External auditory canal and middle ear cholesteatoma and osteonecrosis in bisphosphonate-treated osteoporosis patients: a Danish national register-based cohort study and literature review.

UNLABELLED: Long-term treatment with bisphosphonates against osteoporosis may cause atypical femur fractures and osteonecrosis of the jaw. Eight cases of bisphosphonate-associated osteonecrosis of the external auditory canal area are published. Based on Danish national registers, we report a time- and dose-dependent increased risk of cholesteatoma in osteoporosis patients treated with bisphosphonates. INTRODUCTION: In the recent years, there has been a focus on possible rare side effects of bisphosphonates (BPs). Eight cases of BP-associated osteonecrosis of the external auditory canal have been reported in the world literature. Our aim was to describe the incidence of external auditory canal and middle ear diseases in Danish patients exposed to BPs in the treatment of osteoporosis. METHODS: This register-based nationwide cohort study was conducted on the Danish population of approximately 5.6 million individuals. Patients who were prescribed BP for treatment of osteoporosis from 2003 to 2010 (n = 131,794) were included in the study and compared with the age- and gender-matched controls, unexposed to BP. RESULTS: The overall incidence of cholesteatoma in the ear was low. Only 350 events were seen in 527,176 cases and controls over 2,826,120.73 observation years. Totally, 119 events of cholesteatoma in the ear were recorded after initiation of BP therapy, 34 in the external auditory canal and 85 in the middle ear. Cholesteatoma in the external auditory canal was more frequent in the exposed than in the unexposed group (p < 0.0001). We found a significant dose-event relationship between incidence of cholesteatoma and dose of alendronate (p < 0.0001) and etidronate (p < 0.0001). Furthermore, we found an association between duration of treatment with alendronate and etidronate and risk of cholesteatoma in the external auditory ear canal (log rank, p = 0.002). No cases of bone destruction were observed during the 7-year observation period in either group. CONCLUSION: The use of oral BP is associated with an increased risk of cholesteatoma of the external auditory canal. The risk is small and associated with duration and dosage of BP.

Development of canal cholesteatoma in a patient with prenatal isotretinoin exposure.

PURPOSE: To describe the clinical and radiologic findings in a case of isotretinoin embryopathy-like syndrome and discuss management of hearing loss, congenital external auditory canal (EAC) stenosis, and EAC cholesteatoma. METHODS: Review of medical, audiological, and radiological records. RESULTS: An 8 year old female presented with bilateral moderate conductive hearing loss, bilateral microtia, left EAC stenosis, and right EAC atresia, secondary to prenatal isotretinoin exposure. Comorbidities included developmental delay, ventricular septal defect, hypotonia, and retinal maldevelopment. The left EAC was sharply upsloping with a 2mm-diameter meatus. Computed tomography (CT) scan of the temporal bone demonstrated normal middle and inner ears bilaterally; serial CT scans over 6 years demonstrated progressive development of left canal cholesteatoma. Implantation of a right BAHA system was performed, followed by left canalplasty and excision of cholesteatoma with facial nerve monitoring. An endaural incision was utilized to avoid compromising future microtia repair. Postoperative left-sided hearing improved to mild low-frequency conductive hearing loss rising to normal at 2000 Hz and above. CONCLUSIONS: Despite extensive precautions for its use, isotretinoin remains a cause of major birth defects, including sensorineural, conductive or mixed hearing loss. Congenital EAC stenosis is much less common than congenital atresia or acquired stenosis; optimal surgical approaches vary depending on hearing status and facial nerve anatomy. Close monitoring for development of canal cholesteatoma is necessary.

Cholesterol crystals as an etiological factor in non-resolving chronic inflammation: an experimental study in guinea pigs.

The presence of cholesterol crystals has been suggested to be a factor interfering with periapical healing after conventional endodontic treatment. This investigation addresses the role of cholesterol crystals in impairing healing by studying the tissue response to the crystals, which were implanted in animals. Pure cholesterol crystals, prepared to a mushy form, were placed in Teflon cages that were implanted subcutaneously in guinea pigs. The cage-contents were retrieved after 2, 4 and 32 wk of implantation and processed for light and electron microscopy. The cages revealed delicate connective tissue that grew in through perforations on the cage-wall. The crystals were densely surrounded by numerous macrophages and multinucleated giant cells, forming a well-circumscribed area of tissue reaction. The cells, however, were unable to eliminate the crystals during an observation period of 8 months. The congregation of macrophages and giant cells, known to be major sources of apical inflammatory and bone resorptive mediators, suggest that accumulation of cholesterol crystals can be a factor in the failure of certain apical periodontitis lesions to resolve after conventional root-filling therapy.

Effect of 5-fluorouracil in cholesteatoma development in an animal model.

This investigation was designed to evaluate the effects of the antimetabolite 5-fluorouracil (5-FU) on cholesteatoma formation in a chinchilla model. The animals received middle ear applications of propylene glycol according to a procedure previously shown to produce cholesteatomas in 60% to 70% of animals. A 5% solution of 5-FU was then applied to the lateral surface of the tympanic membrane (TM) and, after 1 month, temporal bones were taken for histologic study. No macroscopically visible cholesteatomas were present in any of the 16 temporal bones included in the study. However, microscopic invasion of epidermis to the medial side of the TM was observed in four specimens; perforations were present in three of these. Although there was considerable variability in the response to 5-FU application, the majority of specimens showed little or no proliferation of connective tissue in the lamina propria of the TM. In the four specimens in which epidermis reached the medial side of the TM, it did so either by migration through microscopic breaks in the fibrous layer or via TM perforations. Thus, 5-FU did not completely inhibit migration of epidermis into the middle ear. However, the results of this study indicate that it does tend to reduce the proliferation of TM epidermis and connective tissue, thereby reducing the likelihood of cholesteatoma formation in the experimental model.

Tympanic membrane microstructure in experimental cholesteatoma.

This study assessed changes in tympanic membrane microstructure associated with cholesteatoma development following middle ear application of 50% propylene glycol in chinchillas. Although the epidermal layer of the tympanic membrane (TM) was destroyed immediately after propylene glycol application, the epidermal basal lamina remained intact and appeared to serve as a substrate for regrowth of epidermis over the TM. During the initial phase of epidermal repair (4 to 7 days after propylene glycol administration), pseudopodial processes from the epidermal cells occasionally penetrated the basal lamina; however, no migration of epidermis into the lamina propria occurred at that time. The basal lamina remained largely intact until about 2 weeks, when it became fragmented in some areas, so that sizable gaps appeared. Hyperplastic epidermal cells then migrated through the gaps into the rapidly proliferating connective tissue of the lamina propria. At 2 to 4 weeks, degenerative changes were observed in portions of the fibrous layer, which underwent phagocytosis by foreign body giant cells. This process created defects in the fibrous layer which permitted invasion of epidermis to the medial portion of the lamina propria. The epidermis subsequently reached the medial side of the TM in areas where there was incomplete repair of the mucosal layer.

The effect of isotretinoin on propylene glycol-induced cholesteatoma in chinchilla middle ears.

Previous studies have shown that propylene glycol causes inflammatory changes and cholesteatoma when applied to chinchilla middle ears. Vitamin A and synthetic analogues are essential for the normal differentiation of epithelial tissues. The purpose of this study was to determine whether the administration of isotretinoin to chinchillas would prevent propylene glycol exposure from inducing middle ear cholesteatomas. Sixteen chinchillas received 90% propylene glycol to the left middle ear and normal saline to the right. Half the animals were placed in the experimental group and received a daily dose of isotretinoin of 2 mg/kg for 7 days prior to propylene glycol administration and then for 6 weeks until killed. At 6 weeks, cholesteatoma was found in six of eight ears treated with propylene glycol in animals receiving isotretinoin. Two animals in the control group died. Three of the remaining eight had cholesteatoma. No ears treated with saline had cholesteatoma. We conclude that isotretinoin, in our chinchilla model, does not prevent propylene glycol-induced cholesteatoma formation.

Experimental cholesteatoma. Epidermal ingrowth through tympanic membrane following middle ear application of propylene glycol.

This study was designed to investigate morphological changes in the tympanic membrane (TM) associated with cholesteatoma formation in experimental animals following application of propylene glycol to the middle ear. A 50% solution of propylene glycol was applied bilaterally to the middle ear cavities of 30 young-adult chinchillas. The animals were sacrificed for light and electron microscopic study at intervals of 2 days to 6 weeks after a single application of 0.2 ml of the propylene glycol solution. At 2 days there was complete destruction of the epidermal and mucosal layers of the TM. The denuded lateral surface rapidly became re-epithelialized by hyperplastic epidermal cells and by 2-3 weeks, keratinizing epidermis penetrated damaged areas of the fibrous layer of the lamina propria to reach the medial surface of the TM. These epidermal cells proliferated in the middle ear cavity, forming cholesteatomas. Our observations indicate that invasion of the intact, but structurally altered, tympanic membrane by hyperplastic epidermis is a primary mechanism of cholesteatoma formation in the animal model.

Experimental induction of middle ear cholesteatoma in rats.

We induced cholesteatoma in two groups of rats by instilling different concentrations of propylene glycol into the middle ear cavity. Fifteen rats were exposed to 50% propylene glycol (group I), while pure propylene glycol was applied to six others (group II). The group I rats were killed 1 month after instillation. Seven of the 15 showed cholesteatoma in the middle ear with accumulation of keratin debris. The group II rats were killed 3 months after instillation. All six animals showed inflammation in the experimental ears, and five of the six experimental ears showed cholesteatoma in the middle ear cavity. Six experimental ears in group I and five in group II revealed retraction of the tympanic membrane, possibly due to eustachian tube obstruction. Bone resorption was seen along with cholesteatoma and inflammatory cells and osteoclasts in the middle ear of all 11 of these rats. The seventh cholesteatoma of group I can be classified as a microcholesteatoma, a pearl-like cyst within the tympanic membrane. The microcholesteatoma was formed by an invasion of basal cells from the tympanic epidermis and the proliferation of these cells in the fibrous layer of the tympanic membrane. Our findings suggest that cholesteatoma in the middle ear cavity is a response to the inflammation produced by high concentrations of propyleme glycol.

Propylene glycol-induced cholesteatoma in chinchilla middle ears.

Propylene glycol is a solvent commonly used in topical otic preparations. This study examines the occurrence of inflammatory changes and cholesteatoma in chinchilla middle ears after the application of propylene glycol in varying concentrations. A total of 32 ears were studied, divided into four treatment groups. Three groups received propylene glycol in concentrations of 10%, 50%, and 90%. One group received normal saline. Six weeks after the application of propylene glycol to the middle ear through the bulla, examination revealed cholesteatoma, tympanic membrane perforations, and middle ear adhesions in most of the ears subjected to 50% and 90% propylene glycol. Only one ear treated with 10% propylene glycol showed a cholesteatoma, while the group treated with normal saline showed only mild inflammation. Histologic preparations confirmed cholesteatoma and revealed replacement of the normal columnar epithelium by keratinizing stratified squamous epithelium, inflammatory infiltration, and granulation tissue eroding underlying bone. We conclude that exposure to propylene glycol in high concentrations will consistently produce cholesteatoma in chinchilla middle ears. Although the effects of propylene glycol in the human middle ear are yet to be investigated, we recommend the avoidance of otic preparations containing high concentrations of propylene glycol in patients with tympanic membrane perforations.

Inflammatory effects of topical antibiotic suspensions containing propylene glycol in chinchilla middle ears.

This study examines the occurrence of inflammatory changes and cholesteatoma in the middle ears of seven chinchillas after the application of topical antibiotic suspensions containing two different concentrations of propylene glycol. The preparations used were Cortisporin otic suspension, which contains neomycin, polymyxin B, hydrocortisone, and 10.5% propylene glycol, and Cortisporin ophthalmic suspension, containing the same ingredients, but only a 2% concentration of propylene glycol. Six weeks after the administration of the Cortisporin preparations, applied to the middle ear through a transbulla approach, no cholesteatomas were found in the seven ears treated with the ophthalmic suspension. Evidence of mild inflammation was present in only two of these ears. In the seven contralateral ears treated with the otic suspension, middle ear adhesions were found in six, cholesteatoma was present in four, serous effusions were found in three, and one had a large tympanic membrane perforation. The ears that showed cholesteatomas also had histologic evidence of squamous metaplasia, granulation tissue, and erosion of the underlying bone. We submit that the pathologic responses of the middle ear mucosa treated with the otic suspension, were due to an inflammatory response to the higher concentration of propylene glycol compared to that of the ophthalmic suspension.

Middle ear cholesteatoma: an animal model.

Topical otic preparations now in clinical use contain a variety of antibiotics and solvents that may produce severe inflammation if they reach the middle ear cavity. This report describes the response of the chinchilla middle ear to direct application of one such preparation that appears to act as a nonspecific irritant. Cortisporin otic suspension (containing neomycin, polymyxin B, hydrocortisone, and propylene glycol) was introduced into the bullae of 32 chinchillas that were kept alive for four days to five months before histologic examination of their temporal bones. All the experimental animals had tissue damage and inflammation within the middle ear. The changes observed included proliferation of ciliated and secretory columnar cells, formation of granulation tissue, bone erosion, and osteoneogenesis. Some areas of the mucosa underwent metaplasia to stratified squamous epithelium; this metaplastic epithelium, however, did not produce keratin. In the majority of animals kept for two months or more, cholesteatoma was identified in the middle ear. The cholesteatomas appeared to develop as a result of penetration of external canal epidermis through intact tympanic membranes or as the result of migration of epidermis through perforations. The experimental cholesteatomas behaved like those seen clinically in humans, with extensive erosion of bony structures within the middle ear.

The formation of ultrastructural epithelial and subepithelial connective tissue changes in experimentally induced cholesteatomas in guinea pigs.

Histoacryl-induced cholesteatomas were produced in guinea pigs at the posterosuperior part of the external ear canal adjacent to the tympanic membrane. Semithin and ultrathin tissue sections were used to study primary formation of the cholesteatoma as well as the influence of the altered epithelium upon the bordering zone of connective tissue. Quantitative and qualitative tissue changes were analyzed by electron microscopy and showed degradation and new formation of collagen and activation of fibroblasts. Our findings are similar to those previously reported on human cholesteatomas and indicate that the bordering subepithelial tissue seems to be influenced through the basilar membrane.

Ototoxicity of otic drops applied to the middle ear in the chinchilla.

Previous studies have shown that the application of topical otic drops to the external ear canals of animals with patent tympanostomy tubes may result in hearing impairment and cochlear hair cell loss. Otic drops are used in patients with tympanostomy tubes or tympanic membrane perforations and could have deleterious effects on the human membranous labyrinth. This report describes the inner ear damage that occurred after direct application of aminoglycoside-containing otic drops to the middle ears of experimental animals. The membranous labyrinths of 25 chinchillas were studied two days to five months after application of Cortisporin otic suspension (which contains neomycin, polymyxin B, hydrocortisone, and propylene glycol) to the middle ear cavity. Application of 0.5 ml of Cortisporin resulted in degeneration of all inner and outer hair cells throughout the cochlea, as well as severe damage to the stria vascularis. Moderate to severe degeneration of the vestibular receptor organs was also observed. The endolymphatic sacs showed dark-staining endolymph, cellular debris, and macrophages in the sac lumina, as well as increased activity of the epithelial lining.

Effects of keratin on bone resorption in experimental otitis media.

Keratin debris is a constant feature in middle-ear cholesteatoma. Keratin prepared from rat skin induced a foreign-body granuloma in the subcutaneous space in the rat. In vitro this granuloma produced high levels of bone-resorbing factors: prostaglandin E2, osteoclast-activating factor, and leucine aminopeptidase. In the in vivo study, keratin-induced granuloma in the rat middle ear caused partial resorption of the cochlear wall. Macrophages, fibroblasts, and osteoclastlike cells were found at bone-resorption areas. These cells appeared to be responsible for bone resorption through production of prostaglandin E2, osteoclast-activating factor, and proteases.

Experimental production of cholesteatoma in rabbits by using non-irritants (skin tolerants).

The application of various skin tolerant substances to the postero-superior and antero-inferior aspects of the external auditory canal of the rabbit, with an intact ear drum and middle ear, produced cholesteatomas in the skin of the auditory canal and in the tympanic membrane, particularly in the pars flaccida. At the antero-inferior insertion of the tympanic membrane, however, cholesteatoma growth could not be induced in these animal experiments. In severe diffuse otitis externa, secondary to the operative closure of the external auditory canal, cholesteatomas also develop preferentially in the pars flaccida. The relatively thick, loose intermediate layer of connective tissue allows a rapid expansion of epithelial ridges, thus favouring the formation of cholesteatoma in this region.

[Experimentally produced cholesteatoma (author's transl)]. / Tierexperimentelle Untersuchungen zur Erzeugung von Cholesteatomen.

Application of various mildly irritant substances to the posterosuperior and anteroinferior aspects of the external auditory canal of the rabbit, with intact ear drum and middle ear, produced cholesteatomas in the skin of the auditory canal, and the tympanic membrane particularly in the pars flaccida. At the anteroinferior insertion of the tympanic membrane, however, cholesteatoma growth could not be induced in these animal experiments. In severe diffuse otitis externa, secondary to the operative closure of the external auditory canal, cholesteatomas also preferably develop in the pars flaccida, the relatively thick, loose intermediate layer of connective tissue allows a rapid expansion of epithelial ridges thus favouring the formation of cholesteatoma in this region.