Univariate and multivariate analyses of risk factors predisposing to auditory toxicity in patients receiving aminoglycosides

Abstract

Risk factors predisposing to auditory toxicity of aminoglycosides were analyzed from records of 187 patients enrolled in three prospective randomized trials comparing the toxicity of netilmicin, tobramycin, and amikacin. Patients were eligible if they received three or more days of therapy and at least two serial audiograms were available. The overall auditory toxicity rate was 9.6% (18 of 187). Auditory toxicity was detected in 4.4, 10.8, and 23.5% of patients given netilmicin, tobramycin, and amikacin, respectively (P = 0.05). In the univariate analysis, patients who developed auditory toxicity were significantly older (P = 0.01) and had a significantly higher (P = 0.04) percentage of trough levels of netilmicin or tobramycin above 2 mg/liter or amikacin above 5 mg/liter. In the final logistic regression model, only age was retained as independently influencing the development of auditory toxicity (P less than 0.00001). Conversely, factors that did not add significantly to the prediction of auditory toxicity were aminoglycoside serum levels, total aminoglycoside dose, duration of therapy, sex, peak temperature, presence of bacteremia, shock, liver cirrhosis, dehydration, previous otic pathology or renal failure, and development of renal toxicity. At least in certain populations, age is the most important predisposing factor for the development of auditory toxicity in patients receiving aminoglycosides.