Radial Wall Strain Assessment From AI-Assisted Angiography: Feasibility and Agreement With OCT as Reference Standard.

ABSTRACT

Background: High-strain spots in coronary arteries are associated with plaque vulnerability and predict future events. Artificial intelligence currently enables the calculation of radial wall strain (RWS) from coronary angiography (RWSAngio). This study aimed to determine the agreement between novel RWSAngio and RWS derived from optical coherence tomography (OCT) followed by finite element analysis, as the established reference standard (RWSOCT). Methods: All lesions from a previous OCT study were enrolled. OCT was automatically coregistered with angiography. RWSAngio was computed as the relative luminal deformation throughout the cardiac cycle, whereas RWSOCT was analyzed using finite element analysis on OCT cross-sections at 1-mm intervals. The luminal deformation in the direction of minimal lumen diameter was used to derive RWSOCT, using the same definition as RWSAngio. The maximal RWSOCT and RWSAngio at healthy segments adjacent to the interrogated lesion were also analyzed. Results: Finite element analysis was performed in 578 OCT cross-sections from 45 lesions stemming from 36 patients. RWSAngio showed good correlation and agreement with RWSOCT (r = 0.91; P < .001; Lin coefficient = 0.85). RWSAngio in atherosclerotic segments was significantly higher than that in healthy segments (12.6% [11.0, 16.0] vs 4.5% [2.9, 5.5], P < .001). The intraclass correlation coefficients for intra- and interobserver variability in repeated RWSAngio analysis were 0.92 (95% CI, 0.87-0.95) and 0.88 (95% CI, 0.81-0.92), respectively. The mean analysis time of RWSOCT and RWSAngio for each lesion was 95.0 ± 41.1 and 0.9 ± 0.1 minutes, respectively. Conclusions: Radial wall strain from coronary angiography can be rapidly and easily computed solely from angiography, showing excellent agreement with strain derived from coregistered OCT. This novel and simple method might provide a cost-effective biomechanical assessment in large populations.