New Prognostic Predictive Models of Mortality of Decompensated Cirrhotic Patients Waiting for Liver Transplantation

The MELD score is a predictive model of cirrhosis mortality used in France since 2007 to prioritize access to liver transplantation for patients enrolled in the national waiting list. The predictive value of this score was recently revised downward with a C index of the order of 0.65-0.67 and 20% of the patients enrolled for decompensated cirrhosis have access to liver transplantation by a subjective system of "expert component" independent of the MELD because of this lack of precision. The use of the MELD score to individually define access to the transplant should so be reconsidered. Recently new predictive models of cirrhosis mortality better than MELD have been developed and new mortality predictors independent of MELD have been published. The goal of this study is to design prognostic predictive models of mortality for decompensated cirrhotic patients enrolled on the national liver transplant waiting list including known (MELD, MELD Na) as more recent (CLIF-C AD, CLIF - CACLF) predictive models and new objective predictors studied in combination in order to optimize the system of allocation of hepatic allografts in France. The expected benefits of this search are twofold: * At the individual level: The possibility for patients at high risk of death but with intermediate MELD score to be transplanted. * Public health plan: * Improving the equity of graft allocation system. * Decreased mortality in the waiting list by improving the fairness and efficiency of the graft allocation system, a major public health issue * An ancillary study to the SUPERMELD study is also proposed, the miR MELD study, whose main objective is to evaluate the value of plasma miRNAs in a cohort of patients with decompensated cirrhosis (acute and chronic, excluding cancer) listed for liver transplantation to predict 3-month mortality on the liver transplant waiting list or drop-out from the waitlist for being too sick. Additional data collection of the vital status 1 year after tra