Otoacoustic emissions analysers for monitoring aminoglycosides ototoxicity

Abstract

Objectives: Aminoglycosides (AG) are widely prescribed despite their notorious toxicity. These antibiotics cause irreversible hearing loss, starting with high frequencies and progressing toward conversational frequencies (0.5-2 kHz), by destroying the acoustic hair cells in the inner ear. The integrity of these cells could be analysed by recording faint sounds that they produce otoacoustic emissions (OAE). The aim of the present study was to monitor and to characterise the acoustic toxicity of the AG using an OAE analyser.

Method: We performed a prospective study of 49 patients receiving gentamicin (G), during 2007-2008. We made serial OAE recordings with an ILO 92 analyser (1-8 kHz) on at least 3 occasions: at the start, during treatment and after the cessation of G therapy (1-6 months). The recordings were performed at the patient's bedside and did not require the active participation of the patient. The method is fast, non-invasive, accurate and does not request an ENT specialist. We included patients presenting OAE. Ototoxicity was defined using adapted ASHA (American Speech Hearing Association) criteria.

Results: We studied 49 patients (24/25 female/male), mean age of 37.24 (3-70 years old), who received G for 4-42 days: less than 10 days--24 patients and more than 10 days--25 patients. Hearing impairment was observed in 10 patients (20.4%) and appeared during the therapy or in the next 3 months, with a loss of one or 2 high frequencies that occurred unilaterally in 9 out of 10 cases. The acoustic damage correlates with the length of G treatment. We did not find a statistical correlation between the degree of impairment and the age, regimen or addition of another ototoxic drug, although the hearing loss was slightly higher in those with concomitant exposure.

Conclusion: OAE monitoring of AG treatment is a very useful way for detecting and preventing acoustic toxicity, because it could warn about hearing loss before damage of the conversational frequencies. The accuracy is similar to the classical methods, but it is faster and easier to perform.