A rat model was utilized to compare the nephrotoxic potential of gentamicin and tobramycin. Gentamicin, 40 mg/kg per day, predictably produced renal failure and morphological evidence of proximal tubular necrosis over 14 days of treatment. An identical dosage of tobramycin was associated with only minimal morphological changes and normal concentrations of serum creatinine and blood urea nitrogen. Similar results were obtained even after the tobramycin dosage was tripled to 120 mg/kg per day. A decrease in urine osmolality, mechanism unknown, was observed in all aminoglycoside-treated rats, but the lowest osmolalities were found in the gentamicin-treated rats. According to both histological criteria and renal function measurements, gentamicin was more nephrotoxic than tobramycin in this animal model. This study was undertaken as a result of our interest in determining whether there were any important differences between the aminoglycoside antibiotics gentamicin and tobramycin. A review of the literature indicated only minor differences in in vitro antibacterial activity (1,6,8,14,20,31). Gentamicin is slightly more active against Serratia species, and tobramycin is slightly more active against isolates of Pseudomonas aeruginosa. In vivo animal studies of antibacterial activity again show similar degrees of activity, except that tobramycin is more active against P. aeruginosa infections (5,30). The pharmacokinetics of the two drugs in humans are felt to be virtually identical (9,12,22,28). Comparative toxicity studies of the effects of these two drugs in humans have not been reported.In a rat model, tobramycin was found to have a lower lethal dose than gentamicin (5). Tobramycin was found to have a lower lethal dose than gentamicin (5). Tobramycin was found to be less ototoxic than gentamicin in guinea pigs (2). Tobramycin was considered less nephrotoxic than gentamicin in rats on the basis of subtle but consistent histological changes and differences in drug-induced urinary lysosomal enzymes (12, 21). Since a clear difference in nephrotoxic potential would have clinical importance, studies were designed to more precisely detect any abnormalities in renal structure or function.Initially, a prospective randomized doubleblind study in humans was considered. A recent report describes such a study comparing gentamicin and amikacin (26). However, it appears virtually impossible to determine the incidence of aminoglycoside-induced nephrotoxicity in humans. There is no way for the investigator to separate disease-induced toxicity from drug-induced toxicity. Also, preillness measurements of renal structure and function are rarely available. Because of these concerns, we chose to use a rat model of aminoglycoside nephrotoxicity. Nephrotoxicity was quantitated by histological changes and alterations in renal fimction. The results demonstrate that, in this model, tobramycin is less nephrotoxic than gentamicin.MATERIALS AND METHODS Animals. Adult male Fischer 344 rats, weighing 175 to 275 g, were used. The rats were kept sing...