Twenty-six patients with the Budd-Chiari syndrome were treated surgically at the Johns Hopkins Hospital. Twenty-one of the patients were female and five were male, with a median age at diagnosis of 37 years. Nine patients had polycythemia vera, 6 were receiving estrogen therapy, 5 had a previous hepatitis A or B infection, and 4 had cirrhosis. There was one case each of hepatic malignancy, paroxysmal nocturnal hemoglobinuria, and idiopathic thrombocytopenic purpura. In five cases no etiologic factors or associated disorders were identified. Ascites was the most common presenting feature in this group of patients. Hepatic function at the time of diagnosis, as measured by standard serum chemistries, was only minimally abnormal. The diagnosis of the Budd-Chiari syndrome was confirmed in all 26 patients by hepatic vein catheterization. Inferior vena cavography was also performed and revealed caval occlusion in 4 patients, significant caval obstruction in 13 patients, and a normal vena cava in 9 patients. Interpretation of the vena cavogram was helpful in selecting the appropriate surgical procedure for each patient. Twenty-three of the twenty-six patients underwent percutaneous liver biopsy before operation, with no morbidity or mortality. Four patients had well-established cirrhosis noted on biopsy. Thirty mesenteric-systemic venous shunts were performed on the 26 patients. In 11 patients a mesocaval shunt was performed and in one instance conversion to a mesoatrial shunt was required as a second procedure. In 15 patients a mesoatrial shunt was performed as the initial procedure. Graft thrombosis occurring in 2 of these 15 patients prompted one revision in 1 patient and 2 revisions in the second patient. After mesenteric-systemic venous shunt, eight of the patients (31%) died before discharge from the hospital. The remaining 18 patients in this series were discharged from the hospital alive and well with patent shunts. Patients were followed for a median of 43 months (range, 9 months to 13 years). Five late deaths occurred between 5 and 84 months after the operation. Three- and five-year actuarial survival rates were 65% and 59%, respectively.
