Diagnostic performance of fractional flow reserve derived from coronary angiography, intravascular ultrasound, and optical coherence tomography; a meta-analysis

Background: Little is known about the overall diagnostic performance of computational fractional flow reserve (FFR) derived from angiography (Angio-FFR), intravascular ultrasound (IVUS-FFR), and optical coherence tomography (OCT-FFR) to detect hemodynamically significant coronary artery disease. The present study aimed to evaluate the diagnostic performance of those novel physiologic indices using conventional FFR as the gold standard. Methods: PubMed and Embase were searched in September 2021 for a systematic review and meta-analysis of studies assessing the diagnostic performance of invasive imaging-derived FFR. The primary outcomes were the summary sensitivity, specificity, correlation coefficients of each index. Results: A total of 6572 records were initially identified and 49 studies were included in the final analysis (7010 lesions from 36 studies for Angio-FFR, 305 lesions from 5 studies for IVUS-FFR, and 667 lesions from 8 studies for OCT-FFR). Invasive imaging-derived FFR had a high diagnostic performance to detect functionally significant coronary lesions using conventional FFR as the gold standard [Angio-FFR, sensitivity 0.87 (95% CI 0.84–0.89), specificity 0.93 (95% CI 0.910.95); IVUS-FFR, sensitivity 0.90 (95% CI 0.84–0.94), specificity 0.95 (95% CI 0.90–0.98); OCT-FFR, sensitivity 0.85 (95% CI 0.78–0.91), specificity 0.93 (95% CI 0.89–0.95)]. The summary correlation coefficients of Angio-, IVUS-, and OCT-FFRs with wire-based FFR were 0.83 (95% CI 0.80–0.85), 0.85 (95% CI 0.79–0.91), and 0.80 (95% CI 0.74–0.86), respectively. Conclusions: This meta-analysis demonstrated that computational FFR derived from invasive coronary imaging has clinically acceptable diagnostic performances irrespective of modalities, supporting their applicability to clinical practice.