Diagnostic performance of the quantitative flow ratio and CT-FFR for coronary lesion-specific ischemia

Abstract

Abstract Fractional flow reserve (FFR) has become the gold standard for evaluating coronary lesion-specific ischemia. However, FFR is an invasive method that may cause possible complications in the coronary artery and requires expensive equipment, which limits its use. Promising noninvasive diagnostic methods, such as computed tomography angiography–derived FFR (CT-FFR) and the quantitative flow ratio (QFR), have been proposed. In this study, we evaluated the diagnostic performance of the QFR and CT-FFR in predicting coronary lesion-specific ischemia, with the FFR serving as the reference standard. Patients with suspected or known coronary artery disease who underwent coronary CT angiography revealing 30–90% diameter stenosis in the main coronary artery (≥ 2.0 mm reference diameter) were enrolled. The FFR was measured during invasive coronary angiography (within 15 days after coronary CT angiography). An FFR ≤ 0.8 was the reference standard for coronary lesion-specific ischemia. A total of 103 vessels from 92 consecutive patients (aged 59.8 ± 9.2 years; 60.9% were men) were evaluated. The diagnostic performance of a QFR ≤ 0.80 for predicting coronary lesion-specific ischemia demonstrated good diagnostic accuracy, sensitivity, and specificity (92.2%, 87.2%, and 96.4%, respectively), with an area under the receiver operating characteristic curve (AUC) of 0.987 (P < 0.0001). The diagnostic performance of a CT-FFR ≤ 0.80 for predicting coronary lesion-specific ischemia also demonstrated good diagnostic accuracy, sensitivity, and specificity (96.1%, 95.7%, and 96.4%, respectively), with an AUC of 0.967 (P < 0.0001). However, there was no significant difference in the AUC between a QFR ≤ 0.80 and a CT-FFR ≤ 0.80 for predicting coronary lesion-specific ischemia (P = 0.319). There was an excellent correlation between the QFR and FFR (r = 0.856, P < 0.0001). The CT-FFR and FFR also showed a good direct correlation (r = 0.816, P < 0.0001). The QFR and CT-FFR are strongly correlated with the FFR and can provide excellent clinical diagnostic performance for coronary lesion-specific ischemia detection.