Controlled trial of vasopressin and balloon tamponade in bleeding esophageal varices

Abstract

In a randomized controlled trial, the effect of continuous intravenous administration of vasopressin was compared with Sengstaken-Blakemore balloon tamponade in 37 episodes of bleeding esophageal varices in patients with cirrhosis. The majority were Group A and B of Child's classification. Bleeding was controlled in 11 of 17 (65%) patients on vasopressin and in 14 of 20 (70%) patients on tamponade. The patients who failed to respond initially (6 episodes on vasopressin and 5 on tamponade) were treated successfully with the alternative method. Overall mortality was similar in both groups: 3 patients in the vasopressin group and 4 in the tamponade group died. Only one patient died of uncontrolled bleeding; 4 patients probably died of complications of treatment, 2 of cardiac ischemia after vasopressin and 2 of pulmonary infection after tamponade. The vasopressin group required significantly fewer blood transfusions than did the tamponade group.