the medial position (lower right). B. The results of interrater and intrarater reliability analyses.
The authors’ analyses found strong correlation of the estimated masseter muscle volume (MMV) between raters and between 2 time points. Results
of the Bland-Altman plots revealed that the difference in the MMV between raters and between time points were consistent, generally with a
standard deviation (SD) in the range of 2. cm3: Cubic centimeters.

JADA ( )
- - http://jada.ada.org - 2017 3
 ORIGINAL CONTRIBUTIONS

TABLE 1 Wilk test. We considered the distribu-

Results of demographic data and clinical metrics. tion of scores to be normal if the P
value of the test was greater than .05.
VARIABLE AGE, Y MP* MMV,† cubic uSFR,‡ milliliters sSFR,§ Owing to the nonnormal distribution
centimeters per minute mL/min
of the results (see the Results section of
Total
this article and Table 1), we performed
N 62 62 62 36 36
nonparametric tests for the statistical
SW¶ test < .001 < .001 .007 .001 .011
analysis. To compare results between
Mean 50.1 71.9 24.9 0.32 1.94
participants’ sex-related or age-related
Standard 17.1 2.8 7.2 0.25 1.19
deviation subgroups, we performed the 2-tailed,
Median 55.5 72.3 22.9 0.26 1.72 Mann-Whitney U test. Using the
Q3# 64.0 73.6 29.0 0.42 2.29 Spearman rank correlation, we assessed
Q1** 30.0 71.1 19.3 0.14 1.33 the strength of the association between
Maximum 74.0 76.5 44.9 1.21 4.93 the clinical metrics. For each hypothe-
Minimum 23.0 63.5 11.7 0.05 0.12 sis, we considered the result of the
Younger
statistical tests to be significant if the
Subgroup a level was .05.
N 31 31 31 26 26
Median 30.0 73.1 27.0 0.30 1.94 RESULTS
Q3 51.0 74.5 31.3 0.49 2.59 Reliability of the MMV assessment.
Q1 25.5 72.0 19.6 0.22 1.61 For the interrater reliability, we found
Older that the results of the ICC test (2-way
Subgroup
mixed model, average measurement
N 31 31 31 10 10
for absolute agreement) had a strong
Median 64.0 71.2 21.8 0.13 1.29
reliability (ICC, 0.996; P < .001).
Q3 67.0 72.6 26.3 0.23 1.52
For the intrarater reliability, we found
Q1 60.0 69.5 19.1 0.06 0.59
††
that the results of the ICC test (2-way
MW test < .001 < .001 .043 .022 .004
mixed model, average measurement
Female
Subgroup
for consistency) had a strong
N 39 39 39 22 22
reliability (ICC, 0.997; P < .001)
Median 56.0 72.0 20.1 0.23 1.60
(Figure 1B).
Q3 62.5 73.1 24.0 0.36 1.96
Demographic profile and clinical
Q1 33.5 70.9 18.6 0.14 0.91
metrics. Tables 1 and 2 show the
Male Subgroup
results of descriptive analysis of the
N 23 23 23 14 14
demographic data and clinical metrics.
Median 54.0 72.7 31.2 0.38 2.38
Consistent with the previous find-
7,27
Q3 64.5 74.5 36.0 0.62 4.02
ings, we found that the correlation
Q1 30.0 71.4 26.3 0.17 1.61
for age was significantly negative
MW test .953 .079 < .001 .098 .014
with MP (r, 0.53; P < .001), stimu-
* MP: Masticatory performance (change in L*a*b* chromatic system).
lated SFR (r, 0.52; P ¼ .001),
† MMV: Masseter muscle volume. and unstimulated SFR (r, 0.48;
‡ uSFR: Unstimulated salivary flow rate. P ¼ .003). We found that the correla-
§ sSFR: Stimulated salivary flow rate.
¶ SW: Shapiro-Wilk. The authors evaluated normality on the basis of the results of the Shapiro- tion between sSFR and uSFR was
Wilk test. The number denotes the P value of the test. significantly positive (r, 0.60; P < .001)
# Q3: Upper quartile. (Figure 2). We found that, among all
** Q1: Lower quartile.
†† MW: Mann-Whitney U test. The authors conducted between-group differences on the basis the participants, the correlation for
of the results of the Mann-Whitney U test. The number denotes the P value of the test. MMV was significant with MP (r, 0.29;
P ¼ .02) (Figure 3), which confirmed
to chew paraffin wax (GC) for 3 minutes during saliva hypothesis 3. We also found that the correlation for
collection. Next, we weighed the tube to calculate the MMV was significantly positive with sSFR (r, 0.33; P ¼
weight of saliva, and then we calculated the volume of .02) but not uSFR (r, 0.15; P ¼ .40) (Table 2), which
saliva, assuming that the density of saliva was 1 gram per confirmed hypothesis 4. In addition, we found that the
milliliter.26 correlation for MP was significantly positive with sSFR
Statistical analysis. We examined all the clinical (r, 0.41; P ¼ .014) but not uSFR (r, 0.26; P ¼ .13)
metrics for normality, by means of using the Shapiro- (Table 2).

4 JADA ( )
- - http://jada.ada.org - 2017
 ORIGINAL CONTRIBUTIONS
Sex- and age-related differences in the clinical
metrics. In terms of age-related effects, MMV was
significantly lower in the older subgroup compared
with the younger subgroup (younger, 27.0 [median]
[19.6 (lower quartile) to 31.3 (upper quartile)] cm3; older,
21.8 [19.1 to 26.3] cm3; P ¼ .043), confirming hypothesis 1.
In addition, we found that the older subgroup had a
significant decrease in MP (younger, 73.1 [72.0 to 74.5];
older, 71.3 [69.9 to 73.0]; P ¼ .005), uSFR (younger, 0.30
[0.22 to 0.49] mL/min; older, 0.13 [0.06 to 0.23] mL/min;
P ¼ .022), and sSFR (younger, 1.94 [1.61 to 2.59] mL/min;
older, 1.29 [0.59 to 1.52] mL/min; P ¼ .004) compared
with the younger subgroup (Table 1).
In terms of sex-related effects, MMV was significantly
higher in the male subgroup compared with the
female subgroup (male, 31.2 [26.3 to 36.0] cm3; female,
20.1 [18.6 to 24.0] cm3; P < .001), confirming hypothesis
2. In addition, sSFR was significantly higher in the male
subgroup compared with the female subgroup (male, 2.38
[1.61 to 4.02] mL/min; female, 1.60 [0.91 to 1.96] mL/min;
P ¼ .014) (Table 1). We also investigated whether an
interactional effect existed between the age- and sex-
related factors. For both age subgroups, we found an age-
related difference (that is, a higher level of MMV in
the male subgroup than in the female subgroup). In
contrast, for both the male and female subgroups, we did
not find a significant difference among the age sub-
groups. The findings revealed a main effect related to
male or female sex in MMV and no evidence of an
interaction between age and male or female sex.
Age-related differences in the association between
the clinical metrics. Table 2 summarizes the results of
age-related differences related to the clinical metrics. In
the older group, we found that the correlation for age
was significantly negative with MP (r, 0.49; P < .005)
(Figure 2A) but not MMV. In contrast, in the younger
group, we found that the correlation for age was signif-
icantly positive with MMV (r, 0.45; P ¼ .011) (Figure 3A)
but not MP. In addition, in the younger group, we found
that the correlations between uSFR and sSFR were
significantly positive (r, 0.69; P < .001) (Figure 2A).
Sex-related differences in the association between
the clinical metrics. Table 2 summarizes the results of
the sex-related differences among the clinical metrics. In
both sex-related subgroups, we found that the correlation
for age was significantly negative with MP (female:
r, 0.57; P < .001; male: r, 0.49; P ¼ .018) and sSFR
(female: r, 0.58; P ¼ .005; male: r, 0.58; P ¼ .029)
(Figure 2B). We found that both groups had a signifi-
cantly positive correlation between uSFR and sSFR
(female: r, 0.50; P ¼ .019; male: r, 0.72; P ¼ .004)
(Figure 2B). In addition, in the male subgroup, we found
that the correlation for age was significantly negative
with uSFR (r, 0.67; P ¼ .008) (Figure 2B). In the female
subgroup, we found that the correlation for MP was
significantly positive with sSFR (r, 0.51; P ¼ .014)
(Figure 2B).
DISCUSSION
Summary of the major findings. With the results of this
study, we have provided new evidence regarding the age-
and sex-related differences in the MMV of healthy
adults. First, we established the voxel-based approach for
assessing MMV, which we based on the T1-weighted
MRI data from scans of the head. We found that, by
means of using the approach, we had an acceptable
interrater and intrarater reliability. Second, related to
MMV, we found a significant sex-related and age-related
difference, which was smaller in the female subgroup
and the older subgroup. Finally, we found a positive
correlation between MMV and MP as well as stimulated
SFR. The findings suggest that there is a substantial as-
sociation between the anatomic signature of stomatog-
nathic system and the clinical metrics of oral functions.
Voxel-based assessment of the MMV. Investigators
of previously published studies have suggested that
MRI data can be used to evaluate the morphology of
masticatory muscle.9,10 Our findings also revealed that
the T1-weighted MRI data from scans of the head, which
are used widely for assessing brain morphology in
neurologic examinations, can be a suitable and conve-
nient source for assessing MMV. The mean MMV (24.9
cm3) that we assessed using our approach is similar to
the previously published findings from investigators in
different geographic regions, including a sample from the
Netherlands (N ¼ 31; mean age, 24 years; mean MMV,
23.8 cm3),14 samples from Japan (N ¼ 25; mean age, 23
years; mean MMV, 24.7 cm3; and N ¼ 10; 20-30 years,
median MMV, 29.2 cm3),15,16 and a sample from
Singapore (N ¼ 12; 20-25 years; mean MMV, 29.5 cm3).17
Also, our results are similar to those of investigators
who assessed the wet contractile muscle volume in post-
mortem samples (20.1 cm3).28 Although ethnic and
regional factors may have contributed to substantial in-
dividual variations, in general, we found that our results
had a convergent result of the estimated MMV.
In terms of the reliability of the approach, our results
show both acceptable interrater and intrarater reliability.
This good level of reliability may be attributed to the
masseter’s homogeneity in terms of its composition of
muscle fibers29 and its lateral position to the teeth. The
latter (the masseter’s lateral position to the teeth) is
crucial to the assessment, because the magnetic reso-
nance signal can be disturbed severely by intraoral metal
restorations. In terms of the validity of the approach,
investigators have reported widely about the convergence
between the MRI-based methods and other methods.
The investigators of previously published studies have
reported a high level of correlation between the muscle
volume assessed using MRI and that assessed using
JADA ( )
- - http://jada.ada.org - 2017 5
 ORIGINAL CONTRIBUTIONS
TABLE 2
Association between clinical metrics.
CLINICAL TOTAL
METRIC
N [ 62 N [ 36
MP* MMV,† cubic uSFR,‡ sSFR,§
centimeters milliliters mL/min
per minute
Age 0.525¶ 0.074 0.476¶ 0.518¶
MP –** 0.293 #
0.257 0.406
MMV – – 0.145 0.334#
uSFR – – – 0.595¶
* MP: Masticatory performance (change in L*a*b* chromatic system).
† MMV: Masseter muscle volume.
‡ uSFR: Unstimulated salivary flow rate.
§ sSFR: Stimulated salivary flow rate.
¶ P < .01.
# P < .05.
** Dashes indicate data not applicable.
ultrasonography12 and from postmortem dissection.13
Researchers using animal models also have reported a
high level of consistency between the volume of the
tongue assessed using MRI and the wet weight of the
tongue.11 Notably, the muscle thickness obtained using
ultrasonography may be smaller than that obtained using
MRI, owing to the compression of muscle during the
ultrasonography scan.12 In contrast, we acquired the MRI
of the head when each study participant was lying in a
fixed position and undisturbed. Therefore, MMV
assessed using MRI may be a better indicator of the
masseter size under a natural and resting condition.
Age- and sex-related differences in the MMV. Our
results are consistent with previously published findings,
which show that there are great individual differences in
MMV. On average, study participants in the older sub-
group had a lower MMV compared with participants in
the younger subgroup (Table 1). However, when we
combined the results of participants from both sub-
groups, we did not find an age-related decline in MMV
(Table 2 and Figure 3A). This result is inconsistent with
the previously published finding in which investigators
noted a negative correlation between age and masseter
muscle thickness.18 A potential methodological explana-
tion is that thickness may increase when the muscle
contracts.12 Therefore, the age-related decline in
muscle thickness may be confounded by the decline of
muscle tension. In contrast, in our study, we assessed
the muscle volume when it was in a relaxed state, at
which time we thought muscle volume would be less
disturbed by muscle tension. Another potential expla-
nation is related to the physical and dental conditions of
the sample. Our study participants were physically
robust and dentally sound in posterior contact. In-
vestigators have reported that a lack of physical fitness
and a loss of posterior contact are associated with muscle
6 JADA ( )
- - http://jada.ada.org - 2017
YOUNGER SUBGROUP OLDER SUBGROUP
N [ 31 N [ 26 N [ 31
MP MMV, cm3 uSFR, sSFR, MP MMV, cm3
mL/min mL/min
0.008 0.452# 0.325 0.307 0.494¶ 0.163
#
– 0.185 0.146 0.231 – 0.214
– – 0.264 0.279 – –
– – – 0.689¶ – –
size and tension.30,31 Our results echoed a finding pub-
lished in 2017 in which investigators compared the oral
functions of people who were elderly and in robust health
with people who were elderly and in frail or near-frail
health. According to the results of that study, investigators
only found age-related decrease in muscle thickness in the
study participants who were in near-frail health, but not
the in the study participants who were in robust health.31
We found that male or female sex was a major
contributing factor to the variation in MMV (Figure 3B).
Investigators found consistently higher masticatory
muscle thickness in men compared with women.19 On a
functional level, investigators also found that male or
female sex was a major contributing factor to individual
differences in maximal bite force.32 We found that such a
sex-related effect did not intersect with age. This finding
from our study results implies that, compared with age,
male or female sex may be a more predominant
contributing factor to individual variations in MMV.
Clinical implications. Among all study participants,
we found a significantly positive correlation between
MMV and MP. Investigators have found that individual
differences in MP are associated with the number of
functional teeth and bite force.33 Because the participants
we recruited for our study had a sound posterior contact
(either with natural teeth or restoration), we reasoned
that the higher MP may have been associated with a
higher MMV. We did not directly assess the participants’
bite force. Nevertheless, the investigators of previously
published studies have revealed that jaw muscle size was
a major factor that explained most of the variation in
bite force6 and that masticatory muscle thickness corre-
lated positively with bite force.5 The evidence we found
in our study suggests that there is a coupling between
structure (muscle size) and function (bite force) in the
stomatognathic system.
 ORIGINAL CONTRIBUTIONS

TABLE 2 (CONTINUED)

OLDER SUBGROUP FEMALE SUBGROUP MALE SUBGROUP

N [ 10 N [ 39 N [ 22 N [ 23 N [ 14
3 3
uSFR, sSFR, MP MMV, cm uSFR, sSFR, MP MMV, cm uSFR, sSFR,
mL/min mL/min mL/min mL/min mL/min mL/min
0.286 0.333 0.571¶ 0.066 0.373 0.580¶ 0.490# 0.172 0.673¶ 0.582#
0.024 0.511 – 0.155 0.049 0.514 #
– 0.306 0.459 0.240
0.659# 0.322 – – 0.353 0.128 – – 0.191 0.095
– 0.122 – – – 0.496# – – – 0.719¶

80 80
PERFORMANCE

PERFORMANCE
MASTICATORY

MASTICATORY
75 75

70 70

65 65
Younger Older Male Female
60 60
20 30 40 50 60 70 80 20 30 40 50 60 70 80
AGE (YEAR) AGE (YEAR)
1.6 1.6
UNSTIMULATED SFR

UNSTIMULATED SFR

1.4 1.4
1.2 1.2
(mL/min)

(mL/min)

1.0 1.0
0.8 0.8
0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0
20 30 40 50 60 70 80 20 30 40 50 60 70 80
AGE (YEAR) AGE (YEAR)
5 5
STIMULATED SFR
STIMULATED SFR

4 4
(mL/min)
(mL/min)

3 3

2 2

1 1

0 0
20 30 40 50 60 70 80 20 30 40 50 60 70 80
AGE (YEAR) AGE (YEAR)
5 5
STIMULATED SFR

STIMULATED SFR

4 4
(mL/min)

(mL/min)

3 3
2 2
1 1

0 0
0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5
A UNSTIMULATED SFR (mL/min) B UNSTIMULATED SFR (mL/min)

Figure 2. The association among the clinical metrics other than masseter muscle volume. A. The results were labeled differentially as the younger
(red) and the older (blue) subgroups. B. The results were labeled differentially as the male (red) and the female (blue) subgroups. Masticatory
performance: Change in L*a*b* chromatic system. mL: Milliliters. SFR: Salivary flow rate.

JADA ( )
- - http://jada.ada.org - 2017 7
 ORIGINAL CONTRIBUTIONS

older subgroup, we did

MASSETER MUSCLE 50 80 not find a positive cor-
relation between uSFR

PERFORMANCE
VOLUME (cm3)
40

MASTICATORY
75 and sSFR or between
30
sSFR and MP. We
70 recommend that this
20 negative finding be
65 interpreted cautiously,
10 because in the older
Younger Older
0 60 subgroup, the sample
20 30 40 50 60 70 80 0 10 20 30 40 50 size for saliva assessment
was relatively small
AGE (YEAR) MASSETER MUSCLE (N ¼ 10), which may
A VOLUME (cm ) 3
have increased the risk of
developing a type II
50 80 error.
MASSETER MUSCLE

Both chewing and

PERFORMANCE
VOLUME (cm3)

40

MASTICATORY
75 saliva secretion are
30 crucial to swallowing. In
70 patients with xerostomia
20 (for example, patients
10
65 with Sjögren syndrome
Male Female or patients who are
0 60 receiving chemoradiation
20 30 40 50 60 70 80 0 10 20 30 40 50 therapy), investigators
AGE (YEAR) MASSETER MUSCLE commonly have seen
B VOLUME (cm ) 3 dysphagia and insuffi-
cient bolus preparation
and a substantially
Figure 3. The association between masseter muscle volume (in cubic centimeters) and age-related and delayed phase of oropha-
masticatory performance. Both panels included all participants (N ¼ 62). A. The results were labeled differentially ryngeal swallowing.36,37
as the younger (red) and the older (blue) subgroups. B. The results were labeled differentially as the male (red)
and the female (blue) subgroups. Masticatory performance: Change in L*a*b* chromatic system. Our results indicate that
as age increases, both MP
and SFR also could
We should note that, apart from MMV, other factors decline. In addition, our findings suggest that it is critical
could contribute to MP, such as the secretion of saliva. for clinicians to evaluate systematically the mastication
Of note in our study, we adopted the chewing-gum and saliva secretion of people who are elderly as part of the
approach for assessing MP, which we thought would assessment of dysphagia.
better characterize one’s “mixing ability” instead of the MRI is not an examination that clinicians primarily
“cutting ability” of mastication.34 As a critical intraoral consider to assess patients’ oral functions. Nevertheless,
lubricant, saliva plays a key role in the mixing procedure. clinicians in geriatric medicine widely recommend that
In our study, we found that the association between patients undergo MRIs, especially when they are diag-
mastication and saliva secretion could be evidenced by nosing a patient’s neurological and psychiatric diseases.
the positive correlation between uSFR and sSFR and These MRI data comprise not only the brain area but
between sSFR and MP, especially in the younger group also the orofacial and neck area; therefore, an MRI scan
(Table 2). Because we collected sSFR levels when the could be an available and convenient source of results for
study participants were chewing paraffin wax, we a clinician to use when assessing a patient’s masseter
determined that this finding implies that the individual muscle. Neuroimaging evidence from study results have
differences in masticatory ability could be a contributing shown that age-related frailty or cognitive impairment
factor to the transition of saliva secretion from the were associated with regional alterations in the brain38,39
resting condition (unstimulated) to the chewing condi- and that chewing ability may be associated with one’s
tion (stimulated). In line with our findings, other in- cognitive functions.40,41 Therefore, for adults who are
vestigators have noted that SFR has correlated positively elderly, especially those who have physical frailty or
with maximal bite force27 and has been associated with cognitive dysfunction, concurrent assessments of MMV
the masticatory–parotid salivary reflex.35 In our study’s (on the basis of available MRI data), MP, and SFR may

8 JADA - - ( ) http://jada.ada.org - 2017

 ORIGINAL CONTRIBUTIONS
provide more information for clinicians to use when
evaluating the patients’ health conditions.
CONCLUSIONS
The evidence from our study results revealed that
the voxel-based approach, which is based on T1-
weighted MRI data, could be a reliable method for
quantifying MMV. Using this method, we identified
both age- and sex-related differences in MMV. Our
findings suggested a close relationship between the
structure (the masseter muscle) and the function
(masticatory performance and saliva secretion) in
study participants’ stomatognathic systems. n
Dr. Chia-Shu Lin is an associated professor, Department of Dentistry,
School of Dentistry, National Yang-Ming University, No. 155, Sec. 2, Linong
St., Taipei, 11221 Taiwan (ROC), e-mail winzlin@ym.edu.tw. Address cor-
respondence to Dr. Chia-Shu Lin.
Dr. Ching-Yi Wu is an assistant professor, Institute of Oral Biology,
School of Dentistry, National Yang-Ming University, and a training
physician, Division of Family Dentistry, Department of Stomatology, Taipei
Veterans General Hospital, Taipei, Taiwan.
Dr. Shih-Yun Wu is a visiting staff member, Division of Family Dentistry,
Department of Stomatology, Taipei Veterans General Hospital, and an
adjunct lecturer, Department of Dentistry, School of Dentistry, National
Yang-Ming University, Taipei, Taiwan.
Dr. Chuang is a graduate student of prosthodontics, Department of
Dentistry, School of Dentistry, National Yang-Ming University, Taipei,
Taiwan.
Dr. Hsiao-Han Lin is a research assistant, Department of Dentistry,
School of Dentistry, National Yang-Ming University, Taipei, Taiwan.
Dr. Cheng is a visiting staff member, Division of Prosthodontics,
Department of Stomatology, Taipei Veterans General Hospital, Taipei,
Taiwan.
Dr. Lo is the division chair, Division of Oral and Maxillofacial Surgery,
Department of Stomatology, Taipei Veterans General Hospital, and an
adjunct associate professor, Department of Dentistry, School of Dentistry,
National Yang-Ming University, Taipei, Taiwan.
Disclosure. None of the authors reported any disclosures.
This study was funded by grant 103-2314-B-010-025-MY3 awarded to Dr.
C.-S. Lin by the Ministry of Science and Technology of Taiwan (Taipei).
This study also was supported in part by the 3T MRI Core Facility in
National Yang-Ming University, Taipei, Taiwan.
1. Iguchi H, Magara J, Nakamura Y, et al. Changes in jaw muscle activity
and the physical properties of foods with different textures during chewing
behaviors. Physiol Behav. 2015;152(Pt A):217-224.
2. Hiraoka K. Changes in masseter muscle activity associated with
swallowing. J Oral Rehabil. 2004;31(10):963-967.
3. Mapelli A, Zanandrea Machado BC, Giglio LD, Sforza C, De
Felicio CM. Reorganization of muscle activity in patients with chronic
temporomandibular disorders. Arch Oral Biol. 2016;72:164-171.
4. McComas AJ. Oro-facial muscles: internal structure, function and
ageing. Gerodontology. 1998;15(1):3-14.
5. Raadsheer MC, van Eijden TM, van Ginkel FC, Prahl-Andersen B.
Contribution of jaw muscle size and craniofacial morphology to human
bite force magnitude. J Dent Res. 1999;78(1):31-42.
6. Sasaki K, Hannam AG, Wood WW. Relationships between the size,
position, and angulation of human jaw muscles and unilateral first molar
bite force. J Dent Res. 1989;68(3):499-503.
7. Ikebe K, Matsuda K, Kagawa R, et al. Association of masticatory
performance with age, gender, number of teeth, occlusal force and salivary
flow in Japanese older adults: is ageing a risk factor for masticatory
dysfunction? Arch Oral Biol. 2011;56(10):991-996.
8. Percival RS, Challacombe SJ, Marsh PD. Flow rates of resting whole
and stimulated parotid saliva in relation to age and sex. J Dent Res. 1994;
73(8):1416-1420.
9. Farrugia ME, Bydder GM, Francis JM, Robson MD. Magnetic reso-
nance imaging of facial muscles. Clin Radiol. 2007;62(11):1078-1086.
10. Schellhas KP. MR imaging of muscles of mastication. AJR Am J
Roentgenol. 1989;153(4):847-855.
11. Lauder R, Muhl ZF. Estimation of tongue volume from magnetic
resonance imaging. Angle Orthod. 1991;61(3):175-184.
12. Raadsheer MC, Van Eijden TM, Van Spronsen PH, Van Ginkel FC,
Kiliaridis S, Prahl-Andersen B. A comparison of human masseter muscle
thickness measured by ultrasonography and magnetic resonance imaging.
Arch Oral Biol. 1994;39(12):1079-1084.
13. Wu EX, Tang H, Tong C, Heymsfield SB, Vasselli JR. In vivo MRI
quantification of individual muscle and organ volumes for assessment of
anabolic steroid growth effects. Steroids. 2008;73(4):430-440.
14. Boom HP, van Spronsen PH, van Ginkel FC, et al. A comparison of
human jaw muscle cross-sectional area and volume in long- and short-face
subjects, using MRI. Arch Oral Biol. 2008;53(3):273-281.
15. Goto TK, Tokumori K, Nakamura Y, et al. Volume changes in human
masticatory muscles between jaw closing and opening. J Dent Res. 2002;
81(6):428-432.
16. Kitai N, Fujii Y, Murakami S, Furukawa S, Kreiborg S, Takada K. Human
masticatory muscle volume and zygomatico-mandibular form in adults with
mandibular prognathism. J Dent Res. 2002;81(11):752-756.
17. Ng HP, Foong KW, Ong SH, et al. Quantitative analysis of human
masticatory muscles using magnetic resonance imaging. Dentomaxillofac
Radiol. 2009;38(4):224-231.
18. Newton JP, Yemm R, Abel RW, Menhinick S. Changes in human jaw
muscles with age and dental state. Gerodontology. 1993;10(1):16-22.
19. Ohara Y, Hirano H, Watanabe Y, et al. Masseter muscle tension
and chewing ability in older persons. Geriatr Gerontol Int. 2013;13(2):
372-377.
20. Lin CS, Wu SY, Wu LT. The anterior insula and anterior cingulate
cortex are associated with avoidance of dental treatment based on
prior experience of treatment in healthy adults. BMC Neurosci. 2015;16(88):
1-17.
21. Hayakawa I, Watanabe I, Hirano S, Nagao M, Seki T. A simple
method for evaluating masticatory performance using a color-changeable
chewing gum. Int J Prosthodont. 1998;11(2):173-176.
22. Hama Y, Kanazawa M, Minakuchi S, Uchida T, Sasaki Y. Reliability
and validity of a quantitative color scale to evaluate masticatory perfor-
mance using color-changeable chewing gum. J Med Dent Sci. 2014;61(1):1-6.
23. Kikutani T, Tamura F, Nishiwaki K, et al. Oral motor function and
masticatory performance in the community-dwelling elderly. Odontology.
2009;97(1):38-42.
24. Schimmel M, Christou P, Herrmann F, Muller F. A two-colour
chewing gum test for masticatory efficiency: development of different
assessment methods. J Oral Rehabil. 2007;34(9):671-678.
25. Navazesh M, Kumar SK; University of Southern California School of
Dentistry. Measuring salivary flow: challenges and opportunities. JADA.
2008;139(suppl):35S-40S.
26. Rohleder N, Wolf JM, Maldonado EF, Kirschbaum C. The psycho-
social stress-induced increase in salivary alpha-amylase is independent of
saliva flow rate. Psychophysiology. 2006;43(6):645-652.
27. Yeh C-K, Johnson DA, Dodds MWJ, Sakai S, Rugh JD, Hatch JP.
Association of salivary flow rates with maximal bite force. J Dent Res. 2000;
79(8):1560-1565.
28. Van Eijden TM, Korfage JA, Brugman P. Architecture of the human
jaw-closing and jaw-opening muscles. Anat Rec. 1997;248(3):464-474.
29. Hannam AG, McMillan AS. Internal organization in the human jaw
muscles. Crit Rev Oral Biol Med. 1994;5(1):55-89.
30. Gaszynska E, Godala M, Szatko F, Gaszynski T. Masseter muscle ten-
sion, chewing ability, and selected parameters of physical fitness in elderly
care home residents in Lodz, Poland. Clin Interv Aging. 2014;9:1197-1203.
31. Watanabe Y, Hirano H, Arai H, et al. Relationship between frailty and
oral function in community-dwelling elderly adults. J Am Geriatr Soc. 2017;
65(1):66-76.
32. Palinkas M, Nassar MS, Cecilio FA, et al. Age and gender influence
on maximal bite force and masticatory muscles thickness. Arch Oral Biol.
2010;55(10):797-802.
JADA ( )
- - http://jada.ada.org - 2017 9
 ORIGINAL CONTRIBUTIONS
33. Hatch JP, Shinkai RS, Sakai S, Rugh JD, Paunovich ED. Determinants
of masticatory performance in dentate adults. Arch Oral Biol. 2001;46(7):
641-648.
34. Weijenberg RA, Scherder EJ, Visscher CM, Gorissen T,
Yoshida E, Lobbezoo F. Two-colour chewing gum mixing ability:
digitalisation and spatial heterogeneity analysis. J Oral Rehabil. 2013;
40(10):737-743.
35. Hector MP, Linden RW. The possible role of periodontal mechano-
receptors in the control of parotid secretion in man. Q J Exp Physiol. 1987;
72(3):285-301.
36. Rogus-Pulia NM, Larson C, Mittal BB, et al. Effects of change in
tongue pressure and salivary flow rate on swallow efficiency following
chemoradiation treatment for head and neck cancer. Dysphagia. 2016;31(5):
687-696.
10 JADA ( )
- - http://jada.ada.org - 2017
37. Rogus-Pulia NM, Logemann JA. Effects of reduced saliva production
on swallowing in patients with Sjogren’s syndrome. Dysphagia. 2011;26(3):
295-303.
38. Bharath S, Joshi H, John JP, et al. A multimodal structural and
functional neuroimaging study of amnestic mild cognitive impairment. Am
J Geriatr Psychiatry. 2017;25(2):158-169.
39. Chen WT, Chou KH, Liu LK, et al. Reduced cerebellar gray matter is
a neural signature of physical frailty. Hum Brain Mapp. 2015;36(9):3666-3676.
40. Lexomboon D, Trulsson M, Wardh I, Parker MG. Chewing ability
and tooth loss: association with cognitive impairment in an elderly pop-
ulation study. J Am Geriatr Soc. 2012;60(10):1951-1956.
41. Ono Y, Yamamoto T, Kubo KY, Onozuka M. Occlusion and brain
function: mastication as a prevention of cognitive dysfunction. J Oral
Rehabil. 2010;37(8):624-640.