Journal of Oral Rehabilitation 2000 27; 754–759
Stress distribution in the temporomandibular joint affected
by anterior disc displacement: a three-dimensional analytic
approach with the finite-element method
TANAKA E., RODRIGO D. P., MIYAWAKI Y., LEE K., YAMAGUCHI K. & TANNE K.
Department of Orthodontics, Hiroshima University School of Dentistry, Hiroshima, Japan
SUMMARY The purpose of this study was to investi-
gate the influences of anterior disc displacement on
TMJ loading during maximum clenching by use of
finite-element analysis. Based on a young human
dry skull, an analytic model of the mandible includ-
ing the TMJ was developed. In addition to the
standard model with normal disc–condyle relation,
two models were designed to simulate various de-
grees of anterior disc displacement. In the standard
model, compressive stresses were induced in the
anterior, middle and lateral areas on the condyle
and glenoid fossa, whereas tensile stresses were
Introduction
TMJ internal derangement (ID) is defined as an abnor-
mal relationship between the articular disc and condyle
(Farrar & McCarty, 1979; Price et al., 1992). When the
disc is displaced anteriorly, disc perforation frequently
occurs at the posterior attachment of the disc (Dolwick
& Riggs, 1983). The evidence was derived from gross
and histological examinations of autopsy material.
Blackwood (1969) observed, in cadavers, tearing of the
articular disc at the posterior attachment and slightly
posterior position of the condyle with anterior dis-
placement of the disc. A possible explanation is that
continuous compression to the post-discal tissues by
the posteriorly displaced condyle may cause some at-
tenuation of the disc attachment and render the disc
more susceptible to rupture. Oberg, Carlsson & Fajers
(1971) found, in an observation of 115 cadavers, 18
discs with arthrotic changes and 14 discs with perfora-
tion in the lateral or central area.
© 2000 Blackwell Science Ltd
observed in the posterior and medial regions. In the
models with anterior disc displacement, compres-
sive stresses were recognized in all the areas of TMJ
components excluding the bilaminar zone. Shear
stresses in the articular disc and bilaminar zone
significantly increased in most areas. In conclusion,
stress distributions in the TMJ with a normal disc
position was substantially different from those
with anterior disc displacement, suggesting that the
progress in disc displacement may have some asso-
ciation with the nature of stress distributions in the
TMJ, in the articular disc in particular.
Clinically, TMJ sound or clicking is regarded as an
initial symptom of TMJ-ID. It is assumed that a simple
clicking proceeds to an intermittent locking, and the
subsequent closed-lock (Dolwick, Katzberg & Helms,
1983). These objective symptoms are associated with
the degree of anterior disc displacement.
Magnetic resonance imaging (MRI) has recently
been used for diagnosis of TMJ-ID. MRI can give infor-
mation for disc hydration and morphology as well as
the disc position, all of which are pertinent to the
severity of TMJ-ID (Helms, 1991). Helms (1991)
classified the degree of anterior disc displacement into
two grades that correlate with the progress in degener-
ative joint disease and duration of the disease history.
Murakami et al. (1993) observed 273 MRIs from TMJ-
ID patients, and assessed disc position relative to the
condyle at the closed position. The disc position was
classified into four types according to the position of
the posterior band in relation to the functional surface
of the condyle. Most patients with TMJ-ID exhibited
754
 STRESS DISTRIBUTION IN THE TMJ WITH ANTERIOR DISC DISPLACEMENT 755

progress in disc displacement, and disc displacement

was not improved without an appropriate treatment.
Among the influences of various variables on TMJ-
ID, the mechanical component appears to be of signifi-
cant importance as the contributing and causative
factor (Arnett, Milam & Gottesman, 1996). Excessive
compressive and shear stresses, especially, are probably
the most common sources of condylar resorption and
disc perforation (Arnett et al., 1996; Lai, Bowley &
Burch, 1998). Progress in disc displacement also seems
to have some association with the nature of stress
distribution and biomechanical equilibrium in the TMJ.
However, there has been little information about
biomechanical components in the TMJ with disc
displacement. Fig. 1. A three-dimensional finite-element model of the
The purpose of this study was to investigate the mandible including the TMJ.
influences of anterior disc displacement on TMJ load-
ing during maximum clenching by use of finite-ele-
on the basis of previous experimental data (Huiskes,
ment analysis.
1982; Carter & Hayes, 1977; Woo, Mow & Lai, 1988;
Tanne, Tanaka & Sakuda, 1991; Chen & Xu, 1994), as
Materials and methods shown in Table 1. The elastic modulus of connective
tissue was used for the bilaminar zone because of no
An analytic model of the mandible including the TMJ information about the mechanical properties of this
was developed from a young human dry skull with structure. The accuracy of stress analysis with this
acceptable occlusion. The dry skull was cut into trans- model has already been confirmed in a preliminary
verse sections of 1 cm thickness in parallel to the study (Tanaka, Tanne & Sakuda, 1994).
Frankfort horizontal plane. Photographs of both the In addition to the standard model, two models were
dorsal and ventral aspects of each section were taken, designed to simulate disc displacement according to the
and traced on acetate paper to represent the actual nature of anterior disc displacement (Helms, 1991;
anatomical structures. These two-dimensional draw- Murakami et al., 1993). The first model represented
ings were divided into a finite number of elements, slight anterior disc displacement, where the border
ensuring that the geometric equivalence of the model between the posterior band and bilaminar zone located
to the object was maintained. All the two-dimensional anteriorly against the most anterior point of the func-
images were then stacked perpendicularly to the tional surface of the condyle. The second one exhibited
Frankfort horizontal plane to develop a three-dimen- severe anterior disc displacement with deformed disc.
sional model of the mandible including the TMJ. The articular disc located antero-inferiorly relative to
The TMJ disc was constructed with solid elements of the condyle, exhibiting a stuck or folded form (Fig. 2).
approximately 2 mm thickness on the basis of anatom-
ical findings (Hansson & Nordstrom, 1977). Further- Table 1. Mechanical properties of different components in the
model
more, the surface of the condyle was designed to be
covered with an articular cartilage layer of 0·2 mm Component Elastic modulus (MPa) Poisson’s ratio
thickness on average (Pullinger, Baldioceda & Bibb,
Cortical bone 13 700 0·30
1990). Finally, a three-dimensional finite-element
Cancellous bone 7930 0·30
model of the mandible including the TMJ was con- Articular disc 44·1 0·40
structed, consisting of 2088 nodes and 1105 solid ele- Bilaminar zone 0·49 0·49
Articular carilage 0·49 0·49
ments. Hereafter, this model is referred to as the
Periodontal liga- 0·79 0·49
standard model (Fig. 1). Six components were included ment
in the model, and the material constants were defined

© 2000 Blackwell Science Ltd, Journal of Oral Rehabilitation 27; 754 – 759
756 E . T A N A K A et al.

Fig. 2. Degree of anterior disc displacement simulated in this

analysis.

These models with two types of disc displacement were

referred to as grade 1 and grade 2 models, respectively. Fig. 3. Means stresses on the surface of the condyle. Standard
The articular discs in these three models were designed ( ), grade 1 (a), grade 2 (
). ** PB 0·01; *** PB 0·001.
to maintain the same volume, although the number of
elements for the bilaminar zone was varied according
obtained from the three principal stresses in each ele-
to the degree of disc displacement.
ment. The articular disc, consisting of 77 elements, was
For loading conditions, the magnitude of muscle
also divided into five areas as described above, each of
forces and the line of action were determined. The
which comprises 16 elements on average. The mean
magnitude of muscle forces was defined as 500 N on
normal and shear stresses were calculated for each area
the basis of such assumption that forces exerted by the
in the articular disc and bilaminar zone. Furthermore,
masticatory muscles were proportional to the cross-sec-
mean stresses in the standard model with normal disc
tional areas of muscles (Maughan, Waston & Weir,
position were compared with those in the grade 1 and
1983). The line of muscle force was determined ac-
2 models with anterior disc displacements by use of a
cording to the anatomical sites of muscle insertion and
Student’s t-test.
origin (Williams et al., 1989); i.e. two nodes located in
the centre of muscle insertion and origin areas were
regarded as the points to construct the force vector. For Results
the boundary condition, the superior area of the tem-
In the standard model, the pattern of stress distribution
poral bone was restrained in all directions to avoid the
on the surface of the condyle was essentially similar to
sliding movement of the entire model.
that on the glenoid fossa, although the absolute values
Stress analysis was executed on a personal computer
were less on the glenoid fossa than on the condyle.
using an analysis program (ANSYS*). Three principal
Compressive stresses were induced in the anterior,
stresses were analysed for the condyle and glenoid
fossa. The surfaces of the condyle and glenoid fossa
within the articular capsule were constructed by 65
and 110 nodes, respectively. One principal stress, of
which the absolute value was larger than those of the
remaining principal stresses, was selected for stress
evaluation. Furthermore, based on anatomical findings
(Rees, 1954), these surfaces were divided into five
(anterior, middle, posterior, medial and lateral) areas.
Each area on the condyle and glenoid fossa consists of
13 and 22 nodes on average. Mean stress was obtained
for the five areas as a mean of the stresses on the nodes
included in each area. For the articular disc and bilam-
inar zone, octahedral normal and shear stresses were Fig. 4. Mean stresses on the surface of the glenoid fossa. Stan-
dard ( ), grade 1 (a), grade 2 (
). * PB 0·05; ** P B0·01;
* Swanson Analysis Systems Co., Houston, TX, U.S.A. *** P B0·001.

© 2000 Blackwell Science Ltd, Journal of Oral Rehabilitation 27; 754–759

 STRESS DISTRIBUTION IN THE TMJ WITH ANTERIOR DISC DISPLACEMENT 757

shear stresses in the bilaminar zone were five times

larger in the disc displacement models than in the
standard model. Although the shear stresses in the
standard model were less than 0·2 MPa in all the areas,
the value became greater than 0·5 MPa in the posterior
area of the articular disc and bilaminar zone in the
grade 1 and 2 models. Meanwhile, no significant differ-
ences in the shear stresses were found in all the areas
between the grade 1 and 2 models (Fig. 6).

Fig. 5. Normal stresses in the articular disc. Standard ( ), grade

Discussion
1 (a), grade 2 (
). * P B 0·05; *** PB 0·001.
The TMJ is one of the load-bearing organs in the
human body. Mechanical loading, if optimal, is of great
middle and lateral areas, whereas tensile stresses were importance for the development of TMJ structures dur-
observed in the posterior and medial regions (Figs 3 ing adolescence and condylar remodelling in the adult
and 4). The maximum tensile stress was observed in
(Mongini, 1977; Copray, Jansen & Duterloo, 1985;
the posterior area on the condyle.
Hinton, 1985). In the present model with normal disc
In the grade 1 model, compressive stresses were
position, compressive stresses were induced in the
induced in all the areas on the condyle and glenoid
anterior, middle and lateral areas, and tensile stresses
fossa (Figs 3 and 4). Mean stresses in the posterior and
were generated in the posterior, medial and post-discal
medial areas varied from tensile to compressive stress.
regions (bilaminar zone). Oberg et al. (1971) and Taka-
Significant differences in the mean stresses were found
mura & Maruyama (1980) observed normal articular
in these areas between the standard and grade 1 mod-
els (Figs 3 and 4). In the grade 2 model, stress distribu- discs in cadavers and demonstrated that such morpho-
tions on the surface of the condyle were similar to logical changes as disc thinning and perforation were
those in the grade 1 model (Figs 3 and 4). On the more frequently found in the anterior and lateral areas
surface of the glenoid fossa, significant differences in than in the remaining regions. Furthermore, Kopp
the mean stresses were found in the middle, posterior (1976) investigated the distribution of glycosaminogly-
and lateral areas between the standard and grade 2 can (GAG), which is associated with compressive
models (Fig. 4). stresses in the human articular disc, and showed that
In the articular disc, tensile stresses were observed in GAG was distributed more frequently in the anterior
the posterior and medial areas for the standard model and lateral areas of the disc. On the other hand, in the
(Fig. 5). In the grade 1 and 2 models, meanwhile, posterior area of the articular disc and the bilaminar
compressive stresses were induced in all the areas of zone, the tensile stress appears necessary for maintain-
the articular disc. In particular, compressive stresses in
the posterior area were the greatest, and significant
differences in the mean stresses were found in the
posterior area between the standard and disc-displace-
ment models. The compressive stress in the middle
area was significantly smaller in the grade 2 model
than those in the other two models. In the bilaminar
zone, the largest tensile stress was observed in the
standard model, and the slight compressive stress was
induced in the grade 2 model (Fig. 5).
Shear stresses in the articular disc and bilaminar
zone exhibited a significant increase in association with
Fig. 6. Shear stresses in the articular disc. Standard ( ), grade 1
anterior disc displacement (Fig. 6). In particular, the (a), grade 2 (
). ** PB 0·01; *** PB 0·001.

© 2000 Blackwell Science Ltd, Journal of Oral Rehabilitation 27; 754 – 759
758 E . T A N A K A et al.
ing normal relations among the condyle, glenoid fossa
and articular disc (Osborn, 1985). Therefore, the na-
ture of stress distributions revealed in the present study
were fully consistent with the previous findings in both
anatomical and histological aspects. It is also suggested
from this study that the articular disc has an important
role in controlling stress distribution and reducing
compressive stresses during mandibular movement, as
reported by Nickel & McLachlan (1994).
The present study has demonstrated that stress distri-
butions in the TMJ are substantially influenced by the
degree of anterior disc displacement. For the mecha-
nisms of clicking, Osborn (1985) demonstrated that an
audible clicking occurs when the energy of compres-
sion is stored into the distorted disc and suddenly
released. Isberg, Widmalm & Ivarsson (1985) investi-
gated an arthrokinetic reflex in the masticatory mus-
cles in association with disc displacement, and
indicated that continuous muscle activity was pro-
voked by disc displacement. An existence of non-phys-
iological stress distribution in the TMJ in association
with disc displacement was revealed in this study,
although the present results are limited in information
about the occurrence of disc displacement and clicking.
In the articular disc, compressive stresses were widely
induced with a disappearance of tensile stresses. Fur-
thermore, in the bilaminar zone, tensile stress was
reduced in the grade 1 model with slight disc displace-
ment, whereas compressive stress was induced in the
grade 2 model with severe disc displacement. An im-
portant function of the bilaminar zone is to maintain
the positional relation between the condyle, disc and
glenoid fossa during jaw movements in terms of pre-
venting a nipping of soft tissues posterior to the
condyle. Scapino (1983) observed human articular disc
with anterior displacement, and reported the existence
of largely dilated venous channels at the posterior
attachment area. These findings were interpreted as a
result of a defect in negative pressure during mandibu-
lar movement which would produce circulatory ob-
struction in the bilaminar zone. Isberg & Isacsson
(1986) investigated TMJ tissue reaction to retruded
mandibular position using Macaca fascicularis monkeys,
and found distinctly dilated blood vessels in the bilami-
nar zone. Given these findings, it can be speculated
that anterior disc displacement produces non-physio-
logical compression in the bilaminar zone, as demon-
strated in this study. It is thus suggested that changes
© 2000 Blackwell Science Ltd, Journal of Oral Rehabilitation 27; 754–759
in stress distributions in the TMJ may have some asso-
ciation with the progress in TMJ-ID.
Progress in disc displacements appears to be caused
by loosening of the bilaminar zone. In the present
study, the bilaminar zone with slight displacement of
the disc experienced tensile stress. However, the shear
stresses in the bilaminar zone are five times greater in
the disc displacement models than in the standard
model. Excessive shear stress tends to increase the
stiffness of the articular disc and the subsequent break-
age of the disc such as perforation (Lai et al., 1998).
Judging from an existence of tensile stress in the bilam-
inar zone for the grade 1 model, some function for
repositioning the dislocated disc to the normal position
may remain. Increase in shear stress in the bilaminar
zone may be regarded as a cause of the progress in disc
displacement. In the grade 2 model with severe disc
displacement, shear stresses were similar to those in
the grade 1 model in qualitative and quantitative as-
pects. A clinical implication from these findings is that
an early approach to disc displacement would be desir-
able to reduce excessive shear stress in the articular
disc and bilaminar zone.
In conclusion, stress distribution in the TMJ with a
normal disc position was substantially different from
those with an anteriorly displaced disc. It is also sug-
gested that progress in disc displacement may have
some association with the nature of stress distribution
in the TMJ, for the articular disc in particular.
References
ARNETT, G.W., MILAM, B. & GOTTESMAN, L. (1996) Progressive
mandibular retrusion-idiopathic condylar resorption. Part II.
American Journal of Orthodontics and Dentofacial Orthopedics, 110,
117.
BLACKWOOD, H. (1969) Pathology of the temporomandibular
joint. Journal of American Dental Association, 79, 118.
CARTER, D.R. & HAYES, W.C. (1977) The behavior of bone as a
two-phase porous structure. Journal of Bone and Joint Surgery,
59A, 954.
CHEN, J. & XU, L. (1994) A finite element analysis of the human
temporomandibular joint. Journal of Biomechanical Engineering,
116, 401.
COPRAY, J.C.V.M., JANSEN, H.W.B. & DUTERLOO, H.S. (1985)
Effects of compressive forces on proliferation and matrix syn-
thesis in mandibular condylar cartilage of the rat in vitro.
Archives of Oral Biology, 30, 299.
DOLWICK, M.F., KATZBERG, M.D. & HELMS, C.A. (1983) Internal
derangements of the temporomandibular joint: fact or fiction?
Journal of Prosthetic Dentistry, 49, 415.
 STRESS DISTRIBUTION IN THE TMJ WITH ANTERIOR DISC DISPLACEMENT
DOLWICK, M.F. & RIGGS, R.R. (1983) Diagnosis and treatment of
internal derangements of the temporomandibular joint. Dental
Clinic of Northern America, 27, 561.
FARRAR, W.B. & MCCARTY, W.L. JR (1979) Inferior joint space
arthrography and characteristics of condylar paths in internal
derangements of the TMJ. Journal of Prosthetic Dentistry, 41,
545.
HANSSON, T. & NORDSTROM, B. (1977) Thickness of the soft tissue
layers and articular disk in temporomandibular joints with
deviations in form. Acta Odontologica Scandinavica, 35, 281.
HELMS, C.A. (1991) Magnetic resonance imaging of the TMJ. In:
Current Controversies in Temporomandibular Disorders (ed. C. Mc-
Neill), pp. 112 – 114. Quintessence Publishing Co., Chicago.
HINTON, R.J. (1985) Adaptive response of the articular eminence
and mandibular fossa to altered function of the lower jaw: an
overview. In: Developmental Aspects of Temporomandibular Joint
Disorders. Craniofacial Growth Series Monograph No. 16. Center for
Human Growth and Development, University of Michigan (eds D. S.
Carlson, J. A. Jr McNamara & K. A. Ribbens), pp. 207 – 234.
Ann Arbor, MI.
HUISKES, R. (1982) On the modelling of long bones in structural
analyses. Journal of Biomechanics, 15, 65.
ISBERG, A.M. & ISACSSON, G. (1986) Tissue reactions of the
temporomandibular joint following retrusive guidance of the
mandible. Journal of Craniomandibular Practice, 4, 144.
ISBERG, A., WIDMALM, S.E. & IVARSSON, R. (1985) Clinical, radio-
graphic and electromyographic study of patients with internal
derangements of the temporomandibular joint. American Jour-
nal of Orthodontics, 88, 453.
KOPP, S. (1976) Topographical distribution of sulphated gly-
cosaminoglycans in human temporomandibular joint disks. A
histochemical study of an autopsy material. Journal of Oral
Pathology, 5, 265.
LAI, W.F., BOWLEY, J. & BURCH, J.G. (1998) Evaluation of shear
stress of the human temporomandibular joint disc. Journal of
Orofacial Pain, 12, 153.
MAUGHAN, R.J., WASTON, J.S. & WEIR, J. (1983) Strength and
cross-sectional area of human skeletal muscle. Journal of Physi-
ology, 33, 37.
MONGINI, F. (1977) Anatomic and clinical evaluation of the
relationship between the temporomandibular joint and occlu-
sion. Journal of Prosthetic Dentistry, 38, 539.
MURAKAMI, S., TAKAHASHI, A., NISHIYAMA, H., FUJISHITA, M. &
FUCHIHATA, H. (1993) Magnetic resonance evaluation of the
© 2000 Blackwell Science Ltd, Journal of Oral Rehabilitation 27; 754 – 759
759
temporomandibular joint disc position and configuration. Den-
tomaxillofacial Radiology, 22, 205.
NICKEL, J.C. & MCLACHLAN, K.R. (1994) In vitro measurement of
the stress-distribution properties of the pig temporomandibular
joint disc. Archives of Oral Biology, 39, 439.
OBERG, T., CARLSSON, G.E. & FAJERS, C.M. (1971) The temporo-
mandibular joint. A morphologic study on a human autopsy
material. Acta Odontologica Scandinavica, 29, 349.
OSBORN, J.W. (1985) The disc of the human temporomandibular
joint: design, function and failure. Journal of Oral Rehabilitation,
12, 279.
PRICE, C.A., CONNELL, D.G., MACKAY, A. & TOBIAS, D.L. (1992)
Three-dimensional reconstruction of magnetic resonance im-
ages of the temporomandibular joint by I-DEAS. Dentomaxillo-
facial Radiology, 21, 148.
PULLINGER, A.G., BALDIOCEDA, F. & BIBB, C.A. (1990) Relationship
of TMJ articular soft tissue to underlying bone in young adult
condyles. Journal of Dental Research, 69, 1512.
REES, L.A. (1954) The structure and function of the mandibular
joint. British Dental Journal, 96, 125.
SCAPINO, R.P. (1983) Histopathology associated with malposition
of the human temporomandibular joint disk. Oral Surgery, Oral
Medicine, Oral Pathology, 55, 382.
TAKAMURA, H. & MARUYAMA, T. (1980) Studies on the structural
changes of the temporomandibular joint — macroscopic obser-
vations. Journal of the Japan Prothodontic Society, 28, 49.
TANAKA, E., TANNE, K. & SAKUDA, M. (1994) A three-dimensional
finite element model of the mandible including the TMJ and its
application to stress analysis in the TMJ during clenching.
Medical Engineering and Physics, 16, 316.
TANNE, K., TANAKA, E. & SAKUDA, M. (1991) The elastic modulus
of the temporomandibular joint disc from adult dogs. Journal of
Dental Research, 70, 1545.
WILLIAMS, P.L., WARWICH, R., DYSON, M. & BANNISTER, L.H.
(1989) Gray’s Anatomy, 37th edn, pp. 459 – 544. Churchill Liv-
ingstone, Edinburgh/London.
WOO, S.L.-Y., MOW, V.C. & LAI, W.M. (1988) Biomechanical
properties of articular cartilage. Chapter 4. In: Handbook of
Bioengineering (eds R. Skalak & S. Chien), pp. 1 – 44. McGraw-
Hill, New York.
Correspondence: Dr Eiji Tanaka, Department of Orthodontics,
Hiroshima University School of Dentistry, 1-2-3 Kasumi, Mi-
nami-ku, Hiroshima 734-8553, Japan. E-mail:
etanaka@ipc.hiroshima-u.ac.jp