Diagnostic performance of coronary computed tomography angiography-derived fractional flow reverse in lesion-specific ischemia patients with different Gensini score levels

BACKGROUND: Coronary pressure-derived fractional flow reverse (FFR) is the standard of the functional assessment of lesion severity. In spite of its strengths in determining ischemia-related coronary stenosis, the invasive operation involved still limits its clinical application. Coronary computed tomography angiography-derived FFR (CCTA-FFR) or computed tomography-derived FFR (CT-FFR) has been indicated as an effective and non-invasive index to evaluate lesion-specific ischemia. However, its diagnostic performance, especially in patients with different severity of coronary stenosis, remains unknown. The current research attempted to demonstrate this problem and provided the foundation for extensive clinical application of CCTA-FFR. METHODS: The design of this study was a diagnostic test. A total of 97 vessels from 91 patients who performed CCTA and coronary angiography (CAG) during a hospitalization collected from two research centers were included in this study. CCTA-FFR and FFR were obtained by CCTA and CAG separately. The Gensini score was calculated according to the CAG in each patient. FFR was indicated as the golden diagnosis of lesion-specific ischemia with a cut-off value of 0.80, which was consistent with most contemporary studies. A receiver-operating characteristic (ROC) curve, simple linear analysis, and Bland-Altman plot were performed to determine the diagnostic performance of CCTA-FFR. RESULTS: CCTA-FFR was well correlated with invasive FFR (R(2)=0.745, P<0.001) and the area under the curve (AUC) was 0.976. The sensitivity was 94.6% and the specificity was 95.1%. The mean difference between FFR and CT-FFR was 0.011, and the 95% confidence interval was −0.173 to 0.196. The AUCs were 0.989 and 0.928 in the low and high Gensini groups, respectively, and there was no significant difference in the diagnostic accuracies between these two groups (Z=0.003, P>0.500). CT-FFR still exhibited a good correlation with FFR (R(2)=0.713, P<0.001 in the low Gensini group and R(2)=0.743, P<0.001 in the high Gensini group). The systematic differences were calculated, and the mean difference between FFR and CT-FFR was −0.005 and 0.025, respectively, in these two groups. CONCLUSIONS: CCTA-FFR exhibited good diagnostic performance in patients with different Gensini score levels. Our results indicate that CCTA-FFR could be an effective tool to screen lesion-specific ischemia in patients with coronary artery disease.