Diagnostic accuracy of photon-counting computed tomography for detecting coronary artery stenosis: a systematic...

Conventional coronary CT angiography (CCTA) is limited by calcium blooming and metallic artefacts, often reducing specificity in high-risk patients. Photon-counting CT (PCCT), with ultra-high spatial resolution and intrinsic spectral sensitivity, may overcome these limitations.

This systematic review and meta-analysis was registered prospectively in PROSPERO (CRD420251126753) and conducted in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. PubMed, EMBASE, Web of Science, Cochrane Library and Scopus were searched from inception to 31 December 2025. Prospective and retrospective studies evaluating PCCT for detection of coronary artery stenosis were included. Diagnostic accuracy was synthesised using a bivariate random-effects model, generating pooled sensitivity, specificity, diagnostic ORs and summary receiver operating characteristic curves.

Out of 1010 records identified, 19 studies were included in qualitative synthesis and 10 were eligible for quantitative meta-analysis. At the patient level, pooled sensitivity was 94.5% and specificity was 81.8%. At the vessel level, sensitivity and specificity were 95.1% and 89.2%, respectively, and at the segment level 95.2% and 94.7%. Heterogeneity was moderate, particularly for specificity. Deeks' funnel plot showed no evidence of significant publication bias. Non-independence of vessel-level and segment-level data may have led to overestimation of precision.

PCCT demonstrates consistently high sensitivity and moderate-to-high specificity for detecting coronary artery stenosis at patient, vessel and segment levels, with improved performance in heavily calcified disease compared with conventional CCTA. These findings support current first-generation PCCT as a high-confidence non-invasive rule-out modality. By mitigating artefacts in complex anatomies, PCCT demonstrates potential to reduce unnecessary invasive angiography, warranting further validation in multivendor, large-scale trials.

CRD420251126753.