Virtual Non-Iodine Coronary Calcium Scoring on Photon-Counting CT: Patient- and Plaque-Level Analysis

Area of Science:

• Radiology and Imaging Science
• Cardiovascular Imaging
• Medical Physics

Background:

• The utility of virtual non-iodine (VNI) imaging derived from photon-counting CT (PCCT) as a substitute for true non-contrast (TNC) scans in coronary artery calcium scoring (CACS) is under investigation.
• Particular concern exists regarding the accuracy of VNI for detecting small, low-density coronary artery plaques.

Purpose of the Study:

• To evaluate the agreement between VNI and TNC for CACS at both patient and lesion levels.
• To quantify the extent of risk category reclassification when using VNI instead of TNC for CACS.
• To assess the performance of different VNI reconstruction algorithms (55 keV QIR 1, 60 keV QIR 4, PureCalcium) compared to TNC.

Main Methods:

• Retrospective analysis of 211 patients with non-zero coronary artery calcium (CAC) on TNC, who underwent PCCT.
• Reconstruction of VNI (55 keV, 60 keV) and TNC images with matched parameters.
• Comparison of Agatston scores and total calcified volumes between VNI and TNC, including lesion-level analysis for false negatives and risk category reclassification.

Main Results:

• Low-keV VNI (55-60 keV) significantly underestimated CAC compared to TNC, with reduced median Agatston scores and total calcified volumes (p < 0.001 for both).
• Over half of the patients were reclassified into lower CAC categories using VNI; notably, 46.9-47.4% of patients with non-zero CAC on TNC were incorrectly assigned a CAC of 0 on VNI.
• At the lesion level, 95% of patients exhibited at least one false-negative plaque on VNI, predominantly small (median 8 mm³) and low-density lesions.

Conclusions:

• In patients with relatively low coronary calcification burden, low-keV VNI (55-60 keV) significantly underestimates CAC and leads to patient down-classification.
• Frequent 'false-zero' CAC assignments on VNI are driven by small, low-density plaques, highlighting limitations for accurate CACS assessment.
• Current low-keV VNI protocols are not a reliable substitute for TNC scans for comprehensive CACS evaluation, particularly in identifying subtle calcifications.