FFR 2.0: A Calculated Run for Supremacy
Fractional flow reserve (FFR) has proven to be a valuable asset in cardiac cath labs for assessing ischemia-causing lesions in patients with coronary artery disease. But it is not without its drawbacks; the procedure is invasive and relies on the use of pharmacologic agents. Two new approaches soon may provide interventional cardiologists with alternatives.
FFR gives interventionalists an important tool for assessing coronary stenosis before and during PCI. The concept is based on the proportional relationship of coronary blood flow to pressure, with a decrease in pressure across a lesion indicative of decreased flow to the myocardium. But the calculation requires that intracoronary resistance remains constant and minimal. To achieve that, physicians administer vasodilator agents, such as adenosine, to pharmacologically induce hyperemia.
Uptake in the use of FFR in the U.S. has been slow, partly because some patients don’t tolerate adenosine well and some physicians find the FFR process cumbersome and time consuming. For instance, in a poster presentation at ACC.12, Rudzinski et al reported that 26 percent of patients infused with adenosine for FFR measurements developed shortness of breath and flushing. Dattilo et al, in a separate analysis of National Cardiovascular Data Registry CathPCI Registry data presented as a poster at TCT.11, reported that FFR was used in only 7 percent of patients who underwent PCI for intermediate coronary stenoses.
“FFR has been widely considered the gold standard but it is an invasive procedure,” says Min. On the other hand, noninvasive coronary CT angiography (CCTA) helps to identify and exclude obstructive coronary stenosis, but it tends to overestimate the presence and severity of occlusions. Min and colleagues sought to meld the benefits of FFR and CCTA by applying the principles of computational fluid dynamics to CCTA scans to model blood flow and pressure data in coronary arteries and calculate lesion-specific FFR measurements. They call the technique FFRCT.
To test their concept, they modeled artery flow based on CCTA data from 103 patients (159 vessels) with suspected or known coronary artery disease who underwent CCTA, invasive angiography and FFR (J Am Coll Cardiol 2011;58[19]:1989-1997). Using FFR as their reference standard, they compared FFRCT with CCTA stenosis, and found that FFRCT had a superior diagnostic accuracy and added to CCTA’s discriminatory capability to identify lesions, mostly by reducing the rate of false positives.
FFRCT showed a good correlation with FFR without the need for additional scans, medications or changes to protocols, prompting the researchers to conclude that computing FFRCT from CCTA scans might provide a noninvasive method for detecting and excluding ischemia-causing lesions.
“FFRCT has the potential to identify lesions by traditional CT that are potentially ischemic and rule out ischemic-causing lesions,” Min says. “In that manner, the combination of CT plus FFRCT may serve as an effective gatekeeper to the cath lab. The CT is much less expensive than an invasive angiogram and FFR.”
“A lot of the FFR pitfalls are associated with administering adenosine,” Davies says. For cath lab personnel and physicians, those issues may include costs, time and variable responses to the drug. While the time and cost per procedure may seem insignificant—in some facilities it may be mere minutes with an incremental cost for adenosine—Davies argues these add up. Additionally, patients with severe lung disease, asthma, very low blood pressure and rhythm disturbances may not be able to tolerate adenosine.
They designed a two-part study to first ascertain a period of iFR and then assess iFR’s diagnostic efficacy compared with FFR (J Am Coll Cardiol 2012;59[15]:1392-1402). In a multicenter study, they identified 131 patients (157 stenoses) scheduled for coronary angiography or PCI who were divided into two groups. In the first group, the researchers measured intracoronary pressure and flow velocity to assess baseline resistance and resistance under pharmacologic vasodilation. In the second group, they measured intracoronary pressure only to assess iFR and FFR.
They found that during a defined diastolic wave-free period, the resting coronary resistance was similar to the resistance of adenosine-assisted FFR and that there was a strong correlation between FFR and iFR, with specificity, sensitivity, and negative and positive predictive values of 91 percent, 85 percent, 85 percent and 91 percent, respectively.
“People fundamentally like things that are easy and techniques that are quicker,” Davies says. “But iFR also takes away many of the steps that could introduce noise into the system.”
FFRCT also is being developed as a user-friendly alternative to FFR, says Min, adding that it can be computed from any CT scan. HeartFlow in Redwood City, Calif., is developing the technology; HeartFlow’s core lab scientists performed the FFRCT in a blinded fashion for the 2011 study.
Both iFR and FFRCT may offer other benefits. Davies sees iFR having a future role periprocedurally by providing real-time assessment of the effectiveness of a stent once it is deployed. FFRCT may be used as a treatment planning tool to predict the therapeutic benefit of PCI .
“Computational fluid dynamics has never been clinically applied to daily practice, so this has the potential to change the game,” Min says.
Dethroning a technology as important as FFR will require compelling results that satisfy operators who want to ensure their patients get the best care. “We are advocating a major change in clinical assessment, and we need to produce more data and build a solid dossier of evidence to support the use of this technique in the cath lab,” Davies says. “We need to hit a very high standard to ensure our technique is worthy of its place in the cath lab.”
FFR gives interventionalists an important tool for assessing coronary stenosis before and during PCI. The concept is based on the proportional relationship of coronary blood flow to pressure, with a decrease in pressure across a lesion indicative of decreased flow to the myocardium. But the calculation requires that intracoronary resistance remains constant and minimal. To achieve that, physicians administer vasodilator agents, such as adenosine, to pharmacologically induce hyperemia.
Uptake in the use of FFR in the U.S. has been slow, partly because some patients don’t tolerate adenosine well and some physicians find the FFR process cumbersome and time consuming. For instance, in a poster presentation at ACC.12, Rudzinski et al reported that 26 percent of patients infused with adenosine for FFR measurements developed shortness of breath and flushing. Dattilo et al, in a separate analysis of National Cardiovascular Data Registry CathPCI Registry data presented as a poster at TCT.11, reported that FFR was used in only 7 percent of patients who underwent PCI for intermediate coronary stenoses.
A surrogate for FFR
James K. Min, MD, of Cedars-Sinai Medical Center in Los Angeles, and Justin E. Davies, MBBS, PhD, of Imperial College London, independently have helped develop alternate approaches to FFR. Min and colleagues have borrowed a computational strategy typically used to design aircraft to find a noninvasive method for calculating an FFR measurement while Davies and colleagues are promoting a technique that eliminates the use of adenosine.“FFR has been widely considered the gold standard but it is an invasive procedure,” says Min. On the other hand, noninvasive coronary CT angiography (CCTA) helps to identify and exclude obstructive coronary stenosis, but it tends to overestimate the presence and severity of occlusions. Min and colleagues sought to meld the benefits of FFR and CCTA by applying the principles of computational fluid dynamics to CCTA scans to model blood flow and pressure data in coronary arteries and calculate lesion-specific FFR measurements. They call the technique FFRCT.
To test their concept, they modeled artery flow based on CCTA data from 103 patients (159 vessels) with suspected or known coronary artery disease who underwent CCTA, invasive angiography and FFR (J Am Coll Cardiol 2011;58[19]:1989-1997). Using FFR as their reference standard, they compared FFRCT with CCTA stenosis, and found that FFRCT had a superior diagnostic accuracy and added to CCTA’s discriminatory capability to identify lesions, mostly by reducing the rate of false positives.
FFRCT showed a good correlation with FFR without the need for additional scans, medications or changes to protocols, prompting the researchers to conclude that computing FFRCT from CCTA scans might provide a noninvasive method for detecting and excluding ischemia-causing lesions.
“FFRCT has the potential to identify lesions by traditional CT that are potentially ischemic and rule out ischemic-causing lesions,” Min says. “In that manner, the combination of CT plus FFRCT may serve as an effective gatekeeper to the cath lab. The CT is much less expensive than an invasive angiogram and FFR.”
Adieu to adenosine
Davies and colleagues also homed in on the mathematical underpinnings of FFR to find a technique that provides similar value without the need for administering adenosine. They hypothesized that they could identify a time period when intracoronary resistance at rest is naturally constant and minimized, providing a window that is equivalent to hyperemia induced by adenosine. They named this the instantaneous wave-free ratio (iFR), which they argued could be used to derive an index of stenosis severity over time.“A lot of the FFR pitfalls are associated with administering adenosine,” Davies says. For cath lab personnel and physicians, those issues may include costs, time and variable responses to the drug. While the time and cost per procedure may seem insignificant—in some facilities it may be mere minutes with an incremental cost for adenosine—Davies argues these add up. Additionally, patients with severe lung disease, asthma, very low blood pressure and rhythm disturbances may not be able to tolerate adenosine.
They designed a two-part study to first ascertain a period of iFR and then assess iFR’s diagnostic efficacy compared with FFR (J Am Coll Cardiol 2012;59[15]:1392-1402). In a multicenter study, they identified 131 patients (157 stenoses) scheduled for coronary angiography or PCI who were divided into two groups. In the first group, the researchers measured intracoronary pressure and flow velocity to assess baseline resistance and resistance under pharmacologic vasodilation. In the second group, they measured intracoronary pressure only to assess iFR and FFR.
They found that during a defined diastolic wave-free period, the resting coronary resistance was similar to the resistance of adenosine-assisted FFR and that there was a strong correlation between FFR and iFR, with specificity, sensitivity, and negative and positive predictive values of 91 percent, 85 percent, 85 percent and 91 percent, respectively.
“People fundamentally like things that are easy and techniques that are quicker,” Davies says. “But iFR also takes away many of the steps that could introduce noise into the system.”
Next steps
Davies holds a patent on processes for identifying the wave-free window and for making the system robust. Imperial College London has licensed the intellectual property to Volcano, a San Diego-based medical device company that supported the two-step study. Algorithms automatically identify the diastolic wave-free period to calculate a pressure-derived index without the need for vasodilator agents. The process is otherwise similar to FFR. “Once the wire is down there, you close the y-connector port, wait for pressure to stabilize and measure it for five beats,” Davies says. “It is very rapid.”FFRCT also is being developed as a user-friendly alternative to FFR, says Min, adding that it can be computed from any CT scan. HeartFlow in Redwood City, Calif., is developing the technology; HeartFlow’s core lab scientists performed the FFRCT in a blinded fashion for the 2011 study.
Both iFR and FFRCT may offer other benefits. Davies sees iFR having a future role periprocedurally by providing real-time assessment of the effectiveness of a stent once it is deployed. FFRCT may be used as a treatment planning tool to predict the therapeutic benefit of PCI .
“Computational fluid dynamics has never been clinically applied to daily practice, so this has the potential to change the game,” Min says.
Dethroning a technology as important as FFR will require compelling results that satisfy operators who want to ensure their patients get the best care. “We are advocating a major change in clinical assessment, and we need to produce more data and build a solid dossier of evidence to support the use of this technique in the cath lab,” Davies says. “We need to hit a very high standard to ensure our technique is worthy of its place in the cath lab.”