Diagnostic performance of the 2019 ESC pre-test probability and Coronary Artery Disease consortium models in estimating obstructive coronary artery disease

Aim . To evaluate and compare the diagnostic performance of the 2019 European Society of Cardiology pre-test probability (PTP) model and the coronary artery disease (CAD) consortium basic and clinical models in predicting obstructive CAD in patients with stable angina. Material and methods . This cross-sectional study included 366 patients (mean age 64.8 years, 62.6% male) with suspected stable angina who underwent coronary computed tomography angiography. Obstructive CAD was defined as the presence of ≥50% stenosis in epicardial coronary artery segments with a diameter of ≥2.5 mm. We assessed clinical characteristics and cardiovascular risk factors. The PTP values from the three models were calculated, and their diagnostic performance was evaluated using area under the receiver operating characteristic curves and the Hosmer-Lemeshow test for calibration. Sensitivity, specificity, and predictive values were also analyzed. Results . Obstructive CAD was detected in 270 (73.8%) patients. Patients with obstructive CAD had higher rates of male sex, hypertension, dyslipidemia, smoking, and typical and atypical angina (all p<0.05). The CAD consortium clinical model provided the most accurate estimate of obstructive CAD prevalence in high-risk patients (76.6% expected vs 84.4% observed), while the 2019 ESC PTP model was more accurate in low-risk patients (2.5% expected vs 0.4% observed). The CAD consortium clinical model demonstrated the best diagnostic performance with an area under the curve (AUC) of 0.760 and good calibration (Hosmer-Lemeshow test, p=0.823). This was followed by the CAD consortium basic model (AUC=0.755), and the 2019 ESC PTP model, which had the lowest performance (AUC=0.701, poor calibration, p=0.001). The CAD consortium clinical model, with a cut-of value >33%, had a sensitivity of 66.7%, specificity of 79.2%, a positive predictive value of 90%, and a negative predictive value of 45.8% in predicting obstructive CAD. Conclusion . The CAD consortium clinical model showed superior accuracy in predicting obstructive CAD in stable angina patients, especially in high-risk groups, compared to the 2019 ESC PTP and CAD consortium basic models. Its strong diagnostic performance and reliable calibration make it a better tool for CAD risk assessment.