Furosemide, 20 mg, was given intravenously as a bolus to seven patients with cirrhotic ascites and a 10-mg intravenous bolus dose was given to three normal subjects. Furosemide concentrations were measured by a specific high-performance liquid chromatographic analytic method. The median plasma elimination half-life (t1/2), volume of distribution at a steady state (VdSS), and VDarea of furosemide were 0.70 hr, 91 ml/kg, and 119 ml/kg in normal subjects. In the cirrhotic patients elimination t1/2 and Vd at steady state doubled and the Vdarea of furosemide was almost double that of the normal. There were no differences in plasma clearance or renal and nonrenal clearance between patients and controls, but five of the seven patients had lower renal clearances than controls. The water and sodium response to furosemide was directly related to the urinary furosemide excretion rate. We identified a subset of cirrhotic patients who responded poorly (125 ml/hr urinary output in the first 4 hr after furosemide compared to 300 ml/hr in the other patients and 400 ml/hr in the controls) to furosemide. These "poor responders" had the lowest renal clearance of furosemide and the lowest urinary furosemide excretion rates. Our data indicate that furosemide kinetics are altered in patients with cirrhotic ascites and lack of response in a subset of these patients is due to lack of delivery of furosemide to the renal site of its action.