Key result
PURE Calcium reconstructions on photon-counting CT deliver near-perfect agreement with true non-contrast calcium scores.
Why the study?
The study aimed to evaluate coronary artery calcium scores obtained from virtual unenhanced ECG-gated and non-gated photon-counting CT images using PURE Calcium and Virtual Non-Contrast algorithms against unenhanced ECG-gated reference images.
Does coronary artery calcium scoring derived from contrast-enhanced photon-counting CT using PURE Calcium and Virtual Non-Contrast algorithms accurately match true non-contrast ECG-gated CT in pre-TAVI patients?
Observational (n=61)
Does coronary artery calcium scoring derived from contrast-enhanced photon-counting CT using PURE Calcium and Virtual Non-Contrast algorithms accurately match true non-contrast ECG-gated CT in pre-TAVI patients?
In pre-TAVI patients undergoing contrast-enhanced photon-counting CT, the PURE Calcium algorithm substantially outperforms Virtual Non-Contrast for estimating coronary artery calcium scores, though systematic underestimation prevents absolute interchangeability with true non-contrast scans.
Answers are grounded in this paper and cite their sources.
Caution against interchangeable use of contrast photon-counting CT calcium scores in pre-TAVI assessment; leaves open algorithm-specific calibration for equivalence.
Purpose To evaluate coronary artery calcium score (CACS) obtained from virtual unenhanced ECG‐gated and non‐gated photon‐counting CT (PCCT) images using PURE Calcium (PC) and Virtual Non‐Contrast (VNC) algorithms, compared to unenhanced ECG‐gated images as reference standard. Materials and Methods Patients scheduled for transcatheter aortic valve implantation (TAVI) were prospectively included. Per patient, CACS was quantified in five datasets: true non‐contrast ECG‐gated scan (TNC, reference); PC and VNC reconstructions of ECG‐gated (PC GATED , VNC GATED ) and non‐gated contrast‐enhanced scans (PC NON‐GATED , VNC NON‐GATED ). Pairwise comparisons with the TNC reference were performed using Bland‐Altman analysis, correlation, intraclass correlation coefficients (ICC), and risk‐category agreement assessed by linearly weighted Cohen's κ. Additionally, CACS was classified into risk categories (0; 1–10; 11–100; 101–400; >400). Results Sixty‐one patients were included (59% female; 78±8 years). Median total CACS on TNC was 659 [222–1183]. Agreement with TNC was highest for PC GATED (ICC = 0.99; bias ‐73), followed by PC NON‐GATED (ICC = 0.98; bias ‐152). In contrast, VNC GATED and VNC NON‐GATED demonstrated substantially larger negative biases (‐615 and ‐687, respectively) and lower agreement (ICC = 0.62 and 0.54, respectively). Compared with TNC, significant differences in total CACS were observed for all reconstruction methods except PC GATED ( p < 0.001 for all significant comparisons). Agreement in CACS category classification was higher for PC reconstructions ( κ = 0.83/0.80 for ECG‐gated/non‐gated) than VNC reconstructions ( κ = 0.41/0.33 for ECG‐gated/non‐gated). Exploratory vessel‐level agreement showed a similar pattern. Conclusion CACS derived from virtual unenhanced contrast‐enhanced PCCT images systematically underestimated the reference CACS. PC substantially outperformed VNC, including non‐gated acquisitions, although underestimation persisted and the absolute scores should not be considered interchangeable with TNC.
No takes yet. Share an insight, caveat, or question.
Busselot et al. (2026) conducted an observational in Severe aortic stenosis scheduled for transcatheter aortic valve implantation (TAVI) (n=61). PURE Calcium (PC) and Virtual Non-Contrast (VNC) algorithms on ECG-gated and non-gated contrast-enhanced photon-counting CT vs. True non-contrast ECG-gated scan (TNC) was evaluated on Coronary artery calcium score (CACS) agreement and risk-category classification compared with true non-contrast ECG-gated CT (ICC 0.99 (PC GATED), ICC 0.98 (PC NON-GATED), ICC 0.62 (VNC GATED), ICC 0.54 (VNC NON-GATED), p=<0.001). PURE Calcium reconstructions on photon-counting CT outperformed Virtual Non-Contrast for CACS agreement with true non-contrast scans.
Synapse has enriched 10 related papers. Consider them for comparative context: