A two center 320 slice CT study for evaluating coronary arteries in subjects with chronic atrial fibrillation: a comparison of prospective and retrospective ECG-gating acquisition

Purpose: To compare diagnostic-accuracy of prospective and retrospective-ECG-gated acquisition in 320-slice-CT for detecting coronary-artery stenosis in subjects with chronic-atrial-fibrillation (CAF) in a two-center study.

Materials and methods: 53 CAF subjects who underwent 320-slice-CT, and invasive-coronary-angiogram (ICA) within 6-months (43 male; 69 ± 9 years; CHADS2 score 2.2 ± 1.3; CHA2DS2-VASc score 3.5 ± 1.6) in the two institutes were analyzed. In Institute-1, prospective-ECG-gated acquisition was routinely performed (N=33). In Institute-2, retrospective-ECG-gated acquisition was routinely performed (N=20). CT and ICA data were transferred to the analysis center and were analyzed by cardiologists blinded to the clinical-data.

Results: Prevalence of >50 and >75% on ICA was 79 and 61% in Institute-1, and 30 and 15% in Institute-2, respectively. In a patient-by-patient analysis, Institute-2 had higher negative-predictive-value (NPV) and accuracy of >75% stenosis on CT in predicting >75% stenosis on ICA. In a vessel-by-vessel analysis, there were no significant-differences of sensitivity, specificity, positive-predictive-value (PPV) and NPV of >50% stenosis on CT in predicting >50% stenosis on ICA between both institutes. But sensitivity, specificity, and NPV of >75% stenosis on CT in predicting >75% stenosis on ICA were significantly higher in Institute-2 than in Institute-1. This is mainly because of more severe coronary-artery disease including calcification in Institute-1; there might also have been an influence of differences in scanning and reconstruction methods.

Conclusions: 320-slice-CT shows relatively high diagnostic-accuracy for the detection of significant coronary-artery stenosis compared with ICA even in CAF subjects, in a two-center analysis. Retrospective-ECG-gated acquisition in 320-slice-CT shows significantly higher diagnostic-accuracy than prospective-ECG-gated acquisition for detection of >75% coronary-artery stenosis.