Can you share some of the background of the Redo-TAVR Registry?

Uri Landes, MD: We initiated the registry 4 years ago and at this point, already have more than 700 transcatheter aortic valve replacement (TAVR)-in-TAVR cases, with almost 50 centers participating. We are continuing to collect data.

Our first report was published in the Journal of the American College of Cardiology in 2020¹, and showed that redo TAVR was fairly uncommon up until about 4 years ago, when roughly one for each 500 TAVR cases was a redo TAVR. We saw that those patients who had a redo TAVR had generally good outcomes and relatively few complications.

The second analysis we did was comparing TAVR in failed transcatheter heart valves (THVs) to TAVR in failed surgical valves using propensity matching.² We found that the outcomes are similar, except that THV in THV ends up with better hemodynamic outcomes compared to a THV in a surgical valve. This makes sense, given that surgical valves are frequently smaller and have a rigid swing ring. We can appreciate every day in the clinic that younger and younger patients are getting TAVR. A THV, as with any other bioprosthetic valve, has a limited lifetime expectancy. When patients are younger and their life expectancy will outlast the valve, then they are going to need some sort of solution. We already know that valve-in-valve is a decent solution for surgical bioprosthetic valves, but we did not fully realize that THV in THV is going to be a little different. There are some advantages in doing a THV in THV. One of them is a better hemodynamic outcome, because you don't have the rigid and restricted swing rings that surgical valves have, and also, THVs tend to be larger than the surgical valves. Thus, hemodynamic-wise, using a THV is quite an advantage, but there are also disadvantages relating to the leaflet position and the morphology. In addition, several THVs are available and are much more heterogeneous as compared to surgical valves in relation to height and frame design. Some have a waist, some don't have a waist, and so on.

Another issue is paravalvular leak (PVL), because although it is much less common these days, PVL is associated with mortality. We recently looked at the outcomes of PVL treatment using 3 different transcatheter strategies.³ We collected 201 patients: 87 had redo TAVR, 79 had plug closure, and 35 had balloon valvuloplasty, all done at the median of 7 months from the index procedure, to treat PVL. In about 10% of patients, even after redo TAVR, we found a significant residual gradient at 30 days. Regardless of the type of treatment, however, those with persistent moderate or greater paravalvular regurgitation had a higher mortality rate at one year.

When considering hemodynamics in redo TAVR according to valve type, you noted it is more vital to finish with a self-expanding valve than to start with a self-expanding valve. Other than that aspect, outcomes between the two types of valves were very similar. Can you talk about why you think that is the case?

Uri Landes, MD: It is important to say that there are numerous limitations to the registry and to any retrospective analysis in general. We try to correct much of the variables by statistical methods. We try to compare data such that the patients would be homogeneous with regard to valve size and type, but we couldn't compare everything, so there is still probably a bias. Having said that, outcomes were indeed very similar with regard to clinical outcomes and also hemodynamic outcomes, except the gradient. It is quite intuitive that the second valve is going to be more important hemodynamic-wise than the first valve, but still, it is important to measure.

Can you tell us about your analysis of valve deployment depth in redo TAVR, and your classification of conventional versus unconventional deployments?

Uri Landes, MD: This is our most novel analysis. This paper is under review and hasn't yet been published. As an operator, we don't have a clue what to do in terms of where to put the valves. All we have is our intuition and experience, but nobody can tell you exactly where to deploy the valve for redo TAVR. In surgical valves, there is the valve-in-valve application, the VinV (valve-in-valve aortic) app, and published experience, but for a redo valve, we don’t yet have the same body of data and experience. I think it is important to have a reference to what is currently happening, what operators have been doing over the last years. Then we can judge if it is good or not, and we can try to make it better. Of course, it is different if you deploy a short-frame THV inside a short-frame THV or a short-frame THV inside of a long-frame THV, and vice versa. So, as I noted, the key was to have a good reference for what people do with each of the 4 main valve combinations.

Then, in regard to your question about conventional versus unconventional deployment, the idea is to use the wisdom of the crowd, to see what most operators do. Let’s take the 50% that are within the middle range, and call it conventional, and take the 25% in the upper range and 25% lower range, and call these the extreme, the unconventional. Then, let’s see if there is a difference. We did find differences because there were more complications with unconventional deployments. Again, there are many limitations, and the numbers are very small. We are only talking about 5 cases that went wrong and made this difference. It makes sense in that, if you deploy a valve extremely low or extremely high, there are only two reasons to do so. Either you did it wrong and just had a bad deployment, or the patient anatomy forced you to do it because there was some concern for annular rupture or more likely, for coronary obstruction. It meant you had to deploy very low or very high as a result. Either way, whether it is a bad deployment or if it is the anatomy that is bad, both scenarios are bad for the patient. This is why it is worth emphasizing. When you do the index TAVR in a young patient, you should make sure that you will have enough space, not to only repeat the procedure if needed, but also to do it safely.

Do we know the best ways to handle difficult anatomy when placing the valve?

Uri Landes, MD: This issue is so complex that one paper or one registry is not going to resolve it. With Redo-TAVR Registry, we hope to add more data that can help. We are working on a Redo TAVR app together with Dr. Vinayak Bapat, and we are also cooperating with some CT modeling platforms that are already on the market and will be more available in this regard as well in the near future. It is case-by-case decision-making, and imaging is going to play a key role.

Right. There is a strong emphasis on operators becoming even more aware and involved with CT and echo imaging. Can you speak to what these types of imaging offer when doing pre-procedural planning?

Uri Landes, MD: You have to know the patient's anatomy and that is where CT plays a role. CT tells us the distance between the current index valve and the coronaries, the sinotubular junction, and, in general, the aorta, because the distance between the aorta and the valve dictates the risk for the amount of disturbance of the blood flow into the sinuses and into the coronaries.

The Redo TAVR Registry can be a useful reference because we now know that when people put a balloon-expandable valve inside a balloon-expandable valve, they usually do it 1 mm below — meaning the inlet of the redo Sapien (Edwards Lifesciences) is usually 1 mm below the failed valve inlet. We have learned that when you do a balloon-expandable valve in a self-expanding valve, usually people put the inlet of the redo valve 6 mm above the inlet of the failed valve. This is completely different. When you put a self-expanding valve inside a balloon-expandable valve, usually what people do is put it 3 mm below the failed valve inlet or even slightly deeper. These numbers are important to have, and then you can customize your deployment and make it suitable for your individual patient. I think interventionalists are going to have to know more and more imaging themselves, but they couldn't possibly know imaging well enough, so collaboration with the imaging specialist is key and is going to continue.  

In your analysis, with an unconventional THV placement, complications led to higher mortality for patients at 30 days. Can you talk more about that?

Uri Landes, MD: Generally speaking, our registry showed that redo TAVR is a safe procedure for those patients who have it. The complication rate was not that high, especially if it was done for valve degeneration (redo TAVR for acute TAVR failure is a different issue). If we are focusing on the current generation, redo TAVR is generally safe. We did see a few complications, and in our analysis, saw one case of coronary obstruction, one case of annular rupture, one case of conversion to surgery, and there was also cardiac tamponade and myocardial infarction. All of these complications happened, except one, in the unconventional deployment subgroup, translating to 8% mortality versus 0% percent mortality for conventional deployment. I don't think the deployment depth is what caused the difference in mortality, but rather the anatomy of the patient, which forced operators to go to extremes, and this didn't always work.

Any final thoughts?

Uri Landes, MD: I would like to say that I am very grateful for all the centers that collaborate with us, because we wouldn't be able to collect this data without a team effort, and many coauthors and friends take part in this effort.

TAVR is getting better all the time. We may be reaching a plateau level because the risk around TAVR is already very low. I think what will improve is valve durability. We are still going to see more and more redo TAVRs, but from generation to generation, the valves continue to get better.

References

1. Landes U, Webb JG, De Backer O, Sondergaard L, Abdel-Wahab M, Crusius L, Kim WK, Hamm C, Buzzatti N, Montorfano M, Ludwig S, Schofer N, Voigtlaender L, Guerrero M, El Sabbagh A, Rodés-Cabau J, Guimaraes L, Kornowski R, Codner P, Okuno T, Pilgrim T, Fiorina C, Colombo A, Mangieri A, Eltchaninoff H, Nombela-Franco L, Van Wiechen MPH, Van Mieghem NM, Tchétché D, Schoels WH, Kullmer M, Tamburino C, Sinning JM, Al-Kassou B, Perlman GY, Danenberg H, Ielasi A, Fraccaro C, Tarantini G, De Marco F, Witberg G, Redwood SR, Lisko JC, Babaliaros VC, Laine M, Nerla R, Castriota F, Finkelstein A, Loewenstein I, Eitan A, Jaffe R, Ruile P, Neumann FJ, Piazza N, Alosaimi H, Sievert H, Sievert K, Russo M, Andreas M, Bunc M, Latib A, Govdfrey R, Hildick-Smith D, Sathananthan J, Hensey M, Alkhodair A, Blanke P, Leipsic J, Wood DA, Nazif TM, Kodali S, Leon MB, Barbanti M. Repeat transcatheter aortic valve replacement for transcatheter prosthesis dysfunction. J Am Coll Cardiol. 2020 Apr 28; 75(16): 1882-1893. doi: 10.1016/j.jacc.2020.02.051

2. Landes U, Sathananthan J, Witberg G, De Backer O, Sondergaard L, Abdel-Wahab M, Holzhey D, Kim WK, Hamm C, Buzzatti N, Montorfano M, Ludwig S, Conradi L, Seiffert M, Guerrero M, El Sabbagh A, Rodés-Cabau J, Guimaraes L, Codner P, Okuno T, Pilgrim T, Fiorina C, Colombo A, Mangieri A, Eltchaninoff H, Nombela-Franco L, Van Wiechen MPH, Van Mieghem NM, Tchétché D, Schoels WH, Kullmer M, Tamburino C, Sinning JM, Al-Kassou B, Perlman GY, Danenberg H, Ielasi A, Fraccaro C, Tarantini G, De Marco F, Redwood SR, Lisko JC, Babaliaros VC, Laine M, Nerla R, Castriota F, Finkelstein A, Loewenstein I, Eitan A, Jaffe R, Ruile P, Neumann FJ, Piazza N, Alosaimi H, Sievert H, Sievert K, Russo M, Andreas M, Bunc M, Latib A, Godfrey R, Hildick-Smith D, Chuang MA, Blanke P, Leipsic J, Wood DA, Nazif TM, Kodali S, Barbanti M, Kornowski R, Leon MB, Webb JG. Transcatheter replacement of transcatheter versus surgically implanted aortic valve bioprostheses. J Am Coll Cardiol. 2021 Jan 5; 77(1): 1-14. doi: 10.1016/j.jacc.2020.10.053

3. Landes U, Hochstadt A, Manevich L, Webb JG, Sathananthan J, Sievert H, Piayda K, Leon MB, Nazif TM, Blusztein D, Hildick-Smith D, Pavitt C, Thiele H, Abdel-Wahab M, Van Mieghem NM, Adrichem R, Sondergaard L, De Backer O, Makkar RR, Koren O, Pilgrim T, Okuno T, Kornowski R, Codner P, Finkelstein A, Loewenstein I, Barbash I, Sharon A, De Marco F, Montorfano M, Buzzatti N, Latib A, Scotti A, Kim WK, Hamm C, Franco LN, Mangieri A, Schoels WH, Barbanti M, Bunc M, Akodad M, Rubinshtein R, Danenberg H. Treatment of late paravalvular regurgitation after transcatheter aortic valve implantation: prognostic implications. Eur Heart J. 2023 Apr 17; 44(15): 1331-1339. doi: 10.1093/eurheartj/ehad146

Find more: 
TVT Newsroom (With a Special Focus on TAVR)

Find Dr. Landes' slides here.