An ex vivo study on adenosine triphosphatase (ATPase) activities of rabbit renal proximal tubules was conducted with a new cephalosporin, cefpirome (HR 810), a positive control, cephaloridine, and a reference third-generation cephalosporin, cefotaxime. Compared with controls, CPH caused a significant time-dependent decrease in ATPase activities [12%, 2 hr after treatment (P < 0.01) and 75%, 48 hr after treatment (P < 0.001)]. This decrease was accompanied by a significant loss in the energy charge of the adenylate pool [27%, 2 hr after treatment (P < 0.001)]. Neither cefotaxime nor cefpirome caused such decreases. The results confirmed those of a previously published in vitro study. The advantages and disadvantages of these two experimental procedures as predictive models for nephrotoxicity are discussed.