III. Comparative Quantitative Analysis

Hypotheses to explain the selective ototoxicity of aminoglycoside antibiotics may be grouped into two categories: 1) a partition effect and 2) innate hair cell selectivity. The results of this study give evidence that, in fact, both mechanisms may interplay. This conclusion is induced from the observations that following the transtympanic administration of streptomycin, dihydrostreptomycin and neomycin: 1) All three drugs elicited nearly the same topographic pattern and degree of damage in the vestibular compartment. 2) All three drugs elicited nearly the same topographic pattern but widely different degrees of damage in the cochlear compartment (neomycin > > streptomycin > dihydrostreptomycin). 3) Neomycin caused the same devastating cochlear hair cell damage as reported for its systemic administration. 4) Dihydrostreptomycin was significantly more vestibulotoxic than cochleotoxic, and streptomycin was both significantly more cochleotoxic than vestibulotoxic and significantly more cochleotoxic than was dihydrostreptomycin. The latter observations are discussed with regard to contrasts with the literature reporting the drug effects following systemic administration.

IV. Influence of Vestibular Stimulation

Guinea pigs were given transtympanic dosages of neomycin followed by three hours of either vestibular stimulation or immobilization to test the hypothesis that inner ear hair cell membrane permeability is increased for drugs as a consequence of the transduction process. It was found, however, that neither the topographic pattern nor the degree of hair cell damage differed between the two groups. It is concluded that if the transduction process is accompanied by cell membrane permeability changes, the effect is not an increased susceptibility to neomycin toxicity. A measurable effect of stimulation was the significant decrease in the standard deviation of the mean dosage-damage response at each dosage level. This increase in reliability probably resulted from mixing of the inner ear fluids (especially in the vestibular compartment) which encouraged consistent dosages to the hair cells with each repeated trial. Finally, the impact of these findings on hypotheses related to the mechanism of selective ototoxicity is discussed briefly.

Ototoxicity of topical otomicrobial agents

Topical antimicrobial solutions are used commonly for prevention or treatment of purulent otorrhea even though most solutions contain ototoxic agents. This study compares the cochlear and middle ear toxicity of cortisporin otic solution (COS), 0.3% gentamicin ophthalmic solution (GOS), benzalkonium chloride (0.026% and 0.05%), and 1.0% Ofloxacin, a new quinolone antibiotic. Saline 0.9% was used as a control. The agent was instilled daily for 7 days into the bulla of juvenile guinea pigs. The animals were sacrificed on the 14th day. The organ of Corti was examined using surface preparation light microscopy and scanning electron microscopy. The tympanic membrane (TM) and adjacent middle ear mucosa were examined with light microscopy. The average cochlear hair cell damage was 66% for COS, 6.5% for GOS, and 1% for Ofloxacin, benzalkonium and saline. COS and benzalkonium 0.05% produced moderate mucosal thickening and inflammation. However, this was not statistically different from the mild mucosal thickening produced by saline, GOS, Ofloxacin and 0.026% benzalkonium. There was statistically significant mild to moderate thickening of the TM for all agents compared to the saline control.

Long-term ototoxic effects of neomycin applied topically in the middle ear: a morphological study in the guinea pig

Neomycin was instilled daily, uni- or bilaterally, into the middle ear of guinea pigs for three months. The cochleae were examined, by light and electron microscopy, six months after the end of treatment. The organs of Corti of the treated ears were completely destroyed, and in the most advanced lesions, were substituted by a single layer of very thin flat cells. In the spiral ganglion only some glial cells and a few neurons could be observed. All surviving neurons were myelinated, and their ultrastructure was greatly altered, with disorganization of the rough endoplasmic reticulum and ribosomes. The basilar membrane almost disappeared, losing it amorphous and filamentous components. The spiral limbus and the stria vascularis were atrophic and were also covered, in the final stages, by flat elongated cells. In view of its morphological characteristics, this epithelium may arise from the displacement of the interdental cells and perhaps from the cochlear surface of Reissner's membrane.