We were greatly interested by the study of Sen et al. (1), which proposes the revolutionary, vasodilator-independent index to assess significance of coronary artery stenosis—instantaneous wave-free ratio (iFR). The investigators identified a period during a cardiac cycle when intracoronary resistance is constant and minimal. The pressure ratio across a coronary stenosis during this period was found to correlate well with the fractional flow reserve (FFR) value obtained after adenosine administration. Good overall agreement between iFR and FFR was demonstrated by Bland-Altman analysis. A cutoff value of iFR of 0.83 corresponding to a FFR of 0.8 was calculated based on a receiver-operating characteristic (ROC) analysis.

Although the idea of performing pressure-derived stenosis assessment without pharmacological intervention is brilliant, we do not think that this work provided enough evidence that iFR correlates well with FFR from the clinician's standpoint.

The correlation plot of corresponding individual iFR and FFR values (Figs. 6 and 8 in Sen et al. [1]) actually showed considerable variability. For example, for the FFR value of 0.8, the iFR value ranged widely from <0.6 up to almost 1.0. Similarly, there was a broad range of FFR values (<0.6 to >0.9) corresponding to the iFR value of >0.83, which was defined as a cutoff value based on ROC analysis.

When looking carefully at the Bland-Altman plot (Fig. 10B in Sen et al. [1]) we notice that: 1) the difference between FFR and iFR exceeds the clinically tolerable threshold of 5% in more than half of the measurements; and 2) the agreement between the 2 methods is better for very high, nonsignificant FFR values (>0.85) than for values that indicate hemodynamically significant stenoses. Therefore, it would be very interesting to see in what percentage of individual cases disagreement between iFR and FFR would result in reclassification of stenosis from nonsignificant to significant and vice versa.

In summary: the investigators should be congratulated for an excellent innovation and a very elegant study. However, we feel that based on the data presented, the satisfactory statistical correlation may not translate into clinical usefulness.