Hemorrhage from esophageal varices is a life-threatening event in patients with liver cirrhosis. About 40% to 80% of the patients surviving the first bleeding suffer from a recurrence of variceal bleeding within one year. This high recurrence rate substantially contributes to the mortality in patients with liver cirrhosis. Therefore, various treatment regimens both in primary and secondary prophylaxis were studied. Most experience in medical primary prophylaxis was collected with beta-blockers, mainly propranolol. Treating patients with esophageal varices with propranolol significantly reduces the incidence of first variceal bleeding. However, the effect on mortality is marginal, and primary prophylaxis is generally not recommended in these patients. Several studies support the hypothesis, that medical prophylaxis with beta-blockers is more effective in reducing the rate of first esophageal bleeding in patients with a high risk of hemorrhage, such as the presence of very large varices with red spots. A score to assess the individual risk of a given patient to suffer a variceal bleeding would be helpful. As long as such a score is not validated, no general rule for this treatment decision can be given. In secondary prophylaxis, both administration of beta-blockers and endoscopic therapy (sclerotherapy or ligation of the varices) are effective in significantly lowering the rate of re-bleeding. However, the effect on mortality was not significant in most studies. Several studies comparing the efficacy of medical prophylaxis and endoscopic treatment showed advantages of the endoscopic therapy with a greater reduction in recurrent bleeding episodes. However, medical prophylaxis with beta-blockers has the important advantage of being immediately effective whereas endoscopic procedures provide the best protection against recurrent bleeding after complete obliteration of the varices. Therefore, in the first weeks and months of endoscopic therapy, the additional treatment with beta-blockers may further reduce the risk of re-bleeding. Only half of all studies on this topic reported a significant advantage of such a combined therapy. Therefore, it seems reasonable to restrict this approach to patients with a high risk of re-bleeding such as patients with large sclerotherapy-derived esophageal ulcers.