Pharmacologic prevention of variceal bleeding and rebleeding.

A review of nine placebo-controlled studies for the prevention of first bleeding and 14 for the prevention of rebleeding in patients with cirrhosis and oesophageal varices indicates that beta-adrenergic antagonists significantly reduced the incidence of both initial and recurrent gastrointestinal bleeding over study durations ranging from 1 to 2 years, provided the patients complied with the regimen. Three meta-analyses concluded that beta-blockers also significantly reduced the risk of fatal bleeding, although this remains controversial. These agents were also effective in patients with portal hypertension from causes other than alcoholic cirrhosis, although not in hepatocellular carcinoma. Side effects occurred in 3-40% of patients and required discontinuation of beta-blocker administration in 5%. In two clinical trials in which beta-blocker therapy was compared with endoscopic sclerotherapy, the drug was at least as effective as sclerotherapy in preventing first episodes of variceal bleeding. In nine studies, the two modalities were comparably effective in preventing rebleeding. Used in combination, beta-blockers and sclerotherapy were more effective in preventing rebleeding than either used alone. However, neither treatment unequivocally prolonged survival relative to placebo.