TCT: Structural Heart Disease and PFO Closure Trials

; Deepak L. Bhatt, MD, MPH


October 19, 2015

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

Robert A. Harrington, MD: Hi, I'm Bob Harrington from Stanford University, here with the Transcatheter Cardiovascular Therapeutics (TCT) conference in San Francisco. I'm joined by my good friend and colleague, Deepak Bhatt, from Brigham and Women's Hospital and Harvard Medical School. Welcome, Deepak.

Deepak L. Bhatt, MD, MPH: Thanks, Bob.

Men With Hammers, Looking for Nails?

Dr Harrington: Deepak, this has been a great week at TCT—as always, a lot of new information, a lot of in-depth discussions around some of the emerging topics in interventional cardiology. There are three things I want to discuss today.

I want to spend some time talking about aortic disease. Now, you'll notice I didn't say "transcatheter aortic valve replacement" (TAVR). I said "aortic disease" because of an interesting registry presented today from Japan on the topic of patients with asymptomatic but severe aortic stenosis.

From there, I want to get into some of the work presented on valve-in-valve procedures and some of the extensions of technology. I want you to think about this question: Are we seeing the maturation of a technology that's moving into new, very justifiable indications, or are we seeing men with hammers looking for more nails to hit?

I will move into another maturing technology, but this time on the drug side, and that's a discussion of bivalirudin. There was an interesting trial on patients randomized to receive bivalirudin vs unfractionated heparin as the anticoagulant during TAVR.

And then the third topic is the conundrum of cryptogenic stroke, and some new data here that maybe aren't so new.

Dr Bhatt: Three good topics.

Dr Harrington: With a little controversy around each of them. Let's start with aortic disease. There are a couple things that caught my attention this week.

First off is that we continue to see expansion into areas that were different from the original PARTNER trial[1] of high-risk patients with aortic disease. I was very interested in the valve-in-valve presentation,[2] replacing the degenerating bioprosthetic valves percutaneously and what that means. I was interested in some of the longer-term follow-ups in intermediate-risk patients presented by Howie Herrmann (SAPIEN 3).[3]

And then the interesting observation—and I'll stress the word "observation"—from Japan (CURRENT AS),[4] that patients with severe but asymptomatic aortic stenosis appear to do quite poorly if you watch them and wait until they develop more indications to do a valve replacement vs replacing the valve initially.

Dr Bhatt: These were all interesting presentations. TAVR, as a field, has been very thoughtful in the United States and even worldwide, but particularly in the United States in conjunction with the US Food and Drug Administration (FDA). I don't think it's a matter of interventionists with a hammer saying, "You know, there's a case; I'm going to do it."

Dr Harrington: I mostly agree with you, and it's also demonstrated a terrific partnership among the regulators, the industry, and the academic community.

Dr Bhatt: That's right—and the professional societies, the American College of Cardiology (ACC) and Society of Thoracic Surgeons (STS). TAVR was done correctly, and that's not to say that everything we've done in interventional cardiology is right.

We're going to come back to patent foramen ovale (PFO) closure later. In that case, there was a proliferation of procedure before data.

Dr Harrington: Right.

Dr Bhatt: But here with TAVR—in the United States in particular, it really was a matter of waiting for the randomized data before jumping.

Dr Harrington: Let me ask you specifically: What do you think of the valve-in-valve? That strikes me as a fairly complex procedure, but I don't do the procedure—you do. Tell me about the valve-in-valve procedure.

Dr Bhatt: For degenerated biological valves (surgical valves), valve-in-valve is essentially the standard of care. Those patients are obviously high-risk for repeat surgery. We're talking a minimum of a redo sternotomy, and at a minimum, the patient is older than when they got the valve initially. In that particular situation, TAVR is definitely the way to go.

Dr Harrington: Would you say that, Deepak, even for a young woman with bicuspid disease who wants to get through her child-bearing years without a mechanical valve and the obligatory need for anticoagulation, who is now coming back beyond her child-bearing age with valve degeneration? For such a patient in her 40s or 50s, would you still do TAVR?

Dr Bhatt: Now you're getting into some tricky areas. I started off with a generality—

Dr Harrington: Because a lot of older people also got bioprosthetic valves to avoid the anticoagulation.

Dr Bhatt: Yes, and it's an easier decision when they're older. But in the younger patient, it's a bit trickier. It really depends. There's been a movement even beyond that young woman who is at child-bearing age to perhaps put bioprostheses in people in whom we used to put in mechanical valves. The thinking is, "Well in 10, 15, or 20 years, we'll just do a TAVR. By then, TAVR is certainly going to be even better than it is now." And that's logical, but that's certainly not a proven train of thought, and we still need to learn more about the durability of TAVR.

We know it has terrific durability out to 5 years. But what happens out to 20 years? On a bench top, we know from simulations that they seem very durable out to millions of cycles of contraction, but how about in a human being? I don't know that I would necessarily rush to put in TAVRs before age 50 years.

The other somewhat related issue is the recent case series of filling defects seen on 4D CT and transesophageal echocardiography (TEE). No gradient was associated with them, so it's not clear whether there is any clinical significance to them.

Dr Harrington: This is the New England Journal of Medicine paper[5] that was just published recently before the TCT. You're one of the coauthors.

Dr Bhatt: That's right. I don't think we need to be alarmist. I don't think this affects the current indications for TAVR, which are very strong and evidence-based. And we know in patients who aren't operative candidates that TAVR is life-saving, and in ones who are high-risk that it's at least as good as surgery, if not better.

SAPIEN 3: TAVR for Intermediate Risk

Dr Harrington: And what are you thinking now about, in light of the Howie Herman data [SAPIEN 3]—what are you thinking now about the intermediate-risk patients?

Dr Bhatt: Well, there were longer-term data on intermediate-risk patients, and I think the indications for TAVR are continuing to expand. It's interesting: In Europe, I spent a month in Paris with Alain Cribier, and there was no one over 80 years of age getting conventional surgery anymore. And of course a lot of octogenerians and nonagenarians are high-risk [for surgery], but not all of them are. A portion of those patients, if they were in the United States, would in fact get surgery.

I think it's very reasonable to go ahead and do TAVR in the octogenarian and above, if they need surgery. Of course, in the United States, there are reimbursement and labeling issues, but in a theoretical world or outside the United States, that's what I would do.

You mentioned the Japanese registry.[4]

Dr Harrington: Yes, let's talk about that. It had 1800 or so asymptomatic patients; 250 or so had early surgery, what they call the "initial strategy." This was not randomized. And the others were followed as per guidelines, replacing their valves when gradients got to a certain point or when they developed symptoms. They did really poorly.

Dr Bhatt: We have a dilemma in cardiology, and in medicine in general, and that is what to do with asymptomatic people. If somebody's got a bad asymptomatic lesion, regardless of what it is, our tendency in medicine is to forge ahead. But with asymptomatic aortic stenosis, because those are often older patients with lots of comorbidities, many times we say, "Let's just leave well enough alone." But this study strongly suggests—-it doesn't prove in a randomized fashion, but it strongly suggests that the wait-and-see strategy means you're going to die or end up in the hospital with heart failure.

Dr Harrington: If you remember the data that plot the natural history of aortic stenosis—they're from a paper by Dr Braunwald,[6] and there is a figure from many years ago which essentially has patients doing okay, doing okay, and then you develop symptoms and you drop off.

Dr Bhatt: Right.

Dr Harrington: We all looked at that and have ingrained in our minds that if patients develop angina, heart failure, syncope—uh-oh, you've got to start to deal with that. The Japanese registry is suggesting that maybe we should rethink that and move it up a little bit, particularly now that the potential procedure is not so morbid.

Dr Bhatt: I agree with you. In that patient population, who are older, probably the wait-and-see strategy is not the way to go. Now, whether that should be moved into younger age ranges requires some thought. A proportion of those folks have bicuspid valve stenosis, so that raises another set of issues. Of course, TAVR can be done in that valve too, but it's a bit more of an elliptical annulus; it's not quite as circular.

Dr Harrington: There are some technical issues that we have to think about.

Dr Bhatt: Yes. Not so much in an older person, but in a younger person, I think we still need more data.

Dr Harrington: You had mentioned a few moments ago about the issues of reimbursement, and that the collaboration among the government, industry, the professional societies, and the clinical academic community has been a very good one. Where are we on the reimbursement for valve-in-valve?

Dr Bhatt: You get reimbursed for that, but for TAVR in general, the reimbursement is really a case of "it depends." It depends on the patient, the insurer, the hospital. It's not often a money-making proposition—not that medicine needs to be, but many times TAVR is viewed as a source of losing revenue for a heart and vascular center.

Dr Harrington: And now that you run a heart and vascular center, you've got to think about that.

Dr Bhatt: It's tough to do the right thing when there are reimbursement consequences that can be in the tens of thousands of dollars. It's one thing losing a hundred bucks on a procedure; I don't think anybody is going to question that if it's a life-saving or quality-of-life–improving procedure. But when it is a potentially $10,000, $20,000, or $30,000 loss.

There are apps available. In fact, a fellow from Beth Israel developed a very nice app[7] [to calculate the reimbursement]. And it depends on the length of stay. It also depends on where you might discharge the patient—whether you discharge them to home, or with visiting nursing services (there can actually be a penalty for using those sorts of services).

Dr Harrington: This is another example of why the heart team is not only important around the management of decisions for the patient, but everyone at the center has to be in tune with this, because there are some significant consequences that have to be worked out.

Dr Bhatt: Moving forward in interventional cardiology and other aspects of medicine, it's something we need to think as we get newer technologies. If the reimbursement penalizes hospitals for doing things that are labeled, that are evidence-based; that's not a good situation.

BRAVO 3: Bivalirudin for TAVR

Dr Harrington: I'll go back to your earlier comment that this alignment of all the interested parties has worked very well in structural heart disease.

Let's move to pharmacology. You and I spent a lot of time over the past 25 years working on clinical trials dealing with antithrombotic therapy, largely in patients with coronary disease. Now we see a randomized trial[8] at this meeting of the direct thrombin inhibitor bivalirudin against unfractionated heparin as the anticoagulant used during TAVR. Sounded like a good idea, didn't it?

Dr Bhatt: It sure did, and the data for percutaneous coronary intervention (PCI) are reasonably consistent that bivalirudin reduces bleeding, no question, if heparin plus a glycoprotein IIb/IIIa inhibitor is the comparator. I also think it reduces bleeding against heparin alone, at least in patients who are high bleeding risk, but that in part depends on the activated clotting time (ACT), the proportion or radial vs femoral procedures...

Dr Harrington: It's complicated.

Dr Bhatt: Yeah, you can't make quite as blanket a statement for bivalirudin vs heparin alone, and there are trials that are ongoing.

Dr Harrington: The Swedish group is doing one of their registry-based trials. Registry-based trials are a whole other topic that you and I have talked about before.

Dr Bhatt: You've got to give them credit for using their registry to move the field forward, and at this meeting, Gregg Stone also announced a HORIZONS ABSORB registry trial, which is going to look at a few different things—

Dr Harrington: Including bivalirudin infusion post-procedure vs during procedure only. That's an area that's gotten a lot of excitement lately because of the notion that you can attenuate some of that potential early hazard of bivalirudin by giving a longer infusion. But this study in patients undergoing TAVR was really different, right? It was not combined with antiplatelet therapy on a routine basis.

Dr Bhatt: No; this was heparin vs bivalirudin for TAVR, and I don't know too many centers that are actually using bivalirudin right now for TAVR, other than in patients with heparin-induced thrombocytopenia (HIT).

Dr Harrington: But with the size of the sheaths you guys use for the procedure, you'd think it might. As you move to radial [PCI], you saw some attenuation of the benefit of bivalirudin.

Dr Bhatt: Right.

Dr Harrington: Just because the bleeding rates overall went down. But—what is it—what did you tell me before we came on the air, 18 F?

Dr Bhatt: Yes. Of course, the manufacturer sheath sizes are often billed as smaller than they really are, and we’re talking about most often bilateral femoral axis with large-bore sheaths in elderly patients, lots of renal failure, lots of calcified vessels. The bane of TAVR right now is vascular complications.

Strokes have become much less frequent, though there's work there to be done as well, I think. But in general, the results are pretty good in the indicated patient groups.

Dr Harrington: And this study saw no benefit of bivalirudin, although the point estimate went in the right direction. So the question is (as always when that happens): Is there just no difference? That's what you'd have to conclude now, but as you know, the type 2 error is the lack of an observed difference is not the same as the lack of a difference.

Dr Bhatt: Absolutely.

Dr Harrington: Was the trial just too small?

Dr Bhatt: Yeah, that's the part that isn't definitive, so right now you can't say to use bivalirudin in TAVR.

Dr Harrington: Right, and the presenters very clearly said that; they said this means that unfractionated heparin has to remain the choice. But they did do a formal noninferiority test that if you have to use bivalirudin for an indication, HIT for example, then maybe you would use it.

Dr Bhatt: Absolutely, it's a good option. If there weren't a price differential, then I would say, "Sure, use bivalirudin. Why not? It's a cleaner anticoagulant." You don't have to worry about HIT.

Dr Harrington: You don't have to measure ACTs or anything.

Dr Bhatt: It has a lot of appeal. But given the price differential, even though it's generic now, that price has come down a bit, but it's not at heparin levels. Probably, it's not going to gain a lot of traction.

There's still that intriguing question that maybe if it had been a larger trial, would we have seen something significant? Of course the counter-argument that some people make is that if the effect size is relatively modest, is it worth that cost?

If you're willing to get into a secondary endpoint, there actually was a significant reduction in early myocardial infarction (MI) in the bivalirudin vs heparin group. In fairness, there was also an increase in early renal events, and neither of those things held up at 30 days. But it's sort of intriguing to think maybe a more potent direct thrombin inhibitor prevents some thrombus from forming, and maybe prevents some embolic MI.

Dr Harrington: But still hypothesis-generating, more work to do.

Dr Bhatt: A secondary endpoint that didn't even hold up at 30 days, and if you're going to believe the good MI finding, you should also believe the renal failure finding.


Dr Harrington: You and I have been involved with trials where a secondary analyses looked pretty intriguing; you have to be cautious, and you have to think about it as hypothesis-generating.

Deepak, we have about 3 minutes left. I'm not sure we can do this justice, but let's dive into cryptogenic stroke. This is an area that is really different from how TAVR developed. The thinking was that if you had a stroke that was otherwise unexplained and there was a PFO, if you put in an occluded device and close that hole, you've got to get better. However, those data have not been good.

Dr Bhatt: The data have certainly not been conclusive. We saw longer-term data at TCT.[9]

Dr Harrington: Talk about the long-term data [RESPECT trial]. What do they do?

Dr Bhatt: Well, it's a complex analysis. There have been intention-to-treat analyses that have been negative.

Dr Harrington: Flat-out nothing.

Dr Bhatt: If you're a purist, statistically speaking. The on-treatment or as-treated analyses have looked better, and then trying to subsegment the type of stroke when it seems to be a purely cryptogenic stroke (not due to atrial fibrillation or some identifiable cause), there seemed to be a reduction in patients who actually got the device. But that's going a couple of layers away from a pure intention-to-treat analysis.

Now in fairness, it's kind of tough with devices to totally do intention-to-treat, because if you do surgery vs meds and half of the patients in the surgery arm don't get the surgery, is intention-to-treat a fair comparison? Maybe it's okay to do on-treatment.

There was also some crossover in the trial, where patients in the medical arm ended up getting closed. But nevertheless, having said all that, it's a relatively small number of events—important events. I'm not saying that a young person with a stroke or a current stroke is not a big deal; it is. But statistically speaking, in order to tease out the data and do post hoc analyses, it's not a lot of events.

I think that there's an intriguing signal. Those in the PFO closure community aren't going to be happy, but I really think it needs a definitive trial before we say, close that PFO.

And an interesting nugget buried in there was that there was an increase in deep vein thrombosis (DVT) and pulmonary embolism (PE). I haven't seen the details of the data—when they're published, we will—but I think what that probably is, is that a proportion of the cryptogenic stroke patients actually have occult thrombus somewhere. Maybe pelvic thrombus; maybe if you had done an MRI, you would have seen it (I'm not saying to do all these expensive tests).

So yes, the PFO was playing a role, but it was actually that thrombus down here that crossed through that PFO that caused the paradoxical stroke, and if you had anticoagulated that patient, then you would take care of both their risk for stroke and their DVT or PE.

Dr Harrington: We have not seen all the details. The control arm here is guideline-directed medical therapy. So, in some patients, that guideline-directed medical therapy includes antithrombotic therapy.

Dr Bhatt: Absolutely.

Dr Harrington: So the question is, do we see differential antithrombotic use? It reminded me of trials that were done years ago comparing rate vs rhythm control, and one of the issues is that if you targeted rate control, you tended to use more anticoagulant therapy, whereas if you targeted rhythm control, you felt safe and you discontinued it, and that might in part have explained the stroke difference. Maybe we're seeing a thrombosis issue here. We've got to see more data.

Dr Bhatt: Yes, and with that potential negative finding, we can't be cavalier and say okay, it looks good in this post hoc subgroup analysis, so let's keep closing these PFOs. If it were totally clean otherwise in terms of safety, then it's just an issue of cost and procedural risk.

Dr Harrington: We're running out of time, but I'd love to have you on another time, where we talk about this notion of when do you really advocate for the wider dissemination of a technology vs the more stepwise use of a technology like we've seen in TAVR. Cryptogenic stroke and PFO closure is one of these examples where the field got way ahead of the data.

Dr Bhatt: Absolutely, and it puts us in a tough position as an interventional or larger medical community, where we're complaining about not getting reimbursed for this or that (as I just complained to you about TAVR)—but then on the same hand, we're just doing lots of other procedures where the data are weak. I understand physicians' desire to do something when they see a young patient [with unexplained stroke], but I think we need to be thoughtful as we were with TAVR.

If everybody comes together—the regulators, the societies (it would have been better if the third-party payers also came together a bit more in the case of TAVR)—then I think it is reasonable to introduce a new technology widely. Without all those steps in place, I think we sometimes shoot ourselves in the foot.

Dr Harrington: Well it's the old discussion that you and I have both had on rounds with our fellows and residents: Is the sin of commission better than the sin of omission? And in our society, people want to do something. As you said: [We think if it's a] young person, we've got to do something. But doing something might not be the right thing to do.

Dr Bhatt: Absolutely, and it might just bankrupt the healthcare system.

Dr Harrington: Well, that's another discussion. I want to thank you, Deepak, for joining me here at TCT. We had some good discussions about aortic disease, anticoagulation during TAVR, and then finally cryptogenic stroke.

Dr Bhatt: Thanks for having me, Bob.

Dr Harrington: And thanks to the audience for listening; we hope you found this informative.


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.