Electron microscope study of bone lesion in cholesteatoma otitis

Presence of osteoclasts in middle ear cholesteatoma: a study of undecalcified bone sections

CONCLUSIONS

Osteoclasts are unlikely to be involved in bone resorption in middle ear cholesteatoma.

OBJECTIVE

The authors searched for osteoclasts in undecalcified bone sections in patients with middle ear cholesteatoma to determine whether and to what extent these cells are involved in this disease.

METHODS

Twelve patients, eight men and four women, aged 30-87 years, who underwent tympanomastoidectomy were enrolled. Six patients had primary acquired middle ear cholesteatoma (cholesteatoma group) and the other six patients had other otologic diseases including otosclerosis, non-cholesteatomatous chronic otitis media, adhesive otitis media, perilymphatic fistula and ossicular malformation (control group). The scutum bone was collected during surgery, fixed with ethanol, stained with Villanueva bone stain, and embedded in methyl methacrylate. Five-micrometer-thick sections were prepared and examined under a polarizing microscope. Images were analyzed using a semiautomatic graphics system.

RESULTS

No osteoclasts were seen in any of the samples in either group. To avoid the risk of under-estimating the presence of osteoclasts, the number of osteoclasts was considered to be <1 in each sample, and the osteoclast density was calculated. The osteoclast densities in both the cholesteatoma and control groups were significantly lower than the sex- and age-matched standard value of the normal iliac cortical bone (p = .028).

Osteoclasts are not activated in middle ear cholesteatoma

It is unclear whether osteoclasts are present and activated in cholesteatomas. We explored the expression of messenger RNA (mRNA) for osteoclast biomarkers and regulating factors in middle ear cholesteatomas to elucidate the level of osteoclast activity in this disease. Bone powder was collected from 14 patients with cholesteatomatous and noncholesteatomatous chronic otitis media during tympanomastoidectomy, separately from cortical bone of the mastoid (clean bone powder), from bone neighboring cholesteatoma (cholesteatomatous bone powder), and from bone of the air cells and antrum of noncholesteatomatous chronic otitis media patients (noncholesteatomatous bone powder). The samples collected were soaked in TRIzol reagent, and total RNA was extracted and purified by the acid guanidinium thiocyanate-phenol-chloroform method, followed by the use of magnetic bead technology. The sample was then subjected to quantitative reverse transcription polymerase chain reaction for receptor activator of nuclear factor κB (RANK), tartrate-resistant acid phosphatase (TRAP), cathepsin K (CTSK), osteoclast-associated receptor (OSCAR), calcitonin receptor (CALCR), matrix metalloproteinase 9 (MMP9), receptor activator of nuclear factor κB ligand (RANKL), and osteoprotegerin (OPG). There was no significant difference in the expression of TRAP, CTSK, OSCAR, CALCR, MMP9, or OPG among the clean, cholesteatomatous, and noncholesteatomatous bone powder. On the other hand, the expression of RANK and RANKL was significantly lower in the cholesteatomatous bone powder than in the noncholesteatomatous bone powder (P = 0.003 and P = 0.028, respectively). The RANKL mRNA/OPG mRNA ratio did not differ among the three samples. These results indicate that osteoclasts are unlikely to be activated in cholesteatomas. Bone resorption mechanisms not mediated by osteoclasts may need to be reappraised in cholesteatoma research in future studies.

Increased permeability of the epithelium of middle ear cholesteatoma

OBJECTIVE

We investigated the electrical impedance of and the expressions of tight junction molecules in the cholesteatoma epithelium to provide supporting evidence for the acid lysis theory of bone resorption in middle ear cholesteatoma.

METHODS

Study subjects were patients with primary acquired middle ear cholesteatoma and those with non-cholesteatomatous chronic otitis media who underwent tympanomastoidectomy. The electrical impedance of the cholesteatoma epithelium was measured during tympanomastoidectomy by loading alternating currents of 320 Hz and 30.7 kHz. The expressions of tricellulin (MARVELD2), claudin-1 (CLDN1) and claudin-3 (CLDN3) were examined by fluorescence immunohistochemistry and quantitative reverse transcription-polymerase chain reaction.

RESULTS

The electrical impedance of the cholesteatoma epithelium was significantly lower than that of the post-auricular skin and external auditory canal skin at both 320 Hz and 30.7 kHz. Immunoreactivity for MARVELD2, CLDN1 and CLDN3 was localised mainly in the granular layer, and to lesser degree, in the horny and spinous layers in both the cholesteatoma tissue and post-auricular skin. Fluorescence intensity was moderate for MARVELD2, weak for CLDN1 and strong for CLDN3. The expressions of MARVELD2, CLDN1 and CLDN3 mRNA were significantly lower in the cholesteatoma tissue than in the post-auricular skin.

CONCLUSIONS

These results indicate the increased permeability of the cholesteatoma epithelium and suggest that this change is, at least partially, dependent on the decrease in the expressions of the tight junction molecules. This evidence supports the acid lysis hypothesis of bone resorption in cholesteatoma.

Bone resorption in human cholesteatoma: morphological study with scanning electron microscopy

The detailed mechanism of bone resorption in cholesteatoma was investigated by means of eroded ossicles obtained during middle ear surgery for cholesteatoma. In the light microscopic study, multinucleate osteoclasts with ruffled borders were found in contact with the eroded bone, which appeared to be osteoclastic lacunae. Scanning electron microscopy revealed the lacunae to be many absorption bays, 30 to 100 microns in diameter, clustered on the surface of eroded areas. Numerous cone-shaped stubs of digested collagen fiber bundles, consisting of scores of collagen fibrils and degenerated extracellular organic substances, were visible at higher magnification on the bottom of the absorption bay. The pattern of fusion and twining of the disarranged collagen fibrils at the top of the partly digested fiber bundles was clearly rendered by the alkali-water maceration method for scanning electron microscopy. We infer from the morphological evidence that osteoclastic resorption may be one of the major mechanisms of bone destruction in cholesteatoma, and that demineralization and degeneration of extracellular organic substances precede disruption of collagen fibrils at the front of bone resorption.

Bone resorption in chronic otitis media. The role of mast cells

Twenty-two surgical specimens of eroded middle ear ossicles were removed from patients with chronic otitis media, with and without cholesteatoma. By using specific mast cell stains, increased numbers of mast cells were found in connective or granulation tissue adjacent to eroded surface of the bone. Mast cells possess the biological machinery necessary for enhancing bone resorption, and the population density of mast cells is increased in a variety of disorders that are associated with bone resorption. It is hypothesized that mast cells contribute to bone resorption in chronic otitis media, and the possible mechanisms by which mast cells exert their action are discussed.

Cellular and subcellular events of bone resorption in human and experimental cholesteatoma: the role of osteoclasts

An ultrastructural study of the interface between aural cholesteatoma and adjacent bone was performed on specimens obtained from human and experimental gerbilline cholesteatoma. When an enlarging cholesteatoma contacted bone, a large number of monocytes and macrophages accumulated in the area of contact. The intervening middle ear mucosa degenerated and bone erosion occurred. Anatomic evidence is provided which indicates that bone erosion in both human and experimental cholesteatoma occurred as a result of the action of multinucleated osteoclasts. These osteoclasts had the ultrastructural appearance of normal osteoclasts with a well developed ruffled border and large numbers of mitochondria. These cells stained readily with acid phosphatase stains. Although many other mononuclear cells were seen in the vicinity of the eroded bone, only the multi-nucleated osteoclasts were associated with disappearance of the bone surface. Osteoclasts were commonly found in experimental cholesteatoma, but were infrequent in human cholesteatoma biopsies, presumable because patients undergoing cholesteatoma surgery ara often treated with topical installations of corticosteroids which inhibit osteoclasts.