Diastolic versus systolic coronary computed tomography angiography derived fractional flow reserve for the identification of lesion-specific ischemia

To investigate the measurement discrepancy of coronary computed tomography angiography (CTA)-derived fractional flow reserve (FFR) between diastolic (CT-FFR-D) and systolic (CT-FFR-S) phases using FFR as the reference standard.Participants, suspected of coronary artery disease and indicated for invasive coronary angiography (ICA) and FFR and coronary CTA and CT-FFR, were enrolled in this study (Clinicaltrials.gov:NCT03692936) from September 2018 to October 2019. For every participant, coronary CTA of both systolic and diastolic phases was postprocessed to calculate CT-FFR-S and CT-FFR-D, respectively. Diagnostic sensitivity, specificity, accuracy, and the area under the receiver operating characteristic (ROC) curve were compared.A total of 181 lesions from 151 participants (mean age 54.5 ± 7.8 years, 113 males) were analyzed. Of these, 129 lesions from 110 participants were successfully measured both in diastolic and systolic phases. Sensitivity, specificity, and accuracy of CT-FFR-D and CT-FFR-S on per-patient level were 88.9%, 91.3%, 90.1% and 66.7%, 87.7%, 76.7%, on per-vessel level were 89.5%, 91.5%, 90.6% and 66.7%, 87.0%, 77.9%, respectively. The ROC curve of CT-FFR-D was significantly higher than that of CT-FFR-S on both per-patient and per-vessel levels (0.938 vs. 0.771, 0.935 vs. 0.772, both p < 0.0001). In severe hemodynamic lesions (FFR ≤ 0.7), the absolute difference between CT-FFR-S and FFR was significantly higher than that between CT-FFR-D and FFR [0.1636, inter-quartile range (IQR): 0.0662-0.2586 vs. 0.0953, IQR: 0.0496-0.1702, p = 0.035].CT-FFR derived in diastole was superior to that derived in systole in detecting coronary ischemic lesions. For lesions with FFR < 0.7, CT-FFR measured in the diastolic phase was noted to be more closely approximated to FFR.