Sclerotherapy after first variceal hemorrhage in cirrhosis. A randomized multicenter trial

At the time of their first hemorrhage from esophageal varices, 187 unselected patients with cirrhosis were randomly assigned to medical treatment including balloon tamponade or to medical treatment supplemented with intensive, paravariceal sclerotherapy. Follow-up period ranged from 9 to 52 months. The overall mortality in the sclerotherapy group (hazard) was 76 per cent (95 per cent confidence limits, 54 to 107 per cent) of that in the medical-regimen group. The relative mortality in the sclerotherapy group, determined by stratifying according to degree of encephalopathy and ascites, was 63 per cent of that in the medical-regimen group (95 per cent confidence limits, 44 to 91 per cent). The main effect of sclerotherapy may be a reduction of long-term mortality, which after Day 40 was only 43 per cent (95 per cent confidence limits, 23 to 79 per cent) of that in the medical-regimen group. Sclerotherapy had no significant influence on the initial hemorrhage as judged from the duration of bleeding or of balloon tamponade, the number of blood transfusions needed, or immediate mortality. Forty-five patients in the sclerotherapy group had 64 episodes of recurrent hemorrhage, as compared with 138 episodes among 51 patients in the medical-regimen group. This difference was due to a pronounced reduction of rebleeding after Day 40 in the sclerotherapy group. We recommend sclerotherapy for patients with cirrhosis who have bleeding from esophageal varices.