Impact of SSF on Diagnostic Performance of Coronary Computed Tomography Angiography Within 1 Heart Beat in Patients With High Heart Rate Using a 256-Row Detector Computed Tomography

Abstract

Objective: The aim of this study was to investigate the impact of a motion-correction algorithm on diagnostic accuracy of coronary computed tomography angiography (CCTA) within 1 heart beat in patients with high heart rate (HR) using a 256-row detector CT. Method: Sixty-four consecutive patients with known or suspected coronary artery disease (symptomatic) and with HR of 75 beats per minute or greater (mean [SD] HR, 82.6 [7.3] beats per minute) undergoing CCTA and invasive coronary angiography within 4 weeks were prospectively enrolled. Coronary computed tomography angiography was performed with a 256-row detector CT (Revolution CT, GE Healthcare) using prospectively electrocardiography-triggered volume scan in 1 heart beat. All images were reconstructed using standard (STD) algorithm and a motion-correction algorithm reconstruction (Snapshot Freeze SSF; GE Healthcare) technique. The image quality of coronary artery segments was evaluated by 2 experienced radiologists using a 4-point scale based on the 18-segment model. Diagnostic accuracy was compared between STD and SSF for significant lumen stenosis (≥50%) of each segment with invasive coronary angiography as the reference standard for determining significant stenosis. Results: The diagnostic sensitivity, specificity, positive predictive value, and negative predictive value with STD and SSF were 93.7%, 85.1%, 50.2%, and 98.8% versus 91.9%, 95.8%, 77.9%, and 98.7% on per-segment assessment; 98.7%, 74.0%, 62.9%, and 99.2% versus 96.2%, 94.4%, 77.9%, and 98.7% on per-artery assessment; and 100%, 14.3%, 70.5%, and 100% versus 100%, 85.7%, 93.5%, and 100% on per-patient assessment, respectively. There was a significant difference in accuracy between STD and SFF on per-patient level 71.9% versus 95.3%, on per-artery level 81.6% versus 94.9%, and per-segment level 86.3% versus 95.3%, respectively. The area under receiver operating characteristics curve analysis also showed a significant improvement on diagnostic performance with the SSF technique versus with the STD algorithm on per-patient level (P < 0.001), with corresponding area under the curve being 0.91 (95% confidence interval, 0.79-1.00) and 0.60 (95% confidence interval, 0.44-0.75). The mean effective dose was 2.0 mSv. Conclusions: Coronary computed tomography angiography can be performed in patients with high HR within 1 heart beat yielding low radiation dose. The use of SSF technique reconstruction for 1 heart beat CCTA achieves significant improvements in image quality and diagnostic value.