One hundred and six episodes of bleeding from esophageal or gastric varices in 72 patients with cirrhosis of the liver were randomized to treatment either with intravenous terlipressin 2 mg initially and 1 mg every four hours for 24 hours together with bolus injection and continuous infusion of placebo, or with somatostatin 250 micrograms as a bolus and continuous infusion of 250 micrograms/h somatostatin for 24 hours and placebo injections. Standard treatment with transfusions, fluid and electrolyte correction, and lactulose was administered in both groups. In the terlipressin group, 48 out of 53 bleeding episodes (91%) and in the somatostatin group 43 out of 53 bleeds (81%) were initially stopped by the vasoactive drugs. Four of the five bleeds not arrested by terlipressin, and nine of the ten bleeds not arrested by somatostatin, were stopped by balloon tamponade. In one patient in each group variceal bleeding could not be stopped initially, and both patients died. The failure rate of the vasoactive treatment alone, including rebleeds within the study period, was 17% in the terlipressin, and 28% in the somatostatin, group. The initial hemostasis, including balloon tamponade, were 98%, and the definitive bleeding control rates were 89% in both groups. The hospital mortality rate was 21% (11/53) in the terlipressin, and 21% (11/53) in the somatostatin, group. Blood transfusions and duration of bleeding did not differ significantly. The study indicates that a large proportion of bleeds from esophageal and fundic varices can be stopped initially (86%) and definitively controlled (77%) by vasoactive drugs alone.