This article is more than 5 years old. Information may no longer be current.
FFRCT: A new paradigm for predicting ischemia
Click Here to Manage Email Alerts
The treatment of patients with CAD would be different if there was a reliable noninvasive method to determine the hemodynamic significance in each case. If cardiologists knew which lesions were likely to cause ischemia just by doing some calculations from a CT scan, the need for invasive coronary angiography could be greatly reduced, resulting in less painful and more accurate treatment for some patients and cost savings to the health care system. Results from two recent studies suggest that time may be near.
The studies — DeFACTO, conducted in 252 patients at 17 sites and published in September 2012, and DISCOVER-FLOW, conducted in 103 patients at four sites and published in November 2011 — assessed the diagnostic performance of fractional flow reserve (FFR) computed from noninvasive CT angiography (FFRCT). As defined in DeFACTO, FFR is “a physiologic measure of coronary stenosis expressing the amount of coronary flow still available despite the presence of a stenosis.” In other words, it can provide cardiologists a good idea of which lesions present the risk for ischemia and which do not.
Daniel S. Berman, MD, said the future of fractional flow reserve computed from noninvasive CT angiography is promising.
Source: Cedars Sinai Department of Medical Media; reprinted with permission
“It is important to identify and diagnose through noninvasive methods coronary lesions that would benefit from revascularization due to ischemia,” said James K. Min, MD, lead author of the DeFACTO study and associate professor at University of California, Los Angeles, School of Medicine at Cedars-Sinai Medical Center. “We can use computational fluid dynamics to do those evaluations, but it has not been applied to coronary lesions because we haven’t had a noninvasive tool to measure it until the introduction of coronary CT angiography.”
Benefits of noninvasive measure
J. Jeffrey Carr, MD, MSc, professor of radiology, cardiology and public health sciences at Wake Forest University School of Medicine, said this line of research is important.
J. Jeffrey Carr
“A noninvasive measure of FFR would dramatically improve our ability to determine the importance of stenotic lesions to the patient and thus optimize interventions,” he said. “Given that 62% of all invasive coronary caths have nonobstructive disease, this would greatly help patients and their physicians avoid unnecessary procedures and result in significant cost savings to the health system.”
Daniel S. Berman, MD, director of nuclear cardiology and cardiac imaging at Cedars-Sinai and professor of imaging at the Cedars-Sinai Heart Institute, also sees great promise.
“I see the use for coronary CT angiography growing rapidly, almost exponentially,” said Berman, who also is a member of the Cardiology Today Editorial Board. “For 30 years, this has been the Holy Grail of noninvasive diagnostics. Being able to combine coronary CT angiography with a functional assessment of coronary stenosis could have a dramatic effect on reducing the number of patients who go to the cath lab.”
That dramatic impact could be as much as 40%, according to Matthew J. Budoff, MD, professor of medicine at UCLA and director of cardiac CT at the Los Angeles Biomedical Research Institute in Torrance, Calif.
Matthew J. Budoff
If FFRCT was widely used, “the number of patients referred for revascularization would be reduced by probably 30% to 40%, as many of these lesions are not functionally significant, but look worrisome anatomically,” Budoff said in an interview. “Further, the number of functional tests and invasive FFR will drop significantly. With use of FFRCT, no additional testing or radiation is necessary.”
Indeed, the idea that FFRCT could be a “one-stop shop” for diagnosing lesions and determining the likelihood of causing ischemia is an idea that the medical community finds appealing, Min said. In fact, he noted, it is one of the reasons he became involved with this line of research in the first place. “I have been interested in identifying anatomic coronary stenoses and the hemodynamic significance of them,” he said.
Recent studies demonstrate success
In both DeFACTO and DISCOVER-FLOW, computation of FFRCT was performed in masked fashion by HeartFlow Inc., using computational fluid dynamic modeling technology. The technology takes data obtained from ICA or a CT scan and performs 3-D blood flow simulations. HeartFlow funded both studies.
DISCOVER-FLOW demonstrated a diagnostic accuracy of FFRCT that was “superior and additive” to evaluation by stenosis as viewed by coronary CT angiography. However, the study was large enough to evaluate only on a per-vessel, not per-patient, level. In particular, FFRCT showed a much lower rate of false positives.
DeFACTO, which was large enough to evaluate on a per-patient level, found that FFRCT diagnosed ischemia much better than reviewing CT scans alone. However, the study did not meet its prespecified primary endpoint, which was diagnostic accuracy of >70% of the lower bound of the one-sided 95% CI. Nonetheless, Min and his co-investigators concluded that the findings “can be considered proof of concept of the feasibility of this novel technology and, to our knowledge, represent the first large-scale prospective demonstration of the use of computational methods to calculate rest and hyperemic coronary pressure fields from typically acquired CT images.”
James K. Min
Results of the studies, particularly DeFACTO, have given prominent cardiologists hope that noninvasive diagnostic tools will soon be used regularly to assess the functional significance of coronary lesions.
“The DeFACTO study showed that this technique improved diagnostic accuracy above standard CTA reading by more than 25%, with a P value of .0001. That is huge,” Budoff said. “That was not the endpoint of the trial, but shows the utility and importance of FFRCT.”
Budoff said a third study is currently being conducted in Europe. Once complete, those data should be sufficient to convince the medical community of the value of FFRCT, he said.
Widespread benefits
Successful use of FFRCT should mean that doctors “will be able to identify cases that would benefit from coronary revascularization, and have a better ability to predict the benefits from revascularization,” Min said. “Not only will they be able to identify a specific lesion that can cause ischemia, but they will be able to computationally model a stent to see if a stent would modify the ischemia. This technology has the ability to do that.”
Another benefit, he said, is that physicians will be able to “identify with high confidence stenoses that will reduce ischemia, thereby avoiding an expensive and unnecessary invasive procedure. In addition, they will also be able to identify patients that they previously missed who would benefit from a stent. These are the patients who do not have high-grade coronary stenosis, but have lesions that cause ischemia.”
Future adoption of the technology
For widespread acceptance to occur, however, the technology will have to prove itself in everyday clinical scenarios, including vastly complex scenarios.
W. Gregory
Hundley
“One will need to examine the utility of this technology in routine clinical situations,” W. Gregory Hundley, MD, professor of internal medicine and radiology and medical director of CV MRI at Wake Forest University School of Medicine, said in an interview. “One of the more important situations will be in advanced CVD where one tries to identify the functionally important lesion from among many different anatomic lesions in a given patient.” Hundley also is the Imaging Section Editor of the Cardiology Today Editorial Board.
Fortunately, the technology is continuing to develop and improve, and the versions being used now are even better than the ones used in the DeFACTO and DISCOVER-FLOW studies. Future studies could produce even more positive results.
“I’d like to see the studies repeated using the newer generation of the technology,” Min said. “Since [DeFACTO was conducted], there have been additional iterations of this technology. So we suspect that it continues to get better.”
Carr agreed. “Both coronary CT and the computational algorithms used to calculate FFRCT will continue to evolve at a rapid pace,” he said. “Following further technical development, a repeat study has the potential to document the clinical viability of this approach.”
Another priority for future study, according to Min, is the comparison of traditional stress testing methods.
Future potential uses
Computational fluid dynamic modeling may be the future of ischemia diagnosis, but some experts said they believe that it may have a future in other medical applications as well.
“Potential uses of a similar technique may be useful outside the coronaries (legs, renals, carotid), as well as aortic disease, but coronary is the big one,” Budoff said.
“These approaches can be applied to many aspects of medical imaging, and we are seeing just the beginning,” Carr said. “Similar computational assets are being used to dramatically lower radiation exposure related to CT through iterative reconstruction algorithms. These are only the first-generation clinical applications, and we should expect more novel applications throughout medical imaging.”
Disclosure: Berman, Budoff, Carr and Hundley report no relevant financial disclosures. Min reports receiving consulting income from HeartFlow.