Controlled trials of sclerotherapy for the prevention of the first variceal hemorrhage in cirrhotics have given conflicting results, in spite of an initial positive controlled trial. We designed therefore a new study in which only 89 of 396 investigated patients were randomized to sclerotherapy (44 patients) or a control group (45 patients). The admission criteria were: no history of variceal bleeding, the presence of high risk varies, i.e., varices of degrees III and IV with minivarices on the surface of them, and portal pressure over 16 mmHg. Sclerotherapy sessions were performed at 0, 7, 14, 21, and 28 days, until the varices were reduced in size and completely covered by fibrous tissue. Follow-up endoscopy was performed at four-month and thereafter at six-month intervals. The control patients underwent repeated clinical investigation and endoscopy at six-month intervals. Bleeding episodes were treated by emergency endoscopic sclerotherapy in both groups, whenever possible. The mean follow-up was 33 months. The results were analyzed using Student's t-test and the log-rank test. Variceal bleeding occured in 11 sclerotherapy patients (25%) and 34 controls (75.6%) (p < 0.05). The overall mortality was 25% (11 patients) among the sclerotherapy patients and 69% (31 patients) in the controls (p < 0.01). Prophylactic endoscopic sclerotherapy was able to prolong survival in Child-Pugh classes A and B, but not in C. It is concluded that prophylactic endoscopic sclerotherapy does reduce the incidence of first variceal bleeding in cirrhotic patients, and is able to prolong survival if only high-risk patients are selected and the treatment is performed by endoscopic experts.