Torasemide is a new loop diuretic with a longer half-life and longer action than furosemide in healthy subjects. In order to evaluate the pharmacodynamic effects, single oral doses of furosemide (80 mg) and torasemide (20 mg), which were equipotent in healthy subjects, were given to 14 patients with cirrhosis and ascites. Before the study patients underwent an equilibration period of 4 days without diuretics. The drugs were alternated following a randomized double-blind cross-over design after a wash-out period of at least 2 days. Urine was collected at defined intervals for 24 h after drug administration and blood samples were taken before, 6 h and 24 h after medication. Torasemide induced greater cumulative 24 h diuresis (2863 ± 343 vs. 2111 ± 184 ml, p < 0.01) than furosemide. Torasemide did not differ from furosemide for cumulative 0–6 h sodium excretion (96 ± 17 vs. 92 ± 23 mmol sodium) but caused a more pronounced cumulative 6–24 h natriuresis (38 ± 11 vs. 17 ± 4 mmol, p < 0.05). Five patients exhibited a weak response to furosemide (0–36 mmol sodium/24 h, median 24 mmol; 690–1460 ml urinary volume/24 h, median 1325 ml). These patients showed significantly higher natriuresis and diuresis following torasemide (26–136 mmol sodium/24 h, median 78 mmol, p < 0.05; 1670–3610 ml urinary volume/24 h, median 2200 ml, p < 0.05). Twenty-four hours after administration of both drugs there were no significant changes in hemodynamic, renal or hormonal parameters. No adverse effects were noted with either treatment. These findings suggest that torasemide might be more advantageous than furosemide in the treatment of ascites due to cirrhosis.