Validation of the European Society of Cardiology pretest probability models for obstructive coronary artery disease...

Area of Science:

• Cardiology
• Diagnostic Accuracy
• Epidemiology

Background:

• The European Society of Cardiology (ESC) updated its pre-test probability (PTP) model for obstructive coronary artery disease (CAD) in 2019.
• External validation of the updated model in high-incidence CAD populations is lacking.
• This study aimed to validate the 2019 ESC-PTP model in a Serbian population with high CAD incidence and compare it with the 2013 ESC-PTP model.

Purpose of the Study:

• To externally validate the 2019 ESC-PTP model for obstructive CAD in a high-risk Serbian population.
• To compare the diagnostic performance of the 2019 ESC-PTP model against the 2013 ESC-PTP model.
• To assess the accuracy of both models in a population with a high prevalence of CAD.

Main Methods:

• Retrospective analysis of 1294 symptomatic patients with suspected CAD (2015-2019).
• Calculation of PTP scores using 2013 ESC-PTP and 2019 ESC-PTP models based on age, gender, and symptoms.
• Invasive coronary angiography (ICA) performed for all patients to confirm obstructive CAD.

Main Results:

• Obstructive CAD was diagnosed in 533 patients (41.2%).
• The 2019 ESC-PTP model classified significantly more patients as low probability (<15%) compared to the 2013 ESC-PTP model (39.8% vs. 5.6%, p < 0.001).
• The 2019 ESC-PTP model underestimated CAD prevalence (calibration intercept 1.15, slope 0.96), while the 2013 ESC-PTP model overestimated it (calibration intercept -0.24, slope 0.73). Both models demonstrated moderate discrimination (similar AUC).

Conclusions:

• Both the 2013 and 2019 ESC-PTP models exhibited moderate accuracy for diagnosing CAD in the high-risk Serbian population.
• The 2019 ESC-PTP model tended to underestimate CAD prevalence, whereas the 2013 ESC-PTP model overestimated it.
• Further research is needed to develop and validate PTP models specifically for high-risk countries.