A scoping review of photon-counting detector computed tomography in cardiovascular imaging

Photon-counting detector computed tomography (PCD-CT) employs direct-conversion detectors that record the arrival and energy of individual photons, enabling high-resolution, multi-energy cardiovascular imaging. We searched MEDLINE, Embase, and Scopus from January 2021 through September 2025 and included 59 studies. Owing to heterogeneity in study designs, protocols, and endpoints, the findings were synthesized narratively across five domains (coronary, myocardial, structural/valvular, pulmonary-cardiopulmonary function, and aortic/visceral/peripheral arteries). In coronary imaging, a routine-practice cohort (n = 7.833) reported a per-patient specificity of 98% vs. 93%, lower invasive angiography of 9.9% vs. 13.1%, and a higher revascularization yield of 43.4% vs. 35.5% [PCD-CT vs. energy-integrating detector CT (EID-CT); ultra-high-resolution protocols achieved a vessel-level area under the curve (AUC) of up to 0.99. Low-dose CCTA was feasible at a CTDIvol of 1.72 mGy, and contrast-saving protocols supported diagnostic studies with a volume of 30 mL. Virtual non-contrast calcium scoring showed an intraclass correlation coefficient of 0.98 vs. true non-contrast. In myocardial tissue characterization, PCD-CT-derived extracellular volume differed from cardiovascular magnetic resonance by ≤2% in selected cohorts, with a kappa of up to 0.956 for late-enhancement agreement; segment-level inflammation classification reached an AUC of 0.95. For structural/valvular assessment, a comparative cohort reported an effective dose of 8.8 ± 4.5 vs. 15.3 ± 5.8 mSv, with visual image quality scores of 4.8 vs. 3.3, respectively, for PCD-CT vs. EID-CT. Lung-perfusion iodine maps for chronic thromboembolic pulmonary hypertension achieved an accuracy of 0.85-0.88 at approximately one-fifth of the dose of single-photon emission CT. For aortic/peripheral applications, thoracoabdominal protocols reported a dose of 4.2 ± 1.4 vs. 7.2 ± 2.2 mGy, with a higher signal-to-noise ratio/contrast-to-noise ratio (PCD-CT vs. EID-CT); infrapopliteal imaging used 60 versus 140 mL of contrast, respectively, with improved vessel sharpness for PCD-CT vs. EID-CT; diagnostic performance for peripheral stenosis reached a sensitivity of 91% and a specificity of 95%, respectively, when compared with digital subtraction angiography. Overall, the evidence-predominantly single center-indicates that PCD-CT may enable dose- and contrast-efficient cardiovascular imaging with strong diagnostic metrics, and confirmation in multicenter outcome and cost-effectiveness studies remains a priority.