Background: Cardiovascular disease remains the leading cause of death and the use of percutaneous coronary intervention (PCI) is steadily increasing. Current guidelines advocate the use of the fractional flow reserve (FFR) to assess coronary stenosis and treatment strategies; however, invasive FFR has some limitations. Angiography-derived FFR is a potential alternative for calculating FFR from two-dimensional (2D) angiographic images, thereby reducing invasiveness and complications. A novel artificial intelligence (AI)-based angiography-derived FFR, named "MPFFR," offers automated operator-independent hemodynamic calculations; this phase 3 trial aims to validate its diagnostic performance against 2D-quantitative coronary angiography (QCA).
Methods and analysis: This pivotal MPFFR trial is a prospective, multicenter, single-blind study. This trial involves patients with coronary artery disease (CAD) from eight cardiovascular centers. Invasive FFR will be performed according to standard guidelines and defined as the reference standard. Angiography-derived FFR will be computed using a proprietary method and 2D-QCA will be performed using validated software. The primary endpoint is the area under the curve for identifying physiologically significant coronary stenosis (FFR ≤0.80), with secondary endpoints including diagnostic accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and correlations between angiography-derived and invasive FFR. This study is designed to demonstrate the superiority of angiography-derived FFR over 2D-QCA and is powered to achieve this with a sample size of 240 patients. Medipixel Inc. supports the trial and is not involved in the data analysis or management.
Keywords: Analytical equation-based model; Angiography-derived fractional flow rate; Artificial intelligence; Fractional flow rate.
Copyright © 2024 Elsevier Inc. All rights reserved.