BACKGROUND: A coronary artery calcium score (CACS) of 0 is associated with a very low risk of cardiac event. However, the Agatston CACS may fail to detect very small or less dense calcifications. We investigated if an alteration of the Agatston criteria would affect the ability to detect such plaques.
METHODS: We evaluated 322 patients, 161 who had a baseline scan with CACS = 0 and a follow-up scan with CACS>0 and 161 with two serial CACS = 0 scans (control group), to identify subtle calcification not detected in the baseline scan because it was not meeting the Agatston size and HU thresholds (≥1 mm2 and ≥130HU). Size threshold was set to <1 mm2 and the HU threshold modified in a stepwise manner to 120, 110, 100 and 90. New lesions were classified as true positive or false positive(noise) using the follow-up scan.
RESULTS: We identified 69 visually suspected subtle calcified lesions in 65/322 (20.2%) patients with CAC = 0 by the Agatston criteria. When size threshold was set as <1 mm2 and HU ≥ 130, 36 lesions scored CACS>0, 34 (94.4%) true positive and 2 (5.6%) false positive. When decrease in HU (120HU, 110HU, 100HU, and 90HU) threshold was added to the reduced size threshold, the number of lesions scoring>0 increased (46, 55, 59, and 69, respectively) at a cost of increased false positive rate (8.7%, 20%, 22%, and 30.4% respectively). Eliminating size or both size and HU threshold to ≥120HU correctly reclassified 9.6% and 12.1% of patients respectively.
CONCLUSION: Eliminating size and reducing HU thresholds to ≥120HU improved the detection of subtle calcification when compared to the Agatston CACS method.