Model for end-stage liver disease (MELD) exception for Budd-Chiari syndrome

Abstract

The Budd-Chiari syndrome (BCS) is an unusual clinical scenario resulting from obstruction to hepatic venous outflow. BCS manifests as ascites, hepatomegaly, and abnormalities in the liver biochemical tests. 1 The clinical presentation of BCS can vary from indolent, slowly progressive liver disease to advanced symptoms of cirrhosis and portal hypertension and, rarely, fulminant liver failure. Liver transplantation (LT) has been shown to be an effective therapy in patients with BCS. However, only a small portion of patients ultimately require transplantation. Patients with advanced symptoms such as ascites, hepatorenal syndrome, encephalopathy, and/or coagulopathy related to liver disease may be the best candidates for LT. A recent publication by Murad and colleagues 2 used the “Rotterdam score” to define those patients at risk for 5-year mortality. Multivariate analysis of 237 patients with BCS demonstrated 4 variables with prognostic value: encephalopathy (risk ratio of 3.58), ascites (risk ratio of 3.83), prothrombin time 2.3 seconds (risk ratio of 2.05), and hyperbilirubinemia (a continuous variable with a risk ratio of 1.004). A linear prognostic formula was developed as follows: encephalopathy (1.04 ascites) (0.72 PT) (0.004 bilirubin) 1.3 Ascites and encephalopathy were scored as present (score 1) or absent (score 0), and prothrombin time (PT) as 2.3 seconds (score 1) or 2.3 seconds (score 0). Bilirubin was included as a continuous variable. Three classes of patients were identified: class I, representing a total score of 0 to 1.1; class II, representing a total score of 1.1 to 1.5; and class III, representing a total score of 1.5. The 5-year survival rates for patients in these 3 groups were 89% for class I, 74% for class II, and 42% for class III patients. This study also demonstrated a trend toward improvement in survival in class II patients who underwent a transjugular interahepatic portosystemic shunt. In reference to LT, the data provide evidence that patients who fall into class III with evidence of advanced liver disease in the form of encephalopathy, ascites, or advanced coagulopathy would be candidates for LT. Class II patients have equivalent results with transplant- or nontransplant therapy (i.e., surgical, shunt or transjugular intrahepatic portosystemic shunt, and anticoagulation therapy). In this large group of patients selected over a 17-year period from several international centers, approximately 20% of the 237 patients studied fell into class III. These patients appear to have a risk-benefit ratio that favors LT over conservative therapy. However, the study addressed only 5-year mortality and did not measure 3-month mortality, which is used in the Model for End-Stage Liver Disease (MELD)-based allocation system for LT. Other smaller studies have provided data that supports the recommendation that only patients with advanced decompensated liver disease would benefit from LT. 3,4 The inherent weakness with the Rotterdam scoring system is the inclusion of the subjective variables of encephalopathy and ascites. Initial treatment of BCS is with diuretics and anticoagulation with intervention, either radiologic or surgical, being carried out only in patients with treatment failure defined as worsening liver function, increasing ascites, hepatorenal syndrome, or hepatic encephalopathy. Surgical options provide excellent long-term outcomes in selected patients 5 ; anticoagulation 6 and tran