Coronary calcium scoring in the evaluation of stable chest pain in general practice: a cluster-randomised trial
Moniek Y Koopman, Thom E J Severijn, Jorn J W Reijnders, Grigory Sidorenkov, Marcel Van Tuinen, Carine J M Doggen, Geert Jan Dinant, Mathijs O Versteylen, Jan-Willem C Gratama, Richard Braam, Bastiaan Kietselaer, Rykel Van Bruggen, Robert Willemsen, Pim van der Harst, Rozemarijn VliegenthartBackground
General practitioners (GPs) have limited tools for chest pain patients to identify (non-)obstructive coronary artery disease (OCAD). This trial evaluates whether GP access to the coronary artery calcium score (CACS) improves diagnostic efficiency and cardiovascular risk management (CVRM).
Methods
COroNary Calcium scoring as fiRst-linE Test to dEtect and exclude coronary artery disease in GP patients with stable chest pain (CONCRETE) is a pragmatic, non-blinded implementation study that cluster-randomised 101 Dutch GP offices into CACS or standard of care (SOC). Patients aged ≥40 (men) or ≥45 (women) years, without known CAD, with possible cardiac or non-cardiac chest pain were included. The primary outcome was the proportional 2-year increase in CVRM registrations at GP office level. Patient-level secondary outcomes included cardiologist referral rate, OCAD diagnosis and CVRM enrolment.
Results
At GP office level, no significant difference in CVRM registration increase was seen for CACS versus SOC (46 CACS practices: +0.3%, 45 SOC practices: +0.2%, p=0.77). Per protocol analysis for patient-level outcomes included 583 patients (57.5% women, mean age 60.6±9.1 years) (CACS arm: 466; SOC arm: 117), recruited between January 2019 and October 2023. The CACS arm showed 47.1% relative reduction in cardiologist referrals compared with SOC (42.5% vs 80.3%; modelled difference 31.3% (95% CI 14.8% to 47.8%)). OCAD detection rates were not significantly different (CACS: 4.9, SOC: 7.7%, p=0.14). Patients with higher CACS had a higher probability of cardiologist referral (from 8.4% for CACS 0 to 94.2% for CACS ≥400, difference +85.7% (95% CI 79.4% to 91.9%)) and of OCAD diagnosis (0% for CACS 0 to 21.2% for CACS ≥400, OR 9.08 (95% CI 4.20 to 19.67)). Compared with SOC, more CACS patients were enrolled in CVRM (36.3% vs 17.0%, p<0.01).
Conclusions
In this pragmatic, primary-care implementation and proof-of-concept trial, CACS was an effective diagnostic test for stable chest pain patients with (very) low likelihood of OCAD. CACS reduced unnecessary cardiology referrals and enhanced individual preventive care, without demonstrable practice-level impact on CVRM registrations.
Registration
The CONCRETE study is registered under CCMO Register NL66821.042.18 and closed for enrolment.