Prevention of variceal rebleeding.

Recurrent variceal hemorrhage occurs in 50% to 80% of cirrhotic patients who survive a variceal bleeding episode. The aim of preventing rebleeding is to improve survival by reducing the mortality associated with rebleeding; however, although shunt surgery is the most effective treatment to prevent recurrent bleeding, it does not increase survival and is associated with an increased incidence of chronic portosystemic encephalopathy. The distal splenorenal shunt is associated with a reduced incidence of encephalopathy, compared with nonselective shunts, but the true magnitude and longevity of this effect is still controversial. beta-Blockers reduce the incidence of rebleeding, but the effect is modest, and there is little or no effect on mortality when compared with no treatment. Injection sclerotherapy reduces the incidence of rebleeding and improves survival when a schedule of both emergency and long-term injection is compared with no sclerotherapy. No technical variation of injection sclerotherapy has been shown to be superior to another. Endoscopic variceal banding may result in fewer complications but the efficacy is similar to that of injection sclerotherapy. Trials of long-term sclerotherapy versus beta-blockers show very similar mortality and rebleeding rates. Addition of beta-blockade to sclerotherapy does not confer any advantages when compared with sclerotherapy alone. Improvements in pharmacologic therapy, such as the addition of isosorbide mononitrate to propranolol, may in the future make drug therapy the first treatment option to prevent rebleeding. Shunt surgery is superior to sclerotherapy in preventing rebleeding and has a similar mortality; however, liver transplantation is technically more difficult in shunted patients, but shunts do not adversely affect overall survival after transplantation. There are few data to allow optimal selection of a particular therapy or sequence of therapies to prevent variceal rebleeding for any individual patient. This will need to be studied in large trials and is a major issue in the current clinical management of cirrhotics who have bled from varices.