Reliability of new scores in predicting perioperative mortality after isolated aortic valve surgery: a comparison with the society of thoracic surgeons score and logistic EuroSCORE

Background: There is still a wide debate concerning the performance of commonly used risk prediction models in assessing the risk of patients undergoing isolated aortic valve surgery. This study was designed to compare the performances of European System for Cardiac Operative Risk Evaluation (EuroSCORE) II and age, creatinine, and ejection fraction (ACEF) score with those of The Society of Thoracic Surgeons (STS) score and logistic EuroSCORE in patients undergoing isolated aortic valve surgery. Methods: Data on 1,758 consecutive patients who underwent isolated aortic valve replacement in a 6-year period were retrieved from 3 prospective institutional databases. Discriminatory power was assessed using the c-index. Calibration was evaluated with calibration curves and associated statistics. Results: In-hospital mortality rate was 1.4%. The discriminatory power was similar in all algorithms (area under the curve 0.80, 95% confidence interval [CI] 0.72 to 0.88 for logistic EuroSCORE; 0.81, 95% CI 0.73 to -0.88 for EuroSCORE II; 0.78, 95% CI 0.68 to 0.88 for ACEF; 0.85, 95% CI 0.78-0.93 for STS score) and not significantly different (p values > 0.05 for all tests). The EuroSCORE II had a better calibration, being the only score with nonsignificant associated statistics (unreliability test, Hosmer-Lemeshow test, and Spiegelhalter Z-test for calibration accuracy). Nonetheless, EuroSCORE II calibration plot highlighted a trend over under-prediction. Conclusions: The EuroSCORE II is a good predictor of perioperative mortality in isolated aortic valve surgery, with lower discrimination if compared with STS and a better calibration when compared with logistic EuroSCORE, ACEF, and STS scores. Its performance is optimal in the lowest tertile of patients, whereas it under-predicts mortality afterward. None of these algorithms seems suitable for risk estimation in mid and high-risk patients that are the ones who might benefit most from transcatheter procedures.