Qualitative overview of randomized trials of aminoglycosides.

A qualitative overview was conducted of randomized clinical trials comparing two or more of the following aminoglycoside antibiotics--amikacin. gentamicin, netilmicin, sisomicin, and tobramycin--with respect to efficacy, nephrotoxicity, or auditory toxicity. Design features considered included number of participants in each treatment arm, proportion of randomized participants available for analysis, blinding of evaluators, and report of statistically significant results. The proportion of participants eligible for analysis was less than or equal to 75 percent for 66 percent of the comparisons regarding nephrotoxicity, 43 percent regarding efficacy, and 19 percent regarding auditory toxicity. Twenty-nine percent of the trials reported that evaluators were blinded to treatment group assignments. The results of most trials showed no significant differences, although sample sizes were, in general, smaller than necessary to detect moderate differences. Quantitative overviews may be a useful means of detecting real differences in risk that are not apparent from the results of several individual small trials.