8.1 iFR (Instantaneous Wave-free Ratio)

0. iFR update (3 May 2017)

This page was originally written in 2014 and there have been significant developments in recent years. Two of the clinical trials comparing iFR to FFR have been completed and published in the New England Journal of Medicine in the 18 March 2017 issue:

DEFINE-FLAIR (NCT02053038) [J.E Davis et al. (2017) DOI: 10.1056/NEJMoa1700445]
iFR-Swedeheart (NCT02166736) [M Götberg et al. (2017) DOI: 10.1056/NEJMoa1616540]

Both studies show that iFR is clearly non-inferior to FFR as a clinical indicator of stenting. Considering the shorter procedure time and the benefits of not using Adenosine, it seems that iFR is on its way to becoming the clinical method of choice.

And now, back to 2014...

1. Introduction to iFR

The most interesting recent clinical application of wave intensity analysis is the development of the instantaneous wave-free ratio (iFR) as a measure of the functional importance of a coronary stenosis. [Sen, et al. (2011)] Since its introduction in 2011, there has been a very rapid development of the technique. At the time of writing (March 2015), iFR has been incorporated into a clinical device which is now installed in more than 2000 catheter labs worldwide and it is being used in several large randomised clinical trials*. Full adoption of this method of assessing the functional effect of coronary stenoses must, of course, await the outcome of these trials. A recent review of iFR [Nijjer et al. 2014a] gives the current clinical status of the method. For up to date information about the rapid development of iFR refer to the Instantaneous wave-free ratio entry in Wikipedia.

A more recent but equally interesting development is the combination of wave intensity analysis and iFR applied to catheter data obtained by pulling the catheter back through the coronary vessel. The technique and preliminary clinical data are described in Nijjer et al. (2014b). This methodology has enormous potential for changing clinical intervention by PCI** because it can determine whether the disease is diffuse (and therefore not ideal for PCI) or focal (and therefore ideal for PCI). The technique has been available commercially for about a year as iFR-Scout and a demonstration is available as a YouTube demonstration of iFR-Scout.

* The name and International Trial Record Numbers of the iFR trials are:
        DEFINE-FLAIR (NCT02053038)
        iFR-Swedeheart (NCT02166736)
        Syntax II (NCT02015832).

**PCI stands for Percutaneous Coronary Intervention which is the clinical name for coronary stenting.

2. Wave intensity analysis in coronary vessels

The quantitative assessment of the severity of a coronary stenosis, i.e. its affect on the perfusion of the myocardium, has been a problem since the inception of coronary stenting. Initially, decisions about which stenoses to stent were made on the bases of angiography, but it quickly became clear that the image of the lumen of the stenosis was not a reliable indicator of its effect on coronary blood flow. Single plane angiography can miss significant narrowing out of the plane of view and multi-plane angiography suffers similar problems when the shape of the stenosis is highly irregular. It seemed clear that a more functional measurement based on either pressure of flow measurements was required to evaluate the functional severity of a coronary stenosis.

The complexity of coronary artery dynamics makes these measurements difficult to interpret. Unlike other arteries where the majority of flow occurs during systole, the contraction of the myocardium compresses the intramyocardial blood vessels, increasing their resistance to flow during systole. When myocardial tension relaxes during systole, the intramyocardial blood vessels decompress, reducing their resistance during diastole. As a result, flow in the coronary arteries occurs predominantly in diastole.

Wave intensity analysis has proven to be a very useful tool in the study of coronary artery dynamics. We have produced a number of papers analysing the forward and backward waves from simultaneous measurements of pressure and velocity in coronary arteries. [Davies et al. I2006)] This is seen in the Figure which shows the pressure, velocity and wave intensity measured in a typical systemic artery (the abdominal aorta) and a coronary artery (the left anterior descending (LAD)) together with the parameters separated into their forward and backward components as discussed earlier [ Wave separation ].

P, U and dI in left abdominal aorta (black) - total values (blue) - forward wave (red) - backward wave	P, U and dI in LAD coronary artery (black) - total values (blue) - forward wave (red) - backward wave

We see from the figure that the measured velocity in the systemic artery increases rapidly during early systole and falls to almost zero throughout diastole whereas in the LAD most of the flow is occurring during diastole with the velocity decreasing during the early part of systole. The reason for this can be seen from the separated wave intensity calculated from the P and U measurements. The pattern of dI in the systemic arteries is a large forward compression wave during early systole, caused by the contraction of the myocardium, followed by a forward expansion wave at the end of systole, caused by the cessation of the myocardial contraction. Relative to these forward waves, there is very little backward wave intensity throughout the cardiac cycle. This pattern is typical of all of the normal systemic arteries in which we have made measurements.

In the LAD, the pattern of wave intensity is very different. At the start of systole we see a peak in both the forward and backward wave intensity; the forward compression wave generated by the rapid increase in left ventricular pressure due to the contraction of the myocardium and the backward peak caused by the compression of the intramyocarial blood vessels distal to the LAD by the contracting myocardium which generates a backward compression wave that decelerates the LAD blood flow. The forward and backward compression waves have opposite effects on the velocity and tend to cancel each other resulting in little change in the velocity at the start of systole. The biggest difference from the systemic artery is the very large backward peak in the wave intensity at the end of systole corresponding to a decompression wave generated by the decompression of the intramyocardial blood vessels. This backward decompression wave decreases the pressure but accelerates the blood, giving rise to the rapid increase in velocity during diastole.

The idea behind all measurements of the functional effect of a coronary stenosis can be described using a simple hydraulic model. For steady flow, the total resistance in the coronary artery can be represented as two resistance elements in series, the resistance due to the stenosis R_s, and the resistance of the microcirculation proximal to the artery R_m. By conservation of mass, the volume flow rate Q through the coronary must be constant Assuming that the resistances obey Ohms law, the difference between the proximal P_a and the distal P_d pressures is simply

P_a - P_d = Q R_s

Similarly, the total pressure drop across both the stenosis and the microcirculation is

P_a - P_v = Q R_T where P_v is the venous pressure, and the total resistance R_T = R_s + R_m. Combining these two equations we get a simple relationship for the resistance of the stenosis relative to the total resistance to flow

R_s/R_T = (P_a - P_d)/(P_a - P_v)

In practice, it is generally assumed that the venous pressure is negligible and so the fractional resistance is usually written as simply

R_s/R_T = (P_a - P_d)/P_a

If everything was as simple as assumed in this model, it would not be difficult to ascertain the functional effect of a coronary stenosis clinically. However, a more realistic model of coronary resistance, illustrated in this figure, highlights some of the practical problems. Most importantly, the whole process is highly unsteady with pressures, flows and resistances changing greatly throughout the cardiac cycle. Another factor which greatly complicates the problem is the existence of anastomoses in the cardiac microcirculation which offers parallel paths for flow. These anastomoses are highly variable in their distribution within and between different coronary arteries and between patients. It is commonly believed that they can develop in response to cardiac pathology, particularly perfusion injury. The temporal variations and the variable parallel paths for flow make it very difficult to find a reliable measure of the functional resistance of the stenosis clinically. [For a full and authoritative discussion of these problems I would recommend [Meuwissen et al. (2002)]

3. Fractional Flow Reserve (FFR)

In the 1990s the fractional flow reserve (FFR) index was suggested as a measure of the influence of epicardial coronary stenoses [Pijls et al. (1995)] The method was based on the idea that the administration of a vasodilator, most commonly adenosine, to the coronary vessel would reduce the temporal variation of the microcirculatory resistance so that the ratio (P_d - P_v)/P_a based on time-average values of P is related to the the fractional resistance of the stenosis. In most applications it is assumed the the venous pressure P_v = 0, so that the FFR is defined as P_d/P_a.

FFR has been used widely and shown in a number of large clinical trials to be the best indicator of the need for coronary intervention (i.e. stenting). [Pijls et al. (2105); Pijls et al. (2010)]. Despite these clinical studies, FFR is used in only a small fraction of stenting procedures; cardiologists do not like it because the administration of adenosine prolongs the duration of the clinical test and the patients do not like it because of the unpleasant side-effects of the drug. Also, because of the length of time necessary for serial administration of adenosine, it is time consuming to assess changes in the FFR over the duration of a procedure.

4. Instantaneous Wave-free Ratio (iFR)

From the analysis of pressure and velocity measurements in coronary arteries, it was observed that the effective resistance, defined as the measured pressure divided by the measured velocity, was relatively constant during much of diastole and that this resistance was fairly constant from beat to beat. This can be seen in the figure (from Sen et al. (2012)) which shows P(t) and U(t) together with R(t) = P(t)/U(t) measured in the right coronary artery of a patient. This gave rise to the idea that the pressure ratio during this time could be used in the same way as in the FFR measurement but, critically, without the need for the administration of a vasodilator. This is shown in the figure where the top row is the velocity U(t), the second row is the proximal pressure Pd in dark blue and distal pressure Pd in light blue, the third row is the total resistance RT = Pa/U. It is evident that the total resistance is both minimal and relatively constant during much of diastole. This is consistent with the fact that the interstitial pressure is smallest during diastole when the myocardium is relaxed and so there is the minimum compression on the intramyocardial vessels that contribute to the microcirculatory resistance.

The problem now is to identify the period in diastole when the resistance is constant, the 'wave-free period. This is most conveniently done using the wave intensity, as shown in the figure (from Sen et al. (2012)). From the separated wave intensities, we see that the wave-free period extends over most of the period between the large negative peak in the backward wave intensity, indicating the backward decompression wave generated by the relaxation of the myocardium, and the start of systole marked either by the R-wave on the ECG or the large forward compression wave generated by the contraction of the myocardium. Thus, the iFR method calculates the ratio of the distal to proximal pressures across the stenosis during the wave-free period. It should be emphasised that although the iFR and the FFR have the same mathematical definition, they do not represent the same thing since iFR is measured during a short period during the normal cardiac cycle and the FFR is based on mean pressures during maximal vasodilatation. They both, however, represent a quantitative measure of the resistance due to the stenosis and the total resistance of the artery and it is reasonable to expect that they will have similar predicative value clinically. The major benefit of the IFR is that it does not require the administration of adenosine. A secondary benefit is that it can be measured beat by beat and provides an easier means to follow the acute changes occurring in a clinical procedure.

Although iFR and FFR have a somewhat different physiological foundation, it has been shown in a small study that their values correlate very well with each other. The figure shows a comparison of iFR and FFR measured in approximately 130 stenosed coronary arteries As can be seen, the correlation between the two measures is extremely good with the line of best fit having a slope that is remarkably close to 1, which represents identity. As a proof of concept study iFR and FFR were compared over a wide range of stenosis severities (Sen et al. 2011). However the majority of patients requiring functional assessment in the cardiac catheter laboratory have lesions that are less likely to be at the extremes of severity. As a result further studies including over 800 stenoses in more clinically relevant populations of stenoses have been performed (Davies et al. 2012; Park et al. 2012). All demonstrate that iFR has close agreement with FFR treatment categorisation. Of course, the clinical efficacy of iFR must be demonstrated in a full clinical trial of outcomes. However, it seems that iFR may prove to be useful clinically and it currently is one of the most exciting of clinical uses of wave intensity analysis. For the most recent developments of iFR see the Instantaneous wave-free ratio entry in Wikipedia.

5. References

Davies JE, Whinnett ZI, Francis DP, et al. (2006) Evidence of a dominant backward-propagating “suction” wave responsible for diastolic coronary filling in humans, attenuated in left ventricular hypertrophy. Circulation 113:1768 –78. [pdf]

Davies JE, Asress K., Echavarria-Pinto M., et al. (2012) Physiological assessment of intermediate coronary stenoses with instantaneous wave-free ratio, an adenosine-free pressure-derived index: results of a multi center international registry. Eurointervention 8: 128.

Meuwissen M, Siebes M, Spaan JA, Piek JJ. (2002) Rationale of combined intracoronary pressure and flow velocity measurements. Z Kardiol 91 Suppl 3:108-12.

Nijjer SS, Sen S, Petraco R, Escaned J, Echavarria-Pinto M, Broyd C, Al-Lamee R, Foin N, Foale RA, Malik IS, Mikhail GW, Sethi AS, Al-Bustami M, Kaprielian RR, Khan MA, Baker CS, Bellamy MF, Hughes AD, Mayet J, Francis DP, Di Mario C, Davies JER, (2014a) Pre-Angioplasty Instantaneous Wave-Free Ratio Pullback Provides Virtual Intervention and Predicts Hemodynamic Outcome for Serial Lesions and Diffuse Coronary Artery Disease. J Am Coll Cardiol Intv. 7: 1386-1396. doi:10.1016/j.jcin.2014.06.015

Nijjer SS, Sen S, Petraco RP, Escaned J, Echavarria-Pinto M, Broyd C, Al-Lamee R, Foin N, Foale RA, Malik IS, Mikhail GW, Sethi AS, Al-Bustami M, Kaprielian RR, Khan MA, Baker CS, Bellamy MF, Hughes AD, Mayet J, Francis DP, Di Mario D, Davies JER. (2014b) Pre-angioplasty instantaneous wave-free ratio pullback provides virtual intervention and predicts hemodynamic outcome for serial lesions and diffuse coronary artery disease. J Am Cardiol Intv. 7: 1386–1396. [link to the paper] on the JACC website.

Park JJ, Yang HM, Park KW., et al. (2012) Diagnostic performance of a novel index, the instantaneous wave-free ratio (iFR), for the detection of functionally significant coronary artery stenosis. Eurointervention 8: 130.

Pijls NH, Van Gelder B, Van der Voort P, et al. (1995) Fractional flow reserve. A useful index to evaluate the influence of an epicardial coronary stenosis on myocardial blood flow. Circulation 92: 3183–93.

Pijls NH, van Schaardenburgh P, Manoharan G, et al. (2007) Percutaneous coronary intervention of functionally nonsignificant stenosis: 5-year follow-up of the DEFER Study. J Am Coll Cardiol 49: 2105–11.

Pijls NH, Fearon WF, Tonino PA, et al. (2010) Fractional flow reserve versus angiography for guiding percutaneous coronary intervention in patients with multivessel coronary artery disease: 2-year follow-up of the FAME (Fractional Flow Reserve Versus Angiography for Multivessel Evaluation) study. J Am Coll Cardiol 56: 177– 84.

Sen, S, Escaned J, Malik IS, et al. (2011) Development and Validation of a New Adenosine-Independent Index of Stenosis Severity From Coronary Wave–Intensity Analysis: Results of the ADVISE (ADenosine Vasodilator Independent Stenosis Evaluation) Study, JACC, 59, 1392-1402. [pdf]