Diagnostic performance of computational fluid dynamics (CFD)-based fractional flow reserve (FFR) derived from coronary computed tomographic angiography (CCTA) for assessing functional severity of coronary lesions

BACKGROUND: Fractional flow reserve (FFR) is the gatekeeper for lesion-specific revascularization decision-making in patients with stable coronary artery disease (CAD). The potential of noninvasive calculation of FFR from coronary computed tomographic angiography (CCTA) to identify ischemia-causing lesions has not been sufficiently assessed. The objective of this study was to evaluate the feasibility and diagnostic accuracy of a novel computational fluid dynamics (CFD)-based technology, termed as AccuFFRct, for the diagnosis of functionally significant lesions from CCTA, using wire-based FFR as a reference standard. METHODS: A total of 191 consecutive patients who underwent CCTA and FFR measurement for suspected or known CAD were retrospectively enrolled at 2 medical centers. Three-dimensional anatomic model of coronary tree was extracted from CCTA data, CFD was applied subsequently with a novel strategy for the computation of FFR in a blinded fashion by professionals. Results were compared to invasive FFR, a threshold of ≤0.80 was used to indicate the hemodynamically relevant stenosis. RESULTS: On a per-patient basis, the overall accuracy, sensitivity, specificity of AccuFFRct for detecting ischemia were 91.78% (95% CI: 86.08% to 95.68%), 92.31% (95% CI: 81.46% to 97.86%) and 91.49% (95% CI: 83.92% to 96.25%), respectively; those for per-vessel basis were 91.05% (95% CI: 86.06% to 94.70%), 92.73% (95% CI: 82.41% to 97.98%) and 90.37% (95% CI: 84.10% to 94.77%), respectively. The AccuFFRct and FFR was well correlated on per-patient (r=0.709, P<0.001) and per-vessel basis (r=0.655, P<0.001). The AUC of AccuFFRct determination was 0.935 (95% CI: 0.881 to 0.969) and 0.927 (95% CI: 0.880 to 0.960) on per-patient and per-vessel basis. CONCLUSIONS: This novel CFD-based CCTA-derived FFR shows good diagnostic performance for detecting hemodynamic significance of coronary stenoses and may potentially become a new gatekeeper for invasive coronary angiography (ICA).