Int. J. Oral Maxillofac: Surg.

1999; 28:464-468 Copyright 9 Munksgaard 1999

Printed in Denmark. All rights reserved
internationalJournal of
Oral&
M ofacial Surgery
ISSN 0901-5027

A n s g a r Gross ~, Axel B u m a n n 2,
Elastic fibers in the human Bodo Hoffmeister 3
1Department of Orthodontics, Georg-August-
University, Goettingen, Germany;
temporo-mandibular joint disc 2Department of Orthodontics, Harvard School
of Dental Medicine, Boston, USA;
3Department of Maxillofacial Plastic Surgery,
Free University of Berlin, Benjamin Franklin
Medical Center, Berlin, Germany

A. Gross, A. Bumann, B. Hoffmeister." Elastic fibers in the human temporo-

mandibular joint disc. Int. J. Oral Maxillofac. Surg. 1999," 28." 464-468.
9 Munksgaard, 1999

Abstract. The elastic fiber content of 12 articular discs, removed from 12 patients
with complete anterior or medio-anterior disc displacement, was examined. Eight
to 12 sagittal sections (3/ma each) were acquired for each disc and stained with
modified resorcin-fuchsin for visualization of elastic fibers. The program "CUE-2
Image Analyzer Morphometry" digitized all sections with 800• magnification,
calculated the elastic fiber content and reconstructed it 3-dhnensionally. The
calculated volume density of the entire disc was a mean xtot=0.339• of
the total tissue. The highest fiber content of 2-4% was found in the posterior
area at the transition to the bilaminar zone. 61% of the elastic fibers were located
in the bilaminar zone, 10% in the posterior, 2% in the intermediary and 26% in
the anterior band. There were substantially more fibers in the cranial part of the
posterior region and at the medial edge as compared to the mean fiber
distribution o f the posterior band and the bilaminar zone. A third of these fibers
were found in the inferior layer, particularly in the lateral part. Exactly the
opposite fiber distribution was seen in the anterior band. The fibers frequently Key words: temporo-mandibular joint; joint
disc; elastic fibers.
appeared in the caudal layer, especially in the lateral periphery, but there were
fewer fibers in the medial cranial layer. Accepted for publication 18 May 1999

There are very few studies on the occur-
rence and distribution of elastic fibers
in the human temporo-mandibular
joint and on the function of these fibers.
The distribution of elastic fibers in the
disc has been described as irregular6.
There are numerous fibers in the an-
terior and posterior disc area, some
parallel to the collagenous fibers, but
none in the intermediate zone s . On the
other hand, many thick elastic fibers are
found in the bilaminar zone in the su-
perior layers'6'~2. There are considerably
fewer of these fibers in the inferior layer
which consists of compact collagen fib-
er bundles of limited elasticity ~~ Fine,
randomly distributed fibers are also
found in the posterior band except in
the bilaminar zone 6. Elastic fibers occur
only sporadically in the intermediary
zone 6. The number of fibers clearly in-
creases in the anterior band, especially
in the inferior part, in relation to the
insertion at the condyle 1~ Other
authors consider both the posterior and
intermediary band to be free of elastic
fibers 5.
Information concerning the distri-
bution of the elastic fibers in the medi-
olateral and cranialcaudal attachment
is scarce in the literature. Increased
numbers of elastic fibers have been de-
scribed in the medial anterior band,
into which the extension of the lateral
pterygoid muscle radiates 1~ but few are
present in the lateral anterior band.
TAKISAWAet al. 13 observed an increas-
ing prevalence of elastic fibers in the
medial region, but a lower prevalence in
the lateral area. They made no mention,
however, of the distribution in the
bands. No parallel orientation with the
collagen fibers could be detected.
The functional importance of elastic
fibers lies in the conversion of expan-
sion energy into potential and kinetic
energy, enabling disc mobility4.
The aim of this study was to deter-
mine the elastic fiber content of com-
pletely displaced articular discs and to
quantitatively analyze these fibers in
different disc regions. The obtained
values were then correlated to selected
patient parameters.
Material and methods
The material consisted of 12 articular discs
of human temporo-mandibular joints which
were removed from patients in the Depart-

m-~terior
/ ~ ~ricntation.
lateral
posterior
Fig. 1. Specimen block with disc (cranial view), showing sections to be cut with the dense-
section microtome, with two bilateral notches each.
0,8
0,7
0.6
0,5
0,4
0,3
0,2
0,1
0
1 2 3 4 5 6
Fig. 2. Elastic fiber content as a percentage of total tissue (corrected fiber content Xk) for each
disc, with the mean of all 12 discs shown as a horizontal line.
ment of Maxillofacial Surgery of the Chris-
tian Albrechts University in Kiel between
1988 1993. All patients were female, between
18 and 48 years old, with a mean of 32.3
years. All patients had complete anterior or
medio-anterior disc displacement.
Discectomy was considered to be indicated
in all patients. The articular discs were ex-
cised with dissecting scissors and put in 10%
formalin solution for histological processing,
subsequently dehydrated and embedded in
rnethyl methacrylate. With a dense-section
microtome (R. JUNG, Nussloch, Germany), 3
/~m-thick sections were cut sagittally in 2 mm
intervals. For spatial orientation, notches
were made in the embedding block which
were visible in the section and transferred to
a slide with a fine glass cutter (Fig. 1). Nine
to I 1 sagittal sections per specimen were ob-
tained from a transverse disc extension of 18
to 22 mm. All sections were stained with re-
sorcin-fuchsin solution according to WEIG-
ERT6, for visualization of the elastic fibers.
The staining times were modified as follows
//notches (anterior)
I ..orientation.
. ~ %'~notches (posterior)
/ I
/ The value x measured for the fiber content of
sections 1-6
7 8 9 10 I1 disc 12
for contrast enhancement in the morpho-
metric evaluation: resorcin-fuchsin staining
according to "V~IGERTat 56~ for 60 minutes,
subsequent rinsing in distilled water for 1
minute, differentiation in 80% alcohol for 5
minutes and then covered with Eukitt.
A computer program, "CUE-2, Image
Analyzer Morphometry" version 3.11 from
the Galai Company, Israel, was used for digi-
talization of histological sections and their
morphological assessment. This system in-
cludes a camera (black/white Olympus CCD
XC-57) mounted to a microscope (Olympus
BH-2) and a second monitor for visualizing
the digitized image, A grid subdivided into
1 • mm fields was used to evaluate the fiber
content. A high magnification (800• was re-
quired for morphometry to obtain an ad-
equately large image of the fine elastic fibers
(ca. 1-5/~m diameter) on the monitor. This re-
sulted in a 0.2• mm measuring field per
morphometric analysis, which corresponds to
a pixel size of 0.4• ,urn. The analysis field
was then digitized with the black-and-white
Elastic fibers in human T M J disc 465
camera fixed to the microscope, in which each
matrix point was assigned a gray value of 0 to
255. Using a given uniform gray scale thresh-
old, all elastic fibers stained black were
marked according to the matrix points. After
interactive manual corrections, the program
calculated the percentage of elastic fibers in
the total image (analysis image). The mean
fiber content for each scanning field was deter-
mined by evaluating several image analyses of
that field with no overlapping.
Error correction
an image analysis corresponds to the surface
density AA(S/B)which, according to DELESSE'S
stereological axiom, can only be equated
with the volume density VV(S/B)we require if
sections are infinitely thin. Correction of data
on the relative fiber distribution in an object
with the same section thickness is not re-
quired. Thus, uniform sections of 3/~m were
selected for this study. Working with data on
the absolute fiber content, (the volume den-
sity Vv(sm)), the correction factor K(vv} must
be calculated for long tubular opaque struc-
tures according to the mean 3 #m diameter of
the elastic fibers. This results in a correction
factor K(vv) of 0.5.
Results
T h e m e a n v o l u m e density Vv(s/m, cor-
r e s p o n d i n g to the m e a n elastic fiber
c o n t e n t as a percentage o f the total
tissue o f the entire disc, varied between
x = 0 . 0 8 4 % (disc No. 6) a n d ?~=0.818%
(disc No. 10), whereas higher values
were f o u n d in the individual bands. T h e
overall m e a n with its s t a n d a r d error for
12 discs was sx=0.339 +0.060% (Fig.
2). T h e m a x i m u m fiber c o n t e n t o f indi-
vidual discs in circumscribed areas
varied between x = 1 . 9 % (disc No. 6)
a n d x = 4 . 2 % (discs No. 10 a n d No. 5).
These h i g h values for volume density
were only f o u n d in the p o s t e r i o r p a r t s
o f the disc a n d in the b i l a m i n a r zone.
Fiber distribution in the bands
(anteroposterior alignment)
I n d e t e r m i n i n g the fiber d i s t r i b u t i o n in
the four b a n d s (anterior, intermediary,
p o s t e r i o r b a n d a n d b i l a m i n a r zone) of
each disc, the percentage of the fiber
c o n t e n t o f o n e b a n d was calculated
f r o m the s u m o f all four fiber c o n t e n t s
(Table 1). Elastic fiber d i s t r i b u t i o n cor-
r e s p o n d e d to the four histologically dis-
tinct areas. T h e m a j o r i t y of elastic
fibers were f o u n d in the b i l a m i n a r zone
a n d a n t e r i o r region a n d the least in the
center o f the i n t e r m e d i a r y b a n d . There
466 G r o s s et al.

Table 1. Volume density of the elastic fibers in the disc bands and in the entire disc, mean of ocaudal fiber distribution, disc thick-
n=12 ness (Table 2).
Region Mean value x Standard error sx In evaluating the distribution of ran-
dom variables, the quotient was calcu-
Anterior band 26.20% 1.94%
Intermediary band 2.34% 0.86% lated from the empirical median and
Posterior band 10.22% 1.16% mean and checked for approximately
Bilaminar zone 61.25% 2.24% normal distribution. The majority of
the random sample distributions could
not be considered normal on the basis
of these calculated values. Random
Table 2. Characteristics of all 12 patients with complete anterior or medio-anterior disc dis- samples were tested for dependence
placement (mean value x, median ~, standard error sx, range R, minimum value Xmm,maxi- using the distribution-independent
mum value Xmax and number of patients n) Spearman's rank correlation coefficient
r s,
X ~ sx R Xmirl Xma x VI
The hypothesis for the two-sided test
Disc age (years) 32.26 30.60 10.38 29.30 18.20 47.50 12 was as follows:
Duration of symptoms 22.67 16.00 20.31 67.00 5.00 72.00 12
(months) HO: ~)S=0
Fiber content Xk,g~s(%) 0.339 0 . 3 4 8 0 . 2 0 8 0 . 7 3 4 0 . 0 8 4 0.818 12 HA: ~)S~a0.
Mediolateral fiber content 57.77 5 4 . 4 4 11.59 4 0 . 5 7 34.81 75.38 12
Fmed/lat ( % ) The hypothesis was tested at a signifi-
Craniocaudial fiber content 59.30 56.62 6.55 21.31 52.17 73.48 12 cance level of c~=0.05. The critical
F o ~ (%) threshold {r~*l was 0.587 for the two-
Disc thickness (mm) 2.21 2.25 0.34 1.23 1.52 2.75 12 sided hypothesis.
The calculated rank correlation coef-
ficients (Table 3) were statistically sig-
nificant (c~=0.05). There was a positive
was a slightly higher density in the pos- ever, the elastic fibers were predominant correlation between mediolateral fiber
terior band. in the caudal layer as shown in Fig. 4, distribution and disc age and a negative
one between fiber content and symptom
duration, as well as with craniocaudal
Fiber distribution in mediolateral Testing the independence of different
fiber distribution.
alignment characteristics

In addition to the assessment of elastic The following characteristics of all 12

Discussion
fiber distribution in the individual patients were more closely evaluated for
bands, fiber distribution was also deter- further analysis: disc age, dental status, The results presented on fiber content
mined in the sagittal sections. When the duration of symptoms, fiber content, and distribution of the elastic fibers in
total fiber content was considered in mediolateral fiber distribution, crani- the human temporo-mandibular joint
each of the 8 sections, there was a clear
increase in elastic fibers over the entire
width of the disc from lateral to medial.
The medial periphery with Fs=16.9%
contained almost one third more fibers
than the entire disc on average and ap-
proximately twice as many as the lateral
periphery. W31-32
a30-31
Fig. 3 shows the fiber distribution as N29,30
a surface relief, for each of the four 1128-29
1127-28
bands separately. Medial shifting of the D26*27
fiber content was observed, especially in 12125-26
the posterior band as well as in the bi- 11124+25
1123-24
laminar zone and intermediary band. rilam, zone
n22+23
The opposite distribution was found 112i-22
Q20-21
laterally, with the highest fiber content N19+20 '
{~nd
in the anterior band. C118-19
017-18 i
116+17
its+16 I
Fiber distribution in craniocaudal ad
alignment

A comparison was made between fiber

distribution in the cranial and caudal
parts of the disc. There was a higher fiber lateral
content in the cranial part o f the distal Fig. 3. Medial lateral fiber distribution as a surface relief. The high relief illustrates the distri-
disc region. In the anterior band, how- bution for each band, 100% corresponding to the total fiber content per band.

F ~6a [%1
...... ................. r................................... ,.............................. ~............. [
60 ...................................... i ...........................
sn
~ X
-"
40 ......................
30 ...............................................
2o
ant. band
Fig. 4. Comparison of the cranial and caudal fiber contents in the four bands and the entire
disc as a percentage of the total fiber content per band, n = 12.
Table 3. Spearman's rank correlation coefficientrs for two random samples
Random sample A
Mediolateral fiber distribution
Fiber content
Fiber content
disc must be regarded critically due to
the relatively small number of patients.
However, the selected patient material
in this study, in comparison with ran-
domized selected joints post mortem, is
the only way of obtaining a sufficient
number of discs with the same verified
diagnosis. It is not possible to obtain
"normal" control discs.
Fiber content
In a study of the elastic fiber distri-
bution in the mandibular capsule in
medi;
Fig. 5. Diagram of elastic fiber distribution
in the human articular disc in the coronal
view. Darker grey values represent a higher
fiber volume density, mpl: lateral pterygoid
muscle; ae: ventral extension of the anterior
band; an: anterior band; in: intermediary
band; po: posterior band; bl: bilaminar zone.
Elastic fibers in human T M J disc 467
and bilaminar zone. The fiber-poor in-
caudalCranial termediate band can be added to the
posterior region on the basis of its fiber
distribution. There is a preponderance
;.................................. of lateral, especially caudal, fibers in the
anterior band, whereas fewer fibers are
~ ................................. F found in the insertion of the medially
radiating lateral pterygoid muscle (Fig.
5). Other authors have also described
~" "~....... i ;,. ~-~- ~_-2..~ increased amounts of caudal fibers, but
these are centrally and medially situ-
', ...................... i ....................... : ........................................... : ated l~ The fibers decrease in the inter-
mediary band by a factor of 9 and are
found predominantly craniomedial.
E i ~ The fibers in the posterior band, in-
interm, band post. band bilam, zone total creasing by a factor of 3, are mainly
located in the medial region, which was
confirmation of the results from other
studies 1~ The fibers are primarily
aligned in the cranial layer. With a
mean fiber content of Xk,bl=0.68%,
there is a higher percentage of elastic
Random sample B r~ fibers in the bilaminar zone (by a factor
Disc age 0.641 of 6) than in the posterior band. The
Symptom duration -0.636 fiber content is lower laterally.
Craniocaudal fiber distribution -0.716 The mediolateral fiber distribution,
favoring the medial area, is even more
pronounced than in the posterior band;
rabbits, the volume density of elastic there are approximately twice as many
fibers in the posterior part of the disc fibers here than at the lateral edge (Fig.
was calculated 11. Correcting the section 5). If one compares the superior layer
thickness using a fiber diameter of 3/~m with the caudal layer of the band, then
led to elastic fiber volume densities of clearly more elastic fibers were found
1.5 to 5.4% of total tissue. In this study, cranially. However, with a ratio of 60%
the volume densities of individual grid cranial to 40% caudal this was a smaller
fields in the posterior disc region were difference than described in the litera-
of the same quantity. However, the ture. The lowest percentage of fibers
mandible of rabbits and humans are was found in the lateral inferior area
only comparable to a limited degree 11, and the highest in the medial superior
stratum. This abundance of fibers in the
since an anterior disc extension can
only be created experimentally and medial disc periphery of the cranial su-
does not exist physiologically. perior layer continues in the posterior
LUDEg & BOBST]~ compared the fiber and, to a lesser extent, in the intermedi-
content of human articular discs with ary band.
that in the skin. They found that there An elastic extension of the lateral
are far fewer fibers in the meniscus than pterygoid muscle passes into the medial
in the skin or in other elastic bands of disc periphery, but cranial muscle fibers
the human body. However, no infor- radiate into the medial region of the an-
mation is available on the absolute elas- terior band in only 60% of the
lateral tic fiber content either in the individual joints 1~ Most of the lateral pterygoid
bands or in the entire disc. muscle courses almost horizontally
from the lateral lamina of the pterygoid
process in a posterior, slightly lateral di-
Fiber distribution
rection to the fovea pterygoidea of the
The measured and calculated fiber dis- mandible, whereas these fibers only ac-
posterior tributions in the 12 articular discs are tively pull the disc in the anteromedian
somewhat different to and more de- direction with the mouth wide open 9.
With a moderately opened mouth, the
tailed than those reported in the litera-
ture. biconcave disc shape and its attachment
A distinction must be made between to the medial and lateral condylar poles
the fiber distribution in the anterior re- mainly lead to anterior shifting and the
gion and that in the posterior region, muscle helps to guide this movement to
which consists of the posterior band some degree 5. The elastic superior layer
468 Gross et aL
of the bilaminar zone acts as an antag-
onist, limiting anterior movement9,13.
According to the results of this study,
the medial disc edge is suited for this
resilience due to its high elasticity. The
region of the highest fiber content in the
cranial part of the medial disc quarter
extends from the posterior band to the
bilaminar zone and its temporal fix-
ation (Fig. 5). Disc repositioning when
closing the jaw is thought to be con-
trolled by the elasticity of the medial su-
perior layer and the lateral pterygoid
muscle 13. There is some doubt about
this assumption, because of the negli-
gible muscle fiber radiation l~ Already
during mouth opening, the disc is re-
turning to neutral position by the lat-
eral and medial disc fixation to the con-
dyle and its biconcave shape 4. A force
is also applied in the posterior direction
by the non-elastic inferior layer when
closing the mouth 4,:~ The lateral in-
ferior layer seems to be the primary
non-elastic component of the bilaminar
zone. When closing the mouth, pos-
terior distal traction of the inferior layer
works together with the posterior me-
dial resilience of the superior layer elas-
tic fibers to move the disc from its ante-
romedian position back to its neutral
position and to center it on the condyle.
According to KLeTT9, the hyperval-
ence of the lateral pterygoid muscle
shifts the disc beyond the physiological
levels in an ~interomedian direction.
This is thought to be caused by muscle
incoordination, e.g. due to oral para-
function or bruxism. Another cause for
a posterior superior shift of the condyle
may be a loss of molars (vertical dimen-
sion) or occlusion disorders. In both
cases, there are unphysiological traction
and compression forces in the posterior
disc region. Under certain conditions,
these forces can be compensated by
muscles or remodeling. An anteromed-
ian shift with disc degeneration only oc-
curs if these forces increase or persist 14.
Macroscopic examination shows sur-
face roughening and fibrillation as well
as subsequent thinning and perforation
of the disc in the lateral regions. These
observations have been confirmed in
this study macroscopically, as well as by
thickness measurements with lateral
thinning in the posterior region.
Microscopic examination shows an
increase in perivascular fatty tissue
(especially in the bilaminar zone), mu-
coid degeneration with liquefaction of
matrix, demasking and fibrillation of
col!agen fibers as well as fluid accumu-
lation in so-called pseudocysts in the bi-
laminar zone2,7.
A decrease in elastic fibers has been
observed in macroscopically degener-
ated discs l-s,7. The negative correlation
found in this study between absolute
fiber content and duration of symptoms
is in line with that finding. On the basis
of the calculated correlations between
the individual characteristics of the 12
specimens, which should be considered
with care because of the small number
of random samples, there was a signifi-
cant negative correlation between the
above-mentioned fiber content of the
entire disc and that of the duration of
symptoms reported by the patients. In
their study of 26 surgery-derived articu-
lar discs, ttALL et al. 7 describe a con-
siderable decrease in elastic fibers in the
bilaminar zone. These observations
were based on histological examination,
but not quantified.
The exact mechanism of elastin disin-
tegration is still unknown, but the pro-
teases, elastase and cathepsin G, have
been discussed as possible causes 1. They
are secreted by neutrophilic granulo-
cytes and synovial fibroblasts, stimu-
lated during degeneration and blocked
by proteinase inhibitors produced in car-
tilage. It is assumed that there is usually
a balance between elastase secretion and
inhibition and that protease production
increases during degeneration.
References
1. ALI A, SHARAWYM, O'DELLNL, AL-BE-
HERYG. Morphological alterations in the
elastic fibers of the rabbit craniomandib-
ular joint following experimentally in-
duced anterior disc displacement. Acta
Anat 1993: 147:159 67.
2. BLAUSTE1ND, SCAI'INOR. Remodeling of
the temporomandibular joint disc and
posterior attachment in disc displace-
ment specimen in relation to glycosamin-
oglycan content. Plast Reconstr Surg
1986: 78: 756-64.
3. CAMPBELLJ. Histopathology of the tem-
poromandibular disc. J Oral Maxillofac
Surg 1987: 45: M4.
4. CARPENTIER P, YUNG J, MARGUELLES-
BONNET R, MEUNISSIERM. Insertions of
the lateral pterygoid muscle: an anatomic
study of the human temporomandibular
joint. J Oral Maxillofac Surg 1988: 46:
477-82.
5. D~XOND. Structure and functional sig-
nificance of the intra-articular disc of the
human temporomandibular joint. Oral
Surg 1962: 15: 48-56.
6. GRIFFIN C J, SHARPE CJ. Distribution of
elastic tissue in the human temporo-
mandibular joint meniscus, especially in
respect to "compression" area. Aust Dent
J 1962: 7: 72-8.
7. HALLMB, BROWNR, BAUGHMANR. His-
tologic appearance of the bilaminar zone
in internal derangement of the temporo-
mandibular joint. Oral Surg 1984: 58:
375-81.
8. HOLMLUND AB, GYNTm~R GW, P~:N-
HOLT FR Disk derangement and in-
flammatory changes in the posterior disk
attachment of the temporomandibular
joint. A histologic study. Oral Surg 1992:
73: 9-12.
9. KLETTR. Zur Biomechanik des Kieferge-
lenkknackens. 2. Diskusverlagerung
durch muskulfireDiskoordination. Dtsch
Zahnfirztl Z 1986: 41: 308-12.
10. LVDER HU, BOBST P. Wall architecture
and disc attachment of the human tem-
poromandibular joint. Schweiz Mon-
atsschr Zahnmed 1991: 101:557 70.
1l. SAVALLEW~ V~EI.ISW, JAMESJ. Elastic and
collagenous fibers in the temporomand-
ibular joint capsule of the rabbit and
their functional relevance. Anat Rec
1990: 227: 159-66.
12. SCAPINO R. Histopathology associated
with malposition of the human temporo-
mandibular joint disc. Oral Surg 1983:
55:382 97.
13. TArdSAWAA, IHARAK, JINGUJIg. Fibro-
architectonics of human temporomand-
ibular joint. Okajimas Folia Anat Jpn
1982: 59: 141-66.
14. V~STESSON P, BRONSTEIN S, LIEDBERG J.
Internal derangement of the temporo-
mandibular joint: morphologic descrip-
tion with correlation to joint function.
Oral Surg 1985: 59: 323-31.
15. WONG G, WEINBERG S, SYMINGTON J.
Morphology of the developing articular
disc in the human temporomandibular
joint. J Oral Maxillofac Surg 1985: 43:
565-9.
Address:
Dr. Ansgar Gross
Kl6resgang 1
19053 Schwerin
Germany
Tel." +49 385 5507611
Fax: +49 385 5507613
e-mail." ansgar.gross@t-online.de