Stuart Watkins, MD, Golden Jubilee National Hospital, Glasgow, Scotland, United Kingdom, presented the results of the Verification of Instantaneous Wave-Free Ratio and Fractional Flow Reserve for the Assessment of Coronary Artery Stenosis Severity in Everyday Practice trial [VERIFY-2] comparing the performance of nonhyperemic indices of stenosis severity (instantaneous wave-free ratio [iFR] or resting distal coronary pressure / aortic pressure ratio [Pd/Pa]) with fractional flow reserve (FFR) for assessing coronary stenosis severity. Lesion-level decision making was evaluated with both a hybrid strategy and a binary cutoff value of iFR and Pd/Pa compared with FFR.

Pressure wire–derived FFR is a validated coronary lesion–level index of functional significance. However, FFR is not widely used, owing to cost, extra procedural time, and the inconvenience of intravenous or intracoronary adenosine administration. The VERIFY [Berry C et al. J Am Coll Cardiol 2013] and RESOLVE [Jeremias A et al. J Am Coll Cardiol 2014] studies confirmed that iFR and Pd/Pa have a similar diagnostic accuracy of ∼ 80% compared with FFR. Researchers of the ADVISE II study [NCT01740895; Escaned J et al. TCT 2013] reported that with a hybrid iFR-FFR strategy, they correctly classified lesions in 94.2% of cases while avoiding adenosine administration in 65.1% of patients.

The prospective VERIFY-2 trial comprised 97 near-consecutive patients with chest pain and moderately severe coronary artery stenoses. Following diagnostic angiography, the Volcano Prestige Pressure Wire was inserted into the distal third of the coronary artery beyond the lesion. Resting Pd/Pa and iFR were recorded. Intravenous adenosine was administered, and FFR was recorded at stable maximal hyperemia.

A total of 120 lesions were studied. The mean Pd/Pa was 0.93 ± 0.06; mean iFR was 0.90 ± 0.08; and mean FFR was 0.82 ± 0.09. Assessment of the concordance of hybrid strategies based on FFR ≤ 0.8 as the gold standard showed that 10.1% of lesions were misclassified with iFR and 6.3% were misclassified with Pd/Pa (Table 1).

Based on this analysis, the rates of inappropriate percutaneous coronary intervention (PCI) and incomplete revascularization would be 8.7% and 10.9% with hybrid iFR-FFR and 0.0% and 7.7% with hybrid Pd/Pa-FFR, respectively.

Use of iFR and Pd/Pa with a predefined cutoff value compared with FFR showed that 18.3% of lesions were misclassified with iFR and 15.0% were misclassified with Pd/Pa (Table 2).

Based on this analysis, the rates of inappropriate PCI and incomplete revascularization would be 8.3% and 10.0% with iFR < 0.9 and 4.2% and 10.8% with Pd/Pa < 0.92, respectively.

Receiver operating characteristic (ROC) curves comparing iFR and Pd/Pa to FFR ≤ 0.8 showed an area under the curve (AUC) of 0.873 (95% CI, 0.805 to 0.941) for iFR and 0.889 (95% CI, 0.82 to 0.958) for Pd/Pa. ROC curves comparing iFR and Pd/Pa to FFR ≤ 0.75 showed an AUC of 0.936 (95% CI, 0.886 to 0.986) for iFR and 0.946 (95% CI, 0.899 to 0.993) for Pd/Pa.

These results showed that using a hybrid strategy with Pd/Pa-FFR or iFR-FFR provides similar levels of misclassification compared with FFR. Using a binary cutoff level for Pd/Pa or iFR results in similar levels of misclassification compared with FFR. VERIFY-2 confirmed that the diagnostic accuracy of iFR is no better than that of Pd/Pa. Whether used as part of a hybrid or binary algorithm, neither resting index is sufficiently accurate to be used as a guide to the need for revascularization.