Accuracy of the BAR score in the prediction of survival after liver transplantation

Abstract

Introduction and aim: The Balance of Risk (BAR) Score, a simple scoring system that combines six independent donor and recipient variables to predict outcome after liver transplantation (LT), was validated in a large U.S./European cohort of patients. This study aims to assess the performance of the BAR score to predict survival after liver transplantation and determine the factors associated with short and long-term survival in Latin-American patients.

Material and methods: A retrospective cohort study was performed in 194 patients [112 (55.4%) males; mean age 52±14 years] who underwent 202 LT during the period 2003-2015. Demographic, clinical, pathological and surgical variables, as well as mortality and survival rates, were analyzed. The BAR score was investigated through a receiver operating characteristics (ROC) curve with the calculation of the area under the curve (AUC) to evaluate the predictive score power for 3-month, 1 and 5-year mortality in a matched donor-recipient cohort. Youden index was calculated to identify optimal cutoff points.

Results: The AUC of BAR score in predicting 3-month, 1-year and 5-year mortality were 0.755 (CI95% 0.689-0.812), 0.702 (CI95% 0.634-0.764) and 0.610 (CI95% 0.539-0.678) respectively. The best cut-off point was a BAR score ≥15 points. In the multivariate analysis BAR score <15 was associated with higher survival rates at 3 months and 1 and 5-years.

Conclusions: BAR score <15 points is an independent predictor of better short and long-term survival in Latin-American patients undergoing LT. The BAR scoring system has an adequate diagnostic capacity allowing to predict 3 and 12-month mortality.

Keywords: Balance of Risk; Liver transplant; Survival analysis.