Introduction

Fractional flow reserve (FFR) has become a gold standard index for the invasive assessment of physiological severity of coronary artery stenosis. Coronary angiography is the traditional imaging modality for visual evaluation of coronary lesion severity and guidance of percutaneous coronary interventions (PCI)s. However, coronary angiography is in reality a two-dimensional shadowgraph of the vessel lumen. It cannot depict the arterial vessel wall or plaque burden. Moreover, it cannot approximate the amount of myocardium subtended by the target vessel. Without a proper assessment of physiological lesion severity, PCIs might mitigate its preferable effects on patient outcome [1].

Considering the importance of discriminating the lesions most likely to derive clinical benefit from PCIs, the Japanese Central Social Insurance Medical Council has changed the requirement for the reimbursement of PCIs in April 2018.

Following this change, the role of physiological lesion assessment in catheter laboratories is expected to increase considering the low penetration rate of the non-invasive physiological test in Japan.

In addition, a new hyperemia-free index as an indication of PCIs, namely instantaneous wave-free period (iFR) has recently been validated and its non-inferiority in the prediction of 1-year outcome of FFR with the results of Functional Lesion Assessment of Intermediate Stenosis to Guide Revascularization (DEFINE-FLAIR) and iFR Swedish angiography and angioplasty registry (SWEDEHEART) has been demonstrated [2, 3].

However, there are many catheter laboratories that do not have adequate experience with physiological lesion assessment using pressure guidewires. In addition, there are many differences in clinical situations between Japan and Western countries (available drugs, the timing of physiological lesion assessment, etc.). In this document we summarize not only the current evidence but also the practical use of physiological lesion assessment in Japan.

Concept of fractional flow reserve

The basic concept of FFR is presented in Fig. 1. The linear relationship between perfusion pressure and blood flow only exists when maximum hyperemia is achieved, although these correlations can be theoretically applicable in diastole [4, 5]. Therefore, the blood flow that passes through the stenosis is proportional to the perfusion pressure under maximum hyperemia.

Fig. 1
figure 1

The concept of FFR. Blue lines represent a case presenting no epicardial stenosis in the coronary artery. In this case the intracoronary pressure is 100 mmHg throughout the coronary artery, which indicates this coronary artery is performing at 100% of its capacity. Red lines represent a case with stenosis causing a hyperemic pressure gradient of 30 mmHg. This lesion caused a 30% decrease in performance relative to the maximum capacity in coronary flow. Pa aortic pressure, Pd distal coronary pressure, Pv central venous pressure

To achieve maximum hyperemia, drugs that have a vasodilatory effect should be used to attenuate the auto-regulatory function of the coronary artery system.

The FFR is defined as the maximum myocardial blood flow in the presence of stenosis divided by the theoretical normal maximum blood flow without a stenosis in the target coronary artery. This index represents the fraction of the performance of the coronary artery that can be achieved despite the coronary stenosis.

The normal value of this index is 1.0, regardless of the patient or the specific vessel studied [6]. This index is independent of changes in systemic blood pressure and heart rate [7]. And most importantly, FFR measurements show excellent reproducibility [8].

Although the threshold of FFR in the presence of ischemia has been reported to be < 0.75 [9], a FFR ≤ 0.80 is widely used as a cutoff value for the indication of PCIs [9, 10]. A FFR between 0.75 and 0.80 is considered the gray zone to avoid overlooking of lesions with ischemia for various reasons (please refer to the section “Pitfalls of FFR measurement”). Furthermore, independent of this explanation, the threshold point for the indication of medical treatment and revascularization has been reported to be 0.80 based on the relationship between the event rate and the FFR value [11, 12], and a FFR ≤ 0.80 is thought to be an indication for revascularization.

FFR and clinical outcome

The Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) Trial [1], which showed no advantage of angio-guided PCIs over optimal medical therapy, provided the opportunity to seriously consider the application of PCIs. The report from the same group also indicated a correct approach. Shaw et al. showed the importance of the presence of ischemia at the target vessel and the importance of the reduction of this ischemia by PCIs to show the effect of PCIs on outcome [13].

Considering the clinical environment mentioned above, the role of physiological lesion assessment using pressure guidewires, which can be used in the catheter laboratories, is expected to increase in Japan.

There have been three landmark studies evaluating FFRs. The first study was the Deferral versus Performance of PTCA in patients without Documented Ischemia (DEFER study) [14]. In this study, the authors clearly demonstrated the safety of the deferral of lesions with an FFR value of equal to or more than 0.75. The concept underlined by the study was that non-ischemic lesions, provided optimal medical therapy is available, have an excellent outcome with rates of death and myocardial infarction of less than 1% per year. Considering the rate of acute stent thrombosis and procedure-related myocardial infarction, even the implantation of a current state-of-the-art stent cannot improve the outcome of these lesions [15]. The validity of this method for extended periods (up to 15 years) has been also demonstrated [16, 17]. The same evidence has been confirmed for the Japanese population [18].

The second landmark study was the Flow Reserve Versus Angiography for Multivessel Evaluation (FAME) study [19]. In this study the superiority of FFR-guide PCI over angiography-guided PCI in patients with multi-vessel coronary artery disease was validated. The composite of death, myocardial infarction, and repeat vascularization was less frequent in FFR-guided PCI compared to angiography-guided PCI (13.2 vs. 18.3%, p = 0.02).

The third and last landmark study was the FAME II study [10]. Unlike the COURAGE trial, which showed no superiority of angiography-guided PCIs over optimal medical therapy, the FAME II trial clearly showed the superiority of FFR-guided PCIs plus optimal medical therapy over optimal medical therapy alone. When the FFR of the target vessel was equal to or less than 0.80, the composite of death from any cause, myocardial infarction, and urgent revascularization were less frequently observed in the PCI plus the Medical Therapy group compared to the Medical Therapy-alone group (4.3 vs. 12.7%, p < 0.001). This difference was mainly driven by the decrease of urgent revascularization requirement in the PCI plus Medical Therapy group compared to the Medical Therapy alone group (1.6 vs. 11.1%, p < 0.001).

Based on these results, the current European guideline recommends FFR-guided revascularization as class I (level of evidence A) in stable patients without clear evidence of ischemia [20].

The validity of FFR-guided revascularization over angio-guided revascularization has also been shown in patients who underwent coronary artery bypass graft surgery [21, 22].

FFR was introduced over 20 years ago [23] and there has been a wealth of evidence reported in almost all subsets of coronary disease and clinical settings. However, the main purpose of this document is to act as a guide for facilities that do not have ample experience with FFR measurement. However, the main purpose of this document is to act as a guide for facilities that do not have ample experience with FFR measurement. Therefore, more detailed evidence supporting FFR should be referred to other reviews.

Use of FFR in real practice

Pressure guidewires

Four wire-type pressure measurement devices (so called pressure-wires) and one monorail microcatheter type device are presently available in Japan. The characteristics of each device are summarized in Table 1. All pressure wires are applicable with a diagnostic catheter. A 5-Fr or larger guiding catheter is required for monorail microcatheter type devices.

Table 1 Characteristics of pressure wire sensors

All devices can now provide resting indexes.

The position of the pressure sensor of the devices should be placed at least 2–3 cm distal to the stenosis to be assessed to avoid the influence of pressure recovery phenomenon. In general, to evaluate whether a target coronary artery is suffering from myocardial ischemia caused by stenosis in the epicardial coronary artery, the pressure sensor should be positioned at the very distal part of the coronary artery [24].

Diagnostic and guiding catheters

One of the large differences in Japan compared to Western countries in terms of FFR measurement is the usage of a small caliber diagnostic catheter. Nevertheless, regardless of the size used, it is not recommended to measure FFR using a diagnostic catheter [24]. The problems associated with the use of a small caliber diagnostic catheter are as follows: (1) unsatisfactory back up force and (2) blunted pressure signal due to the small lumen size. Considering the clinical environment in Japan, measuring the FFR with a 5-Fr diagnostic catheter is widely accepted. There has been a report that showed the feasibility of FFR measurement with a 4-Fr diagnostic catheter; [25] however, the author also described the limitations in selected cases. Therefore, facilities that do not have sufficient experience with FFR measurements are highly recommended to avoid using the 4-Fr diagnostic catheter for FFR measurements.

When using a guiding catheter, it is important to use a guiding catheter without side-holes. The side-holes of the guiding catheter are located 20- to 50-mm proximal from the tip of the catheter. The pressure signals from these side-holes usually derive from ascending aorta and differs to that recorded from the tip of the catheter. The expressed pressure recording in this case is the fusion of the pressure from the ostium of the coronary artery and the ascending aorta. Therefore, the guiding catheter with side-holes should be avoided for the measurement of FFR.

Maximal hyperemia

The drugs available also differ between Japan and Western countries. Regadenoson is not available and adenosine is not reimbursed for use in FFR measurement in Japan. Conversely, nicorandil is available to limited countries (Fig. 2) [26, 27].

Fig. 2
figure 2

Adapted with permission from Jang et al. [27]

Correlation between intravenous administration of adenosine (140 μg/kg/min) and intracoronary administration of nicorandil (2 mg).

Papaverine is an ideal coronary vasodilator [28]. However, considering its arrhythmic side effects, few facilities use this drug in daily practice [29].

Intravenous continuous administration of adenosine-5′-triphosphate (ATP), a precursor of adenosine with the same vasodilator effect, is the current standard in Japan [30].

The list of available drugs for FFR measurement in Japan is summarized in Table 2. It is important to underline that premedication with nitrate, which is a common epicardial vasodilator in the catherization laboratory is essential for the measurement of both resting and hyperemic indexes to eliminate the presence of vasospasm in the epicardial coronary artery.

Table 2 Available agents for FFR measurement in Japan

Although it is available only in a limited number of countries, nicorandil, a hybrid of nitrate and an ATP-sensitive potassium channel opener, is easy to use without the effects of caffeine intake unlike ATP and with less frequency of lethal arrhythmia than papaverine. Nicorandil also causes less hemodynamic changes compared to ATP infusion, which is preferable for patients with severe aortic valve stenosis [31].

Indications for FFR measurement

Although the main target for FFR measurement in real practice is a lesion with moderate stenosis (30–70% stenosis), there are cases where FFR shows a negative value with very tight stenosis (mismatch case) [32, 33]. Conversely, there are also cases where FFR shows a positive value with very mild stenosis (reverse-mismatch case) [32]. The reason for these phenomena seem to depend on the amount of myocardium subtended by the target lesion [34].

Although, the indication for PCIs for mismatch cases tends to be an area of debate, the awareness of the presence of reverse-mismatch cases is more important since these lesions tend to supply blood to a larger extent of the myocardium (ex. left main trunk) and the outcome of these patients is not benign [35, 36].

To avoid overlooking these lesions and applying physiological measurement to as many vessels as possible, iFR [37] and FFRCT, which are thought to be invasive or noninvasive alternatives to FFR, respectively, have some advantages over invasive FFR.

Contraindications for FFR measurement

There are few contraindications for FFR measurement. An obvious contraindication is allergy to vasodilatory drugs such as adenosine, ATP, and so on, bronchial asthma, or COPD. Careful consideration is necessary when vasodilatory drugs are applied in patients with low blood pressure, hypertrophic obstructive cardiomyopathy, and severe aortic valve stenosis.

In this regard, iFR might overcome the problems described above, since there is no need for the use of vasodilatory drugs for measurement [38].

Although, it is not a contraindication, FFR measurement in the target vessel of ST elevated myocardial infarction in the acute setting is not recommended due to the presence of microvascular dysfunction in the targeted area [39]. Conversely, the feasibility of FFR measurement for the evaluation of non-culprit vessels in patients with ST elevated myocardial infarction, and for vessels with a history of ST elevated myocardial infarction has been shown [40,41,42].

The validity of FFR in patients with non-ST elevated myocardial infarction has also been shown in several studies [43, 44]. Nevertheless, evidence is limited, although there has also been a report suggesting that the FFR cutoff value should be higher in this setting [45]. Therefore, the application of FFR in decision-making for patients with non-ST elevated myocardial infarction should be carefully assessed, especially in cases with an FFR > 0.80.

The FFR does not consider the central vein pressure in its simplified formula; although, there is a report suggesting the validity of FFR values even with the presence of high central vein pressure [46]. It is advisable to refrain from FFR measurement in cases in which low blood pressure with increased central vein pressure is expected (e.g., in cases of cardiogenic shock or congestive heart failure).

With regard to anatomical limitations, a lesion with a very tight stenosis, severe winding of the target vessel (a cause of the accordion phenomenon), and the presence of very severe calcification might be sub-contraindications for FFR measurement.

Pitfalls of FFR measurement

The failure to achieve maximum hyperemia may represent a major pitfall of FFR measurement. The major checkpoints are as follows: (1) confirmation of the correct intracoronary injection of vasodilator drug (avoidance of the use of catheters with side-holes, slow injection of the drug to avoid back-flow of the injected drug, placement of the catheter tip in the coronary artery ostium, etc.); (2) recommendation to refrain from caffeine intake more than 24 h before the FFR measurement using adenosine and ATP; [47] (3) avoidance of the use of a vein from the lower extremity and distal forearm for continuous intravenous administration of vasodilator drug, and (4) confirmation of the position of the guiding catheter from the tight engagement to the coronary artery orifice.

The disturbance of the coronary flow merely due to the presence of a large-sized catheter at the ostium of the coronary artery, [48] neglecting to remove the wire introducer from the Y-connector, and the misplacement of the transducer are very simple yet easy to forget pitfalls occurring during the measurement of FFR.

The pressure signal drift in both directions can be the cause of misinterpretation of the FFR value. If one encounters more than a 2 mmHg pressure signal drift, the measurement should be repeated [49]. The careful preparation of the pressure wire should include: (1) flushing the wire tube with a sufficient amount of saline while the pressure wire is in the tube and (2) waiting 20–30 s after the equalization of the electronic pressure wire at the tip of the guiding catheter before advancing the wire into the coronary artery. Avoiding and confirming the presence of a pressure signal drift before the finalization of the measurement is mandatory [50].

The phenomenon is more critical when using resting indexes. Hyperemia usually increases the pressure gradient within a coronary artery by a factor of several times. Therefore, the signal-to-noise ratio during hyperemia is several-fold higher than under resting conditions. In other words, the resting indexes are more vulnerable to the influence of pressure signal drift compared with FFR.

Table 3 summarizes the additional pitfalls compromising the measurement of FFR.

Table 3 Pitfalls in FFR measurement

FFR interpretation in real practice

FFR should be regarded as a continuous variable of risk [11]. This concept is important both for ischemia positive (FFR < 0.75) and negative (FFR > 0.80) lesions.

For ischemia positive lesions, the cutoff value of FFR is different among the indications for PCIs (FFR < 0.75–0.79) and the risk for endpoints like cardiac death and myocardial infarction (FFR < 0.64–0.67) [11, 51].

In this regard, the complexity of the target lesion and the amount of myocardium subtended by the target lesion should be considered [11]. For example, when the lesion is in the ostium of the right coronary artery, where a high rate of restenosis is expected [12], the FFR value should be carefully interpreted. A further example is that if the amount of myocardium subtended by the lesion is small, the threshold of FFR for the indication of PCIs may be lower provided the symptom is manageable with optimal medical therapy (Fig. 3a). Conversely, even if the far distal value of FFR is below the threshold of indication of PCIs, the lesion without a focal pressure gradient might not be suitable for treatment using stent placement (Fig. 3b). The iFR is expected to play a role in discriminating these lesions [52, 53].

Fig. 3
figure 3

a A representative case of focal pressure gradient in both the proximal and distal part of the left ascending coronary artery. The values indicated are iFR. Yellow dots: the delta iFR caused by each lesion. One dot corresponds to a delta iFR of 0.01. White arrows: the focal pressure gradient in the proximal part of the left ascending coronary artery. Considering the large amount of subtended myocardium, these lesions might be suitable for treatment with stent placement. Black arrow: focal pressure gradient in the distal part of the left ascending coronary artery. Considering the small amount of subtended myocardium, these lesions might not be suitable for treatment with stent placement depending on the response of the patient with optimal medical therapy. b A representative case of diffuse pressure across the left ascending coronary artery. The values indicated are iFR (Courtesy of Dr Nishina). Yellow dots: the delta iFR caused by each lesion. One dot corresponds to a delta iFR of 0.01. Considering the widely spread dots without a large focal pressure gradient, these lesions might not be suitable for treatment with stent placement

The last notion is that although FFR > 0.80 is usually considered a lesion without ischemia; however, strictly speaking this understanding is not true. More precisely, an FFR > 0.80 should be considered a lesion that is not likely to be a cause of ischemia; which means that the lesion is not an indication for stent implantation. If maximum hyperemia is correctly achieved for FFR measurement, the impairment of the performance of the vessel by this lesion may be less than 20%. Ischemia can be caused by other reasons such as microvascular dysfunction, vasospastic angina, hypertrophic cardiomyopathy, severe aortic stenosis, anemia, etc [54]. This means that in some patients, additional clinical factors should be considered for the diagnosis of myocardial ischemia.

A recent interesting report showed the importance of optimal medical therapy for deferred lesions based on FFR measurement [55]. The study showed significantly better outcome in patients with higher FFR value than lower FFR values. The author speculated that this paradoxical result was due to the less intensive optimal medical therapy, especially statin use, in patients with higher FFR values than patients with lower FFR values. The concept of deferring stent placement should be considered only when optimal medical therapy is provided to the patients by the physician.

The concept of iFR

Davies et al. invented a wave-free period, which derives from wave-intensity analysis [56]. Resistance is thought to be low and stable during this period [57]. The iFR is calculated by distal coronary pressure (Pd) divided by aortic pressure (Pa) during the wave-free period, where a linear correlation between pressure and flow would be expected, under resting conditions, which means the induction of hyperemia is not required (Fig. 4). Following the results of two large randomized trials [2, 3], the current European guidelines have updated iFR-guided revascularization to class I (level of evidence A) in stable patients without any clear evidence of ischemia [20].

Fig. 4
figure 4

Adapted with permission from Sen et al. [58]

The concept of the wave-free period. The influence of different waves propagating from the proximal and distal ends of the vessel on the coronary flow is shown with separate wave-intensity plots. These waves are absent during the wave-free period. The resistance is low and most stable during the wave-free period. Coronary pressure and flow show a liner relationship during the wave-free period.

Concordance and discordance between the FFR and iFR

The concordance between the FFR and iFR measurements has been reported to be 80–85% [58,59,60]. The main mechanism of discordance between the FFR and iFR is thought to originate from the differences in the patient’s conditions during the measurement (at rest and under maximal hyperemia).

Cook et al. showed a better concordance of iFR with coronary flow reserve (CFR) than with FFR [61]. The lesion showing a low pressure gradient at rest and a high-pressure gradient under maximum hyperemia might represent a case with a higher rate of hyperemic flow to resting flow that is comparable to the concept of high CFR [62].

Lee et al. reported different clinical and angiographical characteristics among the 4 groups classified according to the FFR and iFR (group 1: FFR > 0.80 and iFR ≥ 0.90, group 2: FFR > 0.80 and iFR < 0.90, group 3: FFR ≤ 0.80 and iFR > 0.90, group 4: FFR ≤ 0.80 and iFR < 0.90) [63]. The authors also reported outcomes of concordant and discordant groups of FFR and iFR. Since both the DEFINE-FLAIR and IFR-SWEDEHEART trial did not allow the simultaneous measurement of FFR and iFR, the outcomes of deferred lesions with discordant results between FFR and iFR could not be investigated. Therefore, Lee et al. provide essential information about the outcome of these groups [64]. According to the data, only the group with concordant abnormal results showed a significantly higher risk of clinical events compared with the concordant normal group. The discordant results between the FFR and iFR measurements were not associated with an increased risk of clinical events (Fig. 5).

Fig. 5
figure 5

Adapted with permission from Lee et al. [64]

Outcome of patient with discordant and concordant results from FFR and iFR assessment. FFR fractional flow reserve, iFR instantaneous wave-free ratio, HR hazard rate.

Pullback curve analysis

It was considered difficult to predict the hemodynamic response to stenting in the presence of other flow limiting lesions before the advent of iFR, as hyperemic flow can be affected if the severity of stenosis exceeds 40–50% [65]. In the presence of two flow-limiting lesions, there is an interaction of coronary flow between lesions under hyperemia. The treatment of one lesion changes this interaction and leads to differences in the coronary flow that passes through the untreated lesion (Fig. 6) [66]. This phenomenon makes predicting post-PCI physiological results using FFR difficult. A solution to resolve this phenomenon exists, but it requires a complex formula including the value of wedge pressure that is not practical to measure at the time of diagnostic catheter [67]. Conversely, the resting flow is not affected by the lesion up to the almost sub-occlusion of the vessel. Therefore, without the presence of an extremely tight stenosis, the treatment of one lesion may cause negligible changes in resting flow passing through the residual stenosis. This characteristic of resting flow might allow the iFR to predict the post-PCI physiological outcome [52, 53].

Fig. 6
figure 6

Adapted with permission from Nijjer et al. [66]

The change in coronary flow passing through one lesion pre- and post-treatment of the other lesion. The treatment of one lesion can change the coronary flow passing through the other lesion under hyperemia. The change in coronary flow that passes through the lesion can change the pressure gradient induced by the lesion. This phenomenon is minimum under resting conditions and facilitates the prediction of the coronary flow that will pass through the lesion and the pressure gradient created by the lesion after treatment. PCI percutaneous coronary intervention.

Pitfalls of iFR measurement

The basic pitfalls for the measurement of iFR are almost identical to those of FFR, except that the achievement of maximum hyperemia is not a determining factor.

Instead, there are several important points to consider for the measurement of the resting indexes: (1) the administration of nitrate, which is a vasodilator of the epicardial coronary artery, is also necessary as in FFR measurement and (2) following the injection of any agent having vasodilatory effects (contrast medium, saline, etc.), the operator must wait until the vasodilator effect has subsided. In particular, after the administration of ATP, papaverine, or nicorandil, the measurement of resting indexes should occur after at least a 5-min delay. Although, not associated with the use of vasodilator drugs, hyperemia can be induced by ischemia caused by the occlusion of coronary artery, thus measurement of the resting indexes should take place at least several minutes after the occlusion of coronary artery during the PCI procedure [68]. (3) The resting indexes should be interpreted carefully in patients with left ventricular hypertrophy, in those undergoing hemodialysis, and in severe valvular heart disease, etc., since the validity of resting conditions values for these patients are controversial.

Other resting indexes

Resting indexes are now available with pressure wire systems. Table 4 describes the features of these new resting indexes that have been validated by several reports. Because the coefficients of determination between these new resting indexes with iFR is almost 1.0 (r2 > 0.98), these indexes might be nearly identical to iFR [69, 70]. However, there is no clinical data available for these new resting indexes. Real-world data using these new resting indexes are requested in the near future.

Table 4 Resting indexes