Transcatheter Aortic Valves vs Surgery: Closing the Gap

Susheel K Kodali, MD; Thomas Walther, MD, PhD


April 23, 2015

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

TAVR Durability

Susheel K Kodali, MD: My name is Susheel Kodali. I am here at the American College of Cardiology (ACC) Scientific Sessions. I am joined by Dr Thomas Walther, a cardiac surgeon from Germany. How was the meeting for you? We have heard exciting data on transcatheter aortic valves. What is your takeaway?

Thomas Walther, MD, PhD: The data were fascinating in the hotline session today. Congratulations on the great presentations given by you[1,2] and Mike Mack and Michael Reardon.[3,4] The data with transcatheter aortic-valve implantation (TAVI) are getting better and better.

Dr Kodali: Mike Mack presented 5-year data, and the 5-year durability is comparable with a surgical valve. The mortality rates were comparable; TAVI is noninferior. During that presentation, someone asked whether 5 years was long enough. What are your thoughts? In this patient population, are 5-year data enough?

Dr Walther: From the PARTNER-1 trial, I didn't expect much difference at 5 years because they were similar at 1 and 2 years. Five years is a long time, but in the surgeon community, we talk about long-term durability if you have 10-year data. We need to wait a bit longer, but of course you can expect that results will remain similar, because it's the same pericardium as with conventional valves; it's being crimped, but obviously that doesn't affect the longevity at 5 years.

Dr Kodali: The valve gradients and the valve areas remained consistent throughout the 5 years, which is encouraging, but we will have to wait and see. Maybe the 10-year data will show some differences.

Dr Walther: I was surprised to see the similarity in effective orifice areas and mean gradients with the conventional valves and the TAVI valves, because the stents of the conventional valves are slightly more obstructive than TAVI valves. In the other trials, TAVI valves have had superior hemodynamics, at least in terms of lower mean gradients and slightly larger orifice areas.

Dr Kodali: They did show that at 1 year, but after that the results have been comparable. There may be some loss. We don't know. Some of it is just the challenge of assessment. There is a lot of experience with measuring valve areas with surgical valves. With transcatheter valves, where do you put the Doppler? There have been debates about that. There are challenges.

Dr Walther: There are patient-prosthesis mismatches. Plenty of studies say that this affects survival and others say that it doesn't. It's a question of what you believe. The story will go on for a few years, and that keeps the situation interesting.

The Experience Factor

Dr Kodali: Exactly. Remember, as we look at the PARTNER data, this was the first experience. Half of the patients were done with the RetroFlex 1 device, which didn't have a nose cone. For most centers, this was their first experience [with TAVI]. This was in 2007. So the data are interesting but not really reflective of current practice because we do the procedures differently. We do CTs now. How do you factor that in when you think about the data?

Dr Walther: I believe that by better screening and imaging you will get better size selection of the devices and thus probably have better outcomes with newer data. On the other hand, in the PARTNER trial, owing to the fact that you had the first-generation systems, you had slightly better patients who had better access vessels. If you don't have much experience, then results will be worse. Results get better with more experience on the part of the cardiologists and surgeons on the heart teams that do the procedures. However, these cases were proctored, so this may balance it out. I would expect the results to improve over time.

CoreValve: Did Results Hold Up at 2 Years?

Dr Kodali: This is a nice segue to Mike Reardon's presentation[5] of the 2-year follow-up from the CoreValve pivotal trial. Last year David Adams presented the 1-year results,[6] and they showed a one-sided P-value but significant superiority for CoreValve over surgical aortic-valve replacement. With the 2-year data presented today, he mentioned that the curves continued to separate. This difference was maintained, and, if anything, it widened. Does that change the conversation now? Are we thinking that in this high-risk population (Society of Thoracic Surgeons [STS] risk score in the 7% range) rather than being an alternative therapy, perhaps a catheter valve is preferred?

Dr Walther: The data were very interesting. I was a bit surprised that the curves are spreading further apart. It's hard to explain. Michael couldn't really explain it. There were comorbidities in both groups. On the other hand, these patients had an average STS score of 7%. Basically, in my country, we have been treating these patients with TAVI for years already.

Dr Kodali: That is correct.

Dr Walther: If you look at the [German Aortic Valve Registry] GARY data,[7,8] the mean STS is even lower.

Dr Kodali: Do you think this is a population in whom the preference outside the United States (or at least in Germany) is for TAVR?

Dr Walther: Definitely. There is a divergence of the curves and a difference in outcomes between CoreValve and conventional surgery, whereas this was not observed in the PARTNER trial.

Paravalvular Leak and Other Complications

Dr Kodali: Part of that was a difference in time point. Michael raised an interesting point. He said that we see this in all these patients whether the procedure is done by surgery or with a catheter valve. If you avoid the acute complications, meaning the vascular complications, atrial fibrillation, patient-prosthesis mismatch, and paravalvular leak [PVL], the outcomes are great. So in PARTNER, for our initial results, the major vascular complication rate was 15%, and major bleeding was 15% to 20%. We didn't avoid those early complications because there was a learning curve. And we didn't screen the iliacs very well for access.

There were a lot of complications. In this patient population, once those complications occur, we had the early hit. We never caught up. The CoreValve trial was more contemporary. It was 2011 compared with 2007 for the PARTNER trial. In CoreValve, they did more preoperative screening. They obtained CTs on every patient. They had a slightly lower rate of these initial complications. In surgery, they had higher rates of bleeding, atrial fibrillation, and patient-prosthesis mismatch. There were fewer complications, so they had that early spread, and it was maintained, whereas in the PARTNER trial, we never got that early spread. That's my personal belief.

Dr Walther: That's reasonable.

Dr Kodali: However, we don't know. The likelihood is that there never will be a head-to-head comparison for these devices. Each device is compared with surgery, and that's the only thing we can say about it at this point. My personal belief is that the differences are more related to the time when these studies enrolled patients and how we did the procedure. How I do the procedure now is different from 2 years ago, and I expect that in 2 years it will be very different from now.

SAPIEN 3: On Par With Surgery

Dr Walther: Sure. The teams develop, which is good. You spoke about PVL. You presented very nice data[1,2] on the SAPIEN 3 experience in the United States. We have been using that valve for almost 2 years now in Europe, and we are very happy with it. We have a slightly different sizing strategy based on CT. We don't do too much oversizing. We have the outer skirt and we see low rates of significant PVLs. Do you believe that this is the perfect solution already, or should there be other devices in the future?

Dr Kodali: I don't think it's the perfect solution for every patient, but it's a tremendous leap forward. This was 1600 patients with core-lab echo. We had rates of 3.7% for moderate PVL and 0.1% for severe PVL. This is compared with moderate and severe PVL rates of 12% and 24% in earlier studies. That's a tremendous leap forward. We also are undersizing. Using the sizing algorithm in the trial (and at least in the latter half of the trial) we sized from -5% to -10%. There is a lot of undersizing. Even with the undersizing, you take reduce the risk of having a rupture much more. We still got great results with PVL.

The other thing with the SAPIEN 3 trial that I was surprised about is the stroke rates. I wanted to get your take on that. Disabling stroke rates in these cohorts were 0.9% in high-risk and 1% in intermediate-risk patients. I was shocked by that. What are your thoughts?

Dr Walther: It's probably explained by a pretty flexible system with the RetroFlex that you can nicely pass the arch and thus have less stroke risk. Good screening, of course, of the patients, is another factor. And in our practice we delete one step. We don't do balloon valvuloplasty. We inflate a balloon only once directly with the valve. We do direct stenting in every antegrade or retrograde implant with the SAPIEN 3. This may lead to a faster and better procedure, which probably implies less risk because you only dilate the annular area once. How do you handle that?

Dr Kodali: In the trial we had to do predilatation. A lot of people are doing undersized predilatation. One issue is that it's a smaller footprint. The valve itself is smaller, and we have seen the cerebral Doppler data. Most embolic events occur during crossing the valve or positioning the valve. If you have a smaller footprint, the crossing profile is mainly the balloon and not the exposed frame. Precise positioning and less manipulation of the valve account for the lower stroke rates.

Catching Up With Germany

Dr Walther: Sure. When you talk about those rates, this is still a selective population in the United States. That's why we put quite a bit of effort into looking at GARY data in Germany, which are registry data on more or less all comers. There we saw still some significant complications—up to 4% severe mitral complications, 4% technical complications, 1.5% stroke, and a death rate of 5%.

We have had some discussion in Germany on using TAVI without the heart team, with cardiologists only. The current data, despite all the optimism, are showing a very good direction toward lower event rates. We still need a heart team (cardiac surgeon and cardiology) to work together to have good outcomes.

Dr Kodali: I absolutely agree, and especially for the screening. We can all predict those patients—great femoral access, small left ventricle, very little left ventricular outflow tract [LVOT] calcium, nice-sized annulus—who are at low risk for complications.

But the patient who is at high risk with severe LVOT calcium and borderline access, there has to be a discussion. That patient may be better served by surgery. That's where a discussion plays a role.

I want to ask you about the mortality data. It's 2.2% in high risk, 1.1% in intermediate risk, and in the transthoracic arm (these are the best transthoracic data we have seen in a trial in the United States), mortality is 1.6%. What does this mean going forward? How do you interpret that? Does it change the conversation to say that TAVI is an alternative to surgery, or that it is the preferred therapy for some of these patients?

Dr Walther: It may be. If you look at GARY data, an STS of 5% was more or less your inclusion criterion in the intermediate-risk group. This is a patient who routinely goes to TAVI in Germany already. The mean STS in the GARY registry is 5%. Therefore, this study just supports what we have been doing clinically already for years.

Dr Kodali: Right.

Dr Walther: Once the mortalities drop further, TAVI is competing even in low-risk patients with conventional surgery, but the largest centers who have good experience have consistently low mortality rates for conventional surgery (below 1%); for example, as my center has had over the past 5 years.

Dr Kodali: That's the key. In the United States, it's a very different regulatory environment. Technically these patients receive surgery in our centers. Unless there is something that makes them high risk, we can't treat these patients with TAVR. It's different. We are seeing where things will go. In the United States, there is excitement that this may start the same conversation that already occurred in Germany 2 years ago. The point that you are at now—we are just starting to have that conversation. In 2 years, my feeling is maybe in the United States, we will be at that same point. To that 80-year-old with an STS score of 5% who says, "I don't want open heart surgery," we can offer TAVI.

Dr Walther: I'm very sure that you will get there. Congratulations on this nice randomized trial.

Dr Kodali: Same to you with all the GARY data, and thank you for having this conversation. You have great insights. And thank you all for joining us here at the ACC.


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.