The predictive value of the revised model for end-stage liver disease (MELD) in the clinical early stage after liver transplantation / 中华肝胆外科杂志

ABSTRACT

Objectives To evaluate the predictive value of the revised model for end-stage liver disease in the clinical early stage after liver transplantation.Methods The clinical data of 218 patients were retrospectively analyzed.After calculating the MELD score,ReFit MELD score and ReFit MELDNa score before transplantation,we compared the predictive accuracies of these scoring systems using the area under curve (AUC) of the receiver operating characteristic.The groups were categorized with the cut-offs of the MELD,ReFit MELD and ReFit MELDNa,and the early-stage complications and mortality in the different groups were analyzed.Results The AUC for the MELD,ReFit MELD and ReFit MELDNa were 0.737 (95％CI 0.621～0.854),0.727 (95％CI 0.663～0.785) and 0.735 (95％CI 0.671～0.792),respectively.There was no statistical difference is the AUC among the MELD,ReFit MELD and ReFit MELDNa.Elevated scores in the 3 models predicted higher rates of pulmonary infection,abdominal infection and acute renal dysfunction,as well as a higher mortality.Conclusions The ReFit MELD score and ReFit MELDNa score were relatively useful predictors of short-term survival rates after liver transplantation.The predictive accuracy was similar to the MELD score.Values of the score above the cutoff values indicated higher rates of complication and poorer prognosis.