The bioavailability, pharmacokinetics, and pharmacodynamics of torsemide (10 mg orally and intravenously) and furosemide (40 mg orally and 20 mg intravenously) were determined in a randomized crossover clinical trial in 16 patients with compensated congestive heart failure. Torsemide (time to reach maximum concentration [tmax], 1.1 +/- 0.9 hour) was more rapidly absorbed than furosemide (tmax, 2.4 +/- 2.5 hours), the absorption of which was delayed compared with that in healthy volunteers. Bioavailability of torsemide was also greater and less variable than that of furosemide. All four treatments yielded comparable changes from baseline in 24-hour electrolyte excretion. Based on the relationships between sodium excretion rate and fractional sodium and urinary drug excretion rate, response to both diuretic agents at the level of the nephron was decreased compared with previous studies with healthy subjects. Assessment of the clinical relevance, if any, of the difference in the variability of absorption warrants further study.