Validation of EuroSCORE II in Chinese patients undergoing heart valve surgery.

Background: To assess the performance of the The European System for Cardiac Operative. Risk Evaluation II (EuroSCORE II) in Chinese patients undergoing heart valve surgery at our centre.

Methods: From January 2006 to December 2011, 3479 consecutive patients who underwent heart valve surgery at our centre were collected and scored according to the original EuroSCORE and EuroSCORE II models. All patients were divided into single valve surgery and multiple valve surgery subgroups. The entire cohort and each subgroup were analysed. Calibration of the original EuroSCORE and EuroSCORE II models was assessed by the Hosmer-Lemeshow (H-L) test. Discrimination was tested by calculating the area under the receiver operating characteristic (ROC) curve.

Results: Observed mortality was 3.32% overall, compared to expected mortality 3.84% for the original additive EuroSCORE (H-L: P = 0.013), 3.33% for the original logistic EuroSCORE (H-L: P = 0.08), and 2.52% for the EuroSCORE II (H-L: P < 0.0001). The EuroSCORE II model showed good calibration in predicting in-hospital mortality for patients undergoing single valve surgery (H-L: P = 0.103) and poor calibration for patients undergoing multiple valve surgery (H-L: P < 0.0001). The discriminative power of the original EuroSCORE model (area under the ROC curve of 0.684 and 0.673 for the additive and logistic model, respectively) and EuroSCORE II model (area under the ROC curve of 0.685) for the entire cohort was poor. The discriminative power of the EuroSCORE II model was good for the single valve surgery group (area under the ROC curve of 0.792) and was poor for the multiple valve surgery group (area under the ROC curve of 0.605).

Conclusion: The EuroSCORE II model gives an accurate prediction for individual operative risk in patients undergoing single valve surgery but an imprecise prediction in patients undergoing multiple valve surgery at our centre. Therefore, the use of the EuroSCORE II model for risk evaluation may be suitable in patients undergoing single valve surgery, and the creation of a new model which accurately predicts outcomes in patients undergoing multiple valve surgery is possibly required at our centre in the future.