P855Evaluation of epicardial coronary resistance using computed tomography angiography

Abstract

Abstract Background A Fractional flow reserve (FFR) pullback allows assessing the distribution of pressure loss along the vessel. FFR derived from CT (FFRCT) provides a virtual pullback curve that may also aid in the assessment of epicardial coronary resistance in the non-invasive setting. Purpose The present study aims to determine the accuracy of the virtual FFRCT pullback curve using a motorized invasive FFR pullback as reference in patients with stable coronary artery disease. Methods This is a single centre, prospective study of patients with stable coronary artery disease in whom FFRCT was performed as standard of care for non-invasive assessment. Patients referred to coronary angiography with clinically indicated invasive FFR measurement were included. FFRCT and invasive FFR values were extracted from coronary vessels every 1 mm to generate pullback curves. Invasive FFR pullbacks were acquired using a dedicated device at a speed of 1 mm/s. The area under the pullback curve (AUPC), defined as the sum of areas under the FFR pullback curve, was compared between FFRCT and invasive FFR pullbacks. Lesions were defined based on invasive angiography. FFR gradients in lesions and non-obstructive segments were defined as the difference between FFR values at the proximal and distal edge of the segments. FFR vessel gradient was defined as the difference between the most distal FFR value and the FFR at the ostium of the vessel. Mixed effect model was used to account for the correlation of FFR values within vessels. The agreement between FFRCT and FFR gradients was assessed using the Passing Bablok regression analysis and Bland-Altman methods at the vessel, lesion and non-obstructive level. Results A total of 3172 matched FFRCT and FFR values were obtained in 24 vessels. The correlation coefficient between FFRCT and FFR was 0.76 (95% CI 0.75 to 0.78; p<0.001). The mean difference between the FFRCT and invasive FFR pullback values was 0.07 (LOA −0.11 to 0.24). AUPC was similar between FFRCT and invasive FFR (79.0±16.1 vs. 85.3±16.4, p=0.097); the mean slope of FFRCT pullback curve was steeper compared to invasive FFR (p<0.001). The mean difference in lesion gradient was −0.07 (LOA −0.26 to 0.13) and −0.01 (LOA −0.06 to 0.05) in non-obstructive segments. There were no systematic or proportional differences between FFRCT and FFR gradients either in lesion or non-obstructive segments); however, vessel gradients were overestimated by FFRCT with a bias of −0.12 (LOA −0.35 to 0.12) driven by a higher mean difference in lesion gradients (−0.07; 95% CI −0.26 to 0.13). Conclusions The evaluation of epicardial coronary resistance using coronary CT angiography with FFRCT was feasible. FFRCT pullbacks were accurate in the assessment of lesion and non-obstructive gradients. FFRCT can identify the physiological pattern of coronary artery disease in the non-invasive setting.