Interview with Nico H. J. Pijls, MD, PhD

The FAME study (Fractional Flow Reserve vs. Angiography for Multivessel Evaluation) published in the January 15, 2009 issue of the New England Journal of Medicine, concludes that "Routine measurement of FFR in patients with multivessel coronary artery disease who are undergoing PCI with drug-eluting stents significantly reduces the rate of the composite end point of death, nonfatal myocardial infarction, and repeat revascularization at 1 year."

FFR or Fractional Flow Reserve is accomplished by using a special guide wire with a sensor at the tip to measure the flow pressure inside the coronary artery. By measuring pressures at the proximal and distal ends of a lesion or blockage, FFR can determine if the lesion is in fact flow-limiting and causing ischemia to the heart muscle. In FAME, investigators stented only lesions that measured 80% or less. The control group, where lesions were assessed for stenting using the standard angiographic image, but without the information from FFR, showed significantly higher Major Adverse Cardiac Events (MACE): 18.3% vs. 13.2%, a 28% reduction.

Co-principal investigator for the FAME study is Dr. Nico Pijls, Professor of Cardiology at Catharina Hospital in Eindhoven, The Netherlands. Dr. Pijls has studied FFR extensively, and has led previous trials, such as DEFER to measure its efficacy.

Nico H. J. Pijls, MD, PhD
Catharina Hospital
Eindhoven, The Netherlands

Q: What is the take away message from the FAME study for interventional cardiologists?
Dr. Pijls: We have shown that if we perform PCI in a somewhat smarter way than we did before, in a way that's a bit more sophisticated, that we can improve the result, decrease the rate of complications and help more patients with it. And, if we really have a methodology that makes PCI better, that means more patients can be selected to be helped by it.

What kinds of patients? I believe that patients who are being treated medically, but still have ischemia, can be helped because it will be possible to discriminate which particular lesions are responsible for their complaints. On the other hand, if patients have multiple abnormalities, let's say four or so, and they need to be helped by bypass surgery, but let's say out of the four lesions, only two are responsible for the ischemia, then you could selectively stent them instead. Bottom line is that by using this methodology routinely, that it's possible to refine PCI, to make it to a safer procedure for more patients.

Q: You've previously stated that you think FFR should be an integral part of every interventional procedure. Do you use FFR on every case and are there cases where you don't really need FFR?
Dr. Pijls: If you have a patient who has single vessel disease and he has typical complaints of angina, has a positive exercise test, of course, then 1 + 1 +1 makes 3 and it's not necessary to do FFR. But that is only a minority. In our cath lab today at Catharina Hospital, 60% of all patients have multivessel disease which means that they have multiple abnormalities and it's often difficult to select on the basis of the clinical data and the angiogram which stenosis should be stented and which would better be left alone.

Q: Can’t this information be achieved non-invasively with nuclear stress testing, like SPECT? Is FFR more accurate?
Dr. Pijls: If you are looking at single vessel disease they are similarly accurate, because in single vessel disease, SPECT is a very good methodology. The problem is that, with multivessel disease, the reliability of SPECT non-invasive testing is much less. What you can show by non-invasive stress testing is that there is ischemia. But the non-invasive tests do not tell you which stenosis out of many is responsible for that ischemia. And this has been shown, and is stated by many studies, that if you apply SPECT in patients who have three-vessel disease, only in 30% of them will your test correctly identify the ischemic area. And in 70%, you will miss at least one of the ischemic areas. So, in patients with single-vessel disease, maybe SPECT is as reliable as FFR. But in multivessel disease, FFR is more specific -- because FFR typically gives the information per stenosis.

Q: Currently FFR certainly is not used routinely. What factors do you think will drive increased use of FFR?
Dr. Pijls: I hope that the most important factor to drive its use will be improved clinical outcome and a benefit for patients. That is the most important point. But you are correct in saying that FFR is used in only a minority of cases at the moment. In Holland it is about 15% of all cases. And in the States, it is even less, maybe 5%, and it is focused in a number of centers that believe in it.

If you are using FFR, you have more sophisticated decision-making. You are measuring something and you have to trust what you are measuring. Of course, this is a big step. If people look at an angiogram and they see a stenosis, their first intuitive reflex is, "Well, we have to put a stent there". And they do not realize that there can be many reasons why that is not as mandatory as it seems to be. If you randomly took 100 healthy people, 60 years old, from the street, and you made an angiogram in all of them, you would find that in 25 or 30 of them, there would be at least one significant stenosis by angiography. But the majority of these lesions are harmless for the people walking around with them. And that is, of course, the job of PCI: to select those lesions which are dangerous for the patient -- and by stenting these to be of benefit.

Q: Can you explain how a lesion that looks significant on standard angiography really isn't. What are the tricks that angiography plays on our eyes?

Dr. Pijls in the cath lab at Catharina Hospital

Dr. Pijls in the lab at Catharina Hospital

Dr. Pijls: The first point is that the angiogram is a two-dimensional projection of a three-dimensional structure. If lesions in coronary arteries were nicely circular, then the angiogram would be reliable. However, many of these coronary stenoses are slit-like. They have irregular shapes and therefore the angiogram can be deceiving.

A second point is that on the angiogram, you cannot distinguish what is the original size of the vessel. If there is diffuse disease which is diminishing the entire length of the vessel and you have a focal stenosis on it, you are relating the severity of the stenosis to what you see and consider as the normal size of the artery. But that isn’t necessarily normal at all.

A third point is that, if you have a very smooth stenosis, which is the same severity as an irregular stenosis, with an irregular surface, then the obstruction to blood flow by turbulence can be quite different.

So these are a number of angiographic technical shortcomings. There are many other reasons. If you have a particular stenosis and in an artery that is providing blood flow to a small area of the heart, of course, then it's less significant than in an artery that provides a large area. There is a kind a relation between a coronary artery and the myocardial territory to be supplied. And this information, this coupling, I would say, is not given by the angiogram, but it is given by the blood flow and by the pressure gradient.

Another example is that if you have an artery that is supplying the part of the myocardium which has a previous infarction and which in part has become fibrotic, the same stenosis which could be significant in one case is not significant anymore in the other case. And there can be collaterals supplying the myocardium, and they also make the particular stenosis less significant. So there are many technical and physiological reasons which make the relationship between anatomy and physiology not one-to-one.

Q: I have a question about patients who will be reading this. News reporters will be reporting on the FAME trial and I bet there will be headlines, for example, that proclaim, "Stents Don’t Work", or "Too Many Angioplasties are Being Done".
Dr. Pijls: I hope it will not be that way. I hope we can pick up on the bottom line that we can do something with PCI. It's very helpful, but we could do it better.

Patients should not be disquieted or scared and they should not lose their trust in interventional cardiology. But they should be aware that intervention can be made safer and more to the point by using FFR. Instead of getting three stents, as was the case in the angio group, two stents can be sufficient, even better, and make the procedure safer and cheaper. So I think that stenting in itself is a very good procedure, but by this new technique, we can very nicely refine it, and best define where we have to place the stent in the best possible way. So it's not a matter of stenting or not stenting. But it is a matter of placing the right stent in the right spot.

Q: Why would stenting a lesion that doesn't cause ischemia cause an increase in MACE? Obviously putting in 1/3 more stents raises the risk of restenosis and the need for a repeat PCI. But why would stenting a non-ischemic lesion cause an increase in heart attack and death?
Dr. Pijls: We know that if you have a non-ischemic stenosis and you treat the patient by drugs alone, the chance that patient will die, or have an infarction or heart attack due to that particular stenosis, is less than 1% per year. There are many good studies indicating this. Now if you place a stent in that lesion, you are creating a risk of let's say 3% of death or MI due to stent thrombosis and things like that. That is a problem related to stenting. So, for such a lesion, the benefit of stenting is less than the benefit of medical treatment.

On the other hand, if you have a stenosis that is creating repetitive ischemia, we also know from many studies that these lesions have a much higher risk for death or acute myocardial infarction. So the risk of these lesions causing a heart attack is much higher, and they are causing angina for the patient. So these lesions, which have natural risk of 5 to 10% of creating an infarction or death in the next year, can be reduced to 3%, using a stent.

So suppose you have four stenoses. And two of them are ischemic and two are not. And the two ischemic ones have an intrinsic risk of 5% and the other two of 1%, you can stent them all and you wind up with 12% risk (3% x 4 stents). But if you only stent the two ischemic ones, for those two you are reducing the risk without causing an increased risk to the other two (a total risk of 8%).

That is the conclusion of the FAME study, which shows that if you have an ischemic lesion, you are reducing the risk by stenting, because the intrinsic risk is higher than the risk of stenting. Whereas, if you have a non-ischemic lesion, the intrinsic risk of such a lesion is lower, so it doesn't really make sense to place a stent over there.

Q: There was a recent article in the New York Times stating that too many stents and too much angioplasty is being done; that cardiologists are looking at the wrong things because heart attacks and deaths are NOT caused by blockages but by milder-looking vulnerable plaques that rupture. But that's not exactly what you're saying.
Dr. Pijls: Yes. If you look at history for the last 40 years, what you find in every study and very convincingly, is that for the prognosis of patients with coronary artery disease, the major important factor is the presence and extent of inducible ischemia. That is always the bottom line. And whatever you think about minor plaques, generally they are less dangerous. As we know, they can cause a problem, but if there is the presence of ischemia, then there is a lot of danger. We have seen many studies with many thousands of patients showing that repeatedly.

Q: The FAME trial is sort of a sequel to the DEFER trial that you ran several years ago, and this kind of enhances the message even more. Do you think that the results of the FAME trial are going to change the way interventionalists approach their cases?
Dr. Pijls: Well I do not expect that everybody will be measuring FFR tomorrow, but I hope at least that people will think about it, and they'll be made aware that if we want to maintain PCI as a good treatment option for our patients, we should be sure that the underlying basis is clear. We should make sure that we do not overuse stents, but use them in the smartest way. I would say the word refinement is the best word to use here. If you look at the DEFER study, this was single vessel disease, so it was a matter of stenting or not stenting. And those were only intermediate lesions. In the FAME study, we had an average of almost three stenoses, which were on the average quite severe, more severe than, for example, in the RAVEL study or the SIRIUS study or the TAXUS study. What we have seen in the FAME study is that, at the end, 94% of all patients were stented, but they had less stents. They were stented in a more refined way.

Q: An editorial accompanied publication of the FAME study, written by Dr. Steve Ellis from the Cleveland Clinic. He thought the study was important, but he would have liked to know how the decision to stent in the angio-only cohort was made. He had other questions as well.
Dr. Pijls: The decision-making on what to stent was made in a classical traditional clinical way. We used the clinical data, the results of noninvasive testing, if they were available, the angiogram, and we only included a particular stenosis in the study if it was more than 50% and in a major coronary artery. So even in the angio group we were on the conservative side rather than on the aggressive side. Because I know that in many centers, even lesions of less than 50% are stented, just on the basis of the angiogram. So our selection of patients was completely according to normal routine. That was the purpose of the study: that it reflect the normal routine.

Dr. Ellis made a couple of more comments. He mentioned that the rate of peri-procedural myocardial infarction was high, but that was not the case. He said we don’t know the data and he suggested that it would be 7.5%, but it was actually 3.4% in the angio group and 2.4% in the FFR group. So it was not higher than in the other studies. And he mentioned a couple of other studies, like COURAGE and ISAR, but in those studies, patients with acute coronary syndromes were not included, and in ISAR, diabetes type 1 patients were not included, and the average number of lesions in those patients was only 1.3 or something like that. And in the FAME study, we had a lot of severe disease with an average of three vessels which were stenotic. We also had patients with very low ejection fractions, patients with acute coronary syndrome -- that was 30% of the population -- we had patients with previous PCI, so we had a population typical of the every day population we see in the cath lab.

During the procedure, we used also used thiopyridines, or Plavix: a loading dose of 600mg was used in all patients and IIb/IIIa inhibitors were used in 31% of the patients, which is also the average.

Q: Do you think that if the recommendations of this study were followed, less stents would be used?
Dr. Pijls: No. They would be redistributed. I think that per patient, less stents would be used. But more patients will qualify for PCI – because you have a treatment, and let’s say that you can decrease the risk of the treatment by 30% -- so the threshold to apply that treatment will become lower. What I believe is that you can do it better, by less stents per patient. And because you do it better, there are more patients, either from medical therapy who were previously treated only medically, or patients who need surgery at the moment -- all those patients can also be helped by PCI. So I think that in the end, the number of stents used will be the same, but they will be redistributed.

Q: FFR is done during a diagnostic angiogram. Is this technique hard to learn, is the equipment setup more complicated?
Dr. Pijls: It is really easy. I think that every interventionalist can learn it within seven days. If a particular interventionalist who has never used it does 20 consecutive cases, he will be an expert in it. The technique itself is no more difficult than introducing a guide wire. For the interventionalist, you have to learn a little bit about the hyperemic stimulus, and about the interpretation. There are, of course, some pitfalls. But altogether, it is one of the easiest techniques which we have in the cath lab.

(The FAME Study was supported by unrestricted research grants from Radi Medical Systems and Stichting Vrienden van het Hart Zuidoost Brabant. Medtronic provided limited financial support to some centers by tailoring the price of the Endeavor stents to the local reimbursement system.)

This interview was conducted in January 2009 by Burt Cohen of Angioplasty.Org.