ECG-Edited Middiastolic Phase Reconstruction Improves Image Quality at 64-MDCT Coronary Angiography of Patients with Atrial Fibrillation

OBJECTIVE. The aims of this study were to evaluate image quality at the absolute middiastolic and absolute end-systolic phases of 64-MDCT coronary angiography of patients with atrial fibrillation and to compare the findings with those among patients in sinus rhythm. SUBJECTS AND METHODS. Nineteen consecutively registered patients with atrial fibrillation and 19 patients in sinus rhythm taking heart-rate-lowering agents as needed underwent MDCT. Images were reconstructed with a half-scan reconstruction algorithm after ECG editing (deletion of short R-R intervals, insertion of additional temporal windows into the middiastolic phase of long R-R intervals, and shift of R points). We used a 5-point scale (4, no motion artifacts; 0, unevaluable) to evaluate motion artifacts and coronary artery image discontinuities greater than 1 mm on the curved multiplanar reconstruction images. Each coronary artery image with a motion score of 2 or greater for all segments and with 2 or fewer discontinuities was considered acceptable for diagnosis. RESULTS. Middiastolic images of patients with atrial fibrillation showed fewer motion artifacts and image discontinuities than did end-systolic images of patients with atrial fibrillation. Despite greater heart rate variability under the condition of similar mean heart rates in patients with atrial fibrillation, motion artifacts and image discontinuities on middiastolic images were not significantly different from those on sinus rhythm images. Acceptable quality was achieved on 91% of middiastolic atrial fibrillation images and 93% of sinus rhythm images. CONCLUSION. ECG-edited middiastolic atrial fibrillation images with aggressive heart rate control were of better quality than end-systolic images in patients with atrial fibrillation. The diagnostic image quality of the middiastolic images was comparable with that of sinus rhythm images.