Can coronary artery anomalies be detected on CT calcium scoring studies?

Abstract

Purpose: To determine whether coronary artery anomalies can be detected on noncontrast computed tomography (CT) coronary artery calcium scoring (CCS) studies.

Materials and methods: A total of 126 patients (mean age 62 years; 35 women) underwent noncontrast CCS and contrast enhanced coronary CT angiography (cCTA). Thirty-three patients were diagnosed with a coronary anomaly on cCTA, whereas coronary anomalies were excluded in 93. Two observers (reader 1 [R1] and reader 2 [R2]), blinded to patient information independently evaluated each CCS study for: 1) visibility of coronary artery origins, 2) detection of coronary anomalies, and 3) benign or malignant (ie, interarterial) course. Using cCTA as the reference standard, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of CCS studies for detecting coronary anomalies were calculated.

Results: Of the 33 coronary anomalies, 16 were benign and 17 malignant. Based on noncontrast CCS studies, R1 and R2 correctly identified the left main origin in 123/126 (97.6%) and 121/126 (96%) patients; the left anterior descending origin in 125/126 (99.2%) and 122/126 (96.8%); the circumflex origin in 120/126 (95.2%) and 105/126 (83.3%); and the right coronary artery origin in 117/126 (92.9%) and 103/126 (81.7%), respectively. R1 and R2 identified 34 and 27 coronary anomalies and classified 19 and 15 as malignant, respectively. Interobserver reproducibility for detection of coronary anomalies was good (k = 0.76). Interobserver agreement for detection of malignant variants was even stronger (k = 0.80). On average, coronary artery anomalies were diagnosed with 85.2% sensitivity, 96.4% specificity, 90.5% PPV, and 94.1% NPV on noncontrast CCS studies.

Conclusion: Benign and malignant coronary artery anomalies can be detected with relatively high accuracy on noncontrast-enhanced CCS studies. CCS studies should be reviewed for signs of coronary artery anomalies in order to identify malignant variants with possible impact on patient management.