The objective of this study was to predict postoperative liver failure and morbidity after hepatectomy for hepatocellular carcinoma (HCC) with cirrhosis. The model for end-stage liver disease (MELD) score is currently accepted as a disease severity index of cirrhotic patients awaiting liver transplantation; however, its impact on prognosis after resection of HCC on cirrhosis has never been investigated. One hundred fifty-four cirrhotic patients resected in a tertiary care setting for HCC were retrospectively analyzed. For each patient, the MELD score was calculated and related to postoperative liver failure and complications (morbidity). Hospital stay and 1-year survival was also investigated. MELD accuracy in predicting postoperative liver failure and morbidity of cirrhotic patients was assessed using receiver operating characteristic (ROC) analysis. Eleven patients (7.1%) experienced postoperative liver failure leading to death or transplantation. ROC analysis identified cirrhotic patients with a MELD score equal to or above 11 at high risk for postoperative liver failure (area under the curve [AUC] = 0.92, 95% confidence interval [CI] = 0.87-0.96; sensitivity = 82%; specificity = 89%). Forty-six patients (29.9%) developed at least 1 postoperative complication: ROC analysis identified patients with a MELD score equal to or above 9 at major risk for postoperative complications (AUC = 0.85, 95% CI = 0.78-0.89; sensitivity = 87%; specificity = 63%). Cirrhotic patients with MELD score below 9 had no postoperative liver failure and low morbidity (8.1%). In conclusion, the MELD score can accurately predict postoperative liver failure and morbidity of cirrhotic patients referred for resection of HCC and should be used to select the best candidates for hepatectomy. Liver Transpl 12:966-971, 2006. © 2006 AASLD.