Early Data Hint at Good Results With Angiography-Derived FFR

PARIS, France—Quantitative coronary angiography (QCA) can serve as a stepping stone to measuring fractional flow reserve (FFR”), researchers reported yesterday in a hotline session at EuroPCR. Two methods using specialized computer software were presented.

The idea is that interventionalists can do functional assessment without adding another test to the mix, session moderator Niels R. Holm, MD (Aarhus University Hospital, Aarhus, Denmark), explained to TCTMD.

Also, he said, “you don’t have to use adenosine, which is extremely important in terms of feasibility and price.”

The two presentations, though describing different systems, have the same general idea.

Shengxian Tu, MD (Shanghai Jiao Tong University, Shanghai, China), shared results of the multicenter FAVOR study, which tested the diagnostic accuracy of a method that uses hyperemic flow on two QCA images to calculate quantitative flow ratio (QFR). The researchers refer to the technique as cQFR. Among 88 patients, 15 were excluded due to poor angiographic quality, pressure wire drift, and other technical issues.

In the final study population of 73 patients, Tu said, cQFR showed “good diagnostic accuracy” compared with FFR for the threshold of 0.80 or less: 86% accuracy, 74% sensitivity, 91% specificity, 80% positive predictive value, and 88% negative predictive value.

In addition, Mariano Pellicano, MD (OLV Clinic, Aalst, Belgium), shared data on a technique he called FFRangio. That approach involves using angiograms to create a 3D model of the coronary tree. It “shows a superior concordance with invasive FFR and can be obtained within minutes during a regular coronary angiogram,” he concluded.

From Research to Practice 

Holm explained that the results of both systems are “highly dependent on angiographic quality,” and as such there is some learning curve. Both are faster than previous iterations that relied on computational fluid dynamics, he said. However, he added, the results are early and the enrolled patients do not span the spectrum of what is seen in clinical practice.

According to Holm, who coauthored FAVOR along with Tu, that system “is already now in clinical testing in cath labs.” Medis Medical Imaging is currently seeking CE Mark approval for the software, which Holm predicted would be marketed soon.

“One of the main questions we’ve got is how this should feed into clinical practice” once validated, he said.

For centers that are already doing FFR “then this [newer approach] will be your first step. You’ll see if it’s feasible to do, and if you get measurements in the gray zone you might add FFR,” Holm explained. For those without FFR, it can enable physiologically based lesion assessment at a lower cost.

Justin Davies, MD, PhD (Imperial College London, England), also a panelist at the session, said to TCTMD that the angiography-based FFR is sought after because it’s “quicker and simpler” than standard FFR. “But this is early phase, and these techniques take a long time often to do. . . . You need multiple views, and there’s process and time.”

Five to 10 years from now, that all will have improved, Davies predicted, noting that there are additional systems on the way. “But what my bugbear was there is that they, as often is the case, [seem to have] cherry-picked their distributions,” and this makes it look like the test performs better than it actually does.

While it’s not “cheating,” it’s a consequence of early studies needing to focus on lesions that are less severe or more severe than average, since it is difficult to tease out differences in the middle, where most of the population tends to fall, Davies explained. “But the fact is, to know the power of a test, you need to interrogate it and test it in the distribution [that reflects the real-world range] it’s going to be used in.”

In short, yesterday’s EuroPCR presentations were “slightly over-egging their state of play at the moment,” he commented.

• Pellicano M. Image-based FFR during coronary catheterization. Presented at: EuroPCR 2016. May 19, 2016. Paris, France.
• Tu S. Diagnostic accuracy of a fast computational approach to derive FFR from coronary x-ray angiography: results from the international multicenter FAVOR (Functional Assessment by Various Flow Reconstructions) pilot study. Presented at: EuroPCR 2016. May 19, 2016. Paris, France.

Disclosures: 

• Holm reports receiving honoraria from St. Jude Medical as well as institutional/research
• Tu reports receiving institutional grant/research support from Medis Medical Imaging Systems.
• Davies reports serving as a consultant for Philips Volcano and Medtronic and receiving institutional grant/research support from AstraZeneca, Medtronic, and Philips Volcano.

Related Stories: 

• CT-Derived FFR Alters Angiography-Based Management Plans for Stable Angina 
• NeXT Up: CT-derived FFR Demonstrates Strong Diagnostic Accuracy 
• DISCOVER-FLOW Published: Noninvasive FFR Shows High Diagnostic Accuracy