Diagnostic accuracy of Instantaneous Wave-Free Ratio (iFR) for the assessment of myocardial ischaemia: a comparison with non-invasive stress imaging

Abstract

Abstract Background Instantaneous Wave-Free Ratio (iFR) allows for the assessment of the haemodynamic effects of epicardial coronary stenoses without the need for hyperaemia; iFR is currently recommended as a means to evaluate myocardial ischaemia. Purpose To assess the diagnostic accuracy of iFR with respect to the identification of coronary epicardial stenoses causing ischemia. Therefore, we combined anatomical (% stenosis at invasive coronary angiography, ICA) and functional (non-invasive imaging stress test, NIST) information to obtain a “gold standard” for the identification of stenoses causing ischaemia. Methods We enrolled 71 patients (52 male, 19 female; age mean 68.4±8.1 years) with chronic coronary syndrome or low-risk acute coronary syndrome without ST segment elevation who had at least a NIST and who had at least one vessel with a 50%-85% stenosis at ICA. iFR was measured in all coronary arteries with stenosis >50% and categorised according to the 0.89 threshold for ischaemia. Results iFR was assessed in 122 vessels. In a per-vessel analysis, in 56.7% ischaemia was present both at iFR and NIST, in 21.3% ischaemia was absent in both, while in 23.0% ischaemia was found at NIST but not confirmed by iFR. The overall accuracy of iFR with respect to NIST was 90.1%. However, when considering as the “gold standard” for coronary disease causing ischaemia the contemporary presence of an epicardial stenosis >70% at ICA and a positive NIST, the diagnostic accuracy of iFR greatly improved. The sensibility, specificity, PPV, NPV and accuracy were 96.5%, 75.0%, 73.3%, 96.7% and 84.4%, respectively. In case of discordance between NIST and iFR, revascularization was based on iFR. At a mean follow-up of 23±18 months, the composite endpoint of MACE (major adverse cardiac events, defined as the composite of all-cause death, nonfatal MI and unplanned coronary revascularization) occurred in 16.4%, while death/MI occurred in 11.9%. Stratification according to the per-patient concordance between iFR and NIST showed no significant differences in rates of MACE (p=0.50) and death/MI (p=0.20). Stratification based on iFR showed a higher death/MI rate in iFR-positive patients (11.9% vs. 0%, p=0.047) and a trend to higher MACE rate (11.9% vs. 4.47% p=0.14), Conclusions The diagnostic accuracy of iFR is low when compared with NIST as the reference for myocardial ischaemia, but it is very high when compared with the combined presence of epicardial stenosis and positive NIST. Therefore, iFR can accurately guide the decision to treat or defer revascularization of intermediate coronary stenoses, being most useful in patients with multivessel CAD and when non-invasive functional data are lacking or discordant with anatomy. Funding Acknowledgement Type of funding source: None