iFR Matches FFR for Clinical Outcomes in Patients With Intermediate Lesions

WASHINGTON, DC (UPDATED)—Two late-breaking trials with a combined total of 4,500 patients are making the case for instantaneous wave-free ratio (iFR) measurement to guide decisions on whether or not to revascularize intermediate lesions. The new technology was associated with a noninferior risk of MACE at 1 year compared with fractional flow reserve (FFR), researchers reported here today at the American College of Cardiology (ACC) 2017 Scientific Session.

FFR involves using adenosine—which carries the risk of side effects and costs money, critics say—as well as measuring the pressure gradient across a lesion during hyperemia, whereas iFR is calculated during diastole and does not require use of hyperemic agents. The new results will no doubt add fuel to the heated debate over whether one technology has the edge over the other.

As previewed by TCTMD earlier this week, DEFINE-FLAIR and iFR-SWEDEHEART share much in common. Rather than looking at patients with known disease, investigators studied patients with stable angina or NSTE ACS who underwent coronary angiography and had an indication for physiologic assessment—namely moderate stenosis found on angiography in which the value of stenting was ambiguous. The threshold for revascularization (either PCI or CABG) was 0.89 for iFR and 0.80 for FFR; when values exceeded these prespecified cutoffs, treatment was deferred.

iFR-SWEDEHEART, led by Matthias Götberg, MD, PhD (Lund University, Lund Sweden), is a registry-based randomized controlled trial (RCT) conducted almost entirely in Sweden, whereas DEFINE-FLAIR, led by Justin Davies, MBBS, PhD (Imperial College, London, England), is a conventional RCT with patients in Africa, the Middle East, Asia, the United States, Australia, and Europe. Findings from the two trials were published online March 18, 2017, in the New England Journal of Medicine.

Shifting Away From the ‘FFR Bandwagon’?

Speaking with TCTMD ahead of ACC, Davies deemphasized any debate and suggested that “the big winner,” were iFR to prove noninferior, “should be the patient. Because it should make it easier and cheaper and nicer for patients to have their assessments made.” The win, Davies added, would be an uptick in physiological assessment overall, not moving people from one type of technology to another.

“In FFR we have a good diagnostic method, but people are not using it enough because of the side effects,” Götberg said in a press release. “Our study shows that for cardiologists who currently use FFR, iFR provides a similar clinical benefit but without patient discomfort. For cardiologists who are hesitant to use FFR, iFR might provide an opportunity to increase adoption rates of coronary physiology and benefit more patients.”

At an ACC press conference, Dipti Itchhaporia, MD (Newport Coast Cardiology, Newport Beach, CA), seemed optimistic for iFR’s future considering the growing interest in physiologic assessment as a means to judge procedural appropriateness. Just last week, new appropriate use criteria for stable ischemic heart disease endorsed the use of FFR guidance.

“But we’ve noticed clinically that there’s been some reluctance to jump on that FFR bandwagon,” she noted. “I think that the idea of a technology that can potentially be quicker and as safe as the established FFR is very enticing to a clinician.”

Commenting on the studies for TCTMD, however, FFR expert William F. Fearon, MD (Stanford University Medical Center, Stanford, CA), who served as co-principal investigator for FAME and FAME 2, and now as PI of FAME 3, expressed skepticism that these data would inspire many changes in practice. “It may shift some people from FFR to iFR, because you don’t need to give adenosine,” Fearon predicted, “but the 3- or 4-minute time savings I don’t think is that big of a deal for most people.”

In fact, he suggested, the shift could go in the opposite direction toward FFR, due in part to what Fearon called a “strong trend” toward a higher death rate seen among iFR-guided patients in DEFINE-FLAIR. The “flip side,” he acknowledged, “is that the FFR had [nonsignificant 1.3%] higher unplanned revascularization rate. [But] which do you want?”

Asked by the media about the numerically higher death rate, Davies emphasized that the number cited was “all-cause,” with nearly all of the deaths in the iFR arm being noncardiovascular. “So there’s a lot of cancer,” he said, “and it’s way away from any statistical significance.”

The Outcomes

DEFINE-FLAIR randomized 2,492 patients with lesions of 40-70% stenosis on visual assessment to iFR or FFR guidance. At 1 year, the combined MACE rate of death, nonfatal MI, and unplanned revascularization was similar in the iFR and FFR groups (6.8% vs. 7.0%; P for noninferiority < 0.001), as were the rates of the individual components. However, as mentioned by Fearon, all-cause death trended higher in the iFR group at 1.9% versus 1.1% in the FFR group (P = 0.11). Patients were less likely to report adverse procedural symptoms and clinical signs—ie, chest pain and dyspnea—with iFR than with FFR (3.1% vs 30.8%; P < 0.001). Additionally, median procedure time was shorter for iFR than for FFR (40.5 vs 45.0 minutes; P = 0.001).

iFR-SWEDEHEART, meanwhile, randomized 2,037 patients taking part in the SCAAR registry who had 40-80% stenosis on visual assessment to iFR or FFR-guided care. At 1 year, the MACE rate was 6.7% for iFR and 6.1% for FFR (P for noninferiority = 0.007) and again there were no differences regarding individual endpoints. Moreover, there also was no disparity in the rate of stent thrombosis by 1 year of follow-up.

“These results suggest that the benefits of physiologically guided coronary revascularization with FFR can be achieved with iFR,” Davies et al say.

The Applications

In an editorial accompanying the NEJM paper, Deepak L. Bhatt, MD (Brigham and Women’s Hospital, Boston, MA), points out that the “results of these two trials do not apply to the evaluation of presumed culprit lesions in patients with acute coronary syndromes; for such lesions, current evidence favors early catheterization guided by an anatomical assessment of lesion severity.” Most patients enrolled in the studies had stable disease (approximately 80% in DEFINE-FLAIR and 60% in iFR-SWEDEHEART). In ACS patients, Bhatt explains, “the culprit lesion was treated according to standard practice, and then any additional lesions were assessed with the use of iFR or FFR.”

But for the patients in the cath lab who have a stenosis of intermediate severity, “iFR may be the new standard,” Bhatt concludes.

Fearon for his part remains unconvinced. Another wrinkle to DEFINE-FLAIR in particular, he said, “is that there were significantly fewer PCIs in the iFR group,” at 45.5% compared with 50.0% in the FFR group (P = 0.02). “Whether it’s conscious or subconscious, I think some operators—I mean, we’re interventionalists, we like to intervene—might choose FFR, because they want to do PCI and they might be concerned about leaving lesions that are causing ischemia behind,” Fearon suggested.

The proportion of patients in whom treatment was deferred was higher with iFR than with FFR in DEFINE-FLAIR (53% vs 47%; P = 0.003), though among the deferred patients, there was no difference in event rates between the two technologies.

“So it’ll be interesting to see how it all unfolds, whether it really changes people’s behavior,” he said. “I’m probably biased because I do FFR all the time, but I don’t find that adenosine is the big hindrance.” Side effects typically amount to “a minute or two of heaviness in the chest, shortness of breath. But if you warn the patient, in my experience they tolerate it fine,” he observed.

As for cost, intracoronary adenosine runs about $20 in the United States, Fearon reported, and intravenous adenosine is around $100-200. “But the wires,” which would be used in either technology, “are about $600 or $700,” he noted. “So the wire’s the big expense.” In fact, the lack of reimbursement for physiologic assessment, a factor again shared by FFR and iFR alike, is one of the biggest barriers to its uptake, he stressed.

The Body of Evidence

One big difference between the two technologies is that iFR is proprietary, whereas FFR has never been, Fearon added.

To TCTMD, and in his comments following Davies’ presentation in the Main Tent at ACC, Fearon suggested it would be useful to home in on the approximately 20% of cases for which studies have shown iFR and FFR disagree. Including the vast majority of instances for which the two tests agree in a noninferiority analysis, he explained, “might muddy the water and dilute out any potential difference.”

Asked about this possibility at the press conference, panelists as a whole said it was far more important to look at how these tests would actually be used (the “strategy”) rather than such technicalities of study design, which they thought would matter little to patients in the cath lab. “You act upon your FFR measurement or your iFR measurement—you treat or you don’t treat—so the patient receives a stent or doesn’t receive a stent,” observed Pieter Smits, MD, PhD (Maasstad Hospital, Rotterdam, the Netherlands).

Davies also stressed that the new data do not stand alone. “There’s a body of evidence here,” he said. “And sure, we can’t convince people who’ve built their careers on FFR. I know it’s very difficult for them to see hard clinical endpoints which do appear to show equivalence. But that’s the facts. And we all have to confront things, and particularly when it’s about your own work, it can be very challenging at times.”

Photo Credit: Royal Philips