Performance of the 2019 ESC pre-test probability model in predicting obstructive coronary artery disease in a Chinese population using coronary computed tomography angiography outcomes

Abstract

BackgroundThe 2019 European Society of Cardiology (ESC) guidelines proposed a pre-test probability (PTP) model to determine the likelihood of coronary artery disease (CAD). However, the prediction accuracy of this model has not yet been evaluated in Chinese populations. This study aimed to validate the 2019 ESC-PTP model in predicting CAD using coronary computed tomography angiography (CCTA) outcomes in a Chinese population. MethodsA total of 26,346 consecutive patients with suspected CAD who underwent CCTA were included. The 2019 ESC-PTP model and 2013 ESC-PTP model were calculated for each patient, considering age, sex, and the symptom of chest pain, and the patients were categorized into low-, intermediate-, and high-risk groups. The predictive performance of the 2019 ESC-PTP model was evaluated by comparing it with the 2013 ESC-PTP model and the observed prevalence of CAD from CCTA. ResultsAmong the 11,234 patients analyzed in the study, 1896 (16.9%) patients were found to have obstructive CAD from CCTA. The 2019 ESC-PTP model had better calibration compared to the 2013 ESC-PTP model. After categorization, 80.9% of patients (67.9% in men and 94.4% in women) were in the same risk category as in the 2019 ESC-PTP model, but the risks of younger patients (7.5% versus 2.5%; P ​< ​0.001) and patients with non-anginal chest pain (13.7% versus 8.2%; P ​< ​0.001) were underestimated in the 2019 ESC-PTP model. ConclusionThe 2019 ESC-PTP model demonstrated a good calibration in predicting CAD in a Chinese population who underwent CCTA, but it exhibited an underestimation of CAD probability in younger patients and patients with non-anginal chest pain.