Fractional flow reserve (FFR) was a hot topic at EuroPCR this year, in keeping with its top billing at all recent interventional cardiology sessions. In this article, Cardiology: Interventions reviews the new data presented in Paris and offers some perspectives from experts in the field on the current clinical status of physiology-guided decision making in the catheterization laboratory.

“The utilization of FFR continues to increase year by year,” says William F. Fearon, MD, FACC, in an interview. “It is now used in up to 30 percent of cases in the U.S., but since many patients who are treated in the cath lab present with acute coronary syndrome and single-vessel disease, and some stable patients present with single-vessel disease and an abnormal stress test in the same territory, FFR is certainly not necessary in all cases.”

For his part, Allen Jeremias, MD, MSc, FACC, thinks FFR is probably only used in 15 to 20 percent of cases in the U.S. presently, and is probably warranted in about 30 percent of cases. “Given how solid the data are, it’s probably still underutilized,” he says in an interview, but added that we’re well on track for appropriate levels of usage, aided by “this type of data [such as was presented at EuroPCR] that helps build our confidence that FFR offers substantial benefit and will hopefully lead to greater use among those who have been skeptics or just nonusers.”

FAME 2: The New Data

The majority of an afternoon EuroPCR hotline session was devoted to the newest data on FFR, with the latest data from FAME 2 trial being a highlight. The FAME 2 investigators reported five-year outcomes in 1,220 patients with stable coronary artery disease (CAD) and angiographically-significant stenoses.

In those with at least one lesion with an FFR value ≤0.80, those randomized to FFR-guided PCI plus medical therapy had a lower rate of the primary composite endpoint of death, myocardial infarction (MI) or urgent revascularization compared with those who received medical therapy alone (13.9 vs. 27.0 percent; hazard ratio [HR], 0.46; p<0.001).

The difference was driven by a significant reduction in urgent revascularization (6.3 vs. 21.1 percent; HR, 0.27; p<0.001), with no between-group significant differences noted in the rates of death or MI. But spontaneous MI was numerically lower in the FFR-guided PCI patients compared with the medically-treated group (6.5 vs. 10.2 percent; HR, 0.62; p=0.04), despite the fact that more than 50 percent of medically-treated patients crossed over to PCI.

PCI guided by FFR offered a sustained relief from angina compared with medical therapy alone in up to three years of follow-up, but this statistical difference between groups was lost at the five-year time point.

In his presentation of the FAME 2 findings, which were simultaneously published in the New England Journal of Medicine, Panagiotis Xaplanteris, MD, PhD, noted that by five years, 51 percent of the medical therapy group had crossed over to PCI.¹ “This probably dilutes our finding of angina relief at five years,” he noted, before concluding that the five-year findings “confirm the importance of an FFR-based selection for PCI for both patients and lesions.”

“It is confirmed that when FFR is above 0.8, the outcome is really favorable with medical therapy. What is new is that when FFR is equal to or less than 0.8, PCI with second-generation drug-eluting stents (DES) provides sustained benefits in the need for urgent revascularization, in the rate of spontaneous myocardial infarctions, in symptomatic relief, and these come without late catch-up phenomenon,” said Xaplanteris.

"The utilization of FFR continues to increase year by year. It is now used in up to 30 percent of cases in the U.S., but…FFR is certainly not necessary in all cases. " — William F. Fearon, MD, FACC

“If it ain’t broke, don’t fix it, but if it is, don’t procrastinate I guess would be a take home,” said panelist Philip M. Urban, MD, FACC.

“Now we see that the benefit is not confined to just symptoms alone, but also to potentially reducing MIs, which is obviously a big finding,” says Jeremias. He added, however, that it would be interesting to see if the MI difference “might be even bigger” in the as-treated population, agreeing that the 50 percent crossover rate likely “diluted the signal.”

“Between FAME and now FAME 2, the evidence for FFR to guide revascularization in patients with stable CAD is very solid and clearly better than angiography-guided PCI,” comments Fearon. “FAME and the five-year data from FAME 2 show that it leads to significant reductions in the need for urgent revascularization and lower rates of spontaneous myocardial infarction. As well, the three-year FAME 2 cost-effectiveness analysis showed that FFR-guided PCI is very attractive from an economic standpoint, resulting in significant improvements in quality of life at the same cost as medical therapy for patients with stable CAD.”

Pooled FFR trials

There have been three randomized controlled trials comparing FFR-guided PCI using newer-generation DES to optimal medical therapy (OMT) for stable CAD: FAME 2, DANAMI-3-PRIMULTI and COMPARE-ACUTE. Each showed a clinical benefit for FFR-guided PCI, but none were powered for the endpoint of cardiac death and MI.

At EuroPCR, Frederik M. Zimmerman, MD, presented the first pooled, patient-level analysis to include all three trials and 2,400 patients to address these hard endpoints with increased statistical power.

FAME 2, as just discussed, included patients with stable CAD, while the other two trials included STEMI patients, but randomization was withheld until after successful treatment of the culprit lesion and hemodynamic stability.

For the prespecified primary endpoint of cardiac death or MI, FFR-guided PCI was associated with a 28 percent lower relative risk (p=0.024) and a 4.5 percent lower absolute risk of this endpoint. Taken individually, MI was reduced 29 percent with FFR-guided PCI over OMT (p=0.030), with no significant difference seen in cardiac death (HR, 0.99; p=0.99) or all-cause death (HR, 1.03; p=0.89).

“Critically important, our data show that the benefit for FFR-guided therapy was identical for both presentations,” referring to those with acute coronary syndromes (as seen in DANAMI-3-PRIMULTI and COMPARE-ACUTE, and those presenting with stable CAD, which also justifies the pooling of the data, Zimmerman added.

“These findings imply that appropriately selected patients have a prognostic benefit from PCI independent of its impact on symptoms,” said Zimmermann in his presentation. He expressed hope that the findings will “challenge the American guideline stating that PCI does not lower the long-term risk of MI.”

Gray-zone FFR Values

The optimal FFR cutoff for revascularization has been the subject of some controversy. In FAME 2, a cutoff of ≤0.80 was used, where as other studies used ≤0.75 as the line above which medical therapy was employed over PCI. “This is a common clinical scenario, affecting approximately one in 50 patients attending the cath lab,” reported principal investigator Barry Hennigan, MD.

Another recent series found that 15 percent of lesions, or one in seven, fall into the gray zone.² The recently published IRIS-FFR Registry, noted Hennigan, showed no difference in clinical outcomes in patients with gray-zone coronary stenoses treated medically or with revascularization.³

A single-center, prospective trial, Gray-Zone FFR, included individuals with stable single-vessel CAD and an FFR finding of 0.75 to 0.82. The 0.82 upper limit was chosen based on the test-retest difference in FFR that is between 0.01 and 0.02, and as a means to increase recruitment.

Individuals were randomly assigned to PCI plus OMT or OMT alone. Eligible patients were taken from the cath lab to undergo 3T stress perfusion MRI (with the operator and patient blinded to results) and only then randomized to medical therapy or PCI.

At three months, anginal frequency and quality of life were significantly improved for those who underwent PCI plus OMT, compared with OMT alone. Improvements in quality of life were greater in those with ischemia on stress MRI who underwent PCI compared with those with no ischemia (p=0.046). This difference was not seen in the OMT alone arm.

"Right now, it’s a yes or no test. If our wires pass beyond the blockage, we do the measurement and it’s either treat or not treat [with PCI]. But I think the reality is that there is no black and white in biology, everything is a continuum." — Allen Jeremias, MD, MSc, FACC

Improvements exceeded those seen in ORBITA, although when presenting the findings Hennigan noted the need for cautious interpretation because of the lack of patient blinding. “We must bear in mind that our PCI arm was not blinded to the treatment.”

“The gray zone is exactly the example where you would expect the trade-off to fall between clinical benefit that you can anticipate from revascularization versus clinical benefit that you could anticipate from medical therapy,” noted Emanuele Barbato, MD, PhD, in the discussion of the trial that followed its presentation.

Hennigan noted that while the trial’s sample was too small to detect a treatment effect according to lesion characteristics, he added, “If it’s a proximal LAD [left anterior descending] and a gray-zone patient, particularly if they have a lesion morphology that you’re not comfortable with, I think you need to be very careful before you defer a case.”

Asked if there was any attempt made to move patients out of the gray zone in the cath lab, Hennigan noted that 15 percent of patients were reclassified as ischemic following administration of high-dose (210 mcg/kg/min) adenosine, but he stressed that this strategy is not validated in clinical trials, so “we have to be careful in interpreting the results.”

Asked about his recommendations on how to apply this new information, Hennigan said: “To be totally honest, in my clinical practice, I tend to go ahead with PCI if I think it is clinically feasible and safe to do so, but I think evaluating every case on a case-by-case basis is very important.”

Fearon says, “We chose a cutoff value of 0.80 when designing the original FAME trial because we wanted to make sure FFR caught all significant lesions, even if it meant overtreating some lesions. In my practice, if the FFR is >0.80, I confidently defer revascularization. If it is <0.75, I feel very confident the lesion is functionally significant. And if the FFR falls between these values, I use my clinical judgement. If the patient has typical symptoms and it is a proximal LAD lesion, I will revascularize. But if the patient is asymptomatic and undergoing a cath because of a preoperative evaluation for noncardiac surgery, I will defer, especially if the lesion is not in the proximal LAD.”

Asked whether FFR can be considered a substitute for clinical judgement, Fearon adds: “FFR is never a substitute for clinical judgement, but if you’re not going to follow the FFR result, then there is no need to measure it.”

ORBITA

These days, it’s not a party unless ORBITA attends. Indeed, during the FFR hotline session, Rasha Al-Lamee, MD, ORBITA’s principal investigator, presented a physiology-stratified analysis from the trial.⁴ As a refresher, the main findings of ORBITA, first presented in late 2017, showed no benefit in terms of greater improvements in exercise times or anginal frequency for PCI over a sham procedure in stable CAD patients treated with optimal medical therapy, despite the presence of anatomically and functionally significant stenoses.

In ORBITA, invasive physiology was measured by FFR and instantaneous wave-free ratio (iFR) prior to randomization, but the operator was blinded to the results. This methodology was employed, said Al-Lamee, because the patients were already clinically indicated for PCI and the FFR/iFR cutpoint for angina relief is unknown. So ORBITA was actually designed to identify an appropriate cutpoint for angina relief and therefore had to enroll patients eligible for PCI across a spectrum of FFR/iFR values.

Invasive physiology data were available in 196 (of 200) patients, the majority of whom (76.5 percent) at prerandomization had Canadian Cardiovascular Society class II or III symptoms. Mean FFR and iFR were 0.69 and 0.76, respectively, and 97 percent had at least one positive ischemia test.

PCI improved the stress echocardiography score compared with sham PCI, by 1.07 segment units (p<0.00001). There was a significant interaction with both FFR and iFR values (as continuous variables), with larger improvements in stress echo with lower baseline levels on FFR and iFR (p for interaction <0.00001 for both). However, while PCI resulted in more patient-reported freedom from angina compared with placebo (49.5 vs. 31.5 percent; odds ratio, 2.47; p=0.006), FFR and iFR did not predict the placebo-controlled PCI effect on symptoms or treadmill exercise time.

“…these observations illustrate that the clearly demonstrated physiologic effect of PCI over sham as assessed by FFR and iFR does not easily translate into downstream clinical observations within the trial, highlighting the subjective and multifactorial nature of anginal symptoms,” wrote Ajay J. Kirtane, MD, FACC, in an editorial that accompanied the publication of the new ORBITA physiology-stratified analysis.⁵

“These data emphasize that epicardial coronary physiology is not the sole determinant of clinical symptom relief, particularly among patients whose angina is well-controlled on intensive medical therapy,” he added.

Next steps...

Although the data on FFR in stable CAD is undeniably strong, there are still questions to be answered, namely the role of physiologic testing in clinical contexts outside the stable CAD realm in which they have been validated.

In a recent review, Adrian Banning, MD, and colleagues at Oxford University reviewed the available evidence on the reliability and feasibility of using FFR and iFR in a wider range of scenarios, including in STEMI and NSTEMI, chronic total occlusions, aortic stenosis, cardiomyopathies, and even noncardiac conditions like diabetes and chronic kidney disease.⁶ They concluded that the data “are progressively growing showing a possible extension of application in ‘off label’ clinical contexts.”

FAME 3, which is nearing the end of enrollment, according to study chair Fearon, will compare FFR-guided PCI to CABG in about 1,500 patients with multivessel disease. “This trial should provide further data regarding the use of FFR in patients with complex disease,” says Fearon. Initial findings are expected in August 2019.

For his part, Jeremias wants to better understand how hemodynamic testing can be utilized to allow for more precise PCI in its natural home: the stable patient. “What I’m excited about is to use FFR and iFR in a little bit of a different way than it has been used so far.”

“Right now, it’s a yes or no test. If our wires pass beyond the blockage, we do the measurement and it’s either treat or not treat [with PCI]. But I think the reality is that there is no black and white in biology, everything is a continuum,” Jeremias notes.

" These data emphasize that epicardial coronary physiology is not the sole determinant of clinical symptom relief, particularly among patients whose angina is well-controlled on intensive medical therapy." — Ajay J. Kirtane, MD, FACC

One technique he feels is underutilized is pullback, where the wire is slowly pulled back to evaluate the pressure gradients across all lesions within the vessel. This is particularly helpful in tandem lesions, diffuse disease and ostial stenoses which are sometimes difficult to visualize by angiography alone, but can also be a useful assessment post PCI to ensure optimal treatment.

“Very few people have one single blockage, most have multiple blockages or diffuse disease so it’s not always enough to just have an abnormal FFR or iFR reading,” says Jeremias. “Similarly important questions are where exactly the stents need to be placed, how many stents are needed, how long of a stent is needed, is there diffuse or focal disease, and whether there is residual ischemia post treatment. All these are practical questions in the cath lab that can be answered when physiology is utilized for PCI guidance.”

An FFR pullback is possible but requires intravenous hyperemia and can be difficult to interpret because there can be “a lot of cross-talk between the lesions,” Jeremias notes. A non–adenosine-based test like iFR, can be used to map the ischemic contribution of each lesion without the confounding effect of other stenoses. “It can help you see that sometimes the most severe stenosis is not the most hemodynamically-significant stenosis.”

The DEFINE-PCI trial, for which Jeremias is a co-principal investigator, is a 500-patient pilot study testing the relationship between iFR pullback and the distribution of coronary stenoses as assessed by quantitative coronary angiography after angiographically-successful PCI. Results may be presented as soon as ACC.19.

Perhaps not everyone agrees that iFR is ready to replace FFR.⁷ In Fearon’s view, “iFR and other resting diastolic indices are approximately 80 percent accurate when compared with FFR, although when they are clearly abnormal or clearly normal, one can confidently avoid the need for adenosine, if desired.”

---

References

1. Xaplanteris P, Fournier S , Pijls NHJ, et al. N Engl J Med 2018;379:250-9.
2. Johnson NP, Toth GG, Lai D, et al. J Am Coll Cardiol 2014; 64:1641-54.
3. Kang DY, Ahn JM, Lee CH, et al. Eur Heart J 2018;39:1610-9.
4. Al-Lamee R, Howard JP, Shun-Shin MJ, et al. Circulation 2018;May 22:[Epub ahead of print].
5. Kirtane AJ. Circulation 2018;May 22:[Epub ahead of print].
6. Benenati S, De Maria GL, Scarsini R, et al. Cardiovasc Revasc Med 2018;19:362-72.
7. Ihdayhid AR, Yong A, Harper R, et al. Heart Lung Circ 2018;27:406-19.

Clinical Topics: Acute Coronary Syndromes

Keywords: ACC Publications, Cardiology Interventions, Acute Coronary Syndrome, Exercise Test, Decision Making, Catheterization

< Back to Listings