LAUSR

Bridge to heart transplantation (HTx) with ECMO is associated with poor outcome, but patient status, according to different levels of multiorgan compromise, is generally not considered. The aim of this study was to analyze the prognostic value of acute physiology, age, and chronic health evaluation IV (APACHE IV) score in this setting. Thirty-two patients underwent HTx bridged with ECMO at our institution between 2005 and 2017; they were divided into two groups, according to a cutoff value of APACHE IV score obtained by Receiver operating characteristic curve analysis for 30-day mortality. Kaplan-Meyer survival curves were plotted, and compared through the log-Rank test. Cox regression model was used to estimate which factors were associated with overall survival. The 30-day mortality prediction of the APACHE IV score showed an AUC of 0.98 [95% C.I. 0.84-0.99], with a cutoff value corresponding to a score value of 47 (specificity of 84.6% and sensitivity of 100%) in order to discriminate between a group with low probability (Group A, patients with an APACHE IV score <47), and a group with high probability (Group B, patients with an APACHE IV score ≥47) of 30-day mortality. Median follow up was 26.9 months (range: 0.03-143.8). On overall, patients bridged with ECMO showed a high early mortality (18.7% <30 days, n = 6). All deaths occurred in Group B, where 30-day mortality was 60%. Survival probability among Group B patients was 26.6% at both 1 and 5 years. Conversely, in Group A no early mortality was reported, and the estimated survival was 89.7% at 1 year and 81.5% at 5 years, respectively. Mortality correlated at univariate analysis with recipient age, APACHE IV score and female sex, while the multivariate Cox regression analysis showed that only an APACHE IV score (HR 1.23 [1.08-1.39, 95% C.I.]) and female sex (HR 6.68 [1.42-31.43, 95% C.I.]) had an adverse impact on survival. This study shows that the APACHE IV score could be considered a powerful predictor of survival in patients bridged by ECMO to HTx, and can account for a better selection of patients on ECMO support at time of listing.