The FAME Trial: Measuring FFR Leads to Better Outcomes and Fewer Stents

Could FAME lead the way for PCI to show superiority over medical therapy and CABG for multivessel coronary artery disease?

What were the results of the FAME (Fractional Flow Reserve versus Angiography for Guiding PCI in Patients with Multivessel Coronary Artery Disease) trial¹?
The FAME study was a large, multicenter, prospective, international study, randomizing patients with multivessel coronary artery disease to two different strategies to decide which lesions to stent. One strategy was based on the current, most common method, which is using clinical, non-invasive stress testing data in conjunction with an angiogram and doing so-called “angiographic guidance” to decide where to stent. The other strategy was using a pressure wire to first measure the fractional flow reserve (FFR) and then deciding whether or not to stent, based on the result of the FFR. FAME showed that an FFR-guided strategy resulted in significantly lower rates of death, myocardial infarction, and need for repeat stenting or bypass. In addition, significantly less contrast was used and costs at the time of the procedure were significantly less too, primarily because about one-third fewer stents needed to be implanted in patients with the FFR-guided strategy. FFR measurement did not add any significant time to the procedure, which was also an important finding. The time of the procedure was identical in both arms. More and more often in the cath lab, we are seeing patients who have multiple narrowings. Deciding which ones to stent can be quite challenging. A common scenario is one very tight narrowing which is likely the culprit, and one or two other narrowings in other vessels that are more moderate, maybe 50 to 70% narrowed. The main clinical implication of the FAME study was that basing the decision about the significance of those more moderate lesions on the angiogram is not a good idea, because the angiogram can be misleading. FFR will tell you which of those lesions is actually causing a problem and needs a stent, and which lesions can be better treated medically. So, by doing what we call “functionally” complete revascularization, we are able to maximize the benefit of stenting and minimize the risks.

Were all lesions measured by FFR?
There were about 1,000 patients included in the study, and half were randomized to FFR guidance. In these patients, FFR was measured across all lesions. Before randomization, the operator had to indicate, based on the angiogram, which lesions warranted stenting. Then the patient was randomized. If randomized to angiographic guidance, then the operator just went ahead and stented the previously indicated lesions. If the patient was randomized to the pressure wire, FFR would then be measured, and only if it was significant would a stent be placed.

When you say the angiogram can be misleading, do you mean in terms of the degree of stenosis?
Yes. For example, we broke down the narrowings based on their angiographic severity into three groups: between 50-70% narrowed, 70-90% narrowed, and >90% narrowed. Of the lesions that were >90%, the vast majority, approximately 95%, had a significant FFR. That tells us it is probably not that helpful to measure FFR in this setting, because most of the lesions are going to be significant and need a stent. But when you look at the 70-90% narrowing group, which is generally considered pretty tight, about 20% of those were not significant according to FFR, and could safely be deferred and managed medically. Of the narrowings that were between 50-70%, at least half of those could be deferred and managed medically. We can conclude that the angiogram doesn’t always give you the absolute answer, particularly in moderate lesions, and having further information can be quite useful.

Did FAME use a pressure wire that was integrated into the cath lab or a standalone system?
FAME was sponsored by Radi Medical Systems (now owned by St. Jude Medical) and it used the Radi PressureWire connected to the Radi Analyzer. It was not in any integrated system (St. Jude has partnered with GE to offer an integrated FFR/hemodynamics system), but was the standalone PressureWire system.

Some feel the standalone system is more time-consuming to use, but FAME didn’t show a difference in procedure time.
I think if you use the pressure wire only once a month, it is more time-consuming. Just like rotational atherectomy or intravascular ultrasound, or any other device, if you are not regularly using it, you must first find out which closet it is in, then dust it off, bring it, plug it in and figure out how to work it. So, yes, in that case, it will add extra time. But if you get into the habit of having the pressure wire analyzer mounted on the cath lab table, turned on and ready to go, it will add very little extra time to the procedure.

You also mentioned that less contrast was used in the FFR group.
Yes. Patients who were in the FFR-guided strategy had an average of 270ccs of contrast and those in the angio-guided strategy had an average of 300ccs of contrast. In the FFR-guided strategy, if a lesion does not measure as significant, then you’re done. Whereas in the angio-guided strategy, you put a stent in, take pictures before and after ballooning, and then after stenting, and then put in more stents, requiring more contrast…

Were there any challenges put forward to the trial design or data?
One criticism was that current operators aren’t stenting moderate lesions. It was incorrectly thought that the trial mandated stenting of moderate lesions, not replicating current practice. But FAME didn’t mandate stenting of moderate lesions. The protocol dictated the presence of a 50% or greater narrowing and the interventionalist had to feel that it warranted stenting. The protocol was based on the angiogram, but also based on clinical, non-invasive data. Once the operator decided which lesions they would stent, then the patient was enrolled and randomized. If someone responds, well, I don’t stent moderate lesions like that, so this doesn’t apply to me, of those 50-70% lesions, about 40% of them are causing ischemia and have an abnormal FFR. If you ignore these lesions, 40% of the time, the patient would continue to have symptoms and be at higher risk for events. Or, if you said, well, I stent all lesions in the 50-70% range, so it doesn’t matter, that would also be wrong, because 60% of these lesions don’t need stenting. What it boils down to is that getting more information allows for a more refined approach to stenting. Another criticism was that this study was performed just by people who are familiar with FFR and it’s not broadly applicable. FAME was done at 20 centers, and it was international, involving both Europe and the U.S., so I don’t think that it is fair to say. Others have said that this study needs to be replicated, but it was a large, pivotal study. It wasn’t retrospective. I’m not sure why it would need to be replicated.

How did patients who received FFR-guided PCI in FAME compare to patients who receive medical therapy or who go on to coronary artery bypass graft surgery (CABG)?
If you compare FAME to SYNTAX,² where patients went on to CABG, it was a fairly similar population. SYNTAX did include patients with left main disease, which were excluded in FAME, but otherwise, the number of patients who had prior myocardial infarction with various risk factors, age, etc., was all similar. The angiographic characteristics were similar as well, in terms of the SYNTAX score. Likewise, in comparison to studies of medical therapy, if you compare FAME to COURAGE,³ it had a similar population to those patients treated in COURAGE. We would hypothesize that if SYNTAX were done again with an FFR-guided approach to PCI, and compared to bypass, that the results might be more favorable for PCI. Likewise, if COURAGE were repeated with FFR-guided PCI compared to medical therapy, we would hypothesize that the FFR-guided arm would perform more favorably.

A recently published study⁴ has indicated that computed tomography (CT) does not reliably predict the functional significance of lesions in patients with stable angina and atypical chest pain. Does this underscore the importance of FFR?
Yes, I agree completely. CT is an anatomic assessment, just like angiography and intravascular ultrasound. These are all methods for looking at anatomical characteristics, whereas FFR is a way of getting hemodynamic or physiologic information about the various stenoses. We’ve seen that the angiogram doesn’t correlate well with FFR, so one wouldn’t expect CT to correlate very well, either.

Are there scenarios where FFR could be used as a surrogate for a stress test?
Definitely. Probably around 30 to 40% of patients come to the cath lab without a prior non-invasive stress test. That’s the perfect situation for FFR, because we don’t have the data to know where ischemia is. FFR can give us an online stress test right there in the cath lab, which will allow us to determine which lesions are responsible for the patient’s symptoms. Even in patients who have had stress tests, FFR is quite valuable. Particularly with multi-vessel disease, the stress test may pick up the vessel that has the most severe stenosis, but it may not show ischemia in other regions with more moderate stenoses which may actually be physiologically significant. FFR can help identify those. In addition, if you have a vessel that has two lesions, the stress test may tell you that there is ischemia in the anterior wall, for example, and there may be two lesions in the left anterior descending artery (LAD). With FFR, you can pull the pressure wire back and determine which of the two lesions is more significant and causing more of an impact on flow. It gives you better spatial resolution. It might be that the lesion in the proximal LAD is the one that is causing the poor blood flow that was seen on the stress test. It pinpoints exactly the location of the problem.

Will you be presenting any further data at upcoming meetings?
We hope to be presenting the two-year results of the FAME study at TCT. I am also part of a coronary physiology symposium at the meeting, where we will review the nuts and bolts of FFR from the basics on to more advanced applications.

How do you see the pressure wire technology developing?
Further improvements in the wire technology will occur, so that maneuverability and technical aspects of the wire will continue to improve. The integration with cath lab hemodynamics systems will continue and will improve the ease of use. I expect that the medications we use to cause hyperemia, or maximal vasodilation, in order to measure the FFR will also become better and easier to use. All of these things will continue to make FFR a more user-friendly technique.

Can you tell us more about your research regarding the index of microcirculatory resistance (IMR)?
In general, when we look at the coronary circulation in the heart, we divide it into two areas. There are the epicardial or the large vessels, the ones we see well on the angiogram. FFR is very useful for telling us if there is a problem in the epicardial arteries. Then there are the much smaller vessels, which aren’t easily visualized with traditional angiography, and that we refer to as the microcirculation or microvasculature. Patients can have symptoms because of problems in the microvasculature, even though their epicardial vessel looks fine. It turns out that the pressure wire allows you to measure not only pressure, but to estimate coronary flow using a thermodilution technique. The pressure sensor can also act as a thermistor and measure temperature. By measuring flow as well as pressure, we can calculate what the resistance is in the small vessels, and this index of microcirculatory resistance (IMR) is a reflection of the health of those small vessels. People who have a high resistance have abnormal microvascular function, and that may be responsible for symptoms of chest pain. It’s not too uncommon that a patient will come to the cath lab with typical symptoms and may even have an abnormal stress test, and we do an angiogram and everything looks normal with respect to the epicardial vessels. We can use the pressure wire in that setting to assess the microvasculature to determine if that is actually responsible for the patient’s symptoms. There are other scenarios where the IMR is useful as well, including acute myocardial infarction (MI). It turns out that the amount of damage to the microvasculature is predictive of long-term outcomes. By measuring IMR at the time of the primary intervention for acute MI, we then have an idea of what risk the patient faces for problems in the long run.

Is using the pressure wire in this way part of the regular training?
It is part of the usual package that is commercially available. It’s not used as commonly as FFR, partly because it hasn’t been around as long. We’re just starting to get more and more published data to support its role. We would anticipate in the coming years that its use will continue to grow, and its value be appreciated more and more.

Any final thoughts?
If you have a patient on maximal medical therapy with severe single-vessel disease who continues to have typical symptoms of angina and an abnormal stress test, then there is no reason to measure FFR; just go ahead and stent. Or, if you have completely normal vessels and atypical symptoms, there is also no reason to measure FFR. But those types of cases are probably in the minority. Most patients have more than one lesion. It is in these cases where FFR is most helpful. As we showed in FAME, for mild to moderate lesions of 50-70%, about 40% will have an abnormal FFR and 60% won’t, but the angiogram cannot predict which ones are significant. With stenoses in the 70-90% range, about 20% do not have an abnormal FFR. Ultimately, having FFR at your disposal allows you to perform PCI more judiciously.

Dr. Fearon can be contacted at wfearon@stanford.edu.

References

1. Tonino PA, De Bruyne B, Pijls NH, et al. Fractional flow reserve versus angiography for guiding percutaneous coronary intervention. N Engl J Med 2009 Jan 15;360(3):213-224.
2. Serruys PW, Morice MC, Kappetein AP, et al. Percutaneous coronary intervention versus coronary-artery bypass grafting for severe coronary artery disease. N Engl J Med 2009 Mar 5;360(10):961-972.
3. Boden WE, O’Rourke RA, Teo KK, et al. Optimal medical therapy with or without PCI for stable coronary disease. N Engl J Med 2007; Apr 12;356(15):1503-1516.
4. Sarno G, Decraemer I, Vanhoenacker PK, et al. On the appropriateness on noninvasive multidetector computed tomography coronary angiography to trigger coronary revascularization. J Am Coll Cardiol Intervent 2009; 2:550-557.