Ten years after the introduction of the European system for cardiac operative risk evaluation 2: A single center validation

Abstract

Introduction. The aim of this study was to evaluate the predictive value of the European System for Cardiac Operative Risk Evaluation II in adult patients with acquired heart diseases. Material and Methods. The research included a consecutive series of 6,031 patients who underwent coronary, valvular and combined cardiac surgical interventions in the period from January 15, 2015 to December 31, 2020. Model calibration was assessed by comparing the ratio of actual to expected postoperative mortality and using the Hosmer-Lemeshow test. The discriminative power was examined using the area under the receiver operating characteristic curve. Results. A total of 2,883 patients underwent isolated coronary surgery, 1,841 underwent valvular procedures, while a combined procedure was performed in 1,307 patients. The operative risk was moderately underestimated in the entire group, as well as in the group of patients who underwent surgical revascularization of the coronary arteries. In patients who underwent valvular surgery, the actual mortality rate was slightly overestimated, while in patients with combined procedures it was moderately underestimated. The European System for Cardiac Operative Risk Evaluation II showed excellent discriminative power in the whole group of patients undergoing surgery (area under the curve = 0.825, p < 0.0005). The cut-off value was 2.60, sensitivity 0.757 and specificity 0.750. The discriminative power of the model was excellent in the group of coronary patients (area under the curve = 0.810) as well as in the group with isolated valvular surgery (area under the curve = 0.815). In patients with combined procedures, the discriminatory power was very good (area under the curve = 0.775). Conclusion. The results of our single centre study show that European System for Cardiac Operative Risk Evaluation II predicts hospital mortality with satisfactory results in the entire group, but underestimates it when it comes to combined cardiac surgical procedures. The discriminative power of the model is excellent.