Otitis externa and otitis media are common in rabbits, particularly in lop-eared breeds.1,2 In 1 study of > 2,500 clinically normal rabbits, 32% of adult and 4% of young adult animals had evidence of unilateral or bilateral suppurative otitis media.2 In addition, 34% and 17% of affected adult and young adult rabbits, respectively, had otitis externa.2 Topical medication is the mainstay of treatment for otitis externa in human and veterinary medicine; however, in patients with tympanic membrane rupture, a number of topical medications have been found to have ototoxic effects.3–11 Topical antimicrobial administration is commonly used for treatment of otitis externa and otitis media in patients with tympanic perforation.12 In studies13,14 of human patients with otitis, topical administration of antimicrobials resolved infections more rapidly and effectively than did antiseptic treatment or placebo treatment, and no clear benefit was seen when systemic antimicrobial administration was were used alone. Similar evidence for the use of topical preparations in patients with otitis externa and otitis media is present in the veterinary literature, although the weight of evidence suggests that systemic antimicrobial treatment is also necessary in cases of otitis media in which the tympanic membrane is intact or has healed after myringotomy and lavage.15,16
In practice, topical antibacterial or antifungal medications are frequently administered in combination, often in otic preparations that contain a corticosteroid. When combination products are applied to the middle ear, it is possible that the drugs or other in-gredients (eg, the drug carrier or preservatives) have the potential to be ototoxic.
The purpose of the study reported here was to determine whether an ESS preparation (approved for the treatment of otitis externa in dogs and used extralabelly in cats and exotic species) would have ototoxic effects in healthy rabbits with and without intact tympanic membranes. We hypothesized that ototoxic effects (auditory response deficits as measured by BAER testing and histologic changes) would be observed in patients with ruptured tympanic membranes that received this treatment.
Materials and Methods
Animals
Six healthy adult New Zealand White rabbits of unknown ages (median body weight, 6.34 kg; range, 5.78 to 7.02 kg) were used in the study. Three females and 3 males were used in the study. All rabbits were considered mature adults but not geriatric on the basis of dentition and active reproductive status. Routine physical examination was performed by a veterinarian (FLB) at the start of the study. The external ear canals and tympanic membranes of all rabbits were examined grossly, and a swab specimen of the external ear canal was evaluated cytologically prior to the start of the study (FLB). No evidence of infection was seen, and all tympanic membranes were intact and of normal appearance. According to information provided by the distributor,a the rabbits were free of most routinely monitored infectious diseases (reovirus, lymphocytic choriomeningitis virus, parainfluenza virus type 1, parainfluenza virus type 2, rotavirus, rabbit hemorrhagic disease virus, cilia-associated respiratory bacillus, Bordetella bronchiseptica, Helicobacter spp, Lawsonia spp, Pasteurella spp including Pasteurella multocida, Salmonella spp, Treponema spp, Clostridium piliforme, Cheyletiella parasitovorax, Leporacarus gibbus, Psoroptes cuniculi, Passalurus ambiguus, other helminths, Eimeria spp including Eimeria stiedae, other protozoa, and Encephalitozoon cuniculi) for 18 months prior to the start of the study. The facility of origin reported that 2 of 48 rabbits had tested positive for Pseudomonas aeruginosa by culture within 18 months prior to the start of the study, but this did not specifically include the 6 rabbits of the current study. Rabbits were housed in individual cages with free access to food and water. They were examined twice daily, and water intake, food intake, and body weight were recorded daily. Food was withheld from the animals for 4 to 6 hours prior to anesthesia. The Institutional Animal Care and Use Committee at the University of Georgia College of Veterinary Medicine approved the study.
Procedures
On day 0, rabbits received midazolam (1 mg/kg, IM) and buprenorphine (0.05 mg/kg, IM) as premedicants, and general anesthesia was induced with isoflurane (2% in oxygen) via face mask until a moderate plane of anesthesia was achieved. Anesthesia was maintained with isoflurane in oxygen throughout the testing and surgical procedures. The ears of anesthetized rabbits were reexamined with a 5-mm video otoscopeb prior to BAER measurements.
The BAER measurements were performed with a commercially available electromyography—evoked potential testing system.c For each test, four 1.2 × 0.4-mm (27-ga) electrodesd were used: 1 was placed just ventral to the external auditory meatus (tragus cartilage) on each side of the head, a reference electrode was positioned on the dorsal midline of the head, and a ground electrode was placed at the dorsal midline of the cervical region. Foam inserts with embedded earphones were placed into each external auditory canal (ie, the vertical portion of the ear canal) and allowed to expand to fit the space. Each BAER test consisted of a series of 1,000 clicks from a range of 15 to 95 dB at a standard frequency; normal hearing threshold of rabbits at this frequency is approximately 34.8 dB.17 Clicks were of a 0.1-second duration, and the square wave frequency range was 250 Hz to 7 kHz, centered on 3 to 4 kHz. The left ear was tested first, followed by the right ear in all rabbits. A masking noise of 60 dB above the normal hearing level was delivered into the ear contralateral to the ear being tested. Hearing thresholds were determined by interpretation of BAER-generated graphic data.
After bilateral BAER testing to obtain baseline data, the otoscope was reinserted in 1 ear, and a 5-mm cathetere was inserted through the working channel of the otoscope. The catheter was advanced through the tympanic membrane at the caudoventral aspect. The otoscope was removed, the procedure was repeated for the contralateral ear, and a second BAER test was performed. A catheter advanced through the otoscope was then used to deliver study treatments directly into the middle ear. Sterile saline (0.9% NaCl) solution (0.3 mL) was instilled into the left ear, ESS solutionf was instilled into the right ear, and a third BAER test was performed. The rabbits were then allowed to recover from anesthesia. After recovery, rabbits were monitored for 1 hour and returned to their normal housing. Variables for monitoring throughout procedures and recovery included heart rate, respiration rate, mucous membrane color, capillary refill time, rectal temperature, and O2 saturation as measured by pulse oximetry. Measurements were obtained every 5 minutes throughout the procedures and the 1-hour postprocedure recovery time, and then animals were monitored twice daily for any adverse events during the remainder of the study. No further analgesics were required for any animal. Saline solution (0.3 mL) was administered in the left ear and ESS solution (0.3mL) was administered in the right ear every 12 hours for 7 days by 2 investigators (FLB and LLQ).
On day 8, the rabbits were anesthetized for BAER testing as previously described. When the tests were completed, anesthetized rabbits were euthanized with sodium pentobarbital (150 mg/kg, IV, delivered via a peripheral IV catheter).
Tissue preparation and histologic examination
Within 5 minutes after euthanasia, 2 mL of neutral-buffered 10% formalin solution was instilled via syringe into the external ear canals, and the head was swirled in a circular motion for 1 minute to optimize distribution of the solution to the middle and inner ears. The temporal bones were extracted from all rabbits ≤ 4 hours after euthanasia by soft tissue dissection and use of stainless steel rongeurs. The temporal bones and brain were immediately immersed in the formalin solution (10:1 solution-to-tissue ratio), and all tissues remained in fixative solution for ≥ 48 to 72 hours. The abdominal and thoracic cavities were opened and inspected for gross abnormalities.
The temporal bones, tympanic bullae, and small portions of the occiput were decalcified in Kristenson solution (stock solution: 2,000 mL of 8N formic acid and 2,000 mL of 1N sodium formate) for 48 to 72 hours, with a 10:1 stock solution-to-tissue ratio. Decalcification progress was checked every 24 hours. In preparation for histologic and SEM examination, samples were transected in the dorsal plane through the tympanic bulla and temporal bone, dividing the tissue into serial sections. One half of the cochlea was submitted for histologic examination, and the remaining half was fixed in Karnovsky fixative solution (2% paraformaldehyde; 2.5% glutaraldehyde in 0.1M phosphate buffer; pH, 7.4) at 4°C for at least 7 days in preparation for SEM.
For histologic examination, sections of the brainstem vestibulocochlear nuclei as well as the decalcified cochleae, tympanic bullae, and external ear canals were dehydrated in ethanol, cleared in xylene, and embedded in paraffin wax. Sections (thickness, 4 μm) of the inner, middle, and outer ears were stained with H&E stain. Sections with inflammatory cell infiltrates identified were stained with Brown and Brenn Gram stain.
For SEM, inner ear samples from 4 rabbits (3 ESS-treated ears and 4 saline-treated ears) were selected on the basis of tissue availability and lack of tissue fragmentation secondary to mineralization. Owing to differences in anatomy and rabbit size, inner ear sections varied, and the organ of Corti could not be seen in 4 of the 7 inner ear samples. Three inner ear samples (2 ESS treated and 1 saline treated) included the chosen feature of interest (basal turn of the organ of Corti), and these were included in the SEM evaluations. In each of 4 samples for which the organ of Corti could not be seen, it was thought that slightly more than half of the cochlear tissue was included in the section saved for histology; following tissue processing and paraffin embedding, it was not feasible to go back and retrieve tissue for SEM.
In preparation for SEM, the cochlear samples were removed from Karnovsky fixative solution, washed in 0.1M phosphate buffer, and fixed in osmium tetroxide solution for 90 minutes at 4°C. Specimens were then dehydrated in graded ethanol solutions (concentrations, 30% to 100%), dried with a critical point drier,g and plated with gold by use of a sputter coater.h A qualitative and semiquantitative assessment of the organs of Corti, peripheral nerves, and bony surfaces was performed.18–20,i
Necropsies, histopathologic examination, and SEM were performed by a veterinary pathology resident (SGMK). Final histopathologic and SEM findings were reviewed by a board-certified veterinary pathologist (EWH). Tissues were first evaluated in a blinded manner. The pathologists were then unblinded to the treatment status of each rabbit's ears to enable these individuals to draw appropriate conclusions and to contribute to discussion in the manuscript.
Descriptive data were reported. No statistical analyses were performed.
Results
All rabbits completed the study. Two rabbits were observed to have a moderate, approximately 30° right-sided head tilt; 1 of these 2 rabbits was inappetent on day 3 and was treated with maropitant (1 mg/kg, SC, q 24 h). This animal regained some appetite, although food intake was reduced, compared with that on days 1 and 2. This animal did not lose body weight or condition and regained some appetite; therefore, it remained in the study. Postmortem examination on day 8 revealed a trichobezoar that filled the gastric fundus; this was considered the likely cause of the observed signs. No other apparent adverse effects were observed.
Hearing thresholds in the 6 anesthetized rabbits at baseline (prior to myringotomy) ranged from 15 to 60 dB for left ears and from 15 to 65 dB for right ears. The median difference between left and right ears at this time was 5 dB (range, 0 to 25 dB); 3 rabbits had baseline hearing thresholds that exceeded the expected normal value of 34.8 dB (2 unilaterally [left ear; values of 35 and 40 dB] and 1 bilaterally [60 and 65 dB for the left and right ears, respectively]).
Immediately after myringotomy, most hearing thresholds were altered (with values unchanged for the left ear of 1 rabbit and subjectively increased for all remaining ears). The range of values for left ears was 30 to 85 dB, with the same result for right ears. These values were further altered after instillation of either study treatment, with apparent increases in 11 of 12 ears; the hearing thresholds ranged from 60 to > 95 dB for left (saline solution–treated) ears and from 40 to > 95 dB for right (ESS-treated) ears.
The BAER measurements on day 8 revealed that all rabbits had hearing thresholds that exceeded the respective baseline values in both ears. Results for saline solution–treated and ESS-treated ears ranged from 30 to 85 dB and from 80 to > 95 dB, respectively. The change in threshold from baseline ranged from 5 to 40 dB for saline solution–treated ears, with changes of ≥ 20 dB in 3 rabbits. The change in threshold for ESS-treated ears ranged from > 30 to > 80 dB, and 4 rabbits had no hearing detected in the ESS-treated ear at the test limit of 95 dB. Five rabbits had a hearing threshold ≤ 55 dB in the saline solution–treated ear, and 5 rabbits had a hearing threshold ≥ 85 dB in the ESS-treated ear.
Histologic findings
Saline solution–treated ears—There were few changes observed in saline solution–treated ears. These changes were interpreted as having little clinical relevance and were considered possibly attributable to the myringotomy and sham treatment procedure. One rabbit had a mild accumulation of heterophils and foamy macrophages at the entrance to the saline solution–treated middle ear. Another rabbit had a small, hyperplastic focus of heterophilic epidermitis, with a thin, serocellular crust in the external ear canal of this ear. One saline-treated ear had mild, multifocal perineural mineralization adjacent to the external ear canal. The vestibulocochlear nuclei and inner ears that received this treatment were histologically unremarkable in all rabbits.
ESS-treated ears—Examples of histologic abnormalities identified in ESS-treated ears are shown (Figures 1–3). Six of 6 ears that received this treatment contained a large, serocellular plug in the external ear canal and had severe heterophilic otitis externa with various degrees of epithelial hyperplasia. There was occasional mineralization of the lamina propria (n = 3) or perineural tissue (4) near the external auditory meatus. When portions of the healing tympanic membrane were included in tissue sections (n = 3), the outer portion was expanded by edema and granulation tissue, and infiltration by low numbers of heterophils was evident.
Pathological findings of the middle ear following ESS treatment consisted of various degrees of mucoperiosteal edema and mild to moderate periosteal new bone formation (6/6 ears), typically characterized by thin, short, protruding, osseous spicules; in 1 rabbit, the latter finding was severe, forming a solid, mass-like protrusion of immature trabecular bone into the tympanic cavity (Figures 1 and 2). In 3 of 6 ears, moderate to marked heterophilic and histiocytic otitis media were also present with inflammatory infiltrates in the lumen or within the wall of the thickened mucoperiosteum. There was multifocal necrosis of the mucoperiosteum in the ear of the rabbit with the mass-like protrusion of granulation tissue and bone.
Four of 6 ESS-treated inner ears were histologically indistinguishable from those treated with saline solution. The 2 rabbits with right-sided head tilts and otitis media also had mild ipsilateral heterophilic and histiocytic otitis interna with scant hemorrhage within the scala vestibuli. One rabbit had periganglionic calcification within the nerve fibers of the cochlear modiolus, and this change was adjacent to partially mineralized modiolus bone; this finding was not present in the contralateral, saline solution–treated ear. The vestibulocochlear nuclei of all ESS-treated ears were histologically unremarkable.
SEM
Three inner ears (2 ESS treated and 1 saline solution treated) containing the basal cochlear turn were evaluated by SEM. Within the organ of Corti of 1 ESS-treated ear, hair cells were obscured by extracellular debris and could not be reliably evaluated. Additionally, evaluation of the basal cochlear turns of the remaining 2 ears (1/treatment) revealed that the inner hair cells of both ears were similarly obscured by accumulation of extracellular debris. Outer hair cells of the basal cochlear turn in the saline solution–treated ear were uniformly intact, whereas most outer hair cells in the ESS-treated ear lacked stereocilia or were absent; those that lacked stereocilia contained cell membrane depressions and occasional rounded membrane blebs (Figure 4). A dendritic cell covered the basilar membrane where outer hair cells of the second and third rows were largely devoid of stereocilia.
Discussion
The normal hearing threshold in rabbits has been previously determined. Stieve et al17 obtained normative BAER data for hearing thresholds in healthy rabbits for both click and tone-pip stimuli with bone and air conduction. A potential threshold of 34.8 dB was determined when air-conducted click stimuli were used,17 and these same stimuli were used for the present study. We found substantial variation in the baseline hearing thresholds of healthy study rabbits, with some having unilateral or bilateral thresholds at or exceeding the reported normal range of 34.8 dB.
Ototoxic effects of many compounds have been described in people and in nonhuman animals. Examples include aminoglycosides, which diffuse across the fenestra rotunda (ie, round window) membrane and concentrate in the inner ear.21,22 Gentamicin has been reported to have vestibular toxic effects,3 and topically applied neomycin has also been found to cause toxicosis affecting the cochlear and vestibular regions in human patients.3,23,24
Fluoroquinolone antimicrobials are available in topical preparations and have not been reported to have ototoxic effects when applied directly to the middle ear.25 Fluoroquinolones are broad-spectrum bactericidal drugs with dose-dependent effects. Ciprofloxacin does not cross the round window membrane and is not found in the inner ear following application in people.21 Topical ciprofloxacin or ofloxacin treatments are commonly used for human patients, and the results of studies with laboratory animals revealed no meaningful change in hearing thresholds after a ciprofloxacin-dexamethasone combination was applied directly to the middle ear.9,26–28 Enrofloxacin is a fluoroquinolone that has been used topically in treatment of otitis externa and otitis media in companion animals. The spectrum of enrofloxacin and its active metabolite ciprofloxacin includes gram-negative and some gram-positive organisms (especially Staphylococcus spp). Typically, the inject-able product is used extralabelly as topical medication,16 and to the authors' knowledge, no reports of ototoxic effects exist when the formulation is administered in this manner.
An ototoxic effect is variably defined as damage to either the cochlear or vestibular portion of the auditory system.29 According to 1 review,30 ototoxic effects are typically considered to have occurred if there is an increase in the hearing threshold ≥ 15 dB at any 2 frequencies or ≥ 20 dB at ≥ 1 frequency, compared with baseline values. According to the latter definition, all 6 rabbits had evidence of ototoxic changes in the ESS-treated ear (with numeric increases of > 30 to > 80 dB from baseline), and no hearing threshold was detected in 4 ESS-treated ears. Three rabbits had similar evidence of ototoxic effects in the saline solution–treated ear; however, these had changes of ≤ 40 dB. The alteration in hearing thresholds in these 3 rabbits may have been attributable to the myringotomy procedure, as the degree of change was less than that of the ESS-treated ears.
On histologic examination, pathological changes were detected in the outer and middle ear compartments, and occasionally, milder changes were present in the inner compartment of ESS-treated ears. There were various degrees of epithelial hyperplasia throughout external ear canals of ESS-treated ears, although similar, albeit milder, changes were occasionally seen in saline solution–treated ears. There was substantial variability in the thickness of the external ear canal epithelium, depending upon where the sample was sectioned, so an assessment of degree of hyperplasia was difficult given the small sample size in this study. However, the ESS-treated ears consistently had more epithelial degeneration and necrosis as well as a robust heterophilic inflammatory response; grossly, these ears were occluded by a serocellular plug. The lamina propria calcification that was seen in 4 of the 6 ESS-treated ears was seen in 1 of the 6 saline solution–treated ears. We speculated that this may have been related to localized changes in tissue pH associated with the ESS treatment. Alternatively, this may have represented an artifact of tissue decalcification.31
Pronounced changes in the middle ear following ESS treatment included inflammation (3/6 ears), bony proliferation (6/6), and necrosis of the mucoperiosteum (1/6). Gram staining did not reveal bacteria in the inflammatory exudates, and the rabbits were free of the majority of common bacterial, viral, and parasitic infectious agents for ≥ 18 months prior to the start of the study. We considered that the inflammatory response in ESS-treated ears most likely occurred secondary to localized cell injury in the middle ear cavity, as bony proliferation of the tympanic bulla is a well-documented response to chronic otitis media in domestic animals.32 We could not completely exclude the possibility that the mild to moderate mucoperiosteal new bone formation seen in some ears was a preexisting lesion, possibly related to prior middle ear infection, or a background, age-related change, especially for those lesions that lacked corresponding nearby inflammatory changes. However, this was considered less likely, given the immaturity of the new bone and the lack of new bone formation in saline solution–treated ears. Additionally, we considered it possible that active foci of inflammation may have been outside the plane of sectioning in those cases where bone remodeling was observed without an accompanying inflammatory component.
In addition to profound otitis media, the 2 rabbits with right-sided head tilts had mild heterophilic and histiocytic otitis interna. The inner ear infiltrates in these rabbits were likely the result of direct extension through the round window or the fenestra vestibuli (ie, oval window).33 Although 4 of 6 ESS-treated inner ears were histologically unremarkable, in some cases of toxicant-induced cochlear inflammation, hemorrhage and inflammation may be present in only gravity-dependent areas of the cochlea33; thus, it was possible that affected areas were present in other ears but might not have been examined histologically. It was difficult to interpret the clinical relevance of perineural calcification that was occasionally seen in the middle ear and external ear canal, given that this was seen, although rarely, in 1 saline solution–treated ear. Additionally, the clinical relevance of cochlear periganglionic calcification in 1 ESS-treated ear was unclear. It is possible that these lesions may have been present prior to the study, potentially related to age or a prior pathological process, or that they represented an artifact of decalcification.31
The condition of hair cells of the organ of Corti was investigated by SEM for 1 ear that received each type of treatment. There was outer hair cell degeneration and stereocilia loss in the ESS-treated ear, with only scattered stereocilia tufts remaining in the second row of these cells. Following most types of cochlear injury, degenerated hair cells are replaced by expansion of nonsensory (support) cells of the phalangeal plate, and the result is termed a phalangeal scar.34 The function of phalangeal scars is debated, but it is thought that these allow the organ of Corti to continue functioning despite loss of hair cells.34 In our study, the sites of outer hair cell loss in the ESS-treated cochlea may have represented phalangeal scarring. Outer hair cell susceptibility to toxic compounds can vary among the basal, middle, and apical turns of the cochlea. In studies33,35 involving laboratory animals, aminoglycosides were found to target the high-frequency outer hair cells of the basal cochlear turn. The outer hair cells of the basal cochlear turn were affected in the ESS-treated ear in the present study, although the full spectrum of adverse cochlear effects would require further investigation. Age-associated hair cell degeneration could not be completely excluded as a contributing factor; however, given the striking disparity between the number of outer hair cells in the saline solution–treated and ESS-treated cochleae, age was considered to be a minor variable in this evaluation. All rabbits were considered to be mature (not geriatric), although the precise ages were not known.
In the deepest portion of ESS-treated ear external ear canals, epithelial hyperplasia and lamina propria edema most likely represented an irritant reaction. Heterophilic epidermitis is a common reaction to irritants in toxicology studies involving rabbits, with necrosis, ulceration, and ceruminous gland hyperplasia resulting from some prolonged treatments.33 The cause of occasional perineural and lamina propria calcification in ESS-treated ears was unclear; this may have been related to localized changes in perineural pH or ionic concentration resulting from the topically applied ESS compound. The vestibulocochlear nuclei of both ears for all rabbits were histologically normal, so a central cause of hearing loss was considered unlikely in these rabbits.
Taken together, the subjectively greater degree of hearing loss after ESS treatment, compared with that after saline solution administration, and the outer hair cell stereocilia degeneration in the examined ESS-treated ear supported that this product had ototoxic effects when applied to the middle ear of rabbits. Additionally, histopathologic changes in the middle ear and external ear canals of ears that received this treatment suggested that the compound caused irritation. Given the aforementioned safety of the topically applied quinolones, it seemed unlikely that enrofloxacin alone was the cause of the local adverse effects observed. The product used in the present study also contained silver sulfadiazine (10 mg; 1.0% w/v), benzyl alcohol (a preservative), and cetyl stearyl alcohol (a stabilizer) in a neutral oil and purified water emulsion. Silver sulfadiazine has broad-spectrum antibacterial activity and exerts this effect by bacterial wall damage leading to osmotic changes and by impairment of DNA replication.36 Silver sulfadiazine has been used since 1968 in human medicine and has a good safety record. Sporadic cases of allergic contact dermatitis37,38 as well as localized39 and systemic argyria38 have been reported, but the ototoxicity of this compound is unknown.
Benzyl and cetyl stearyl alcohol also have the potential to be ototoxic. Whereas the authors were aware of no specific data for these 2 alcohols, Perez et al40 determined that a topically applied 70% solution of ethyl alcohol in water induced loss of vestibular and auditory potentials in several laboratory rats. However, the authors of that study40 noted that some of the alcohol-treated animals developed granulation tissue that covered the round window and thus may have protected the inner ears from toxic effects, and they speculated that the ototoxicity of the alcohol solution was greater than their results suggested. One rat in that study40 also had a normal vestibular response after alcohol application but did not respond to auditory stimuli. This might imply that alcohol has greater cochlear than vestibular ototoxicity, but additional studies are needed to investigate this.
The exact formulation of the neutral oil component of the vehicle in the test compound was undisclosed (no further description was provided on material data safety sheets or the product label). Although the ototoxicity and irritant potential of the compound could not be investigated further, we considered it less likely than other components to have been the cause of the adverse effects observed in the present study, given the safety of other lipid products such as olive oil, baby oil, almond oil, and mineral oil when applied to the middle ear.41–44 Finally, the purified water component of the emulsion has long been considered safe in the middle ear.
In the present study, rabbits were successfully used for in vivo testing of the effects of a topical otic product on the outer, middle, and inner ear. It is important to note that there are anatomic and physiologic differences between the middle ears of people and small mammals in this regard. For example, diffusion of molecules across the round window is the main mechanism by which ototoxic compounds access the inner ear and organ of Corti. In people, this membrane is recessed into the triangular fossa and protected by a bony prominence. With the head in an upright position, fluid placed in the middle ear will naturally fall ventrally and away from the round window. In contrast, animals have a prominent and large round window (described in rodents and chinchillas) that is not recessed and is positioned in such a way that prolonged contact with a fluid placed in the middle ear is more likely than it is in most human patients.45–47 In addition, the thickness of the round window membrane varies among species, with rodents having the thinnest membrane (10 to 14 μm) and humans having the thickest (40 to 60 μm).4,45,47 This may indicate that rodents are more susceptible to diffusion of an ototoxic compound across the round window membrane (and therefore ototoxic effects on the inner ear). However, the thickness of the round window membrane in rabbits is unknown. Although care must be taken when interpreting results of tests in healthy laboratory animals and extrapolating the findings to human patients or other animal species, to our knowledge, the in vivo evaluation of topical otic products in rabbits and other laboratory animal species remains the best way to investigate their potential for ototoxic effects.
The present study had several limitations, including the small number of rabbits, evaluation of only 1 ESS-treated ear by SEM, and lack of statistical analyses. We did not have a population size that would allow for the inclusion of a positive control, such as an aminoglycoside, for ototoxic effects. Ideally, this study should be repeated with a larger sample size and include positive and negative controls as well as testing the individual components (silver sulfadiazine, enrofloxacin, and excipients) to further isolate the ototoxic compound or compounds in the ESS evaluated.
Our results supported our hypothesis that the ESS product would have ototoxic effects when applied to the middle ear of rabbits. The histopathologic finding of occasional perineural mineralization and the ultrastructural finding of outer hair cell degeneration in the organ of Corti suggested that sensorineural deficits contributed to hearing loss in ESS-treated ears. Additionally, external ear canal luminal plugging by serocellular debris may have contributed to this finding through conductive hearing loss. The degree of changes in hearing thresholds as assessed by BAER testing and the histopathologic changes (in particular calcification observed in the external ear canal epithelium, periosteal new bone formation in the tympanic bulla, and loss of outer hair cells in the organ of Corti) were unexpected within a 7-day treatment period. It is important to note that the commercial ESS product tested in the present study was not licensed for use in rabbits, and the manufacturer clearly states on the package insert48 that the product is not to be used in the case of a ruptured tympanum. The study results suggested that clinically relevant changes can occur in a short period of time if product is applied inappropriately. Further studies are needed to confirm these findings and clarify the precise mechanism of hearing impairment caused by the ESS product used in this study.
Acknowledgments
Not funded by any external source. The authors have no financial conflicts of interest or disclaimers to report.
ABBREVIATIONS
BAER | Brainstem auditory-evoked response |
ESS | Enrofloxacin–silver sulfadiazine emulsion |
SEM | Scanning electron microscopy |
Footnotes
C23 location (Canada), Charles River Laboratories, Wilmington, Mass.
Karl Storz, Tuttingen, Germany.
Nicolet VikingQuest EMG/NCS/EP System, Natus Medical Inc, Pleasanton, Calif.
Cadwell Industries Inc, Kennewick, Wash.
Coviden, Mansfield, Mass.
Baytril Otic, Animal Health Division, Bayer Healthcare LLC, Shawnee Mission, Kan.
Tousimis, Rockville, Mary.
SPI Supplies, West Chester, Pa.
FEI Teneo SEM, FEI Co, Hillsboro, Ore.
References
1. Csomos R, Bosscher G, Mans C, et al. Surgical management of ear diseases in rabbits. Vet Clin North Am Exot Anim Pract 2016;19:189–204.
2. Flatt RE, Deyoung DW, Hogle RM. Suppurative otitis media in the rabbit: prevalence, pathology and microbiology. Lab Anim Sci 1977;27:343–347.
3. Matz G, Rybak L, Roland PS, et al. Ototoxicity of ototopical antibiotic drops in humans. Otolaryngol Head Neck Surg 2004;130(suppl 3):S79–S82.
4. Roland PS, Rybak L, Hannley M, et al. Animal ototoxicity of topical antibiotics and the relevance to clinical treatment of human subjects. Otolaryngol Head Neck Surg 2004;130(suppl 3):S57–S78.
5. Brummett RE, Harris RF, Lindgren JA. Detection of ototoxicity from drugs applied topically to the middle ear space. Laryngoscope 1976;86:1177–1187.
6. Jakob T, Wright CG, Robinson K, et al. Ototoxicity of topical ticarcillin and clavulanic acid in the chinchilla. Arch Otolaryngol Head Neck Surg 1995;121:39–43.
7. Paterson S. Brainstem auditory evoked responses in 37 dogs with otitis media before and after topical therapy. J Small Anim Pract 2018;59:10–15.
8. Jang CH, Park H, Cho YB, et al. Evaluating the ototoxicity of topical piperacillin-tazobactam. Int J Pediatr Otorhinolaryngol 2008;72:1815–1821.
9. Kavanagh KR, Parham K, Schoem SR. Auditory function after a prolonged course of ciprofloxacin-dexamethasone otic suspension in a murine model. Arch Otolaryngol Head Neck Surg 2009;135:238–241.
10. Phillips JS, Yung MW, Burton MJ, et al. Evidence review and ENT-UK consensus report for the use of aminoglycoside-containing ear drops in the presence of an open middle ear. Clin Otolaryngol 2007;32:330–336.
11. Serin GM, Ciprut A, Baylançiçek S, et al. Ototoxic effect of Burow solution applied to the guinea pig middle ear. Otol Neurotol 2007;28:605–608.
12. Harris AS, Elhassan HA, Flook EP. Why are ototopical aminoglycosides still first-line therapy for chronic suppurative otitis media? A systematic review and discussion of aminoglycosides versus quinolones. J Laryngol Otol 2016;130:2–7.
13. Macfadyen CA, Acuin JM, Gamble C. Systemic antibiotics versus topical treatments for chronically discharging ears with underlying eardrum perforations. Cochrane Database Syst Rev 2006;25:CD005608.
14. Macfadyen C, Gamble C, Garner P, et al. Topical quinolone vs. antiseptic for treating chronic suppurative otitis media: a randomized controlled trial. Trop Med Int Health 2005;10:190–197.
15. Bruyette DS, Lorenz MD. Otitis externa and otitis media: diagnostic and medical aspects. Semin Vet Med Surg (Small Anim) 1993;8:3–9.
16. Morris DO. Medical therapy of otitis externa and otitis media. Vet Clin North Am Small Anim Pract 2004;34:541–555.
17. Stieve M, Mojallal H, Winter M, et al. Normative auditory brainstem response data for hearing threshold in the rabbit. Audiol Neurootol 2006;11:310–317.
18. Mujica-Mota MA, Bezdjian A, Salehi P, et al. Assessment of ototoxicity of intratympanic administration of Auralgan in a chinchilla animal model. Laryngoscope 2015;125:1444–1448.
19. Rask-Andersen H, Schrott-Fischer A, Pfaller K, et al. Perilymph/modiolar communication routes in the human cochlea. Ear Hear 2006;27:457–465.
20. Zine EA, Romand R. Expression of alpha-actinin in the stereocilia of hair cells of the inner ear: immunohistochemical localization. Neuroreport 1993;4:1350–1352.
21. Becvarovski Z, Bojrab DI, Michaelides EM, et al. Round window gentamicin absorption: an in vivo human model. Laryngoscope 2002;112:1610–1613.
22. Helal A. Aminoglycoside eardrops and ototoxicity. CMAJ 1997;156:1056–1058.
23. Bath AP, Walsh RM, Bance ML, et al. Ototoxicity of topical gentamicin preparations. Laryngoscope 1999;109:1088–1093.
24. Marais J, Rutka JA. Ototoxicity and topical eardrops. Clin Otolaryngol Allied Sci 1998;23:360–367.
25. Macfadyen CA, Acuin JM, Gamble CL. Topical antibiotics without steroids for chronically discharging ears with underlying eardrum perforations. Cochrane Database Syst Rev 2005;19:CD004618.
26. Samarei R. Comparison of local and systemic ciprofloxacin ototoxicity in the treatment of chronic media otitis. Glob J Health Sci 2014;6:144–149.
27. Lemke LE, McGee DDH, Prieskorn DM, et al. Safety of ciprofloxacin and dexamethasone in the guinea pig middle ear. Arch Otolaryngol Head Neck Surg 2009;135:575–580.
28. Daniel SJ, Munguia R. Ototoxicity of topical ciprofloxacin/dexamethasone otic suspension in a chinchilla animal model. Otolaryngol Head Neck Surg 2008;139:840–845.
29. Lanvers-Kaminsky C, Zehnhoff-Dinnesen AA, Parfitt R, et al. Drug-induced ototoxicity: mechanisms, pharmacogenetics, and protective strategies. Clin Pharmacol Ther 2017;101:491–500.
30. Brummett RE, Fox KE. Aminoglycoside-induced hearing loss in humans. Antimicrob Agents Chemother 1989;33:797–800.
31. Rastogi V, Puri N, Arora S, et al. Artefacts: a diagnostic dilemma—a review. J Clin Diagn Res 2013;7:2408–2413.
32. Njaa BL. The ear. In: Zachary J, ed. Pathologic basis of veterinary disease. 6th ed. St Louis: Elsevier Health Sciences, 2017;1223–1264.
33. Schafer KA, Bolon B. Special senses—ear. In: Wallig MA, Haschek WM, Rousseaux CG, et al, eds. Fundamentals of toxicologic pathology. 3rd ed. St Louis: Elsevier Health Sciences, 2018;729–747.
34. Raphael Y, Altschuler RA. Scar formation after drug-induced cochlear insult. Hear Res 1991;51:173–183.
35. Martin L. Models related to hearing loss. In: Suckow MA, Stevens KA, Wilson RP, eds. The laboratory rabbit, guinea pig, hamster, and other rodents. St Louis: Elsevier Health Sciences, 2012;1017–1023.
36. Lansdown AB, Williams A, Chandler S, et al. Silver absorption and antibacterial efficacy of silver dressings. J Wound Care 2005;14:155–160.
37. García AA, Rodríguez Martin AM, Serra Baldrich E, et al. Allergic contact dermatitis to silver in a patient treated with silver sulphadiazine after a burn. J Eur Acad Dermatol Venereol 2016;30:365–366.
38. Fuller FW. The side effects of silver sulfadiazine. J Burn Care Res 2009;30:464–470.
39. Fisher NM, Marsh E, Lazova R. Scar-localized argyria secondary to silver sulfadiazine cream. J Am Acad Dermatol 2003;49:730–732.
40. Perez R, Freeman S, Sohmer H, et al. Vestibular and cochlear ototoxicity of topical antiseptics assessed by evoked potentials. Laryngoscope 2000;110:1522–1527.
41. Emami N, Daniel SJ. Ototoxicity of olive oil in a chinchilla animal model. Laryngoscope 2013;123:2009–2012.
42. Al-Jarallah A, Akinpelu OV, Citra D, et al. Ototoxicity of baby oil in a chinchilla animal model. Int J Pediatr Otorhinolaryngol 2012;76:564–568.
43. Peleva E, Mourad S, Citra D, et al. Assessment of the ototoxicity of almond oil in a chinchilla animal model. Laryngoscope 2011;121:2661–2664.
44. Nader ME, Saliba I. Ototoxicity of intratympanic docusate sodium and mineral oil in the guinea pig. Otolaryngol Head Neck Surg 2012;146:455–460.
45. Morizono T. Toxicity of ototopical drugs: animal modeling. Ann Otol Rhinol Laryngol Suppl 1990;148:42–45.
46. Schuknecht H, Gulya A. Anatomy of the temporal bone with surgical implications. Philadelphia: Lea & Febiger, 1986.
47. Brown OE, Wright CG, Meyerhoff WL. Workshop on Chronic Suppurative Otitis Media: Etiology and Management. August 29–30, 1985, Pittsburgh, Pennsylvania. Proceedings. Ann Otol Rhinol Laryngol Suppl 1988;131:38–40.
48. Baytril Otic [package insert]. Shawnee Mission, Kan: Bayer HealthCare LLC, 2017.