Reliability of a new method for coronary artery calcium or metal subtraction by 320-row cardiac CT

© 2015, European Society of Radiology. Purpose: To investigate the feasibility and diagnostic accuracy of subtraction CTA on patients with highly calcified coronary artery disease (CAD) or previous implanted stents, in comparison with invasive coronary angiography (ICA). Materials and methods: Twenty-three patients were recruited. All conventional and subtraction CTA exams were performed using a 320-row CT. Subjective image quality score was assessed for each segment using a 4-point scale: 1-uninterpretable to 4-good image quality. Results: A total of 129 calcified or stented coronary segments were studied. Mean coronary image quality with conventional CTA was 2.73 ± 0.97 and in subtracted CTA 3.3 ± 0.92 (p < 0.01). After metal subtraction, image quality in stented coronary segments with >3 mm of diameter improved from 2.69 ± 0.97 to 3.34 ± 0.89 (p = 0.01) and in those with <3 mm of diameter from 2.11 ± 0.78 to 2.67 ± 0.87 (p = 0.17). There was an improvement in diagnostic accuracy to detect ICA stenosis >50 % by subtraction CTA compared with conventional CTA (AUC 0.93 to 0.87; p = 0.02). Conclusion: Subtraction CTA is promising in overcoming limitations of conventional CTA due to calcium or metal artefacts, especially if no motion artefact is present or when stents > 3 mm are studied. Key Points: •Calcium and metal artefacts are still a limitation for conventional coronary CTA •Diagnostic accuracy is improved by subtraction as compared with conventional CTA •Subtraction CTA is a promising tool to overcome limitations of conventional CTA