Can the EuroSCORE predict the early and mid-term mortality after off-pump coronary artery bypass grafting?

Background. This study evaluated the role of the European System for Cardiac Operative Risk Evaluation (EuroSCORE) in the prediction of early-term and mid-term mortality in patients undergoing isolated off-pump coronary artery bypass grafting (OPCAB). Methods. From January 2002 to August 2006, 757 consecutive patients underwent isolated OPCAB. The patients' operative risks were calculated according to the standard and logistic EuroSCORE models. The cohort was classified into four subgroups according to both EuroSCORE scales. To evaluate the predictability, the expected mortality was compared with the observed mortality. The receiver operating characteristic curves were plotted and calibration was assessed. Mean follow-up was 32.8 +/- 13.9 months. Results. Ten (1.3%) in-hospital deaths occurred. The predicted total numbers of deaths by the EuroSCORE models were 34.2 (4.5%) for the standard EuroSCORE and 37.8 (5.0%) for the logistic EuroSCORE. The expected mortality rates were significantly higher than the observed mortality rates in all subgroups, except one. The area under curve (AUC) in in-hospital mortality was 0.72 for the standard EuroSCORE and 0.71 for the logistic EuroSCORE, but the tests of calibration for both EuroSCORE models were significant. Mid-term mortality was 3.6%. The AUC curve in mid-term mortality was 0.71 for the standard or logistic EuroSCORE. The calibration in both EuroSCORE models for mid-term mortality was nonsignificant, indicating good calibration. Conclusions. Both EuroSCORE models overestimated the in-hospital mortality; however, both models showed good predictability for mid-term mortality. The EuroSCORE could be helpful in planning resource allocation and tailoring follow-up for patients undergoing isolated OPCAB.