Beta-adrenergic-antagonist drugs in the prevention of gastrointestinal bleeding in patients with cirrhosis and esophageal varices. An analysis of data and prognostic factors in 589 patients from four randomized clinical trials. Franco-Italian Multicenter Study Group

Abstract

Background: The value of beta-adrenergic-antagonist drug therapy for the prevention of initial episodes of gastrointestinal bleeding in patients with cirrhosis and esophageal varices is uncertain, both positive and negative study results having been reported.

Methods: In this study, we analyzed data on individual patients from four randomized, controlled trials to assess the efficacy of this treatment. Of the 589 patients studied, 286 received a beta-adrenergic-antagonist drug (propranolol in 203 and nadolol in 83) and 303 received placebo.

Results: After two years, the mean (+/- SE) percentage of patients who had had no upper gastrointestinal bleeding was 78 +/- 3 percent in the beta-adrenergic-antagonist treatment group and 65 +/- 3 percent in the control group (P = 0.002). The percentage of patients without fatal bleeding was 90 +/- 2 percent in the treatment group and 82 +/- 3 percent in the control group (P = 0.01). The percentage of patients surviving after two years was 71 +/- 3 percent in the treatment group and 68 +/- 3 percent in the control group (P = 0.34). After age and severity of cirrhosis were taken into account, the survival rate was better in the treatment group (P = 0.09). The percentage of surviving patients who had had no bleeding after two years was 62 +/- 3 percent in the treatment group and 53 +/- 3 percent in the control group (P = 0.04). Both propranolol and nadolol prevented a first episode of bleeding. Severe cirrhosis and especially the presence of ascites were associated with bleeding (P less than 0.001) and death (P less than 0.001) in both groups. The efficacy of beta-adrenergic-antagonist therapy in the prevention of bleeding (P less than 0.001) and of fatal bleeding (P = 0.004) and in the prevention of bleeding or death (P = 0.005) was the same after adjustment for cause and severity of cirrhosis, ascites, and size of varices.

Conclusions: Propranolol and nadolol are effective in preventing first bleeding and reducing the mortality rate associated with gastrointestinal bleeding in patients with cirrhosis, regardless of severity.