Computed Tomography Coronary Angiography and Computational Fluid Dynamics Based Fractional Flow Reserve Before and After Percutaneous Coronary Intervention

Chandola, Gaurav; Zhang, Jun-Mei; Tan, Ru-San; Chai, Ping; Teo, Lynette; Allen, John C.; Low, Ris; Huang, Weimin; Leng, Shuang; Fam, Jiang Ming; Chin, Chee Yang; Kassab, Ghassan S.; Low, Adrian Fatt Hoe; Tan, Swee Yaw; Chua, Terrance; Lim, Soo Teik; Zhong, Liang

Invasive fractional flow reserve (FFR) is recommended to guide stent deployment. We previously introduced a non-invasive FFR calculation (FFR(B)) based on computed tomography coronary angiography (CTCA) with reduced-order computational fluid dynamics (CFD) and resistance boundary conditions. Current study aimed to assess the feasibility and accuracy of FFR(B) for predicting coronary hemodynamics before and after stenting, with invasive FFR as the reference. Twenty-five patients who had undergone CTCA were prospectively enrolled before invasive coronary angiography (ICA) and FFR-guided percutaneous coronary intervention (PCI) on 30 coronary vessels. Using reduced-order CFD with novel boundary conditions on three-dimensional (3D) patient-specific anatomic models reconstructed from CTCA, we calculated FFR(B) before and after virtual stenting. The latter simulated PCI by clipping stenotic segments from the 3D coronary models and replacing them with segments to mimic the deployed coronary stents. Pre- and post-virtual stenting FFR(B) were compared with FFR measured pre- and post-PCI by investigators blinded to FFR(B) results. Among 30 coronary lesions, pre-stenting FFR(B) (mean 0.69 ± 0.12) and FFR (mean 0.67 ± 0.13) exhibited good correlation (r = 0.86, p < 0.001) and agreement [mean difference 0.024, 95% limits of agreement (LoA): −0.11, 0.15]. Similarly, post-stenting FFR(B) (mean 0.84 ± 0.10) and FFR (mean 0.86 ± 0.08) exhibited fair correlation (r = 0.50, p < 0.001) and good agreement (mean difference 0.024, 95% LoA: −0.20, 0.16). The accuracy of FFR(B) for identifying post-stenting ischemic lesions (FFR ≤ 0.8) (residual ischemia) was 87% (sensitivity 80%, specificity 88%). Our novel FFR(B), based on CTCA with reduced-order CFD and resistance boundary conditions, accurately predicts the hemodynamic effects of stenting which may serve as a tool in PCI planning.