A simple scoring model for predicting early graft failure and postoperative mortality after liver transplantation

Introduction and objectives: Graft failure and postoperative mortality are the most serious complications after liver transplantation. The aim of this study is to establish a prognostic scoring system to predict graft and patient survival based on serum transaminases levels that are routinely used during the postoperative period in human cadaveric liver transplants. Patients and methods: Postoperative graft failure and patient mortality after liver transplant were analyzed from a consecutive series of 1299 patients undergoing cadaveric liver transplantation. This was correlated with serum liver function tests and the rate of reduction in transaminase levels over the first postoperative week. A cut-off transaminase level correlating with graft and patient survival was calculated and incorporated into a scoring system. Results: Aspartate-aminotransferase (AST) on postoperative day one showed significant correlation with early graft failure for levels above 723 U/dl and early postoperative mortality for levels above 750 U/dl. AST reduction rate (day 1 to 3) greater than 1.8 correlated with reduced graft failure and greater than 2 with mortality. Alanine-aminotransferase (ALT) reduction in the first 48 h post transplantation also correlated with outcomes. Conclusion: A scoring system with these three variables allowed us to classify our patients into three groups of risk for early graft failure and mortality