The metabolic fate of furosemide was studied in kidney transplant patients after oral and intravenous administration of the diuretic at therapeutic doses. Serial urine samples were collected over a 24 h period and furosemide was analyzed by a specific high performance liquid chromatographic method using fluorescence detection. We found no evidence of the putative furosemide metabolite, 2-amino-4-chloro-5-sulfamoylanthranilic acid (CSA), in any of the samples analyzed. The amount of furosemide excreted as the glucuronide metabolite accounted for 8% of the available dose, whether administered orally or by intravenous infusion. In addition, the significant positive correlation observed between the percent of the available dose excreted as furosemide glucuronide and the renal clearance of furosemide (r = 0.581, p less than 0.02) suggests that the glucuronidation process for furosemide may be occurring in the kidney. Furosemide and its glucuronide metabolite accounted for only 45% of the intravenous dose recovered in the urine. Biliary excretion of unchanged furosemide and/or furosemide glucuronide into the feces probably accounts for the remainder of the dose not recovered.