Should portosystemic shunt be reconsidered in the treatment of intractable ascites in cirrhosis?

Fifty-seven cirrhotic patients with intractable ascites had a portosystemic shunt. In 35 patients, a peritoneovenous shunt had previously failed. Forty-six patients were in Pugh's class B and 11 were in class C. There were three operative deaths (5.3%). Fifty-three (98.2%) of the 54 survivors were cleared of ascites. In one patient, ascites persisted because of postshunt heart failure that resulted in a marked increase of caval pressure. Twenty-seven patients (50%) had late encephalopathy, which was severe and disabling in 12 (22%). One- and three-year survival rates were 72% and 36%, respectively. These results suggest that although portosystemic shunts are remarkably effective in dealing with ascites, the high rate of postoperative encephalopathy is a strong argument against their routine use in the management of intractable ascites in cirrhosis.