Twelve patients with cirrhosis (seven mild and five severe) were administered intravenous and oral furosemide in random order to assess its absorption and disposition. Total serum clearance (113 +/- 49 ml/min), volume of distribution (11.9 +/- 4.5 L), and elimination half-life (166 +/- 149 minutes) were similar to those reported previously in both healthy control subjects and patients with cirrhosis. Bioavailability of 58% +/- 17% (range, 37% to 82%) was comparable to that of previous studies, and there was no difference between patients with mild and those with severe cirrhosis. In 9 of 12 patients the mean absorption time was longer than the mean residence time determined after intravenous administration (mean for all patients, 203 +/- 86 versus 134 +/- 101 minutes; p less than 0.05), indicating that furosemide followed a "flip-flop" model in these patients. In all patients the mean absorption time was prolonged relative to normal subjects irrespective of the presence of edema. As such, the slower absorption of furosemide in edematous states, such as congestive heart failure and cirrhosis, does not appear to be a consequence of edema per se. Moreover, because similar changes occur in patients with congestive heart failure, it seems that diseases with diverse pathophysiology can slow furosemide absorption.