Development of bleeding gastroesophageal varices is a serious consequence of portal hypertension secondary to cirrhosis. Nonselective beta-blockers have been used to reduce portal pressures and prevent primary and secondary bleeding episodes. However, up to two thirds of patients may not respond appropriately to these agents. Nonselective beta-blockers combined with vasodilatory drugs result in enhanced lowering of portal pressures by targeting several mechanisms involved in this process. Unfortunately, this practice is associated with increased adverse effects, such as hypotension, and minimal reductions in mortality. Carvedilol possesses both nonselective beta-antagonist and alpha1-receptor antagonist activity. Given its combined mechanism of action, carvedilol presents a potential option for lowering portal pressures. Its effects on lowering portal pressures and its role in therapy are undefined. Using MEDLINE (1966-2003) and International Pharmaceutical Abstracts (1970-2003), the English-language literature was searched to identify human studies assessing carvedilol's effects on lowering portal pressure. In general, carvedilol therapy was associated with mean reductions of 16-43% in portal pressure, assessed by the hepatic venous pressure gradient (HVPG) after single and multiple doses. Studies comparing carvedilol with propranolol revealed equal or enhanced efficacy in lowering HVPG. Large percentages of patients had significant HVPG reductions to levels that prevent variceal bleeding. Carvedilol also was associated with substantial symptomatic hypotension, especially in patients with ascites or Child-Pugh class B or C cirrhosis. Efficacy and adverse effects generally seem to be dose related. Carvedilol appears to be a potentially viable option for treating portal hypertension. Further multiple-dose trials comparing carvedilol with standard therapy are needed to assess the agent's long-term safety and effectiveness in preventing variceal bleeding.