Fractional Flow Reserve (FFR) has been established as the gold standard for determining the functional significance of coronary artery stenosis. Current guidelines have classified FFR as a Class IA recommendation for the assessment of intermediate coronary artery lesions. However, FFR remains underused in daily clinical practice, due to requirement for pressure wire use, hyperemia induction, or prolonged procedural time.
To overcome these limitations, angiography-derived computation of FFR have been widely adopted as wire-free alternatives. These technologies enable functional assessment of coronary stenosis without pressure wires, providing a less invasive and more comfortable alternative to wire-based FFR. Multiple modalities have shown reasonable diagnostic accuracy to predict FFR≤0.80. Among them, Quantitative Flow Ratio (QFR)-guided percutaneous coronary intervention (PCI) demonstrated superior clinical outcome than angiography-guided PCI. Based on these results, QFR-guided PCI is supported by class 1B recommendation from European Society of Cardiology guideline. Nevertheless, angiography-derived FFR also has limitations, primarily related to the technical and workflow demands of the process. Computation of angiography-derived FFR typically requires vessel segmentation, correspondence marking, and 3-dimensional reconstruction from angiographic images, which are time-consuming and subject to operator-dependent variability.
Indeed, recent data shows limitations of angiography-based FFR computation. Study by Ninomiya et al. evaluated five different angiography-derived FFR methods (QFR, vFFR from Pie Medical Imaging, caFFR from Rainmed Ltd, 2D-µFR, and 3D-µFR from Pulse Medical Imaging Technology). Although these angiography-derived FFR methods provided higher discrimination than angiographic stenosis severity to discriminate functionally significant stenosis defined by FFR≤0.80 or instantaneous wave-free ratio≤0.89, the AUC ranged from 0.65 to 0.75. Furthermore, recent FAVOR III Europe trial showed that QFR-guided strategy did not meet non-inferiority to FFR-guided strategy in terms of a composite of death, myocardial infarction, and unplanned revascularization at 12 months. These results support invasive FFR-guided strategy is gold standard method.
Recent advances in Artificial Intelligence (AI) have led to development of automated tools for cardiovascular diagnostics, improving both accuracy and workflow efficiency. The AI-driven angiography-based FFR (Medipixel FFR [MPFFR]) has been developed utilizing AI-based fully automated quantitative coronary angiography (AI-QCA). MPFFR utilizes automated frame selection, AI-based contouring, and real-time modeling, allowing for rapid and accurate physiological assessment without manual segmentation. In previous validation study conducted in Korea (599 vessels from 452 patients who underwent clinically indicated FFR measurement from 5 university hospitals in Korea), Mean analysis time of MPFFR was 12.5±1.7 seconds and manual correction was needed in 32 vessels (5.3%). MPFFR showed similar diagnostic performance with QFR (correlation with FFR; MPFFR vs. QFR: R=0.885 vs. R=0.860, P for comparison=0.011; area under curve to predict FFR≤0.80; 0.949 vs. 0.953, P for comparison=0.631). At a median follow-up of 2 years (interquartile range, 1.6 to 2.6 years), patients with MPFFR≤0.80 had higher risk of target vessel failure than those with MPFFR>0.80 (4.5% vs. 0.8%; adjusted HR, 5.94; 95% CI, 1.27-27.91; P=0.024). C-index to predict target vessel failure was comparable between MPFFR and QFR (0.770 vs. 0.753, P for comparison=0.469).
However, whether MPFFR-guided PCI can be used in daily practice still needs to be validated by randomized controlled trial using invasive FFR-guided PCI as reference standard. On this background, the current trial aims to compare clinical outcomes between MPFFR-guided PCI and invasive FFR-guided PCI in patients with coronary artery disease.
