Coronary computed tomography angiography in primary care patients with chest pain or dyspnea – a cross-sectional study

TL;DR

This study examines how well coronary CT angiography rules out heart artery blockages in primary care patients with chest pain or shortness of breath.

Contribution

The study evaluates the accuracy of pre-test probability estimates by primary care physicians when referring patients for CCTA.

Findings

CCTA ruled out significant coronary stenosis in 88% of patients with pre-test probability of 5–15%.

Primary care physicians underestimated significant stenosis in patients with pre-test probability above 15%.

Pre-test probability estimates below 5% had low validity in predicting CCTA results.

Abstract

Coronary Computed Tomography Angiography (CCTA) is recommended as a first-line investigation to exclude significant coronary artery stenosis in case of low to intermediate pre-test probability (PTP). The aim was to investigate CCTA findings in relation to the PTP of patients referred directly from primary health care centres.

In this retrospective cohort study consecutive primary care CCTA referrals in a Swedish county 1st of June 2021 until 30th Nov. 2022 were included. CCTA reports were obtained for 483 patients ≥ 30 years old, without known CAD and stratified as no CAD, with atheromatosis or with suspected significant stenosis. For the 381 patients with eligible PTP data, the mean age was 60 years and 70% were women. While the median PTP was 11%, significant stenosis was suspected on CCTA in 18%. Among patients with PTP ≤ 15%, CCTA exposed no significant stenosis in 88%. No significant stenosis was found in patients with PTP < 5% true to patient age and gender in a sensitivity analysis (n = 25).

CCTA ruled out coronary stenosis as the cause of chest pain and dyspnea in 88% of patients referred from primary care with PTP 5–15%. PTP estimations by primary care physicians in CCTA referrals agreed with the occurrence of suspected significant stenosis among patients with PTP 5–15%, but underestimated it in PTP > 15%. The validity of PTP estimates < 5% was low.

The online version contains supplementary material available at 10.1186/s12875-025-02877-z.

Figures