Patients who have survived hemorrhage from esophageal varies have a 50-70% risk of rebleeding within a 2 year period. The mortality for each bleeding episode is from 30 to 50% depending on how the patients are selected. The present therapeutic proposal will compare in a controlled clinical trial the three most promising, current forms of therapy (portal-systemic shunt surgery, endoscopic sclerotherapy and beta-adrenergic blockade with propranolol) for the prevention of recurrent variceal hemorrhage. In controlled clinical trials each of these therapies has been shown to be superior to conventional, non-specific therapy in reducing the frequency of recurrent hemorrhage and in decreasing mortality. There are, however, no published controlled investigations that compare any two of these three forms of therapy with each other. The present investigation is undertaken to compare in a prospective, randomized, controlled clinical trial these three treatments in patients who have survived the initial episode of variceal hemorrhage. We will compare portal-systemic anastomoses (primarily distal splenorenal anastomoses) with endoscopic sclerotherapy (a series of three to five injection sessions) and with propranolol (dosage determined by a measured decease in portal pressure after trial propranolol administration). Patients who fail to satisfy admission criteria for one of the three treatments will be randomized to the other two treatments. Thus, there will be one three-armed investigation and, in addition, three two-armed subinvestigations which will be conducted and analyzed separately. A prior investigation comprising portal-systemic and distal splenorenal shunts by this same group of investigators at the same hospitals provides a reliable estimate of the number so patients admitted and eligible for these procedures. Although it may not be possible to demonstrate statistically significant differenes in survival, we will be able to show differences in the frequency of recurrent hemorrhage from varices, in transfusion requirements, in the complications of therapy, in the cost of therapy, and in the quality of life for the three treatments under investigation.
