Comparison of the EuroSCORE II, STS and Ambler Scoring Systems in Predicting In-hospital Mortality Among Patients Undergoing Heart Valve Surgery

Abstract

Background: Risk models are helpful in planning perioperative care in patients undergoing heart valve surgery. In the absence of a local risk model, the European System for Cardiac Operative Risk Evaluation (EuroSCORE II), Society of Thoracic Surgery (STS) risk score and Ambler scores are frequently used in our institution to predict in-hospital mortality after heart valve surgery. Objective:To compare the accuracy of the EuroSCORE II, STS and Ambler scoring systems in predicting in-hospital mortality among patients who underwent heart valve surgery. Methods: From January 2008 to December 2015, the data of 162 patients who underwent heart valve surgery were collected through medical record review. EuroSCORE II, STS and Ambler scores were calculated for each patient. Calibration of these scores were assessed by the Hosmer–Lemeshow test and discrimination was tested by calculating the area under the receiver operating characteristic (ROC) curve. Results: The overall observed mortality was 6.17%. The EuroSCORE II, STS and Ambler scores underestimated in-hospital mortality (4.13%, 3.47%, and 4.12%, respectively; p<0.05 for all) but had good discriminative power (ROC area 0.82, 0.88 and 0.77, respectively) in patients who underwent single heart valve surgery. In the subgroup analysis of those who underwent isolated mitral valve surgery, the EuroSCORE II, STS and Ambler scores underestimated in-hospital mortality but had good discriminative power. In the subgroup analysis of patients who underwent aortic valve surgery, the EuroSCORE II, STS and Ambler scores showed good calibration in predicting in-hospital mortality, but only the STS score showed excellent discrimination. For dual valve surgery (observed mortality 4.0%), the EuroSCORE II and Ambler score showed good calibration (2.58% and 2.6%, p>0.05 for both) and discrimination (ROC area 0.83 and 0.85, respectively) in predicting in-hospital mortality. Conclusion: In this single-center retrospective study, the EuroSCORE II, STS and Ambler scores underestimated overall in-hospital mortality after heart valve surgery. All three scores had good discriminative power; however, the STS score showed better performance compared to the EuroSCORE II and Ambler score in identifying high surgical risk individuals who underwent heart valve surgery.