ACC15 for Interventional Cards: TOTAL, MATRIX, DANAMI

David E. Kandzari, MD; Gilles Montalescot, MD, PhD


March 26, 2015

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

TOTAL Thrombectomy Trial: Totally Negative?

David E. Kandzari, MD: I'm Dr David Kandzari from Atlanta, Georgia, and with me is Dr Gilles Montalescot from Paris, France. Welcome to San Diego for the American College of Cardiology (ACC) Scientific Sessions.

We have just left the late-breaking clinical trial session on interventional studies. Gilles, you were a panelist, along with myself and other colleagues, and some very exciting trials were presented here and have already been published in some of our leading medical journals. We have talked about studies in thrombectomy, anticoagulant strategy, treatment of nonculprit lesions, and acute myocardial infarction, among others.

Let's begin with thrombectomy and the TOTAL trial.[1,2] Tell me your perspective on the results of this study.

Gilles Montalescot, MD, PhD: This study was awaited because of a meta-analysis[3] telling us beautiful things about thrombectomy—its effectiveness, and how we can reduce death with thrombectomy. The Swedish TASTE study[4] saw no benefit, but there was also no downside, so people have continued to use thrombectomy. Therefore, the TOTAL trial is very important. This is another negative study—a completely flat study involving more than 10,000 patients with ST-segment elevation myocardial infarction (STEMI).

Dr Kandzari: I agree with you. The previous studies included the modest-sized TAPAS trial[5] showing a mortality reduction of about 50% (almost unheard of), and the TASTE trial,[4] a much larger study, showing no difference. This was the tie-breaker in some ways.

It is very much the case that physicians still use thrombectomy because it facilitates the procedure or reduces periprocedural complications. We have seen in large databases and registries that 20% still use thrombectomy in percutaneous coronary intervention for acute coronary syndrome. But now we see the emergence of a potential safety hazard with stroke. What are your thoughts on that?

Dr Montalescot: It's not only acute strokes. We also see the curves diverging over 30 days, so there is an excess of stroke in this group. Is it a play of chance? Is there a good explanation? Is there more atrial fibrillation in this group? I don't know. We don't have the answer to that question, but we have to look at the data showing that there is an excess of stroke, which is not good.

Up until now, there was no downside to thrombectomy, but now we have one, which is a limiting factor for the use of thrombectomy. Even if there are some signals of benefit—ST resolution is better, stenting is easier, distal emboli occur less with thrombectomy—we saw all these benefits in the TOTAL study, but at the end of the day, there is no benefit for the patient.

Dr Kandzari: That is well stated. In this trial, the stroke rate was 0.7% early on with thrombectomy and 0.3% without. In the TASTE trial, in both arms it was 0.5%, so we are right around the margins, but with a sample size of 10,000 patients, you have statistical significance. If it's something that we associate with thrombectomy, then there is an issue. If it's a play of chance or an epiphenomenon, then perhaps that is a different story, and there might still be a role for thrombectomy for bailout examples. Is that right?

Dr Montalescot: Exactly. It's the same story as with glycoprotein IIb/IIIa inhibitors quite a few years ago. We tended to use them systematically in STEMIs, and now they are used primarily in bailout.

MATRIX: Bivalirudin Back, or Play of Chance?

Dr Kandzari: That is a great transition to the next topic, which is the use of anticoagulant therapy, and few people would be better than you to discuss this.

We have seen in the MATRIX study[6] now more than 7000 patients with acute coronary syndrome (non-STEMI and STEMI) who were randomly assigned to receive bivalirudin vs unfractionated heparin, with about 25% use of glycoprotein IIb/IIIa inhibitors. This study showed no overall reduction in net composite events, but of interest, a significant reduction in bleeding (especially fatal bleeding according to the Bleeding Academic Research Consortium [BARC] criteria) and also a reduction in mortality that we have seen in previous (but not all) studies. You have been involved in that. Tell me your thoughts.

Dr Montalescot: First, it is a double-negative study on the two primary endpoints, so we have to consider this. The major adverse cardiac events (MACE) endpoint was negative, and the net clinical benefit endpoint was negative, so everything else is hypothesis-generating.

I was not surprised by the slight excess of stent thrombosis with bivalirudin. I was not surprised by the reduction in bleeding. We know that the safety of this drug is better than what we achieve with unfractionated heparin (UFH) with or without glycoprotein IIb/IIIa inhibitors.

I was surprised by the reduction in deaths because we have two other studies, EUROMAX[7] and HEAT-PPCI,[8] which also studied a nonsystematic use of glycoprotein IIb/IIIa inhibitors with UFH and showed no signal on mortality.

So MATRIX is different, but is it true, or a play of chance? It's difficult to say.

Dr Kandzari: Do you think this study will change practice in Europe and in other geographies for the treatment of acute coronary syndrome? Will more bivalirudin be used, or will people still hold to whichever trial they prefer to follow?

Dr Montalescot: Given that bivalirudin is going to be generic very soon (in Europe), combined with these results, that might push physicians to use more bivalirudin than they have done before, so it's possible.

MATRIX: Radial Trumps Femoral in ACS

Dr Kandzari: The other component of this trial was a comparison of radial vs femoral access. This was a timely presentation because the TOTAL study, performed in multiple countries, showed that roughly 70% of the procedures were performed by radial access, and that was a thrombectomy trial. In MATRIX, the investigators compared the radial with the femoral approach in these patients, and once again showed significant reductions in access, bleeding complications, ischemic events, and perhaps even mortality. We have seen that in other trials too. What is your takeaway there?

Dr Montalescot: The data are quite clear now that radial is better than femoral to prevent bleeding, but also with an impact on overall mortality when you look at the global picture, with all the other studies. It makes sense.

We have to think twice before recommending systematic use of the radial approach. I am a radialist, as are my colleagues, but this study was conducted in patients with acute coronary syndrome and STEMIs. This is where you would expect the largest benefit, because you use a lot of antithrombotic medications at high doses, and you may have more bleeding. This is where we see the benefit of the radial approach.

If you were a physician working in a small center doing mostly elective cases, as a femoralist, what would be the benefit of moving to the radial approach? The benefit in stable patients may not be as big as what we see in patients with acute coronary syndrome. If you are in a small center, the benefit of the radial approach is not as significant as it is in a large centers. We shouldn't push them too far.

Dr Kandzari: I had many of the same thoughts during the presentation. As a radial operator myself, what we have seen in this trial and also in the previous RIVAL study[9] is that the benefit is greatest with the radial approach in reducing adverse events with high-volume, experienced operators. That said, being a radial operator also makes you a better femoral operator and vice versa, so there is a crossover of benefit, but there is more to be learned about the experience. This trial mandated that radial operators have fairly good experience. Recall that 25% of the patients in this study were women, and that is sometimes a more difficult group of patients to treat by radial access.

Dr Montalescot: If you want to use radials for STEMIs, you have to start with stable patients and elective cases.

DANAMI-3: Culprit vs Complete Revascularization

Dr Kandzari: You brought up STEMI, so let's transition to our last trial for discussion: the DANAMI-3 study.[10] The DANAMI-3 study randomly assigned more than 600 patients at two centers to either treatment of the culprit artery alone or, alternatively, treatment of the culprit artery plus either simultaneous or staged treatment of nonculprit arteries. These nonculprit arteries had fairly high criteria for significance. They either had to be 90% stenosed angiographically or have an abnormal fractional flow reserve (FFR) of less than 0.8. The study showed a reduction in the composite endpoint, but it was driven mostly by repeat revascularization. What are your thoughts there?

Dr Montalescot: As always, when you have such a difference on soft endpoints, you have to look very carefully at the control arm. The question I have—and I don't have the answer—is what about those patients in the control arm with 90% stenosis in a nonculprit artery who leave the hospital like that? What happened to those patients? Were they scheduled for an ischemic stress test a week or two later and they came back to the hospital for an angioplasty (and that was an endpoint), or did they wait for urgent revascularization? It was only 40% of the cases. It's a soft endpoint.

I think both strategies are possible. What happens in the control groups in these studies is not exactly how we manage patients in most institutions. In the patient with a nasty, tight lesion on a nonculprit artery, you are usually going to stent this lesion before hospital discharge.

Dr Kandzari: There are some analogies with the thrombectomy experience, when we had set-up trials leading to another study, and that has been the case here as well. Two previous studies[11,12] were performed, one suggesting a mortality benefit that is probably not reproducible. This study probably represented the refinement of these trials, in the sense that they were a little bit more restrictive about treating nonculprit lesions. They were staged, and not necessarily done at the time of primary percutaneous coronary intervention. Patients had to have an FFR value < 0.80 or a very tight stenosis, but it still leaves us with this unanswered question about how these patients are followed and treated, and cared for afterward.

Dr Montalescot: In an open-label study, bias is also possible.

Dr Kandzari: Thank you so much for your insights. I always learn more from you every time we are together.


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.