FFR’s Future: Negative Trials Spur Debate Over Technology’s Role

It hasn’t been a good couple of years for fractional flow reserve (FFR), whether it’s being tested in stable patients with multivessel disease undergoing PCI, against traditional coronary angiography in STEMI patients, or in the workup of chest pain patients in the catheterization laboratory.

In light of several negative trials, some physicians have questioned whether FFR can still play a meaningful role in the management of patients who may need coronary revascularization. Is it time to forego FFR? According to physicians who spoke with TCTMD, the decision to skip FFR may depend on the setting and purpose for the test.

“I think it’s a valid question and central theme among most interventionalists now,” said Kevin Bainey, MD (Mazankowski Alberta Heart Institute/University of Alberta, Edmonton, Canada). “I think the enthusiasm for FFR-guided PCI has kind of dampened a little bit with the recent studies. FAME 3 was surprising and RIPCORD 2 was also negative. Those took away some of the enthusiasm. The way I look at FFR is to look at it in the ACS and the non-ACS setting. In the ACS setting, I am not convinced at all that there are benefits to FFR-guided PCI.”

FFR in ACS

In ACS, several major trials have delivered a blow to invasive functional tests. In COMPLETE, in 2019, complete revascularization versus treatment of the culprit lesion only in STEMI patients with multivessel disease was associated with significantly lower risks of cardiovascular death or MI, as well as lower rates of ischemia-driven revascularization. Importantly, use of FFR in the trial was trivial, with treatment of the nonculprit lesions largely guided by angiography alone.

Similarly, the FLOWER-MI trial, published earlier this year, showed that a physiological assessment with FFR was no better than angiography alone for guiding complete revascularization in STEMI patients with multivessel disease undergoing primary PCI.

Both studies were published in the New England Journal of Medicine.

Add to those trials the FUTURE study showing that FFR-guided treatment in patients with relatively complex multivessel disease was associated with a doubling in the risk of death within the first year, with no beneficial impact seen on other outcomes. Roughly half of the patients in FUTURE presented with ACS.

I think the enthusiasm for FFR-guided PCI has kind of dampened a little bit with the recent studies. Kevin Bainey

FUTURE and FLOWER-MI have “made me pause the use of FFR” in ACS, said Bainey. “I think it comes down to the physiology. I’m not sure we exactly know how to read the flow across the stenosis when there is infarcted or acutely ischemic territory down below. The validation of FFR-guided PCI in ACS settings is very ambiguous and unclear. We need a larger clinical trial appropriately powered for hard endpoints to help address this question in the ACS setting.”

William Fearon, MD (Stanford University School of Medicine, CA), who led the FAME series of trials, has also heard rumblings—particularly on social media—that FFR has fallen out of favor, but believes that message is amplified by physicians not familiar with the details of the various studies.

Both FLOWER-MI and FUTURE, for example, included lesions where FFR was not necessary. As a result, patients in the FFR-guided PCI arm were treated nearly in the same manner as those who underwent angiography-guided PCI, “so it is not surprising that there was no difference in outcomes,” said Fearon.

With respect to FLOWER-MI, for example, 66% of the FFR-guided patients and 97% of the angiography-guided patients received PCI of a nonculprit vessel, which suggests that most nonculprit lesions were severe as opposed to intermediate-grade where FFR would have had more of a benefit. With only one-third of patients treated differently, and a low event rate, the study was underpowered, he said.

“The take-home message I have from FLOWER-MI is that with FFR-guided nonculprit PCI, you can avoid unnecessary PCI in one-third of patients with equivalent outcomes,” said Fearon.

Stable CAD

FFR has weathered some blows in chronic coronary syndromes as well. Just last month at the TCT 2021 meeting, FAME 3, also led by Fearon, threw yet more cold water on the use of FFR-guided PCI, showing that it failed to match CABG surgery for the reduction of major adverse cardiovascular and cerebrovascular events at 1 year in stable patients with three-vessel coronary artery disease.

Cardiologists largely proclaimed the study a major win for surgery over the less-invasive procedure, particularly since the benefit was apparent so early on. But for William E. Boden, MD (Boston University School of Medicine/VA New England Healthcare System, MA), FAME 3 also exposed the limitations of physiologic coronary flow assessment.

“My take-home from FAME 3 was that if you can’t show a benefit of FFR-guided PCI at 1 year, you’re never going to show a benefit beyond that time point because in all the studies that have compared PCI with bypass surgery, the longer the duration of follow-up, the greater separation there is in the survival curves in terms of outcomes [in favor of surgery],” he said. 

Fearon, on the other hand, sees FAME 3 differently. While researchers intended to enroll patients with low-to-intermediate SYNTAX scores, the patient population was quite complex, which is reflected by the fact that FFR could be measured in only 82% of patients. In addition, 76% of those lesions were abnormal, meaning they only deferred PCI in 24% of lesions deemed not physiologically significant. In the subgroup of patients with less-complex disease—a subgroup in which PCI outperformed CABG—the percentage of FFR-negative lesions was higher, he said.

“The take-home message to me from FAME 3 is that it is not a failure of FFR, but that in very complex disease, CABG remains the treatment of choice. In less-complex disease, FFR-guided PCI is very reasonable,” said Fearon.

If you can’t show a benefit of FFR-guided PCI at 1 year, you’re never going to show a benefit beyond that time point. William E. Boden

Sukhjinder Nijjer, MBChB, PhD (Imperial College London, England), agreed that a growing body of evidence challenges the value of FFR in ACS, but in those with stable ischemic heart disease, he argued, FFR has its advantages.

“FFR is essentially a tool or a technology that helps us guide intervention and is almost certainly most useful in patients who have stable coronary artery disease and in those with multivessel disease, particularly for those patients that haven’t had some form of ischemia testing before they get to the cath lab,” he said. “Here in the UK, where we do a lot of CT coronary angiography, you can find very significant disease and patients may end up in the cath lab without ischemia testing.”

At their center, such patients will routinely have a pressure-wire assessment at the stage of angiography, said Nijjer. That helps make a diagnosis of coronary artery disease that may or may not require revascularization. “We use it truly as a tool in order to help risk-stratify and assess these stable patients,” he said. “Where the fallacy comes is when people start to use FFR in a wider spectrum of patients, where the evidence for FFR has never been strong.”

Overall, ACS patients have a much higher event rate than patients with stable coronary artery disease, he said. Additionally, studies have consistently shown that event rates in ACS patients managed with FFR are consistently higher than in similarly managed stable patients. Even among patients who are deferred for revascularization, MACE rates on follow-up are higher in ACS patients than in stable patients. Given that, it’s not surprising that recent studies like FUTURE, with its mix of stable and unstable patients, were negative.

The true value of FFR, said Nijjer, is as an aid to decide on deferral. He offered a hypothetical case: a patient with classic angina symptoms and myocardial ischemia revealed on cardiac MR perfusion imaging, but only a moderate lesion on angiography. In that setting, where it’s not a “slam dunk severe stenosis where you’re going to do PCI,” a pressure-wire assessment can be useful to help make a decision.

“In other cases, you may find anterior ischemia on stress MRI but a severe stenosis in the RCA [right coronary artery] and only mild disease in the LAD [left anterior descending artery],” he said. “The FFR can help you assess that,” he said. He noted that some of the earliest trials that showed the effectiveness of FFR, such as FAME, included stable patients treated with first-generation drug-eluting stents, a group in which it was probably best to avoid revascularization as much as possible.

Another setting where FFR can be useful is with select ACS patients with multivessel disease, said Nijjer. For example, in an unstable patient without a rise in cardiac troponin, which suggests the microcirculation is working well, an FFR value less than 0.80 tells operators the lesion is significant and the patient may benefit from revascularization. “If the value is close to 1, or 0.95 or so, the benefit is going to be very modest, even if there has been some instability in that territory,” said Nijjer.

Fishing for a Vessel to Stent?

But even in patients with stable ischemic heart disease, Boden questioned why such patients would even need physiologic testing given that the ISCHEMIA trial, as well as COURAGE, showed that optimal medical therapy was superior to an invasive strategy. Why would FFR be needed as a deferral tool in this setting? In fact, Boden believes there are operators who fall back on FFR just to find a lesion below the 0.80 cut point.

“They do repeated measures looking for the physiologic-‘tool’ justification to proceed with PCI,” said Boden. “I think it can cut both ways. It can be a deferral tool, but it can also be a strategy for trying to identify patients for whom you think there might be a benefit of doing PCI. I think it’s gaming the system to a degree.”  

While FFR undoubtedly measures the pressure reduction across the flow-limiting stenosis, Boden believes there are still some questions around the amount of downstream tissue perfusion.

“In other words, is the segment of myocardium subtended by the narrowed artery vulnerable to low tissue perfusion?” he asked. “We know that FFR doesn’t necessarily measure that. We see that all the time in the cath lab. You can have a CTO but because collaterals exist, the myocardium distal to the occlusion is still functioning, still contracting. There’s no regional wall motion abnormality. So even in the most severe case of a chronic total occlusion, it doesn’t necessarily mean there is no perfusion of the segment. The problem with FFR is that it measures a pressure gradient across an epicardial coronary stenosis—it doesn’t really measure the presence or absence of ischemia or regional tissue perfusion.”

And even if the pressure reduction across the stenotic lesion correlated with reduced perfusion, the ISCHEMIA trial “detonated” the ischemia hypothesis, said Boden. In a subgroup analysis, there was no signal that the magnitude and severity of ischemia identified a subset of patients who benefited from revascularization, said Boden.   

Additionally, Boden highlighted a 2017 study by the IRIS-FFR researchers, which was led by Jung-Min Ahn, MD (Asan Medical Center, Seoul, Korea), showing that revascularization of lesions with an FFR less than 0.64 reduced the risk of cardiac death or MI. This FFR threshold is more stringent than what is routinely used in clinical practice, which is 0.80. Similar data, which was published by Lance Gould, MD (University of Texas Health Science Center, Houston), showed that a stricter FFR threshold was needed to identify significant coronary stenoses.

“Why are we doing a pressure-wire determination on patients to essentially justify undertaking PCI for which there would be no expected clinical benefit of death or MI reduction?” said Boden. “It would seem to me that if you’re going to do it for that purpose, then the true cut point would be an FFR of less than 0.64.”

Nijjer conceded that the current cut point of 0.80 is arbitrary, adding that “there is no such thing as a straight line in biology.” Early on, a cut point of 0.75 or less was shown to predict myocardial ischemia in noninvasive testing with good accuracy, but it was later bumped up on the basis of the earlier FAME trials where the 0.80 cut point was shown to predict clinical outcomes.

Locate Lesions Causing Symptoms

Bainey agreed that ISCHEMIA showed that the invasive strategy did not reduce hard clinical events in patients with stable ischemic heart disease, but it did show there were advantages in terms of improving patient symptoms. For that reason, FFR can be helpful in selected patients, he argued. For example, in a symptomatic patient who has tried medical therapy, FFR can help find the truly ischemic lesion where revascularization can improve quality of life, something it’s important to explain to patients.

“I’m not making a difference in terms of death or MI,” he said. “I’m clear with them up front about that.”

Nijjer, for his part, doesn’t quite understand the criticism of FFR given that the technology is intended to reduce the number of patients who end up with stents they don’t need.

“I have not found, in my experience, interventionalists taking patients to the cath lab and performing FFR to put in extra stents,” he said. “Most of the time patients come to the cath lab because there’s a clinical indication. They’ve had symptoms. These are people who are worried about significant coronary artery disease. They’ve usually had a number of investigations before they get to the lab and then you use FFR to guide the intervention if you have multiple lesions and you don’t quite know which one is causing the problem.”

In his experience, operators aren’t “fishing” for lesions to stent, said Nijjer. He argued that for stable patients who are relatively healthy, a negative FFR can assure patients that blood flow is still decent and that they can be managed with medical therapy instead of receiving a stent. Even in the FAME 2 study—a trial that showed PCI-guided by FFR was superior to medical therapy for reducing the need for repeat revascularization—the event rate was quite low in those treated with medication.

I have not found, in my experience, interventionalists taking patients to the cath lab and performing FFR to put in extra stents. Sukhjinder Nijjer

“There are two sides to this coin, and how you interpret the data depends on your internal biases,” Nijjer said. “I think the medical therapy people often try to paint this as a tool used by interventionists to deploy unnecessary stents. I think they could champion FAME 2 and say, ‘Look, some 70% or 75% of the medical therapy arm had no events at all despite having a significant FFR value.’ I think it’s something they’re missing out on.”

Nevertheless, Boden counters that “if FFR does not identify patients who can derive cardiac event reduction with FFR-guided PCI and we know from both COURAGE and ISCHEMIA that optimal medical therapy is equally safe and effective in stable patients, why would we add to the cost, increased procedure time, and potential risk of complications with FFR-PCI”? 

Bainey said that in stable ischemic heart disease, the FAME studies have clearly shown there is an indication for FFR-guided PCI, and the clinical revascularization guidelines reflect those data. However, he noted that the reduction in MACE observed in those trials was driven by a reduction in repeat revascularization. In the FFR-guided PCI studies, there has been no demonstration of a reduction in hard clinical endpoints, such as cardiovascular death or MI. Also, in the landmark ISCHEMIA trial, coronary anatomic severity, and not ischemic burden, had the biggest impact of prognosis.

“It all comes back to the angiogram,” said Bainey. “At this time, I’m still very much reliant on what I see on the angiogram.”

For Bainey, one of the most informative analyses to come out of  the COMPLETE trial was the optical coherence tomography (OCT) substudy, which suggested that the benefits of complete revascularization might be attributable to stenting nonculprit vulnerable plaques. In that substudy, roughly half of angiographically significant lesions had a thin-cap fibroatheroma, lesions that are prone to rupture and causing clinical events. 

“You could have that 70% lesion with a thin-cap fibroatheroma, and you put an FFR wire across and there’s no significant pressure difference across that stenosis,” said Bainey. “That’s where I think that imaging is important. Right now, we rely on angiographic imaging because the significant stenoses, almost half the time, have a thin-cap fibroatheroma, which are stented with angiographic complete revascularization. The next step in my opinion is veering away from physiology and imaging the nonculprit lesions in depth to truly identify the thin-cap fibroatheroma that need to be treated with PCI.” 

What The Guidelines Say

In the latest revascularization guidelines released just a couple weeks back, the American College of Cardiology/American Heart Association/Society for Cardiovascular Angiography and Interventions recommend the use of FFR or instantaneous wave-free ratio (iFR), which is an alternative to FFR and doesn’t require hyperemia, in stable patients with angina, undocumented ischemia, and intermediate stenoses to help make a decision about PCI (class I, level of evidence A). For stable patients with an intermediate-grade lesion, PCI should not be performed if the FFR exceeds 0.80.

Similarly, the European Society of Cardiology recommends FFR or iFR to assess hemodynamic significance of an intermediate-grade stenosis when evidence of ischemia is not available (class I, level of evidence A). FFR-guided PCI should also be considered in patients with multivessel disease undergoing PCI (class IIa, level of evidence B).

“I think the data supporting FFR and coronary physiology for guiding PCI is well established and reflected in all of the major guidelines, which give it a class I recommendation,” said Fearon. “When you push a test to its limits and evaluate it in extreme situations, the benefit may be less. However, in everyday practice where we are constantly faced with the question in the cath lab, ‘Is that lesion significant?’, that is where FFR has its main value. And this is particularly common in patients with multivessel CAD, where one lesion may be obviously significant, but another one or two are less clear.”

Despite the recommendations, which were in place with previous guidelines, adoption of FFR in clinical practice has historically been quite low, and this has been attributed to the extra cost and time it requires. More than 10 years ago, numbers from the CathPCI Registry showed FFR was used in just 6.1% of PCIs for intermediate-grade lesions. Last year, in a study published in the Journal of the American College of Cardiology, Rushi Parikh, MD (University of California, Los Angeles), reported that the utilization of FFR in stable ischemic heart disease patients with an intermediate-grade lesion increased from 14.8% to 18.5% between 2009 and 2017.

The research field, however, remains active with numerous ongoing trials testing FFR, particularly in the setting of ACS, hinting that this door is not entirely closed.

In the registry-based FULL-REVASC study of STEMI patients, investigators are testing whether complete revascularization with FFR-guided PCI is superior to conservative management of nonculprit lesions during the index hospitalization. Similarly, the SLIM trial is comparing FFR-guided complete revascularization versus usual care in NSTEMI patients with multivessel disease. In the OPTION-STEMI trial, researchers are looking at the optimal timing of FFR-guided PCI of the non-infarct-related artery in STEMI patients, while the FRAME-AMI investigators are comparing an FFR-guided approach to complete revascularization versus an angiography-guided strategy in acute MI patients with multivessel disease.

In the stable setting, there are fewer trials, although the unique DANANGINA study is comparing FFR-guided PCI against a sham PCI procedure in patients with stable angina. Results from this trial, which includes 450 patients, are expected next year.

For now, for physicians using FFR in practice and wondering if they should stop, Fearon offered a litmus test to apply when deciding whether FFR might prove helpful—as FAME 3 made crystal clear.

“It is hard to show a benefit of a strategy,” he said, “if the test being done doesn’t alter therapy.”