The results of a prospective, randomized controlled trial of chronic esophageal variceal sclerotherapy conducted over a 38-month period are presented. One-hundred twenty patients were randomized following variceal bleeding, 63 to esophageal variceal sclerotherapy and 57 to control. Mean follow-up was similar in both groups (esophageal variceal sclerotherapy, 12.5 +/- 8.8 months; control, 14.9 +/- 6.6 months). Twenty-one percent of the patients in each group were lost to follow-up. Esophageal variceal sclerotherapy decreased rebleeding as evidenced by a decrease in the mean bleeding risk factor, transfusion requirement and by an increase in bleeding free interval; differences between the treated and control groups in these parameters were especially significant after variceal obliteration. A high incidence of asymptomatic ulceration and low frequency of strictures were notable effects of esophageal variceal sclerotherapy. Cumulative life table analysis revealed no differences in survival between esophageal variceal sclerotherapy and control groups. However, when patients who received portal-systemic shunt surgery (esophageal variceal sclerotherapy, 6%; control, 28%) were removed from the analysis at the time the shunt surgery was performed (defining the shunt as an endpoint, a significant difference in survival (p less than 0.05, F ratios) in favor of esophageal variceal sclerotherapy was observed.