Journal of

Dental Research
http://jdr.sagepub.com/

Architecture of the Human Pterygoid Muscles

T.M.G.J. van Eijden, J.H. Koolstra and P. Brugman
J DENT RES 1995 74: 1489
DOI: 10.1177/00220345950740080901
The online version of this article can be found at:
http://jdr.sagepub.com/content/74/8/1489

Published by:
http://www.sagepublications.com

On behalf of:
International and American Associations for Dental Research

Additional services and information for Journal of Dental Research can be found at:
Email Alerts: http://jdr.sagepub.com/cgi/alerts
Subscriptions: http://jdr.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
Citations: http://jdr.sagepub.com/content/74/8/1489.refs.html

Downloaded from jdr.sagepub.com by guest on July 12, 2011 For personal use only. No other uses without permission.

J Dent Res 74(8): 1489-1495, August, 1995

Architecture of the Human Pterygoid Muscles

T.M.G.J. van Eijden, J.H. Koolstra, and P. Brugman
Department of Functional Anatomy, Academic Centre for Dentistry Amsterdam (ACTA), Meibergdreef 15, 1105 AZ Amsterdam, The
Netherlands

Abstract. Muscle force is proportional to the physiological

cross-sectional area (PCSA), and muscle velocity and
excursion are proportional to the fiber length. The length of
the sarcomeres is a major determinant of both force and
velocity. The goal of this study was to characterize the
architecture of the human pterygoid muscles and to evaluate
possible functional consequences for muscle force and
muscle velocity. For the heads of the lateral and medial
pterygoid, the length of sarcomeres and of fiber bundles, the
PCSA, and the three-dimensional coordinates of origin and
insertion points were determined. Measurements were taken
from eight cadavers, and the data were used as input for a
model predicting sarcomere length and active muscle force
as a function of mandibular position. At the closed-jaw
position, sarcomeres in the lateral pterygoid (inferior head,
2.83 0.1 pm; superior head, 2.72 0.11 pm) were
significantly longer than those in the medial pterygoid
(anterior head, 2.48 0.36 pm; posterior head, 2.54 0.38
pm). With these initial lengths, the jaw angle at which the
muscles were capable of producing maximum active force
was estimated to be between 5 and 10. The lateral
pterygoid was characterized by relatively long fibers
(inferior, 23 2.7 mm; superior, 21.4 2.2 mm) and a small
PCSA (inferior, 2.82 0.66 cm2; superior, 0.95 0.35 cm2),
whereas the medial pterygoid had relatively short fibers
(anterior, 13.5 + 1.9 mm; posterior, 12.4 + 1.5 mm) and a
large PCSA (anterior, 2.47 0.57 cm2; posterior, 3.53 0.97
cm2). The mechanical consequence is that the lateral
pterygoid is capable of producing 1.7 times larger
displacements and velocities than the medial pterygoid,
whereas the medial pterygoid is capable of producing 1.6
times higher forces. The model showed that jaw movement
had a different effect on active force production in the
muscles.
Key words: masticatory muscles, muscle architecture,
sarcomere length, physiological cross-section, length-force
relationship.
Received November 29, 1994; Accepted June 15, 1995

Introduction
The force a muscle can produce depends on the length and
shortening velocity of the sarcomeres. The length changes to
which the sarcomeres are subjected during joint motion are
closely related to muscle architecture (Gans, 1982; Otten,
1988). Parameters of architecture-such as muscle fiber
position relative to the axis of rotation, pennation angle,
fiber length, sarcomere length, tendon length, and
physiological cross-sectional area-differ between muscles
(Wickiewicz et al., 1983; Friederich and Brand, 1990). In
addition, these parameters can vary within anatomically
complex muscles. For example, the anteriormost and
posteriormost regions of the human masseter have different
fiber and moment arm lengths (Van Eijden and Raadsheer,
1992). Hence, it is likely that the effect of joint motion upon
sarcomere length-and, as a consequence, upon the
magnitude of force and the velocity of shortening-differs
between and within muscles.
The human pterygoid muscles have a complex
architectural design. The lateral pterygoid is considered as
being fan-shaped with relatively long muscle fibers and the
medial pterygoid as being multipennate with short fibers
(Schumacher, 1961; Williams et al., 1989). In addition, the
lateral pterygoid is composed of two separate heads, which
are fused in front of the temporomandibular joint. Such a
division is less clear in the medial pterygoid, although its
anterior and posterior muscle fibers have distinct origins.
The action lines of the muscle heads differ in orientation
and position, suggesting a variety of functions in selective
activation (Van Eijden et al., 1988). The superior head of the
lateral pterygoid is believed to be activated during jaw
closing, whereas the inferior head is activated during
opening (Juniper, 1981; Wood et al., 1986). The medial
pterygoid is considered one of the main elevators of the
mandible.
The aim of the present study was to determine
quantitatively the structural characteristics of the pterygoid
muscles and to examine possible functional consequences.
1489

Downloaded from jdr.sagepub.com by guest on July 12, 2011 For personal use only. No other uses without permission.

1490

j Dent Res 74(8) 1995

van Eijden et al.

of the mandible.

Table 1. Characteristics of cadavers used

Sex

Age

F
F
F
M
F
M
F
M

56
69
56
89
55
82
83
74

Number of Teeth Present

lower dentition
upper dentition
12
5
6
8
9
9
6
3

12
10
8
10
5
11
8
5

Since sarcomere length and muscle fiber length are

important determinants of the active force and excursion a
muscle is capable of producing, we investigated sarcomere
and fiber lengths in a number of defined muscle regions. To
obtain an estimate of the maximum muscle force of each
region, we obtained physiological cross-sectional areas.
During jaw movements, the sarcomeres in the muscle
regions undergo length changes, depending on initial
sarcomere and fiber lengths, the nature of jaw movement,
but also on the three-dimensional coordinates of the
attachment locations. The spatial coordinates of attachment
points in the closed-jaw position were determined and used
in a three-dimensional rigid body model, which predicts
sarcomere length changes and the concomitant effects of
active muscle force during a number of defined jaw
movements.

Materials and methods

Cadaver heads
Eight Caucasian cadavers were used. Table 1 lists their sex, age,
and their condition of dentition. The material had undergone
routine perfusion with a mixture of formalin, glycerol, alcohol,
and phenol. Although the mouth was closed in all cadavers,
there was an average interincisal separation of 4.2 0.9 mm
(mean SD), determined from a lateral-view radiograph with a
film-focus distance of 4.5 m. All measurements were taken from
the right side of the head.

Definition of muscle portions

After the upper anterior half of the ramus of the mandible was
removed, the lateral surfaces of the pterygoid muscles were
freed from overlying tissue. The lateral pterygoid was
subdivided into an inferior and a superior head. The inferior
head originates from the lateral surface of the lateral pterygoid
plate, the superior head from the infratemporal surface of the
sphenoid bone. The heads converge laterally and posteriorly to
their insertion on the front of the anterior aspect of the mandible
and the articular capsule. The medial pterygoid was subdivided
into an anterior head, arising from the lateral surface of the
pyramidal process of the palatine bone and the tuberosity of the
maxilla, and a posterior head, which originates from the
pterygoid fossa and the medial surface of the lateral pterygoid
plate. The muscle passes laterally, inferiorly, and posteriorly
and inserts on the lower part of the medial surface of the ramus

Sarcomere lengths
From each cadaver, 16 samples (length, 5 mm; diameter, 2 mm),
eight per muscle and four per muscle head, were obtained. They
were taken halfway from each muscle belly, by means of a
sharp scalpel. The long axes of the samples were parallel to the
fiber direction of the muscle head and at equal craniocaudal
(inferior head, lateral pterygoid), mediolateral (superior head,
lateral pterygoid), and anteroposterior (medial pterygoid)
distances from each other. The samples were dehydrated in
alcohol and embedded in hydroxyethyl methacrylate
(Technovit). Three groups of four longitudinal sections (2 pm)
from each sample were stained with phosphotungstic acidhematein (PTAH); the distance between the groups was 200 pm
(i.e., 100 sections of 2 pm thickness). Fifteen randomly selected
rows, five per group, of ten sarcomeres were measured at a
magnification of xlOOO, by means of a Zeiss microscope with a
micrometer eyepiece (Weijs and Van der Wielen-Drent, 1982).

Fiber lengths and physiological cross-sectional areas

The muscle portions were detached from their origins and
insertions, and the centers of attachment were marked by holes
(diameter, 0.9 mm) drilled into the bone. To estimate the
pennation angle in the medial pterygoid muscle, we traced
tendon plates and a number of fiber bundles on acetate paper
from a longitudinal section through the muscle's heads; fibers of
the lateral pterygoid had no pennated arrangement. From each
muscle head, 20 small fiber bundles were carefully isolated by
dissection, and fiber bundle length was measured to the nearest
1 mm. After being blotted dry and weighed (total wet muscle
weight, MW), tendinous tissue was separated from the
contractile tissue and weighed (wet tendon weight, TW). The
physiological cross-sectional area of each muscle portion,
according to Weber (1846), was calculated as (MW-TW)/FL,
where FL is average fiber bundle length. To be able to
determine dry fiber and tendon weights, we dried the material
at 60C for two weeks (data not reported).
Three-dimensional coordinates of attachment sites
A 3-D digitizing system (3space digitizer, Polhemus, Colchester,
VT, USA) was used to register the spatial coordinates of origins
and insertions of the muscle heads and of several skull markers.
The system consists of an electromagnetic six-degrees-offreedom digitizer interfaced with a PC-compatible computer
running a custom-made anatomical data acquisition program.
Before dissection started, small holes (diameter, 0.9 mm) were
drilled into the bony cortex of the skull. They provided three
sets of markers: (1) to define a skull-related coordinate system,
(2) to provide reference points in the course of the dissection
procedure for determination of the position of the skull
coordinate system relative to the external coordinate system
defined by the 3-D digitizing system, and (3) to estimate
measurement error.
The following Cartesian coordinate system was defined: the
origin halfway between the temporomandibular joints, the x-y
plane parallel to the Frankfort Horizontal plane (FH) and the xz plane mid-sagittally; and the positive x-, y-, and z-axes,
pointing, respectively, anteriorly, laterally to the left, and
cranially. On each side of the head, holes were drilled in the

Downloaded from jdr.sagepub.com by guest on July 12, 2011 For personal use only. No other uses without permission.

J Dent Res 74(8) 1995

Architecture of the Pterygoid Muscles

lateral pole of the mandibular condyle (markers 1, 2) and in the

point orbitale along the margo infraorbitalis (markers 3, 4); in
addition, the point tragion was marked on the skin at the upper
rim of the meatus acusticus externus (markers 5, 6). The y-axis
was running through markers 1 and 2, and the FH plane was
computed as best fit between markers 3, 4, 5, and 6.
Four reference points were defined: (1) on the right mastoid
process, (2) on the top of the skull near the mid-sagittal plane,
and (3, 4) on, respectively, the left and right margo
infraorbitalis. The position of the reference points relative to the
skull coordinate system was fixed, and thus these points could
be used to determine the position of the skull reference system
relative to the external coordinate system of the measuring
apparatus. During the dissection procedure, the head was
placed in the apparatus several times, to register one or more of
the muscle attachment positions. After the head was placed in
the device, the coordinates of the reference points were
registered to calculate a rotation matrix and translation vector
(in accordance with Koolstra and Van Eijden, 1992), to
transform the muscle coordinates into the skull coordinate
system.
The following four markers were used to estimate
measurement error: (1) the right mastoid process, (2) the top of
the skull, (3) the left margo supraorbitalis, and (4) the mandible
near the mental foramen. The coordinates of these markers were
measured each time the head had a new position in the
apparatus. During the entire dissection and digitizing
procedure, including registrations on the masseter, temporalis,
and suprahyoid muscles (data not reported here), each head
was placed in the measuring apparatus 28 times. For each
marker, mean and standard deviation values of x, y, and z
coordinates were determined (n = 28), and the average value of
the standard deviations of the eight subjects was taken as an
estimate of measurement error. It appeared that the
measurement error did not differ significantly between the four
markers and the three coordinates. Mean SD of measurement
error was 0.44 0.12 mm (range between 0.26 and 0.72 mm; n =
12, 4 markers x 3 coordinates).

Model simulations
For each muscle portion, average sarcomere length, fiber length,
and the three-dimensional attachment coordinates were
measured for the closed-jaw position. For all other mandibular
positions, sarcomere length and concomitant active muscle
force were calculated from the changed distance between origin
and insertion with respect to the closed-jaw position by means
of a three-dimensional rigid-body model (Weijs et al., 1987). The
muscle tendons were modeled as stiff structures. The active
length-force relationship that was incorporated into the model
was a third-degree equation (Van Ruijven and Weijs, 1990) and
was based on experimental data from rabbit digastric muscle
(Muhl et al., 1978). Optimum sarcomere length was assumed to
be 2.73 jpm, and active force ranged between 1.89 pm and 4.00
pm. Since the three-dimensional displacements of the insertions
of the muscle portions are not known for normal activities (such
as chewing), open\close excursions were chosen as rotations of
the jaw (Van Eijden and Raadsheer, 1992). The following
rotations, in 10 increments, were simulated: (1) about a
transverse axis situated 2.0 cm below the temporomandibular
joint, i.e., about halfway up the ramus of the mandible through

1491

Sarcomere Length

Sarcomere Length

3.00

2.80
2.60
2.40

2.20

2.00

ant

post

Fiber Length

Fiber Length

E~
f

-i

post

ant

Physiological Cross-section

Physiological Cross-section

E
'a

inf

sup

Lateral Pterygoid

'ezez,,,ant

Medial

,
post

Pterygoid

Figure 1. Measured sarcomere lengths (top panels), fiber bundle

lengths (middle panels), and physiological cross-sectional areas
(lower panels) of the pterygoids. inf, inferior head; sup, superior
head; ant, anterior head; post, posterior head.

the mandibular foramen ("ramus axis"); and (2) about a

transverse axis running through the joint ("condylar axis").
Since, in our sample of cadavers, the jaw was not completely
closed, a rotation to the closed position was simulated. The
amount of this rotation was estimated to be 4.

Statistics
For survey of differences in sarcomere length within each
muscle, one-way analysis of variance (ANOVA) was used.
Paired Student t tests were used for comparison between
different muscle heads. In all tests, the level of significance was
set to p < 0.05.

Results
Sarcomere lengths
In the eight regions of the lateral pterygoid, the average
sarcomere length for all subjects ranged from 2.68 to 2.90
jm. The difference in sarcomere length among the eight
regions of the muscle was not significant. Average
sarcomere lengths in the regions of the medial pterygoid
muscle ranged between 2.43 and 2.60 pm and did not differ
significantly from each other. Fig. 1 shows grand means of
sarcomere length of the muscle heads. For each subject,
average value of the 4 samples per muscle head was
calculated (n = 60, 4 samples x 15 fibers); the standard
deviation bars are a measure of interindividual variability.
Sarcomeres in the lateral pterygoid (inferior head, 2.83 pm;

Downloaded from jdr.sagepub.com by guest on July 12, 2011 For personal use only. No other uses without permission.

1492

J Dent Res 74(8) 1995

van Eijden et al.

Table 3. Muscle length and directiona of action lines

Table 2. Variation of sarcomere and fiber bundle length

Sarcomere Length
Mean
SD

Lateral pterygoid
inferior head
superior head

8.0%
7.1%

2.2%
1.9%

Fiber Bundle Length

Mean
SDa
8.5%
9.6%

3.4%
3.0%

Medial pterygoid
anterior head
7.3%
0.8%
10.2%
5.6%
7.2%
1.2%
posterior head
13.4%
5.6%
a
Mean SD of the coefficient of variation of the eight muscles.

superior nead, 2.72 pm) were about 0.25 pm longer than

those in the medial pterygoid (anterior head, 2.48 pm;
posterior head, 2.54 pm), and the difference in length
between the muscles was significant. The variability of
sarcomere length within each muscle head is given in Table
2. It was estimated by the co-efficient of variation (CV =
X/SD * 100%) for the 60 fibers (4 samples x 15 fibers). This
variability did not differ significantly between the heads of
each muscle.

Fiber lengths and physiological cross-sections

Grand means of fiber bundle length are also given in Fig.
1 and variation of fiber bundle length within each muscle
head in Table 2. Average length of the fibers of the
inferior and superior head of the lateral pterygoid was
23.0 2.7 mm and 21.4 2.2 mm, respectively. The fibers
of the anterior and posterior head of the medial pterygoid
were 13.5 1.9 mm and 12.4 1.5 mm long, respectively.
These fibers were significantly shorter than those of the
lateral pterygoid. The total muscle length, calculated from
the three-dimensional coordinates of origins and
insertions, is given in Table 3. The ratios of fiber bundle
length to total muscle length for the inferior and superior
head of the lateral pterygoid were 0.74 0.07 and 0.70
0.06, respectively, and for the anterior and posterior head
of the medial pterygoid, 0.35 0.05 and 0.28 0.04,
respectively.
Physiological cross-sections of the muscle heads are
given in the lower panels of Fig. 1. The total cross-sectional
areas of the lateral and medial pterygoid muscles were 3.78
0.71 cm2 and 6.00 1.24 cm2, respectively. The inferior
head (2.82 cm2) of the lateral pterygoid was significantly
larger than the superior head (0.95 cm2), and the anterior
head (2.47 cm2) of the medial pterygoid was significantly
smaller than the posterior head (3.53 cm2).
Spatial orientation
The direction of the action lines of the muscle heads in
angles is given in Table 3. The sagittal angle is defined as the
angle between the action line and the vertical axis in the
sagittal plane; the frontal angle is defined as the angle
between the action line and the vertical axis in the frontal
plane; and the transverse angle is defined as the angle
between the action line and the anterio-posterior axis in the
transverse plane.

Mean

SDb

Lateral pterygoid
inferior head

length
sagittal angle
frontal angle
transverse angle
superior head
length
sagittal angle
frontal angle
transverse angle

32.6 mm
120.70
28.40
47.40

3.1 mm
11.70
8.40
7.30

31.3 mm
76.40
16.70
41.70

2.9 mm
7.80
11.50
8.00

Medial pterygoid
anterior head

length
sagittal angle
frontal angle
transverse angle
posterior head
length
sagittal angle
frontal angle

37.9 mm
33.80
31.30
42.10

6.0 mm
6.30
5.80
9.50

44.4 mm

5.5 mm
8.20
8.40
4.20

39.20
42.80
48.80

transverse angle
See text for definition of angles.
b Mean SD of the eight muscles.
a

Model simulations
During jaw opening, the distance between origin and
insertion became shorter in the lateral pterygoid and longer
in the medial pterygoid. This is illustrated in the top panels
of Fig. 2, where sarcomere length positions on the
normalized active length-force curve are shown for the
closed and open jaw, after a rotation about the ramus axis.
Note that the lateral pterygoid is mainly on the ascending
limb of the curve and the medial pterygoid mainly on the
descending limb. The amounts of sarcomere shortening in
the inferior and superior heads of the lateral pterygoid were
0.70 pm and 0.87 pm, respectively. In the anterior and
posterior heads of the medial pterygoid, the amounts of
sarcomere lengthening were 1.57 pm and 0.88 pm,
respectively. Differences in sarcomere excursion between
the heads of the muscles were primarily the result of
different moment-arm lengths relative to the axis of rotation.
In the middle and bottom panels of Fig. 2, the relationship
between jaw angle and active force is compared between
open/close rotation around the ramus axis and the joint
axis. The model predicted that, in the case of jaw opening
around the axis situated within the ramus, active force
decreased in both pterygoid muscles. Change of force was
relatively small in the inferior head of the lateral pterygoid
and in the posterior head of the medial pterygoid. When the
axis was located in the temporomandibular joint, the
moment-arm length of the medial pterygoids became
relatively large, while that of the lateral pterygoids was
negligible. Consequently, sarcomere excursions in the

Downloaded from jdr.sagepub.com by guest on July 12, 2011 For personal use only. No other uses without permission.

j Dent Res 74(8) 1995

Architecture of the Pterygoid Muscles

medial pterygoids increased, and the range of movement in

which the medial pterygoid contributed to active force
decreased. In the lateral pterygoid, no significant shortening
of sarcomeres occurred. Thus, the muscle could not
contribute to the opening moment of the jaw, despite the
favorable optimal position of its sarcomeres on the lengthforce curve throughout the entire movement.

1493
Medial Pterygoid

Lateral Pterygoid
--O interior head

0.80

040 open

pen

0L0.40

0.200

0.20

clos
Xn

0.60

o.oo

1.50

2.00

Discussion

2.50

3.00

3.50

4.00

.,50

4.50

2.00

The pterygoids in the muscle-joint system

The mechanical effects of the pterygoids depend not only on
their force-generating capabilities, determined primarily by
architecture, but also on their three-dimensional position
relative to the temporomandibular joints and/or axes of
rotation. To understand the muscle-joint interaction, one
must realize that the joint angle-sarcomere length
relationship depends on both moment-arm and muscle fiber
length. The ratio between the length of the fiber and the

inferior head

1.00

2.50

3.00

3.50

4.00

4.50

sarcomere length (pm)

sarcomere length (pm)

Mechanical consequences of pterygoid muscle architecture

The main purpose of the present study was to characterize
the architectural properties of the pterygoid muscles. Muscle
force is proportional to the physiological cross-sectional
area, and muscle velocity and excursion to the fiber length
(equivalent to the number of sarcomeres in the series). By
determining these architectural parameters, we can estimate
how much isometric force a muscle can produce, how fast it
contracts, and how far it shortens. The present results show
that these parameters vary between the pterygoids. The
lateral pterygoid is characterized by relatively long fibers
and a small physiological cross-sectional area, whereas the
medial pterygoid has relatively short fibers and a large
physiological cross-sectional area. The mechanical
consequence is that the lateral pterygoid can produce 1.7
times larger displacements and velocities than the medial
pterygoid (ratio of fiber lengths: 1.71), whereas the medial
pterygoid can produce about 1.6 times higher forces (ratio of
physiological cross-sections: 1.58). Similar architectural
specializations have been found in other human skeletal
muscles [(leg) Wickiewicz et al., 1983; Friederich and Brand,
1990; (arm) Lieber et al., 19901.
The lateral pterygoid has often been classified as a jaw
opener and the medial pterygoid as a closer. In the cadavers,
sarcomeres in the lateral pterygoid were, on average, 0.25
pm longer than those in the medial pterygoid. Note that, in
our sample of cadavers, the amount of jaw opening was
about 4. Consequently, the sarcomere length change that
occurred in the heads of the lateral and medial pterygoids
was estimated to be about 0.1 to 0.15 pm (ramus axis
rotation). The present results are in line with those of
another study (in preparation) in which we found that
sarcomeres of the openers were significantly larger than
those of the closers. These longer sarcomeres seem to be an
adaptation to the functional requirements imposed on the
muscles. During opening rotation and anterior translation
movements of the jaw, sarcomeres in the lateral pterygoid
will shorten, and a longer initial sarcomere length enables
the muscle to produce force throughout the entire
movement range (see Fig. 2, length-force curve).

A posterior head

cis

1.00

0.60

-0- anterior head

-a- superior head

close

z2S

1.00 -

superior head

anterior head

posterior head

00
1)

.------------

o.so

ab0.60
LL

0.60/

0.40.

U-

0.20

0.40.\"
\

0 20

o.oo

"

o.oo

20
25
5
t0o
t5
open angle (c) ondylar axis

0.80

superior head

inferior head

(-

1.00

0.80

LL

L0. 40

0.20

0.00

o.oo

15
25
1o
20
open angle () ramus axis

.\

0.40

0.20

--posterior head

tD0.60

."

anterior head

30

:'

e 0.60

5
10o
25
t5
20
open angle () condylar axis

30

30

.
o

10

15

20

25

30

open angle () ramus axis

Figure 2. Range of sarcomere lengths (top panels) on the normalized

active length-force curve (Van Ruijven and Weijs, 1990). As in our
sample of cadavers, the jaw was not completely closed. A rotation of
40 to the closed position was simulated first. From this position, a
rotation of 300 was carried out to open the jaw; the axis of rotation
was located halfway up the ramus. Jaw angle-force relationships
were predicted with the model for open rotations about the
condylar (middle panels) and ramus axis (lower panels). Vertical
dashed line indicates the estimated jaw angle (40) present in the
cadavers.

moment arm will determine the amount of sarcomere

lengthening/shortening during jaw movement. With a long
moment arm, fiber excursion will be larger than with a short
moment arm. The length change is absorbed by the
sarcomeres, and the number of sarcomeres in the series
determines the length change per sarcomere. The longer a
muscle fiber, the larger the number of sarcomeres and the
smaller the length change per sarcomere. The two pterygoid
muscles have different three-dimensional positions, with
different moment arms and different fiber lengths. In the
case of a rotation axis situated in the joint, the lateral
pterygoid fiber length is long relative to its moment arm,
and sarcomere length and thus muscle force will change
little during jaw opening (Fig. 2, condylar axis). The
drawback, however, is that the contribution of the muscle to
the opening moment will be small, because of its small
moment arm. The medial pterygoid has short fibers
compared with its moment arm, and the sarcomeres will
change length a great deal, as will muscle force (Fig. 2,
condylar axis). A more caudally situated axis (Fig. 2, ramus
axis), which is more realistic, seems more appropriate, since
then both pterygoids are capable of contributing to the
production of moment throughout a relatively large range of
motion.

Downloaded from jdr.sagepub.com by guest on July 12, 2011 For personal use only. No other uses without permission.

1494

J Dent Res 74(8) 1995

van Eijden et al.

Because the pterygoid muscles have relatively large

attachment areas, different muscle regions have different
moment-arm lengths. Hence, during jaw movement, muscle
fibers in the various muscle regions will undergo different
length changes. Theoretically, an equivalent change of
sarcomere length, and consequently of relative tension, in
the different muscle heads can be attained if the regions lie
parallel and the ratio of sarcomere numbers (fiber length) is
proportional to the ratio of moment-arm lengths (Gans and
de Vree, 1987). Obviously, this was the case neither in the
pterygoid muscles nor in the masseter muscle (Van Eijden
and Raadsheer, 1992). We did not find a significant
intramuscular difference in fiber length. For this reason,
different sarcomere excursions within each pterygoid are
primarily due to different moment-arm lengths.
Consequently, after jaw opening, the positions of the
sarcomeres on the length-force curve are not similar, either
for the anterior and posterior heads of the medial pterygoid
or for the superior and inferior heads of the lateral
pterygoid. This means that during jaw movement, the
distribution of active tension is not uniform but will vary
continuously throughout the muscle, due to differential
lengthening/shortening of sarcomeres. In addition, and in
contrast to muscles with small-circumscript attachment
sites, the maximum forces that can be produced by the
pterygoids become less dependent on muscle length (jaw
position).
Other factors that possibly affect the shape of the
length(angle)-force relationship of a particular muscle
(region), and that have not been taken into account in the
model that we used, are variation in pennation angle (Otten,
1988), and distribution of fiber lengths (Ettema and Huijing,
1994) and sarcomere lengths (Van Eijden and Raadsheer,
1992). Because of fiber angulation, the extension in the
sarcomeres is actually less than that of the muscle. In the
medial pterygoid, average pennation angle was smaller than
150, and for this angle the estimated decrease in sarcomere
excursion is relatively small: approximately 3.5%. If, in a
particular muscle region, fiber and sarcomere length vary,
the sarcomeres will have different optimum lengths.
Consequently, the active working range of the muscle region
increases, and the maximum forces that can be produced
become less dependent on muscle length (jaw angle). In
addition, we modeled the tendon plates as stiff structures,
but there is evidence (e.g., Ettema and Huijing, 1989) that
they can undergo large extensions (> 5%), increasing the
muscle's working range.

Comparison with data from the literature

With respect to sarcomere lengths, no data have thus far
been reported for the human pterygoid muscles. Weijs and
Van der Wielen-Drent (1982) found, in the rabbit medial and
lateral pterygoid, lengths of, respectively, approximately
2.15 and 2.65 pm (their Fig. 2), which is in line with the
present results.
The most complete set of data on fiber lengths and
physiological cross-sectional areas comes from the work of
Schumacher (1961) and Weijs and Hillen (1984, 1985). The
fiber lengths agree fairly well, but the cross-sectional areas
reported in the present study are significantly larger than

those that Schumacher (1961) and Weijs and Hillen (1984)

found in cadavers; however, they agree well with those
determined in vivo by means of computed tomography
(Weijs and Hillen, 1985). Although it is known that, with
increasing age, the cross-sectional area of the medial
pterygoid can be reduced by about 35% (Newton et al.,
1987), age alone cannot explain the observed differences. An
explanation might be that in our sample and in the in vivo
study by Weijs and Hillen, the number of teeth present was
much higher than in Schumacher's study, and that loss of
teeth is more important in jaw muscle atrophy, and thus for
force, than age.
With respect to the three-dimensional orientation of the
muscles, information has recently become available from
reconstructions using Magnetic Resonance Imaging
(Koolstra et al., 1990; Van Spronsen, 1993). For the lateral
pterygoid, the spatial orientations given in their and our
studies agree well. However, in their studies, the medial
pterygoid is oriented more vertically: about 200 in the
sagittal plane and about 150 in the frontal plane. This might
suggest that, for vertically directed muscles, determination
of action lines from MRI reconstructions is susceptible to
error.

Acknowledgments
We are grateful to W.A. Weijs for his constructive criticism,
L. van Ruijven for help with various technical problems, and
H. Korfage for help with histological processing of the
muscle samples. Institutional funds supported this project.

References
Ettema GJC, Huijing PA (1989). Properties of the tendinous
structures and series elastic component of EDL muscletendon complex of the rat. J Biomechan 22:1209-1215.
Ettema GJC, Huijing PA (1994). Effects of distribution of muscle
fiber length on active length-force characteristics of rat
gastrocnemius medialis. Anat Rec 239:414-421.
Friederich JA, Brand RA (1990). Muscle architecture in the
human lower limb. J Biomechan 23:91-95.
Gans C (1982). Fiber architecture and muscle function. Exerc
Sport Sci Rev 10:160-207.
Gans C, De Vree F (1987). Functional bases of fiber length and
angulation in muscle. i Morphol 192:63-85.
Juniper RP (1981). The superior pterygoid muscle? Br J Oral
Surg 19:121-128.
Koolstra JH, Van Eijden TMGJ (1992). Application and validation
of a three-dimensional mathematical model of the human
masticatory system in vivo. J Biomechan 25:175-187.
Koolstra JH, Van Eijden TMGJ, Van Spronsen PH, Weijs WA,
Valk J (1990). 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 35:549-556.
Lieber RL, Fazeli BM, Botte MJ (1990). Architecture of selected
wrist flexor and extensor muscles. J Hand Surg 15:244-250.
Muhl ZF, Grimm AF, Glick PL (1978). Physiologic and
histologic measurements of the rabbit digastric muscle.
Arch Oral Biol 23:1051-1059.
Newton JP, Abel RW, Robertson EM, Yemm R (1987). Changes
in human masseter and pterygoid muscles with age: a

Downloaded from jdr.sagepub.com by guest on July 12, 2011 For personal use only. No other uses without permission.

j Dent Res 74(8) 1995

Architecture of the Pterygoid Muscles

study by computed tomography. Gerodontics 3:151-154.

Otten E (1988). Concepts and models of functional architecture
in skeletal muscle. Exerc Sport Sci Rev 16:89-137.
Schumacher GH (1961). Funktionelle Morphologie der
Kaumuskulatur. Jena: Gustav Fischer Verlag.
Van Eijden TMGJ, Raadsheer MC (1992). Heterogeneity of fiber
and sarcomere length in the human masseter muscle. Anat
Rec 232:78-84.
Van Eijden TMGJ, Klok EM, Weijs WA, Koolstra JH (1988).
Mechanical capabilities of the human jaw muscles studied
with a mathematical model. Arch Oral Biol 33:819-826.
Van Ruijven LJ, Weijs WA (1990). A new model for calculating
muscle forces from electromyograms. Eur J Appl Physiol
61:479-485.
Van Spronsen PH (1993). Masticatory muscles in normal and
long-face humans (dissertation). Amsterdam: Free
University Amsterdam.
Weber E (1846). Muskelbewegung. In: Handworterbuch der
Physiologie. Wagner S, editor. Braunschweig: Bieweg.
Weijs WA, Hillen B (1984). Relationship between the
physiological cross-section of the human jaw muscles and

1495

their cross-sectional area in computer tomograms. Acta

Anat 118:129-138.
Weijs WA, Hillen B (1985). Physiological cross-section of the
human jaw muscles. Acta Anat 121:31-35.
Weijs WA, Van der Wielen-Drent TK (1982). Sarcomere length
and EMG activity in some jaw muscles of the rabbit. Acta
Anat 113:178-188.
Weijs WA, Brugman P, Klok EM (1987). The growth of the skull
and jaw muscles and its functional consequences in the
New Zealand rabbit (Oryctolagus cuniculus). J Morphol
194:143-161.
Wickiewicz TL, Roy RR, Powell PL, Edgerton VR (1983). Muscle
architecture of the human lower limb. Clin Orthop Rel Res
179:275-283.
Williams PL, Warwick R, Dyson M, Bannister LH, editors
(1989). Gray's anatomy. 37th ed. Edinburgh: Churchill
Livingstone.
Wood WW, Takada K, Hannam AG (1986). The
electromyographic activity of the inferior part of the
human lateral pterygoid muscle during clenching and
chewing. Arch Oral Biol 31:245-253.

Downloaded from jdr.sagepub.com by guest on July 12, 2011 For personal use only. No other uses without permission.