BACKGROUND: The model for end-stage liver disease is the best available predictor of waiting list mortality among liver transplant candidates. OBJECTIVES: To validate the applicability of MELD in Israel. METHODS: All candidates awaiting liver transplantation in our institution were followed prospectively since 2002. We measured the concordance (c-statistic) equivalent to the area under the receiver operating characteristic curve in order to assess the predictive power of MELD. Other independent mortality risk factors were identified by a separate multivariate analysis. Mortality rates within different MELD and Child-Pugh-Turcotte scores were compared to the original (United States) MELD data. RESULTS: Of 86 patients listed for transplantation, 40 were transplanted (36 in Israel and 4 abroad). Of the other 46 patients, 24 are alive and still listed, and 22 died (25%, approximately 7%/year). The area under the ROC curve for MELD score was 0.79 (0.83 USA) compared to a CPT score of 0.71 (0.76 USA). High MELD scores, occurrence of spontaneous bacterial peritonitis, and diagnosis of hepatocellular carcinoma were independent risk factors of mortality. Death rates per mid-MELD score (20-29) were significantly higher than the USA results. CONCLUSIONS: MELD is valid in Israel and superior to CPT in predicting waiting list mortality. Although longer waiting time due to organ scarcity is a key factor, death rates in the mid-range (10-29) MELD groups indicate further audit of the care of patients with end-stage liver disease.
BACKGROUND: The model for end-stage liver disease is the best available predictor of waiting list mortality among liver transplant candidates. OBJECTIVES: To validate the applicability of MELD in Israel. METHODS: All candidates awaiting liver transplantation in our institution were followed prospectively since 2002. We measured the concordance (c-statistic) equivalent to the area under the receiver operating characteristic curve in order to assess the predictive power of MELD. Other independent mortality risk factors were identified by a separate multivariate analysis. Mortality rates within different MELD and Child-Pugh-Turcotte scores were compared to the original (United States) MELD data. RESULTS: Of 86 patients listed for transplantation, 40 were transplanted (36 in Israel and 4 abroad). Of the other 46 patients, 24 are alive and still listed, and 22 died (25%, approximately 7%/year). The area under the ROC curve for MELD score was 0.79 (0.83 USA) compared to a CPT score of 0.71 (0.76 USA). High MELD scores, occurrence of spontaneous bacterial peritonitis, and diagnosis of hepatocellular carcinoma were independent risk factors of mortality. Death rates per mid-MELD score (20-29) were significantly higher than the USA results. CONCLUSIONS: MELD is valid in Israel and superior to CPT in predicting waiting list mortality. Although longer waiting time due to organ scarcity is a key factor, death rates in the mid-range (10-29) MELD groups indicate further audit of the care of patients with end-stage liver disease.