Diagnostic accuracy of a novel optical coherence tomography-based fractional flow reserve algorithm for assessment of coronary stenosis significance

Background: This study aimed to propose a novel computational approach for assessing physiological significance of coronary lesions from optical coherence tomography (OCT), and to evaluate its diagnostic performance. Methods: A novel fractional flow reserve (FFR) algorithm (AccuFFRoct) based on the fusion of OCT and coronary angiography was developed to evaluate functional ischemia of coronary stenosis. Thirty-four consecutive patients underwent coronary angiography, OCT and FFR were included, and AccuFFRoct was used to calculate the FFR for all patients. The diagnostic performance of AccuFFRoct was compared with the wire-measured FFR reference standard. Results: Per vessel accuracy, sensitivity, specificity, positive predictive value and negative predictive value for AccuFFRoct in identifying hemodynamically significant coronary stenosis were 93.8%, 94.7%, 92.3%, 94.7%, and 92.3%, respectively. Good correlation (correlation coefficient r = 0.80, p < 0.001) between AccuFFRoct and FFR was observed. The Bland-Altman analysis showed a mean difference of -0.037 (limits of agreement: -0.189 to 0.115). The area under the curve (AUC) of AccuFFRoct in identifying physiologically significant stenosis was 0.94, which was higher than that of minimum lumen area (AUC = 0.91) and significantly higher than diameter stenosis (AUC = 0.78). Conclusions: The study demonstrated high efficiency and accuracy of AccuFFRoct for clinical implementation in functional assessment of coronary artery disease. It could provide additional insights beyond current coronary imaging-based anatomical assessment, aiding in clinical decision-making.