Diagnostic performance of quantitative flow ratio in prospectively enrolled patients: An individual patient-data meta-analysis

Objectives We aimed to provide robust performance estimates for quantitative flow ratio (QFR) in assessment of intermediary coronary lesions. Background Angiography-based functional lesion assessment by QFR may appear as a cost saving and safe approach to expand the use of physiology-guided percutaneous coronary interventions. QFR was proven feasible and showed good diagnostic performance in mid-sized off-line and on-line studies with fractional flow reserve (FFR) as reference standard. Methods We performed a collaborative individual patient-data meta-analysis of all available prospective studies with paired assessment of QFR and FFR using the CE-marked QFR application. The main outcome was agreement of QFR and FFR using a two-step analysis strategy with a multilevel mixed model accounting for study and center level variation. Results Of 16 studies identified, four studies had prospective enrollment and provided patient level data reaching a total of 819 patients and 969 vessels with paired FFR and QFR: FAVOR Pilot (n = 73); WIFI II (n = 170); FAVOR II China (n = 304) and FAVOR II Europe-Japan (n = 272). We found an overall agreement (mean difference 0.009 +/- 0.068, I-2 = 39.6) of QFR with FFR. The diagnostic performance was sensitivity 84% (95%CI: 77-90, I-2 = 70.1), specificity 88% (95%CI: 84-91, I-2 = 60.1); positive predictive value 80% (95%CI: 76-85, I-2 = 33.4), and negative predictive value 95% (95%CI: 93-96, I-2 = 75.9). Conclusions Diagnostic performance of QFR was good with FFR as reference in this meta-analysis of high quality studies. QFR could provide an easy, safe, and cost-effective solution for functional evaluation of coronary artery stenosis.