In clinical practice, the patient-to-patient variability in furosemide response is a well known fact. To study whether changes in furosemide metabolism may contribute to this variability, 72 rats received furosemide in a dose of 5 mg/kg i.p. There was a good correlation between the percentage of furosemide recovered in the urine and the diuretic effect. Accordingly, the rats were separated into groups corresponding to good responders (GR) and poor responders (PR) based on the extent of diuresis and the percentage of furosemide recovered in the urine. In an in vivo experiment, 12 rats received 30 mg/kg of chloramphenicol i.p. 30 min before furosemide. In PR rats, the average percentage of furosemide (+/- S.D.) in urine increased from 24.1 +/- 6.3 to 67.1 +/- 8.9 (P less than .001) and urine volume increased from 13.5 +/- 4.6 to 28.6 +/- 3.5 ml/24 hr (P less than .005). No such changes were observed in the group of GR rats. In a subsequent experiment, 11 GR rats received 80 mg/Kg of phenobarbital i.p. for 5 days before furosemide. The percentage of furosemide recovered in the urine decreased significantly (53.4 +/- 10.2 to 39.8 +/- 7.2, P less than .01), although the diuretic effect was not modified. In vitro studies confirmed that the T 1/2 of furosemide biotransformation in liver was different in GR and PR rats: 63.6 +/- 5.5 vs. 43.5 +2- 2.8 min, respectively (P less than .001). After receiving chloramphenicol, PR rats showed a longer T 1/2 of biotransformation in liver of 54.8 +/- 6.4 min (p less than .05 when compared to PR controls). After the administration of phenobarbital, GR rats showed a short T 1/2 of 35.3 +/- 7.0 min (P less than .005 when compared to GR controls). It is concluded that in the rat, the variability in diuretic responses to furosemide is related to the capacity of the individual rat to metabolize furosemide and that furosemide metabolism may be inhibited by chloramphenicol and increased by phenobarbital.