This article is more than 5 years old. Information may no longer be current.
Fractional Flow Reserve: From Case Examples to Clinical Research
Click Here to Manage Email Alerts
With the publication of the FAME II trial in August 2012, interest in fractional flow reserve-guided PCI has grown steadily during the following months to the extent that many clinicians are now wondering whether FFR has finally reached primetime status. In order to address this clinically meaningful question, the following is a brief review of this potentially practice-changing technology.
FFR in Practice
FFR was developed in the early 1990s as a tool to invasively determine the presence and extent of inducible myocardial ischemia within a coronary artery. The technique requires advancing a 0.014-inch wire across an indeterminate coronary stenosis, achieving maximal hyperemia with adenosine, and measuring the distal pressure in relation to proximal pressure. This pressure ratio mathematically and conceptually corresponds to a flow ratio.
For example, if the FFR value is 0.70, it means that the maximum blood flow is only 70% of what is expected if there was no stenosis. FFR takes into account the viability of the subtended myocardium. If the myocardium subtended by a coronary stenosis has significant scar tissue, inducible ischemia may not be present since less flow is required to this territory. It is important to point out that FFR is testing for inducible ischemia, which a patient might experience as a result of increased heart rate and/or BP during stress/exertion. FFR is capable of determining the presence of rest ischemia; however, this would normally be evident clinically.
FFR to Defer Revascularization
FFR is a valuable tool to select patients who can safely be deferred from revascularization. The DEFER trial found that it was safe to defer revascularization of non-left main coronary arteries with an FFR ≥ 0.75. The threshold of 0.75 was selected because it was previously found to correlate with an abnormal finding on at least one modality of non-invasive stress test. Other studies have documented the safety of deferring revascularization of borderline left main coronary arteries. Figure 1 shows a left main coronary artery with a proximal 60% to 70% stenosis. This patient had been referred for CABG. FFR was performed and found to be 0.83; therefore, revascularization was deferred.
Anthony A. Bavry
Another patient was referred to our institution for CABG and aortic valve replacement. The patient was aged 82 years, had a history of severe aortic stenosis, two-vessel CAD, prior MI, diabetes, atrial fibrillation, and left ventricular dysfunction (LV ejection fraction, 35%). A coronary angiogram from the outside institution was interpreted as a 50% to 60% diffuse stenosis in the proximal to mid-right coronary artery (Figure 2, panel A) and a 60% to 70% diffuse stenosis in the proximal to mid-left anterior descending artery (Figure 2, panel B). Due to the questionable significance of these lesions, a coronary angiogram was repeated with FFR (right coronary artery FFR=0.97; left anterior descending artery FFR=0.91). Surgical aortic valve replacement was performed without revascularization.
Figure 1. Ostial 60% to 70% left main stenosis (FFR=0.83).
Images: Anthony A. Bavry, MD, MPH
Figure 2. Panel A reveals moderate diffuse disease in the mid-right coronary artery (FFR=0.97). Panel B reveals moderate-severe diffuse disease in the mid-left anterior descending artery (FFR=0.91).
FFR to Select Patients for Revascularization
FFR is equally useful to select lesions for appropriate revascularization. The following patient in Figure 3 (see p. 8) had a congenitally absent circumflex artery that was supplied by collaterals from the right coronary artery. Due to a high-grade stenosis in the left anterior descending artery, he had previously undergone CABG to this vessel; however, his angina persisted. A subsequent coronary angiogram revealed the right coronary artery to be mild to moderately diffusely diseased with no high-grade stenosis (Figure 4, panel A). Since the right coronary artery additionally supplied the circumflex territory, more blood flow was needed to this hyperdominant vessel. Such a patient is particularly prone to mild reductions in flow. FFR was 0.56, which revealed profound inducible ischemia. Revascularization was performed with two everolimus-eluting stents (3 mm × 38 mm and a 3 mm × 32 mm; Xience, Abbott) with subsequent high-pressure balloon inflation (Figure 4, panel B). The patient had complete resolution of symptoms.
In another example, a patient with a history of gastrointestinal hemorrhage had significant dyspnea (and tachypnea) with minimal exertion, which resulted in multiple hospitalizations for what was attributed to panic attacks. He denied any angina. A coronary angiogram was ultimately performed and revealed mild diffuse disease in the proximal left anterior descending artery (Figure 5, panel A). FFR of the left anterior descending artery was 0.72. IVUS documented a diffusely diseased artery with a minimal luminal area of approximately 2.7 mm2. Revascularization was performed with a 2.5-mm × 28-mm bare-metal stent, post-dilated with high-pressure to 2.75 mm (Figure 5, panel B). The patient had no additional angina-equivalent episodes.
In 2012, the FAME II trial found that patients with an FFR ≤0.8 who were randomly assigned to PCI had a reduction in adverse events (death, MI or urgent revascularization) compared with medical treatment without revascularization. The trial was terminated early due to benefit from PCI in the presence of inducible ischemia, which was driven by a reduction in urgent revascularization.
Figure 3. Congenitally absent circumflex artery.
Figure 4. Panel A reveals moderate diffuse disease in the right coronary artery (FFR=0.56). Continuation of the right coronary artery to the circumflex territory is seen resulting in a hyper-dominant vessel. Panel B is the final angiographic result after PCI with two DES.
Limitations to Widespread Use
Although FFR is effective at selecting patients who benefit from revascularization by reducing adverse events and by selecting patients who can safely be deferred from revascularization, it is still not commonly used. A large sample of Medicare patients from 2001 to 2009 documented the use of FFR to be climbing during this period; however, in 2009, it still represented less than 1% of total catheterizations. There are several reasons for this.
First, FFR may be perceived as consuming too much time. If a lesion is assumed to be obstructive or non-obstructive, the use of FFR might be viewed as unnecessary. However, the FAME trial reminded us that 20% of angiographically severe lesions (71%-90%) are found to not be flow-limiting when interrogated, and, conversely, 35% of moderate lesions (50%-70%) can be ischemic. This was highlighted in these case examples. The reality is that we do not know what the flow is in the artery unless the steps are taken to obtain this information.
Second, FFR is a relatively simple procedure. Although this is good for ease of use, it may not carry the same appeal that other areas of interventional cardiology enjoy; among them are revascularization of chronic total occlusions/bifurcation lesions, use of atherectomy/thrombectomy devices and mechanical support devices (ie, Impella), and advanced imaging catheters (ie, optical coherence tomography and infrared imaging).
Figure 5. Panel A reveals mild diffuse disease in the proximal left anterior descending artery (FFR=0.72). Panel B is the final angiographic result after PCI with one bare-metal stent.
Third, FFR will reduce the number of PCIs. The FAME trial documented a 30% reduction in PCI with a FFR-guided PCI strategy vs. an angiography PCI strategy. Other studies and personal experience support this notion. Our goal is to revascularize patients in whom benefit is expected to outweigh risk. PCI might make a coronary artery look better angiographically, but if there was no flow limitation to begin with, the stent may just add to patient risk.
Conclusion
In summary, FFR is an important tool in the cath lab that can precisely determine the location and degree of myocardial ischemia in a given coronary artery. It is useful for determining lesions that can safely be deferred from revascularization. On the other hand, FFR is useful in selecting appropriate lesions for revascularization. The use of FFR will likely continue to increase in the future, especially as it becomes incorporated into guidelines and appropriateness criteria. Unresolved questions include the optimal threshold to define inducible ischemia since 0.75 to 0.8 is currently considered a “gray-zone” area. It is also unknown how best to medically treat patients who have mildly impaired coronary blood flow, but are deferred from revascularization (ie, patients with an abnormal FFR, but >0.8). Presently, FFR is not quite at primetime status, but its day is rapidly approaching.
Disclosure: Bavry reports receiving research support from Eli Lilly and Novartis Pharmaceuticals; and serving as a consultant for the American College of Cardiology’s CardioSource.