King's College London

Research portal

Revisiting the Optimal Fractional Flow Reserve and Instantaneous Wave-Free Ratio Thresholds for Predicting the Physiological Significance of Coronary Artery Disease

Research output: Contribution to journalArticlepeer-review

Original languageEnglish
Pages (from-to)e007041
JournalCirculation: Cardiovascular Interventions
Issue number12
Early online date14 Dec 2018
Accepted/In press12 Oct 2018
E-pub ahead of print14 Dec 2018
PublishedDec 2018

King's Authors


BACKGROUND: There has been a gradual upward creep of revascularization thresholds for both fractional flow reserve (FFR) and instantaneous wave-free ratio (iFR), before the clinical outcome trials for both indices. The increase in revascularization that has potentially resulted is at odds with increasing evidence questioning the benefits of revascularizing stable coronary disease. Using an independent invasive reference standard, this study primarily aimed to define optimal thresholds for FFR and iFR and also aimed to compare the performance of iFR, FFR, and resting distal coronary pressure (Pd)/central aortic pressure (Pa).

METHODS AND RESULTS: Pd and Pa were measured in 75 patients undergoing coronary angiography±percutaneous coronary intervention with resting Pd/Pa, iFR, and FFR calculated. Doppler average peak flow velocity was simultaneously measured and hyperemic stenosis resistance calculated as hyperemic stenosis resistance=Pa-Pd/average peak flow velocity (using hyperemic stenosis resistance >0.80 mm Hg/cm per second as invasive reference standard). An FFR threshold of 0.75 had an optimum diagnostic accuracy (84%), whereas for iFR this was 0.86 (76%). At these thresholds, the discordance in classification between indices was 11%. The accuracy of contemporary thresholds (FFR, 0.80; iFR, 0.89) was significantly lower (78.7% and 65.3%, respectively) with a 25% rate of discordance. The optimal threshold for Pd/Pa was 0.88 (77.3% accuracy). When comparing indices at optimal thresholds, FFR showed the best diagnostic performance (area under the curve, 0.91 FFR versus 0.79 iFR and 0.77 Pd/Pa, P=0.002).

CONCLUSIONS: Contemporary thresholds provide suboptimal diagnostic accuracy compared with an FFR threshold of 0.75 and iFR threshold of 0.86 (cutoffs in derivation studies). Whether more rigorous thresholds would result in selecting populations gaining greater symptom and prognostic benefit needs assessing in future trials of physiology-guided revascularization.

View graph of relations

© 2020 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454