Stenosis specific assessment
The current iteration of FFR measurement does not distinguish between the impact of collateral flow and as such, FFR does not provide stenosis specific information, rather a measurement of total ischemia in a given myocardial bed. To calculate stenosis specific information, (FFRcor) requires coronary wedge pressure (Pw) to be measured by performing balloon occlusion of a stenosis; the measured distal pressure during balloon occlusion estimates the flow contribution from collateral vessels. The final equation is: (Pd–Pw/Pa–Pw). Clearly this is not desirable for most moderate stenoses where balloon occlusion of a moderate stenosis essentially risks converting a potential deferral of PCI, into bailout stenting after endothelial injury or dissection. This is a limitation, and it means that FFR values reflect not only pressure loss generated by the stenosis, but the capacity of the myocardial microcirculation to dilate; vasodilatation in other territories may contribute to additional pressure loss, over and above caused by the stenosis.
Thresholds for significance and evidence to support
A cut‐off FFR value of <0.75 across an epicardial stenosis is accepted to be indicative of myocardial ischemia [1,2]. There is a close correlation between FFR <0.75 and different non‐invasive indices of reversible myocardial ischemia [6,10,11]. Whilst there may be discrepancy with individual functional tests, when multiple tests are performed, agreement is more likely; early work suggests FFR≤0.75 detects 97% of ischemia. Meta‐analytical work suggests the match with non‐invasive testing is typically 70% [10]. The DEFER study demonstrated it was clinically safe to defer revascularization with FFR values over 0.75 [11].
FFR values over 0.80 have over 90% sensitivity of excluding ischemia. The FAME and FAME II studies used a 0.80 threshold, ostensibly to ensure potentially ischemic stenoses were not missed. FAME randomized patients to either FFR‐guided revascularization in multi‐vessel disease or an angiographic approach where all stenoses ≥50% were stented; FFR reduced the number and length of stents placed and reduced the number of lesions considered significant. This lead to an improvement in composite outcomes of death, myocardial infarction, and repeat revascularization. FAME‐II randomized those patients with confirmed FFR≤0.80 to PCI with optimal medical therapy or optimal medical therapy alone. Two‐year follow‐up suggested the composite of death, non‐fatal MI and revascularization was significantly lower with PCI than medical therapy, although the majority of the events were urgent revascularization, and the trial was stopped early which limits its statistical power. Longer‐term follow‐up confirms the advantage noted at two years is sustained over a five year period.
Therefore, it is typical for revascularization to be offered when stenoses have FFR≤0.75, while deferral is more likely when FFR>0.80. For values in the “grey‐zone” (between 0.75–0.80), clinical judgement is required, combining knowledge of the patient’s clinical presentation with other testing. It is common practice to give additional doses of hyperemic agent when values are close to the threshold: higher doses can give confidence that maximal hyperemia has been achieved, but since flow is not being directly measured, caution is required since some changes can result from the hemodynamic disturbance of higher doses.
An area of great potential remains the use of FFR during routine clinical angiography. At present, FFR is used at the time of coronary intervention, either to defer a lesion when others are being treated, or to provide confirmation of ischemia when other tests are not available. Greater utility may be gained by performing three‐vessel physiological assessment at the time of angiography to objectively delineate the clinical significance of any stenoses; this will reduce the number of patients needing to return for further procedures and will offer more definitive diagnoses at the first procedure. The findings of studies suggest there may significant changes in medical decision making with the additional information [42,53].
FFR outside typical scenarios
FFR has been widely assessed in a number of different settings to assist clinical decision making. Whilst there are no randomized studies, many observational studies have demonstrated the safety of using FFR to determine revascularization decisions for left main stem disease, side‐branch vessels, and for guiding surgical graft placement. There remains modest data for the use of FFR within surgical grafts but the principal concept should be valid. As graft intervention is associated with a higher failure rate, native vessel intervention where feasible may provide a more durable result.
FFR in Acute Coronary Syndromes
FFR has been used in patients with acute coronary syndromes with reasonable reproducibility, despite concerns that the microvasculature is not optimally responsive to adenosine shortly after infarction. In STEMI, it should not be used in the culprit artery due to the impact of thrombus embolization, stunning of the myocardium and microvascular dysfunction. Non‐culprit vessels have been assessed with FFR in patients with STEMI. This has been performed during the index procedure for STEMI in the COMPARE‐ACUTE study [54], and after an interval within the same inpatient stay in the DANAMI3‐PRIMULTI study [55]. Both have shown utility in reducing later revascularization although there was no difference in major events. Meta‐analysis of studies considering multi‐vessel non‐culprit intervention vs culprit‐only intervention, have shown a benefit for revascularizing non‐culprit lesions. The benefit is seen by a reduction in re‐infarction rates, repeat revascularization, and cardiovascular mortality [56]. Assessing non‐culprit vessels remains an area of significant interest with many different modalities available [57].
In NSTEMIs, FFR has been used in non‐culprit lesions and these patients have been included in the original FAME study. However, in some cases the culprit lesions can be more challenging to identify. In the FAMOUS‐NSTEMI study, FFR was used to guide revascularization decisions in all vessels, which will have included the culprit vessel [58]. Although the study was underpowered to derive clinical outcomes and the primary outcome measure was not clinically meaningful, the event rate in patients deferred for PCI appeared to be high.
Algorithmically determined FFR
A number of studies have sought to correlate intravascular ultrasound (IVUS) or optical coherence tomography (OCT) parameters with FFR. There is also moves into predicting FFR based upon angiographic data alone [59]. A number of computational algorithms exist with good correlation with invasively measured FFR but larger outcome studies are awaited [60]. Advances in computational power now allow more accurate modelling of flow conditions but anatomical models are limited by their inability to predict the functional status of the distal microvasculature ‐ which is an important determinant of the functional significance of a given stenosis.
The instantaneous wave‐free ratio (iFR) and other non‐hyperemic pressure ratios (NHPRs)
The instantaneous wave‐free ratio (iFR) is a resting index of stenosis severity which quantifies the impact a stenosis upon the coronary circulation. A pressure‐only index, it is measured during the wave‐free period in diastole. This is a period where microcirculatory resistance is low, and importantly, stable, compared to other components of the cardiac cycle. When resistance is stable, pressure and flow are linearly related, enabling a resting pressure index to assess stenosis severity without the need of an exogenous vasodilator. iFR can be computed either as a single heartbeat or as an average over five heart beats. The wave‐free period is not significantly affected by beat‐to‐beat variability or atrial arrhythmias. Heart rates typical for performing physiological assessment (40–130 bpm). Automated algorithms can remove incorrectly calculated beats showing wide variability. iFR can also be plotted throughout the vessel during a gentle pressure wire pullback – performed over 30 seconds, it can identify focal and diffuse disease.
It is also possible to measure iFR during administration of a vasodilator such as adenosine (referred to as iFRa). This consistently produces lower