We measured the hemodynamic effects of intravenous vasopressin, ketanserin (a 5-hydroxytryptamine-2 receptor blocker), and vasopressin plus ketanserin in 33 patients with hepatitis B-related cirrhosis. Thirteen patients received vasopressin alone (0.66 units/min), ten patients ketanserin alone (10 mg), and ten patients vasopressin followed by vasopressin plus ketanserin. Vasopressin alone reduced the hepatic venous pressure gradient (from 18 +/- 5, mean +/- S.D., to 9 +/- 3 mmHg, p less than 0.0001) and cardiac output (p less than 0.0001), but increased mean arterial pressure (p less than 0.005), mean pulmonary arterial pressure (p less than 0.0001), pulmonary capillary wedge pressure (p less than 0.0001), and systemic vascular resistance (p less than 0.001). There was no significant change in heart rate. Ketanserin alone produced a significant fall in the hepatic venous pressure gradient (from 16 +/- 4 to 13 +/- 3 mmHg, p less than 0.0001), mean arterial pressure (p less than 0.005), mean pulmonary arterial pressure (p less than 0.005), and pulmonary capillary wedge pressure (p less than 0.005). Heart rate, cardiac output, and systemic vascular resistance were not significantly changed. The addition of ketanserin to vasopressin corrected most of the systemic hemodynamic disturbances produced by vasopressin. This combination did not lead to a further reduction in the hepatic venous pressure gradient. We conclude that intravenous ketanserin reduces portal pressure in patients with hepatitis B-related cirrhosis. The addition of ketanserin to vasopressin improves the detrimental systemic hemodynamic effects of vasopressin without further reducing the portal pressure.