The preliminary analysis of a multicentre, randomised, single-blind trial of propranolol for prophylaxis of first bleeding in cirrhosis is reported. 174 consecutively-chosen patients with large oesophageal varices were randomly assigned to either propranolol, in doses which reduced the resting heart rate by 25% (85 patients), or to vitamin K (89 patients). 25 patients had to be withdrawn from treatment with propranolol because of poor tolerance. The 30-month cumulative proportion of patients free of bleeding was 74% in the propranolol group and 63% in the vitamin K group; corresponding survival figures were 59% and 74%, respectively. These differences were not statistically significant. A retrospective analysis, according to the presence of ascites at randomisation showed that a significantly higher proportion of patients without ascites in the propranolol group were free of bleeding compared with those in the control group (87% vs 64%; p = 0.023). No significant differences were found in patients with ascites at randomisation. Length of survival was not significantly affected by treatment in any subgroup, although it was shorter in ascitic patients given propranolol than in controls (33% vs 63%; p = 0.07). If confirmed on a longer follow-up, these results suggest that propranolol could prevent primary variceal haemorrhage in patients with well-compensated cirrhosis.