Full text of DOI:10.1097/00005537-200209000-00025

The Laryngoscope
Lippincott Williams & Wilkins, Inc., Philadelphia
© 2002 The American Laryngological,
Rhinological and Otological Society, Inc.
Tympanosclerosis in the Rat Tympanic
Membrane: An Experimental Study
John D. Russell, FRCS (ORL); J. J. Giles, PhD
Objectives/Hypothesis: Tympanosclerosis is
a
media progressing to chronic otitis media with effusion.1
This disease continues to be a costly drain on the health
resources and is associated with delayed speech develop-
ment, temporary or permanent hearing loss, and the more
serious life-threatening disease, cholesteatoma. Tympano-
sclerosis is a frequent sequel of persistent otitis media
with effusion and/or its treatment. Tos and Poulsen2 re-
ported that 19% of 252 children reevaluated between 3
and 8 years after treatment for otitis media with effusion
had tympanosclerosis. Maw3 showed that tympanosclero-
sis developed in 37% to 39% of ears treated with one
ventilation tube compared with 47% to 49% of ears treated
with more than one tube. In the latter study, no evidence
of a resolution of the condition with time or demonstrable
effect on hearing ability was found even in cases with
severe tympanosclerosis.
The two main pathologic responses of the tympanic
membrane to prolonged otitis media with effusion appear
to be atrophy and/or tympanosclerosis. Tympanosclerosis
is a pathologic condition affecting the middle ear and
tympanic membrane. It is characterized by white plaques
of varying consistency depending on the amount of extra-
cellular calcium deposition.4 Histologically there is an in-
crease in collagen fibers as well as hyaline degeneration
within the lamina propria.5–8 The precise etiology of this
condition is unclear, but associated factors include inflam-
matory disease of the middle ear,9 ventilation tubes,10 and
raised partial pressure of oxygen within the middle ear
after ventilation tube insertion.11 Maw3 reported a rela-
tionship between tympanosclerosis after ventilation tube
insertion and intraepithelial hemorrhage at the time of
myringotomy, and the presence of the tube in the pars
tensa for greater than 6 months. Kuijpers et al.9 demon-
strated in an experimental animal model that tympano-
sclerosis occurred in the pars tensa of rats with sterile
prolonged otitis media with effusion. This suggested that
tympanosclerosis was triggered by a negative pressure
induced by eustachian tube obstruction or that some com-
ponent in the accumulated fluid precipitated it. Although
many studies on tympanosclerosis have been carried out
mainly using animal models,9,12 the underlying mecha-
nisms of this disease are still unknown. The aim, there-
fore, of this study was to examine the histologic sequelae
leading to tympanosclerosis in the pars tensa of the rat
pathologic condition affecting the middle ear and
tympanic membrane. It is a common condition in hu-
mans, most notably after grommet insertion. How-
ever, the pathogenesis of this disease is unclear. The
aim of this study was to investigate the development
and progression of tympanosclerosis in the pars tensa
of rat tympanic membranes at various time intervals
after inducing sterile middle ear effusions. Methods:
Fifty-six male-specific, pathogen-free CD Wistar rats
(100–170 g) underwent unilateral eustachian tube ob-
struction. Contralateral ears served as controls. Only
specimens from animals with sterile effusions were
included in this study. Light and electron microscopic
analysis was performed on the pars tensa of rats with
induced effusions ranging from 1 to 12 months. Re-
sults: Histologic changes consistent with tympano-
sclerosis were seen in the majority of animals with
effusions from 3 months and greater. The process
started in the submucosal connective tissue layer and
progressed to involve all connective tissue sublayers.
The extent of calcium deposition and fibrosis across
the membrane was related to the duration of otitis
media with effusion. Atrophy was not found in any
specimens. Conclusions: Tympanosclerosis is a pro-
gressive disorder, which appears to be the main re-
sponse of the rat pars tensa to prolonged sterile otitis
media with effusion. The extent of involvement of the
different sublayers of the pars tensa was closely re-
lated to the duration of otitis media with effusion. We
propose that this new rat model for prolonged otitis
media with sterile effusion is therefore a suitable an-
imal model for the study of tympanosclerosis in the
pars tensa. Key Words: Tympanosclerosis, otitis me-
dia with effusion (OME), pars tensa.
Laryngoscope, 112:1663–1666, 2002
INTRODUCTION
Otitis media remains a common problem in children
with approximately 5% to 10% of cases of acute otitis
From the Departments of Otolaryngology/Head & Neck Surgery
(J.D.R.) and Human Anatomy and Physiology (J.J.G.), University College,
Dublin, Ireland.
Editor’s Note: This Manuscript was accepted for publication March
11, 2002.
Send Correspondence to John D. Russell, FRCS, ORL, Consultant
Ear, Nose and Throat Surgeon, Suite 3, 69 Eccles Street, Dublin 7, Ireland.
E-mail: russellj@indigo.ie
Laryngoscope 112: September 2002
Russell and Giles: Tympanosclerosis in the Rat Tympanic Membrane
1663

tympanic membrane using a new experimental model for
persistent otitis media with effusion.13
One month after eustachian tube obstruction (N ϭ 7)
light microscopic examination showed discrete focal out-
growths along the inner mucosal layer of the membrane
(Fig. 2). Ultrastructural analysis showed that a thickened
submucosal layer had developed invested with fibroblasts,
blood vessels, and cuboidal cells containing large electron-
dense inclusions (Fig. 3). The outer epidermal and middle
fibrous layers appeared normal. In two of the specimens
from this group the tympanic membranes appeared
normal.
Three months after obstruction (N ϭ 8) focal thick-
enings, and in some cases a uniformly thickened submu-
cosal layer containing darkly stained granules, was seen
by light microscopy. Examination by transmission elec-
tron microscopy (TEM) showed the latter to be consistent
with calcium plaque deposition at various stages of devel-
opment. Fully formed, electron-dense, shell-like deposits
were noted primarily in the submucosal layer, whereas in
some specimens (N ϭ 2) the inner circular layer contained
less dense shells and the outer radiate layer showed the
presence of matrix vesicles (Figs. 4 and 5).
Specimens examined 6 months after eustachian tube
obstruction (N ϭ 6) showed similar changes as those seen
at 3 months. However, calcium plaque deposition was now
consistently seen in the thickened submucosal layer and
both sublayers of the lamina propria. Two specimens at
this stage appeared normal.
Nine months (N ϭ 7) to 1 year (N ϭ 6) after obstruc-
tion, the pars tensa appeared greatly thickened and fi-
brotic. The normal arrangement of the lamina propria into
outer radiate and inner circular layers was lost and re-
placed by haphazardly arranged collagen fibers. A definite
subepidermal fibrous layer had developed. Discrete cal-
cium plaques were found throughout the lamina propria
(Fig. 6). Concomitantly with the increase in calcification,
the mucosal cells showed decreased activity and were
often absent. No inflammatory cells were seen.
MATERIALS AND METHODS
Fifty-six male-specific, pathogen-free CD Wistar rats (100–
170 g) were used in this study. Persistent otitis media with
effusion was induced using the procedure of Russell and Giles.13
Briefly, the midportion of the eustachian tube was excised and
heated gutta percha was molded into the defect. Only one eusta-
chian tube in each rat was obstructed, allowing the contralateral
tympanic membrane and middle ear to serve as a control. All
animals underwent otoscopy and microscopic examination of the
ears 48 hours after the procedure. This examination was per-
formed once weekly for the first 4 weeks and then once monthly
until the rats were killed. Five groups consisting of 10 rats each
were killed at 1, 3, 6, 9, and 12 months after the procedure.
Standard microbiologic analysis for aerobic and anaerobic organ-
isms was performed on the effusion material. The presence or
absence of gutta percha was also noted. Of the 56 rats that
underwent the procedure, sterile effusions were recorded in 40
rats at the time they were killed. Only those rats with sterile
effusions were used in this study. The middle ear and tympanic
membranes were dissected out and fixed in 2.5% phosphate-
buffered glutaraldehyde for 24 hours. The specimens were then
decalcified for 24 hours in Calci-Clear (National Diagnostics,
U.K.) and processed routinely into araldite. Postfixation in os-
mium tetroxide was avoided to facilitate the correct orientation of
specimens within the resin blocks. Serial transverse sections
through the pars tensa from at least six specimens in each group
were examined by light and electron microscopy.
RESULTS
Light and electron microscopic examination of control
specimens (N ϭ 6) confirmed the findings of Lim.14 The
pars tensa consisted of a thin trilaminar membrane com-
posed of the following: 1) an outer epidermal layer con-
taining stratified squamous cells, 2) a middle fibrous layer
organized into outer radiate and inner circular sublayers,
and 3) an inner mucosal lining layer of flattened epithelial
cells (Fig. 1).
DISCUSSION
This study describes the histologic changes leading to
the development of tympanosclerosis in a new rat model
for chronic otitis media with effusion. The results support
the findings of Kuijpers et al.9 The sequelae leading to
tympanosclerosis started within 1 month of eustachian
tube obstruction in the submucosal connective tissue layer
Fig. 1. Electron micrograph of a cross-section through the pars
tensa of a control tympanic membrane. Note the outer epidermal
layer (large arrow), the radiate (R) and circular (C) fibrous layers of
the lamina propria, and the inner mucosal lining layer (small arrow)
(magnification ϫ15,000).
Fig. 2. Light micrograph of pars tensa from specimen 1 month after
eustachian tube obstruction showing focal thickening (arrow) on the
inner mucosal side of the membrane (magnification ϫ80).
Laryngoscope 112: September 2002
1664
Russell and Giles: Tympanosclerosis in the Rat Tympanic Membrane

Fig. 4. Electron micrograph of pars tensa from specimen 3 months
after eustachian tube obstruction showing the presence of calcium
plaques throughout the lamina propria. Note the outer epidermal
layer (arrow) (magnification ϫ7000).
Fig. 3. Electron micrograph showing thickening of the submucosal
connective tissue layer from a specimen 1 month after eustachian
tube obstruction. Note cells containing large electron-dense inclu-
sions (arrows). The circular fibrous layer is seen in the top right-hand
corner (magnification ϫ8000).
that the extent of the histologic changes described varied
both within and between groups.
of the lamina propria and spread to involve the circular
and radiate layers by 1 year after obstruction. The histo-
logic features of fiber formation, degeneration, and calci-
fied plaque formation were similar to those described in
the human tympanic membrane.8,15,16 The extent of cal-
cium plaque deposition across the thickness of the pars
tensa was closely related to the duration of the otitis
media with effusion. Thus, in specimens examined after 1
month, no evidence of calcium deposition was seen. In 3- to
6-month specimens, calcium deposition was mainly con-
fined to the submucosal and inner circular fiber layers. In
9-month to 1-year specimens, the structure of the pars
tensa became profoundly changed. The membrane was
greatly thickened and the normal arrangement of the
lamina propria into inner circular and outer radiate layers
was lost. Evidence of calcium deposition was consistently
seen throughout the entire thickness of the lamina propria
and adjacent submucosal layers. It is important to note
Wielinga and Kuijpers17 showed that early re-
aeration resulted in complete disappearance of the cal-
cium deposits but that the fibrous tissue remained. In this
study the normal ultrastructure and orientation of the
fibers in the lamina propria at 9 months to 1 year was
significantly disrupted and distorted. The repair process
did not reproduce the radial and circular orientation of the
fibers, but consisted of the laying down of haphazardly
arranged mature collagen fibers. It is interesting that the
repair mechanism cannot regenerate the same orientation
of radiate and circular fibrous layers. Thus, it seems
likely, as proposed by McMinn,18 that the fibroblasts,
whether native or bloodborne immigrants, are incapable
of reconstituting the exact fiber orientation in the process
of repair. This implies that the precise and well-structured
lamina propria of the pars tensa may be decided during
embryonic life.
Fig. 6. Electron micrograph of the pars tensa from a specimen 1
year after eustachian tube obstruction showing the presence of
calcium plaques throughout the lamina propria and submucosal
layers. Note fibrosis and loss of normal fiber arrangement in the
circular fibrous layer. Inner mucosal layer (arrowhead), outer epider-
mal layer (arrow) (magnification ϫ1600).
Fig. 5. Electron micrograph showing the different stages of calcium
plaque development. Note the presence of fully formed, electron-
dense deposits in the submucosal layer, less dense shells in the
circular fibrous layer (arrow), and matrix vesicles in the radiate layer
(arrowhead) (magnification ϫ30,000).
Laryngoscope 112: September 2002
Russell and Giles: Tympanosclerosis in the Rat Tympanic Membrane
1665

In the present study no cases of atrophy, perforation,
or cholesteatoma were seen in the rat pars tensa in pro-
longed otitis media with effusion. This finding is in agree-
ment with previous reports by Stenfors et al.19 and Kui-
jpers and Vennix.20 In contrast, however, atrophy and
retraction were seen in the pars flaccida as previously
described.13 Several possibilities exist as to why the pars
tensa of the rat does not develop atrophy with prolonged
otitis media with effusion. There are differences between
humans and rats in the ultrastructure and composition of
the fibers in the lamina propria.21 This may predispose to
atrophy and/or retraction pocket formation in the human,
particularly in the posterior part of the pars tensa where
the circular fibrous layer is more delicate and less obvi-
ous.22 The rat pars tensa being relatively smaller may be
less susceptible to negative pressure and strain than the
human tensa. In addition, the rat pars tensa appears to
depend on the process of diffusion for its nutrition,23 un-
like the human pars tensa, which has a rich blood supply.
It is conceivable, therefore, that any disruption of the
blood supply could lead to atrophy. The most susceptible
regions of the tympanic membrane to atrophy, therefore,
would be those that are most vascularized, i.e., the pars
flaccida and the posterosuperior part of the pars tensa.22
Alternatively, it may be that longer periods of persistent
otitis media with effusion in the rat are needed for atrophy
and/or atelectasis to develop.
The human pars tensa appears to respond to pro-
longed otitis media with the development of tympanoscle-
rosis, atrophy, or both. Atrophy can then progress to per-
foration, atelectasis, and/or cholesteatoma. The reason
why one human tympanic membrane will respond with
the development of tympanosclerosis and another with the
development of atrophy, and yet another with both tym-
panosclerosis and atrophy, is still unclear. The fact that
atrophy was not seen in the rat pars tensa suggests that
this animal model may not be suitable to study posterior
retraction pockets or mesotympanic cholesteatoma. It
would appear, however, to be an excellent model for the
study of tympanosclerosis in the pars tensa and retraction
and atrophy in the pars flaccida.
BIBLIOGRAPHY
1. Teele DW, Klein JO, Rosner BA. Epidemiology of otitis media
in children. Ann Otol Rhinol Laryngol 1980;89(Suppl 68):
5–6.
2. Tos M, Poulsen G. Changes of the pars tensa in secretory
otitis. Otorhinolaryngology 1979;41:313–328.
3. Maw RA. Glue Ear in Childhood. London: Mac Keith Press,
1995:103–117.
4. Gibb AG. Tympanosclerosis. Proc R Soc Lond B Biol Sci
1976;69:155–162.
5. Igarashi M, Konishi S, Alford BR, Guilford FR. The pathology
of tympanosclerosis. Laryngoscope 1970;80:233–243.
6. Sheehy JL, House WF. Tympanosclerosis. J Laryngol Otol
1962;76:151–157.
7. Zollner F. Tympanosclerosis. J Laryngol Otol 1956;70:77–85.
8. Chang IW. Tympanosclerosis: electron microscopic study.
Acta Otolaryngol 1969;68:62–72.
9. Kuijpers W, Wielinga E, Tonnaer E, Jap HK. Experimentally
induced tympanosclerosis. In: Lim DJ, Bluestone CD,
Klein JO, Nelson JD, eds. Recent Advances in Otitis Media.
Toronto: Decker Periodicals, 1988:468–471.
10. Lesser THJ, Williams KR, Skinner DW. Tympanosclerosis,
grommets and shear stresses. Clin Otolaryngol 1988;13:
375–380.
11. Helstrom S, Goldie P, Magnuson K, Fau C. Oxygen level
influences the middle ear tissue reaction in ears with per-
forated tympanic membranes. In: Lim DJ, ed. Recent Ad-
vances in Otitis Media. Philadelphia: B.C. Decker, 1993:
334–335.
12. Wielinga EWJ, Kuijpers W, Tonnaer E, Jap PHK. An exper-
imental model for tympanosclerosis: a preliminary report.
Acta Otolaryngol (Stockh) 1988;105:537–542.
13. Russell J, Giles SJ. Persistent otitis media with effusion: a
new experimental model. Laryngoscope 1998;108:
1181–1184.
14. Lim DJ. Tympanic membrane. Acta Otolaryngol 1968;66:
181–198.
15. Friedmann I, Hodges GM, Graham M. Tympanosclerosis: an
electron microscopic study of matrix vesicles. Ann Otol
Rhinol Laryngol 1980;89(Suppl 68):241–245.
16. Moller P. Tympanosclerosis of the ear-drum in secretory oti-
tis media. Acta Otolaryngol (Stockh) 1984;Suppl 414:
171–177.
17. Wielinga EWJ, Kuijpers W. The influence of re-aeration of
the middle ear on tympanosclerotic lesions in otitis media
with effusion. In: Tos M, Thomsen J, Balle V, eds. Otitis
Media Today. The Hague: Kugler Publications, 1997:
629–630.
18. McMinn RMH. Electron microscopic observations on the re-
pair of perforated tympanic membranes in the guinea-pig.
J Anat 1975;120:207–217.
19. Stenfors LE, Carlsoo B, Winbald B. Structure and healing
capacity of the rat tympanic membrane after eustachian
tube occlusion. Acta Otolaryngol 1981;91:75–84.
20. Kuijpers W, Vennix PPCA. Origin and fate of spontaneous
tympanic membrane perforations. In: Lim DJ, Bluestone
CD, Klein JO, Nelson JD, Ogra PL, eds. Recent Advances in
Otitis Media. Toronto: Decker Periodicals, 1993:453–455.
21. McMinn RMH, Taylor AM. Ultrastructure of fibrils in devel-
oping human and guinea-pig tympanic membrane. J Anat
1978;125:107–115.
22. Ars B. Tympanic membrane retraction pocket. Acta Otorhi-
nolaryngol Belg 1995;49:163–171.
23. Albin N, Hellstrom S, Salen B, Stenfors LE, Wirell S. The
vascular supply of the rat tympanic membrane. Anat Rec
1985;212:17–22.
CONCLUSION
Tympanosclerosis in the rat pars tensa is a progres-
sive disorder. It begins in the inner submucosal connective
tissue layer and spreads over time to involve the lamina
propria. The extent of calcium deposition and fibrosis
across the sublayers of the rat pars tensa appears closely
related to the duration of the effusion. This rat model for
prolonged otitis media with effusion is therefore a suitable
animal model for the study of tympanosclerosis in the pars
tensa.
Laryngoscope 112: September 2002
1666
Russell and Giles: Tympanosclerosis in the Rat Tympanic Membrane