Title: Diagnostic accuracy of 320-slice multi-detector computed tomography in patients with coronary artery bypass grafts
Abstract: Background: Non-invasive assessment of patients with prior coronary artery bypass grafts (CABG) by coronary CT angiography has been clinically applicable, however, it is hampered by some artifacts in the era of 64-slice multi-detector CT (MDCT). A new generation of MDCT with 320-row enables us to scan the entire heart in one rotation within one heartbeat, thus minimizing stair-step artifacts. However, there have been few prospective studies which investigate the diagnostic accuracy of 320-slice MDCT in patients with prior CABG. Method: From October 2010 to September 2012, 72 consecutive patients who underwent 320-slice MDCT coronary angiography (CTA) and invasive coronary angiography (ICA) within 2 months were prospectively enrolled. The CTA was visually analyzed for presence of significant stenosis (greater than 50%) or occlusion in grafts and coronary arteries. An independent observer blinded to MDCT data performed quantitative coronary angiography analysis as standard of reference. Agatston calcium scoring method was used to quantitate coronary calcification in MDCT. Result: Seventy-two patients (65 men, 68±11 years), with a total of 263 grafts (124 artery grafts, 139 venous grafts) were investigated. A total of 40 graft disease (24 grafts occlusion and 16 grafts stenosis) was confirmed by ICA. The mean amount of contrast media was 67±9 ml. Agatston score averaged 1513±1496 in native coronary arteries. The prevalence of evaluable grafts, distal runoff arteries, non-grafted arteries was 93.2%, 90.3%, 93.0%, respectively. The reasons for non-evaluablity were motion artifacts (52%), metal-clip (18%), calcification (18%), stair-step artifacts (7%) and poor opacification (5%). On evaluable grafts, including assessment of anastomoses, the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and diagnostic accuracy (DA) to detect significant stenosis or occlusion were 86.8%, 97.6%, 86.8%, 97.6% and 95.9% respectively. Predominant cause of false interpretation was small vessel diameter less than 1.5 mm in distal anastomoses of internal thoracic artery. On evaluable runoff arteries and non-grafted arteries analysis, the sensitivity, specificity, PPV, NPV and DA were all 100%. On a per-patient basis, classifying patients with at least 1 detected stenosis or occlusion or non-evaluable vessel as positive, the sensitivity, specificity, PPV, NPV and DA were 96.1%, 81.0%, 92.5%, 89.5% and 91.7% respectively. Conclusion: Assessment of patients with prior CABG by 320-slice MDCT seems to be useful in clinical practice.