Abstract Objectives Coronary artery calcium (CAC) scoring is a well-established method for cardiovascular risk assessment but has limited reproducibility. For energy-integrating detector (EID)-CT, a new, multivendor validated protocol has been proposed. This study aimed to investigate the variability of photon-counting detector (PCD)-CT-based CAC scoring and propose a new protocol with decreased variability. Materials and methods A chest phantom containing nine calcifications was scanned on a PCD-CT using various settings: tube voltages (90 kVp, 120 kVp), tube currents (100% to 25% dose), slice thickness (3 mm, 1 mm), quantum iterative reconstruction (IR, 1–4). To evaluate interscan variability, phantoms were scanned five times per protocol with slight translational (5 mm) and rotational (2°) movements. The standard PCD-CT protocol used 120 kVp, 100% dose, 3 mm slices. CAC scores, image noise, and calcification detectability were assessed. Results were compared to the standard PCD-CT, and standard and proposed EID-CT protocols. Results Compared to the standard PCD-CT protocol, score variability decreased by 37% using a thin-sliced protocol at 120 kVp, 25% dose reduction and IR2. Compared to the proposed EID-CT protocol, variability was 66% lower. The optimized PCD-CT protocol met noise targets, eliminating the risk of false-positives. While 6.0 ± 0.0 and 7.0 ± 0.4 calcifications were detected using the PCD-CT standard and the proposed EID-CT protocol, respectively, 7.1 ± 0.7 calcifications were detected with the optimized PCD-CT protocol. Volume and mass scores were closer to physical reference. Conclusions A thin-slice, 25%-dose-reduced PCD-CT protocol at 120 kVp improves CAC score reproducibility and outperforms the proposed EID-CT protocol, possibly offering more reproducible CAC quantification at lower radiation doses. Key Points Question Coronary artery calcium scoring is used for cardiovascular risk stratification. However, the current standard method lacks score reproducibility. Findings A thin-slice, 25%-dose-reduced photon-counting detector CT protocol at 120 kVp significantly reduces score variability compared to previous protocols, including the proposed energy-integrating detector CT protocol. Clinical relevance Improved reproducibility of coronary artery calcium scoring may enable more consistent cardiovascular risk prediction and provide a robust technical basis for further in vivo studies. Graphical Abstract
Fink et al. (Thu,) studied this question.