Virtual coronary artery bypass grafting

Abstract Coronary artery bypass grafting (CABG) offers superior long-term survival over percutaneous coronary intervention (PCI) or medical therapy in patients with complex coronary artery disease (CAD). This prospective proof-of-concept study aims to develop and validate a non-invasive computational platform that integrates coronary computed tomographic angiography (CCTA) and computational fluid dynamics (CFD) to predict post-CABG hemodynamics, including virtual grafting and fractional flow reserve (FFR) estimation. Four patients with stable multi-vessel CAD undergoing elective CABG were included. Pre-CABG CCTA was used for 3D reconstruction of coronary anatomy. Virtual bypass grafting was performed using both patient-specific graft sizes, derived from post-operative imaging and mixed-specificity graft sizes using patient-specific LIMA and standardized non-LIMA graft sizes, derived from population averages. CFD simulations were used to estimate post-CABG FFR and validated against invasive FFR measurements. Computational FFR showed strong correlation with invasive FFR (patient-specific: r² = 0.92; mixed-specificity: r² = 0.88). Bland-Altman analysis demonstrated minimal bias (patient-specific: 0.006 ± 0.027; mixed-specificity: -0.007 ± 0.029). Agreement with invasive FFR was 90% for patient-specific grafts (κ = 0.74, p = 0.016) and 80% for mixed-specificity grafts (κ = 0.41, p = 0.107). This virtual CABG model represents a significant advancement over existing non-invasive systems by accurately predicting post-operative hemodynamics and FFR, offering potential to optimize graft strategies and reduce reliance on invasive FFR. Future studies should explore clinical integration and large-scale validation to enhance CABG surgical planning and improve patient outcomes.