Use of the model for end-stage liver disease score for guiding clinical decision-making in the selection of patients for emergency cardiac transplantation.

OBJECTIVES The outcomes of emergency cardiac transplantation remain controversial, but recipient selection is essential for success. With a shortage of organs, it is essential to determine an objective method, such as a risk score, for choosing patients who are at too great a risk to undergo cardiac transplantation. In this study, we analysed the model for end-stage liver disease in terms of predicting operative mortality after emergency cardiac transplantation. METHODS We analysed the Nancy University database of heart transplantation and selected all patients who underwent emergency heart transplantation between January 2005 and January 2012. The calibration and discriminatory power were evaluated to determine the model for end-stage liver disease (MELD) score. Preoperative and peri-operative variables regarding the prediction of operative mortality were analysed by univariate and multivariate logistic regression models. RESULTS Forty-three patients underwent emergency cardiac transplantation. The operative mortality was 20.9% (n = 9). The Hosmer-Lemeshow test demonstrated a calibrated model for predicting operative mortality (P = 0.15), and the MELD score presented an excellent discrimination between survivors and non-survivors (AUC: 0.89 ± 0.05; 95% CI: 0.79-0.99). In the univariate analysis, an MELD score of ≥ 16 and bilirubin concentration were predictive markers of operative mortality. Multivariate logistic regression tested the contribution of the univariate risk predictors (P < 0.15) and confirmed that an MELD score of ≥ 16 was predictive of operative mortality. CONCLUSIONS The MELD score appears to be adequate for predicting operative mortality among patients who undergo heart transplantation. The MELD score could therefore be used to guide clinical decision-making for emergency transplantation.