MELD and other predictors of survival after liver transplantation

Abstract

BACKGROUND: This study examined how reliable is the pre-transplant model for end-stage liver disease (MELD) score in predicting post-transplantation survival and analyzed variables associated with patient survival. METHODS: A cohort study was conducted. Receiver operating characteristic curve c-statistics were used to determine the ability of MELD score to predict mortality. The Kaplan-Meier (KM) method was used to analyze survival as a function of time regarding the MELD score and Child-Turcotte-Pugh (CTP) category. The Cox model was employed to assess the association between baseline risk factors and mortality. RESULTS: Recipients and donors were mostly male, with a mean age of 51.6 and 38.5 yr, respectively (n = 436 transplants). The c-statistic values for three-month patient mortality were 0.60 and 0.61 for MELD score and CTP category, respectively. KM survival at three, six and 12 months were lower in those who had a MELD score > or =21 or were CTP category C. Multivariate analysis revealed that recipient age > or =65 yr, MELD > or = 21, CTP C category, bilirubin > or = 7 mg/dL, creatinine > or = 1.5 mg/dL, platelet transfusion, hepatocellular carcinoma, and non-white color donor skin were predictors of mortality. CONCLUSIONS: Severe pre-transplant liver disease, age > or = 65, non-white skin donor, and hepatocellular carcinoma are associated with poor outcome.