Revascularization With PCI or CABG in the Post-SYNTAX Era

Robert Widmer, MD, PhD; Lyle Joyce, MD, PhD; David Holmes, Jr, MD


April 24, 2017

Editorial Collaboration

Medscape &

This feature requires the newest version of Flash. You can download it here.

Jay Widmer, MD, PhD: Howdy, I'm Jay Widmer, interventional cardiology fellow at Mayo Clinic Rochester. During today's recording, we'll be discussing percutaneous coronary intervention (PCI) vs coronary artery bypass grafting (CABG) in the post-SYNTAX era. I'm joined by two mentors: Dr David Holmes, a pioneer in interventional cardiology, and Dr Lyle Joyce, cardiovascular surgery, here at Mayo. Welcome to you both.

We'll start with you, Dr Joyce. Based on two of the recent trials and the SYNTAX trial, how can there be such striking differences in terms of the results and the reporting of such well-designed trials?

Reconciling Results of EXCEL, NOBLE, and SYNTAX Trials

Lyle Joyce, MD, PhD: We need to consider several facts related to the design of the trials and also their timing. The SYNTAX trial,[1] for example, completed its enrollment in 2007, whereas the EXCEL and NOBLE trials are much more recent.[2,3]

In that period of time, there has been a marked improvement in PCI and stent technology. So, to compare technology from 10 years earlier with what we have now is likely going to give varying results. There are also some differences in the definitions—for example, what constitutes a myocardial infarction (MI). Also some differences in what groups of patients were studied; the EXCEL trial focuses more on the left main with low-risk SYNTAX scores compared with the others.

Dr Widmer: Exactly, and regarding that left main subpopulation, Dr Holmes, who do you think is a candidate for left main coronary PCI based on the results of these trials?

David Holmes, Jr, MD: That's a really important question. It depends on the specific metrics you discuss with the patient. Those patients who are optimally suited for PCI of left main disease would still be those with ostial lesion and trunk lesion. Distal bifurcation left main disease is clearly treatable [with PCI], although it is much more complex, as you know. Those PCI results are not as good as with more isolated left main disease. Isolated left main disease—in the absence of other complicating factors of coronary artery disease—is very well treated with PCI as well as with surgery.

There is a difference between [PCI and surgery]. One important end point is that the data on surgery, except for the NOBLE trial, indicate that stroke [risk] is increased with surgery as compared with PCI. Now, the difference is statistically significant, but it is small.

The next piece of information is that as the lesions get more complex, the SYNTAX trial indicated that more extensive disease is better treated with coronary bypass graft surgery. I do not think the latest trials have changed that or moved the needle. Those patients who have extensive multivessel disease that cannot be treated with PCI and patients who have LV [left ventricular] dysfunction are still better served by coronary surgery in terms of the overall composite end point.

Incorporating New Data into Practice

Dr Widmer: Dr Joyce, in terms of the newer trials, have you changed your practice based on NOBLE and EXCEL vs what we saw in SYNTAX?

Dr Joyce: No, I think our practice has continued pretty much the same. If you consider the details of the EXCEL trial, for example, it focused on patients with a low SYNTAX score. The results of the SYNTAX trial in the low-SYNTAX-core group are not that much different from the EXCEL trial.

Dr Holmes: There is an interesting issue. We have relied on the concept that arterial revascularization would be better than anything. We have reasoned that if there is one good artery, it is a [left internal mammary artery] LIMA. If you were to add a second artery, the [right internal mammary artery] RIMA, then the results would be much better than a LIMA plus vein grafts.

So, it was surprising to see the literature on the recent trial documenting no difference between one mammary and two mammaries.[4] We always said the reason that SYNTAX surgeons did not do as well as [predicted] was that, most of the time, they just used a single mammary. The same was true in NOBLE and EXCEL. We thought, if you had used two mammaries, they would do much better. Yet, the ART trial showed no difference if you use one vs two. Can we talk about that?

Dr Joyce: We have to go back to some of the original studies looking at what keeps vessels open. There is no question that a graft to the left anterior descending (LAD) coronary artery has a higher long-term patency rate than any of the other circumflex or the right coronaries. So, to expect a RIMA to the right or to the circumflex to be equal to the LIMA to the LAD probably is not justified. Part of it is the runoff and the makeup of the vessel and the disease of the vessel. We certainly believe that all arterial grafting is the way to go.

Dr Widmer: Do you think that we are not quite at the end point for that? We were looking at 5-year results for the recent trial that you noted. At 10 years, do you think we will see any difference?

Dr Joyce: Certainly, it has been shown that the LIMA to the LAD has a striking difference in ≥15-year patency rates. So I think you are right. If we get out past the perioperative period the first year and look down the road 10 and 15 years, we probably will see an improvement in bilateral mammaries and perhaps all arterial. The radial is a little less dependable than the right anterior mammary artery.

Dr Holmes: One of the problems, as you have talked about, is by the time we look at 10 and 15 years, all the technology has changed. That was seen in SYNTAX. In SYNTAX, for example, we used a drug-eluting stent that is no longer manufactured, but we did not have a better stent when the SYNTAX trial was started. The interpretation of long-term data is always complicated by that.

Surgical techniques and interventional techniques also improve—does that mean that what we should focus on concepts of complete revascularization or concepts of using an arterial graft irrespective of whether it is single or dual mammaries or radials vs using percutaneous techniques that treat only a small segment of the artery? There are a lot of different questions to address.

Dr Joyce: There are. In fact, in the SYNTAX trial, one problem with PCI was the inability to completely revascularize. Actually, a subgroup focusing on the residual SYNTAX score found that if it was over 8, then the prognosis was worse.[5]

So yes, I think those elements are important, but you are absolutely right—it is very difficult to compare 10-year data from a stent that is not even around anymore. It is perhaps a bit more reliable [to compare] surgical technique, because the only variables there at the present time that we have really made are: 1) whether we use all arterial or not and 2) and whether it is on- or off-pump. Otherwise, our techniques are pretty similar to what was done 10 to 15 years ago.

Using the SYNTAX Score in Practice

Dr Widmer: I would like to talk about some of the technologies in a little bit, but returning to the residual SYNTAX score, we saw that even in EXCEL, the on-site calculated or estimated SYNTAX score was much lower than the core lab. It is difficult to score, and sometimes it is done on the fly. Do you think that the SYNTAX score is something that we can integrate into our practice easily? What are some ways we can use that, or should we be looking somewhere else?

Dr Holmes: It is terribly important that the practice of interventional cardiology has changed and continues to change. The concept of PCI at the time of diagnostic angiography was very popular with patients, but the professional societies have backed away from that, and I think that that is important.

In fact, Europe has led the way in eliminating ad hoc PCI. If you have a patient with elective symptoms whom you are considering for elective revascularization (the patient has relatively stable angina), if they have multivessel disease, by mandate, they are taken off the table and then the heart team is brought together to review the case. In that system, then you can calculate a SYNTAX score. It is not like you have to do it on the fly; you have time to do that.

Now, there was some pushback from patients. They would rather not come back to the laboratory. They would rather not have two procedures. However, it is such an important decision that I think that the heart team to truly invoke that and bring that to play is critical. Lyle, you can talk about that.

Dr Joyce: I agree. It is complex and difficult for the patient to understand, but the most important point for the patient to know is that there is a heart team, comprising both surgeons and cardiologists, looking at it. That assures patients that when both groups have looked at it and decided, on the basis of SYNTAX or other elements that are involved, that the recommendation is a combined one. The surgeon, for example, is going to look more at, for instance, any comorbidities such as chronic obstructive pulmonary disease (COPD) or renal failure. The cardiologist will probably look more at features of the anatomy—how much calcification is, chronicity of occlusion—that make a technical difference. It is important that the patient knows that all of those things have been factored into the recommendation that was provided to him [or her].

Making Patient-Centered Decisions

Dr Holmes: We talk about patient-centered care. We have the heart team, and then we have the patient as part of that. As we talk with patients, it is very hard for them to keep in mind the issue that there is a difference in that there may be a hierarchical end point of stroke vs repeat revascularization vs death vs myocardial infarction.

There are some patients, when you talk to them and say the survival at 5 years is the same, and that is true in most of the trials; but there is a difference in repeat revascularization and a difference in myocardial infarction (MI) and a difference in stroke. The patient then typically says, "Well, tell me about stroke, because that is the thing that I would fear the most." While you can present [facts] to the patient, the patient's attitude winds up coloring the decision. How do you approach that?

Dr Joyce: When I discuss it with a patient, I try to get them to focus on two differences: 1) the long-term benefit vs 2) the short-term benefit. I explain to them that, yes, when you [pursue the surgical route in treating] stroke and MI, then you are taking more of the risk up front. Whereas if you are looking at the long-term outcome, then you may consider, "Okay, is it more important for me to get over this hump and then know that I am not going to have to worry about anything for a longer period of time, or is it more important to me to just have the least invasive approach taken right now and I will worry about that later?" And it is interesting—patients are different, aren't they?

Dr Holmes: Extremely.

Dr Joyce: One patient is adamant that there is no way I am ever coming back: "You better do whatever you are going to do that is going to fix me for the rest of my life." And the next one says, "If I do not have to go to surgery, just fix me up as simply as you can and we will worry about that later."

Dr Holmes: That [former] patient, having been in that circumstance of taking short-term option tends to say, "Well in 5 years, I will have been exercising like crazy. I will have stopped smoking, and my blood pressure is going to be perfect." Having said that, most of the time they do nothing, because in 5 years, we are still about as heavy as we are now.

Dr Widmer: You bring up a good point about optimal medical therapy—smoking cessation, exercise, and the like; and you coauthored a paper with Iqbal and colleagues about optimal medical therapy in this SYNTAX trial.[6] Curious if you have thoughts if optimal medical therapy in some of these newer trials will or could have an impact on the outcome, and how we weigh that into some of our patient-centered decisions.

Dr Holmes: The concept of optimal medical therapy is incredibly easy to say. But the implementation is incredibly difficult because it involves multiple challenges. It entails the economics of taking pills for a long time. It involves the issue of willpower, which, as we know, lasts for 30 days and is soluble in alcohol, and then there are side effects for drugs.

When we put patients on optimal medical therapy that includes losing weight, exercise, stopping smoking, and taking medications every day, at 5 years, the number of patients who are able to do that is really limited. It would be just like saying the Ornish diet is a "great" diet. The number of people who follow it do incredibly well, but it's small number. Optimal medical therapy is a great goal, terribly important, but we do not have very good strategies for that.

Maybe you have better strategies than the surgical ones.

Dr Joyce: Unfortunately, I am disappointed in our surgical colleagues in the SYNTAX trial. The failure to continue with optimal medical treatment was far too high in that surgical arm. We know it is beneficial. There is no question about it if they are on at least a single platelet inhibitor and a statin and a beta-blocker and an angiotensin-converting enzyme inhibitor (ACE) inhibitor. But you are right, it just is not followed as completely as it needs to be, even knowing that the outcomes are so much better.

Dr Holmes: One of the problems is that patients come in with a heart problem, and they think you are going to save their life. Then you revascularize them with a surgery or PCI. At that point in time, it is very hard to get the message across that we have not eliminated the need for medications and the procedural benefits are not going to last forever.

Looking Ahead in Stent and Surgical Technologies

Dr Widmer: Looking down the road 5 to 10 years, in terms of stent technology, surgical technologies, what [advancements] might turn the corner for some of these or be game changers? Dr Holmes?

Dr Holmes: For some of the new medications, for example, the [proprotein convertase subtilisin/kexin type 9] PCSK9 [inhibitors], compliance may not be as big a problem because it could be [taken] once a month or once every 2 weeks. Or it could be a vaccine. If indeed that works, that will be a huge game changer. Now nobody knows whether that is going to work or not, but that is the sort of thing that could 1) improve compliance if you can afford it, and 2) may be incredibly effective in reaching target levels—whatever those are.

In terms of the technical approach to revascularization, I think we are continuing, from a percutaneous standpoint, to make iterative changes. We are continuing to make the stents thinner with thinner struts and better in terms of being able to resist the compressive forces, but those are iterative changes. I do not see a home run as yet.

Dr Joyce: I agree. Many of the advancements will be in the medical realm. From a surgical standpoint—particularly if there continues to be equipoise between the two technologies—we will swing toward hybrid approaches. The best treatment for any patient is probably an internal mammary down to the LAD. Maybe some of those other lesions, particularly if the technology improves, will be better treated in the cath lab. And then we can offer a truly minimally invasive approach surgically to get the important lesion taken care of in the LAD and finish it up in the cath lab.

Dr Holmes: Tell us about the future, of a mammary artery that could be "on the shelf" using biotech genetic engineering. You could then, in the operating room, have three mammary arteries and put them in every place and minimally invasively. Is that reality? Is that coming? Is it feasible? Because that would be a true game changer.

Dr Joyce: I think that is the future. My concern is whether it will always be the future. We have tried so many ways to engineer artificial vessels, and none of them have panned out as of yet. But that is the direction that we need pursue from a surgical standpoint. If we could do that, it would be so much better for the patient.

Dr Widmer: Dr Holmes, Dr Joyce, thank you both so much for agreeing to do this. Thank you all for joining us here on on Medscape for these very important insights.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.