Abstract
The study evaluates the impact of body mass index (BMI), heart rate and rhythm on coronary artery calcium scoring (CACS) derived from calcium-sensitive virtual non-contrast (VNC) series of photon-counting detector (PCD) computed tomography angiography (CTA) compared to true non-contrast (TNC) series. Patients who underwent cardiac imaging with TNC and CTA on a PCD-CT were included. Agatston scores from TNC and VNC images were used to assign CACS risk category. Analyses considered BMI, heart rhythm and heart rate. Distributions were tested for differences between TNC and VNC derived scores and their correlation was assessed. The final cohort included 88 patients. CACS on VNC showed an underestimation of TNC derived values on median Agatston score TNC = 542 (IQR 200-1294), on median Agatston score VNC = 449 (IQR 130-1183), p < 0.001, percentage difference - 11%. However, linear correlation coefficient was high (r(2) = 0.95), and the CAC severity was categorized equivalent in 80%. In approximately 11% of the study cohort, the misclassification of CAC severity could have potentially led to inappropriate treatment following established guidelines. An impact on the significance and extent of the difference in CACS for BMI > 28 kg/m(2) and heart rate groups > 69 bpm was found. VNC reconstructions from PCD-CT reliably estimates TNC CACS for BMI ≤ 28 kg/m(2) and heart rate ≤ 69 bpm in patients with severe coronary artery disease. Potential underestimation of risk category, especially with increased BMI and heart rate, must be considered for clinical decision making.