PLATFORM: FFRCT-Guided Strategy in Suspected CAD ‘Durable’ Out to 1 Year

Long-term follow-up of patients in PLATFORM—a study showing that a diagnostic strategy using fractional flow reserve (FFR) derived from computed tomography (CT) significantly reduces the number of patients requiring invasive coronary angiography—suggests the results are durable beyond the initial 90 days. FFRCT also was associated with significantly less use of hospital resources and cost compared with planned invasive angiography.

Overall, care of stable CAD patients guided by FFRCT was also associated with equivalent clinical outcomes and quality of life when compared with a strategy of planned coronary angiography. 

“The reason for the 1-year follow-up was predominately safety,” said lead investigator Pamela Douglas, MD (Duke Clinical Research Institute, Durham, NC). “We wanted to make sure we didn’t miss a significant lesion—that people weren’t having late urgent revascularizations, late heart attacks, late deaths due to cardiovascular reasons—in the study arm where we canceled 60% of scheduled catheterizations.” She said that in altering the planned clinical pathway with CT and FFR, the goal was to determine the durability of that decision. “Do you just limp along for 3 months,” said Douglas, “or can we get safely out to 1 year?”

In the patients who had their invasive coronary angiography canceled on the basis of the CT/FFRCT findings, there were no late clinical events during 1 year of follow-up and only one late coronary revascularization for a patient with a lesion that had progressed in the follow-up period. Among the patients in the usual care arm who proceeded directly to invasive coronary angiography, there were a total of two major adverse cardiovascular events and three late coronary revascularizations during follow-up.  

“There were really few clinical events in either arm, but certainly no more in the CT/FFRCT group than there were in the usual-care arm,” Douglas told TCTMD. “There was only one revascularization, which was essentially a lesion that had progressed rather than a missed lesion.”

Ronald Karlsberg, MD (Cardiovascular Research Foundation of Southern California, Beverly Hills), who wrote an editorial accompanying the study along with René Sevag Packard (UCLA Medical Center, Los Angeles, CA), told TCTMD that he and others previously showed that the use of coronary CT angiography in their cardiology practice reduced the annual utilization of invasive coronary angiography by approximately 50% without changing the number of patients requiring PCI. “In our experience, even 10 years ago it got the right patient to the lab, compared to all the other testing available” noted Dr. Karlsberg.

He added that CT angiography alone—even without FFR—is “very powerful” in terms of predicting “what an interventional cardiologist would do in the cardiac catheterization lab.”

“FFRCT really helps, but it helps in the intermediate group and in patients with calcification; it is not needed for every case. said Karlsberg. “If you have a coronary narrowing less than 30%, or you have a critical narrowing on CT, you don’t need FFR. It’s a very helpful technology and it can strengthen the ‘soft underbelly’ of CT angiography, which is the intermediate-range vessels and calcium. [FFR] is important technology and it’s valuable, but it must be implemented correctly and used selectively.”    


Published online ahead of the August 2, 2016, issue of the Journal of the American College of Cardiology, the 584-patient PLATFORM study included two arms: 204 patients undergoing planned noninvasive testing and 380 patients undergoing planned invasive testing with diagnostic angiography. In both study groups, patients were assigned to usual care or an FFRCT-guided strategy.

As reported previously by TCTMD, the purpose of the study was to determine if FFRCT is safe and feasible as a triage strategy in patients with new-onset chest pain. The FFRCT system (HeartFlow, Redwood City, CA), which quantifies FFR using data obtained from a standard CT scan, is intended to provide anatomic and functional data about the identified ischemic lesion in patients. The CT data is forwarded to HeartFlow, where the FFR is estimated using computational flow dynamics and proprietary software, and FFR results are returned to the hospital in approximately 24 to 48 hours.

The primary results of PLATFORM—the percentage of patients without obstructive coronary artery disease at 90 days assessed by coronary angiography—were first presented at the European Society of Cardiology Congress 2015. Investigators showed that 60% of angiograms were canceled in the FFRCT-guided arm after receiving functional and anatomic data from the imaging test. For patients who did undergo angiography, just 12% did not have obstructive coronary disease. In contrast, 73% of patients sent directly to invasive angiography had no obstructive disease. This translated into a 61% decrease in the finding of obstructive coronary disease at the time of invasive angiography.

In addition to assessing the 1-year clinical outcomes, the researchers performed an economic analysis in the planned invasive and noninvasive cohorts. Among the patients scheduled for a planned invasive procedure, the mean 1-year cost of medical care per patient was significantly lower with the FFRCT-guided strategy versus usual care ($8,127 vs $12,145; P < 0.0001). The reduction in cost was driven by the canceled invasive coronary angiograms in the FFRCT arm, although when the costs of the initial tests were excluded, the mean per-patient downstream costs were still lower in the FFRCT arm than in the usual-care arm ($7,831 vs $9,864; P < 0.0001).

In the patients scheduled for noninvasive testing, clinical event rates were low at 1 year, with just one major adverse cardiovascular event occurring in the usual-care group and none in the FFRCT-guided group. Mean 1-year per-patient costs were similar between the two strategies ($3,049 with FFRCT-guided care vs $2,579 for usual care; P = 0.82).

To TCTMD, Douglas said that in patients scheduled for noninvasive testing, the FFRCT-guided strategy didn’t appear to add much in terms of clinical benefit. She noted that PLATFORM was conducted at 11 European sites, as well as at the Duke Clinical Research Institute, and these centers frequently use coronary CT as part of usual care, which made it difficult to differentiate outcomes between the two arms.

Treatment Protocol

In the editorial, Karlsberg and Packard acknowledge the difference in local practice patterns between Europe and the United States, drawing attention to the planned invasive cohort and noting that patients would not likely be sent directly to invasive coronary angiography in many centers under current guidelines. They note that in PLATFORM, approximately 50% of the patients in both study arms had noninvasive testing before the decision to proceed to invasive angiography and that the results of those noninvasive tests are not known. 

Speaking with TCTMD, Douglas said that in patients without known coronary artery disease, the clinical guidelines and appropriate use criteria do allow for physicians to send patients directly to coronary angiography without having a noninvasive test first. Although this should occur relatively infrequently, “when all the stars point to coronary disease,” such as in patients with multiple risk factors or typical chest pain and those in the “right demographic,” physicians will send these individuals for angiography without any other testing first.

Still, the researchers see a significant value in using CT/FFRCT to limit the number of patients sent for diagnostic angiography.

“If you don’t need the cath lab to make the diagnosis of coronary artery disease, I think most people would say, in the absence of extreme amounts of coronary calcification, CT is really pretty good,” said Douglas. “And if you’re not a candidate for revascularization with PCI because you don’t have a hemodynamically significant lesion, then what good is the cath lab? It’s not good for reassurance. It’s not good for diagnosis. It’s not good for planning treatment. So if you can keep those people out, you markedly reduce the cost, as we noted, of a chest-pain evaluation.”

For Karlsberg, the use of FFR for patients with chest pain who undergo CT angiography may also serve as a high-level secondary opinion with the added benefit of artery specific physiological data. “Everybody thinks they do a great job, however interpretations do vary from expert to expert. When an FFRCT is performed a very detailed analysis of the artery is performed. That may alone enhance the interpretation. In addition we also get a measure of blood flow in the artery which is the most powerful predictor of the need for revascularization”

For FFRCT to be used more widely, though, there is a need for clinical outcomes trials, similar to the FAME and FAME 2 studies, showing the technology can safely and effectively manage patients with coronary artery disease, he added. “I think PLATFORM is a step in that direction,” said Karlsberg, but an outcomes study comparing CT angiography alone, CT with FFR, and CT with other functional studies, such as myocardial perfusion testing, is needed.

Michael O’Riordan is the Managing Editor for TCTMD. He completed his undergraduate degrees at Queen’s University in Kingston, ON, and…

Read Full Bio
  • Douglas PS, De Bruyne B, Pontone G, et al. 1-year outcomes of FFRCT-guided care in patients with suspected coronary disease. J Am Coll Cardiol. 2016;68:435-445.

  • Packard RR, Karlsberg RP. Integrating FFRCT into routine clinical practice. J Am Coll Cardiol. 2016;68:446-449.

  • The PLATFORM study was funded by HeartFlow.
  • Douglas reports receiving grants from HeartFlow and previously received grant support from GE Medical Systems.
  • Packard and Karlsberg report being supported by a grant from the National Institutes of Health. Packard reports being supported by a grant from the UCLA STAR program and Karlsberg reports being supported by a grant from the Cardiovascular Research Foundation of Southern California.