Jan 27, 2020
Dr Carolyn Lam: Welcome to Circulation On the Run, your weekly podcast summary and backstage pass to the journal and its editors. I'm Dr Carolyn Lam, Associate Editor from the National Heart Center and Duke National University of Singapore.
Dr Greg Hundley: I'm Dr Greg Hundley, also Associate Editor, the Director of the Pauley Heart Center at VCU Health in Richmond, Virginia.
Dr Carolyn Lam: Say, Greg, you know the feature paper this week talks about the perennially hot topic now and that is transcatheter aortic valve replacement or TAVR or TAVI. It's actually data from the France TAVI Registry comparing balloon expandable versus self-expanding transcatheter aortic valve replacement.
I'm sure you want to hear more about it, but first I'm going to tell you about another paper in the same issue, this time also comparing a balloon expandable versus a self-expanding transcatheter aortic valve implantation, but data from a nationwide analysis and from corresponding author Dr Fauchier from Centre Hospitalier Universitaire Trousseau. He and his colleagues basically did a head to head comparison of the two competing transcatheter aortic valve replacement technologies that have been published but have not really been followed for long-term clinical outcomes. This was comparing balloon expandable versus self-expanding technology.
They collected information from more than 31,000 consecutive patients treated with Tavern in France between 2014 and 2018 and based this on the French administrative hospital discharge database. They did propensity score matching, which was used for the analysis of outcomes according to the Sapien 3 balloon expandable versus the Evolut R self-expanding TAVR technology and studied this as nationwide level in France.
Dr Greg Hundley: Wow. Carolyn, 31,000 patients. That's a really large study. What did they find?
Dr Carolyn Lam: They basically found that balloon expandable TAVR was associated with lower mortality rehospitalization heart failure and pacemaker implantation compared with the self-expanding TAVR. Now, that's of course a pretty big finding and this is discussed along with the feature paper that we're going to hear about in an editorial by Drs. Abdel-Wahab and Thiele from Heart Center Leipzig.
I want to tell you about another paper before I let you tell you about yours, okay?
Dr Greg Hundley: Sounds great, Carolyn.
Dr Carolyn Lam: Greg, what is your clinical impression of Impella use in the United States among patients undergoing PCI? Do you think it's increasing or decreasing over time? As a reminder, Impella was approved for mechanical circulatory support in 2008, so from then, what do you think?
Dr Greg Hundley: You know, Carolyn, I really think it's increasing, especially used more frequently rather than an intra-aortic balloon pump. How about you? What's going on in your area of the world?
Dr Carolyn Lam: My impression too, but you know, you're lucky because we now have data looking at the trends in Impella use, but in the United States, and this comes from the corresponding author, Dr Amit Amin from Washington University School of Medicine and colleagues who describe clinical outcomes and costs across U.S. hospitals in PCI patients treated with mechanical circulatory support, which is either the Impella or the intra-aortic balloon pump.
They found that among more than 48,300 real world patients undergoing PCI with mechanical circulatory support at 432 hospitals between 2004 and 2016 in the Premier Healthcare Database, Impella use was indeed found to be rapidly increasing with marked variability across hospitals and not only its use, but also in its associated adverse outcomes. When analyzed by time periods or at the level of the hospitals or at the level of the patients, Impella use was associated with higher rates of adverse events and higher hospital costs.
Dr Greg Hundley: You know, I wasn't thinking about the higher rate of adverse events. You wonder sometimes, are we using a technology in a sicker group of patients? Did this study shine any light on that?
Dr Carolyn Lam: Those are great, great thoughts. The authors concluded that the variability in Impella use, the variability in its associated outcomes, and the association of Impella use with higher adverse events and costs really, really underscore the need for better defining of the appropriate use of mechanical circulatory devices and that was what you indicated as well, Greg, and what we need there is adequately powered randomized clinical trials and prospective real world evidence, which we don't quite have yet. Until then, perhaps a more measured approach is needed in clinical practice that balances risks versus benefits in complex patients undergoing PCI who require mechanical circulatory support.
Dr Greg Hundley: That's going to be really needed, I think in this era, especially with the results from this study. Well, Carolyn, I'm going to switch over to the world of basic science and the first study I'm going to talk about is from Dr Richard Lee from Harvard University and it's a very interesting study. Just as some background, current differentiation protocols to produce cardiomyocytes from human induced pluripotent stem cells are capable of generating highly pure cardiomyocyte populations, but these cardiomyocytes remain immature and they really more closely resemble the fetal state.
As a result, they have a lower maximum contractile force, slower upstroke velocity, and immature mitochondrial function compared to adult cardiomyocytes. Also, they're prone to ventricular arrhythmias. During development, cardiomyocytes undergo a shift from a proliferative state in the fetus to a more mature but quiescent state after birth. The mechanistic target of Rapamycin mTOR signaling pathway plays a key role in nutrient sensing and growth, and Dr Lee and colleagues hypothesized that transient inhibition of the mTOR signaling pathway could lead cardiomyocytes to a quiescent state and enhance cardiomyocyte maturation.
Dr Carolyn Lam: Wow Greg, I really love the way you explained that. That's so interesting. What did they find?
Dr Greg Hundley: Among human induced pluripotent stem cell lines, transient treatment with Torin 1, an inhibitor of the mTOR pathway, shifted cells to a quiescent state and enhanced their cardiomyocyte maturity. Also, the investigative team suggests that further testing will be necessary to evaluate whether delivery of Torin 1 treated cardiomyocytes could reduce the risk of ventricular arrhythmias in newly differentiated myocytes derived from pluripotent stem cells. Really an important advance in this whole area of developing mature cardiomyocytes from our own pluripotent stem cells.
Well, Carolyn, my second basic science paper comes from Dr Calum MacRae from Brigham And Women's Hospital, also at the Harvard Medical School. Carolyn, this study used both highly purified human pluripotent stem cell derived cardiomyocytes displaying physiological and molecular characteristics of atrial cells with human MYL4 mutations in a zebrafish MYL4 knockout model, which exhibited molecular, cellular, and physiologic abnormalities that parallel those in humans bearing the cognate mutations associated with definitive genetic causes of atrial fibrillation.
Dr Carolyn Lam: Oh, that's really interesting. Is this new genetic predispositions that they discovered?
Dr Greg Hundley: I think the answer's yes. They found there was evidence of increased retinoic acid signaling in both human pluripotent stem cell derived cardiomyocytes and zebrafish mutant models, as well as abnormal expression and localization of cytoskeletal proteins and loss of intracellular NAD and NADH, and thereby established a mechanistic link between the transcriptional, metabolic, and electrical pathways previously implicated in the atrial fibrillation substrate of MYL4. In the future, these data could lead to novel therapies for some patients with atrial fibrillation.
Dr Carolyn Lam: Wow. That really is fascinating, Greg. Well, let me round up by telling you about some of the other things in the issue. There is a research letter by Dr Parish on the effects of Omega-3 fatty acid supplements on arrhythmias and here, these authors reported more comprehensively on atrial fibrillation and other arrhythmias using additional data extracted from linked electronic health records in the ASCEND trial, remember, which was 1 gram of Omega-3 fatty acid supplementation daily in people with diabetes but without known atherosclerotic cardiovascular disease.
Dr Greg Hundley: Oh wow. That's fantastic, Carolyn. I've got a couple other really interesting articles in the issue. First there's an In Depth review from Dr Yvan Devaux from Luxembourg Institute of Health, and he discusses regulatory RNAs in heart failure. In a perspective piece, Dr Alejandro Lucia from Universidad European de Madrid discusses the role of aerobic and resistance training as a therapy in addition to prescribed medications in patients with resistant hypertension. Really interesting.
Then finally, Dr Ify Mordi from University of Dundee examines metformin use and clinical outcomes among patients with diabetes, with or without heart failure, kidney dysfunction observations from the SAVOR-TIMI 53 trial in which Dr Bergmark and colleagues found that metformin use was associated with a reduction in all-cause mortality and cardiovascular death, but not due to myocardial infarction or stroke, particularly in patients without a prior history of heart failure.
What could the mechanism be, if not related to presumed atherosclerosis? Dr Mordi and colleagues proposed possibilities, and Dr Brian Bergmark from the TIMI study group and the cardiovascular division of Brigham and Women's hospital at Harvard Medical School and colleagues, they write a very nice response. It's really interesting listening to how could metformin reduce events but not related to atherosclerosis? How about onto our feature article?
Dr Carolyn Lam: You bet.
Dr Amit Khera: This is Amit Khera, digital strategies editor for Circulation from UT Southwestern Medical Center joined by my colleague, Dr Dharam Kumbhani who's also an associated editor at Circulation and we're pleased to have Dr Eric Van Belle, Professor Van Belle, from Lille University Hospital to discuss the featured article today, "Transcatheter Aortic Valve Replacement Propensity Match Comparison From the France TAVI Registry." Welcome to you both.
Dr Van Belle, I'm going to start with you and we always like to hear a little bit. Perhaps you can tell us some of the background, what led up to this investigation, what led to your group pursuing this manuscript?
Dr Eric Van Belle: Nowadays, the TAVI procedure, the TAVR procedure, is becoming very prominent kind of way of treating patients without stenosis, and basically we have two different type of devices that are available to treat the patients. Once series is based on the balloon expandable concept and the other one on the self-expandable concept. These two type of devices are considered to be used primarily in every kind of patient. Theoretically, we can use any of these two devices, any kind of patient, if we follow the recommendation of the manufacturer and I'll just say that that'd been done.
These two devices are being validated against surgery, so basically, we could potentially use any kind of them. In today, there is no direct and there was no direct comparison between the two different kinds of concept, although they are very different. Again, the device is different. The way we implant the device is different. The major question that we had behind was to say, okay, what is the outcome? If it's a mean patient get one of the devices or can we expect or should we expect a different outcome? That was the main question behind it.
Dr Amit Khera: Okay, so essentially there's two valves, they're both being used fairly regularly and without any kind of direct comparisons. Tell us a little bit about the study design and what you found in this project.
Dr Eric Van Belle: For methodology, we used what is called a French study registry, basically nationwide registry with almost all patients treated in France included, and we used it as a database of patients between 2013 and 2015 with an overall group of 12,000 patients treated with either of these two kinds of devices. This is one of the aspect of this registry. The other very important aspect of this registry, and that's the mortality data survival that was obtained in all the patients in 2008 through 2016, so we have a set of 12,000 patients. It was a cool kind of device with complete mortality data by April 2016 so basically, this is the main methodological aspect. On top of this, we did the best to do some matching on the older clinical variables and all the matching valuables that we had to create pairs of patients that could be matched to one to one. We had, at the end, a group of almost 4,000 patients.
Dr Amit Khera: Okay and tell us a little bit about some of your main findings of this study.
Dr Eric Van Belle: The two main findings were those differences between the two groups of patients, that is a patient treated with self-expandable devices at a higher risk of valvular regurgitation. This was mainly a confirmation because this finding was already reported previously in previous studies trying to compare the two devices, but what was more striking was the difference in mortality. It was a difference mostly in hospital mortality but also in mortality after two years. That was significant with an absolute difference in mortality around 3% by two years.
Dr Amit Khera: Well, obviously important, as you mentioned that paravalvular leak had been seen before and this now a long-term mortality difference. Certainly an important finding and one of the main findings of your study. One of the concerns about comparative effectiveness research, essentially you're using observational data such as this is that there still could be residual confounding. There still may be patient characteristics or decisions made by interventionalists that aren't fully accounted for. How did you all really try to account for some of these components, this residual confounding to try to get the best answer that you could?
Dr Eric Van Belle: That's going to be a major comment, and everything you can do, every best way to try to control for this, there is no better answer than to do a randomized study, and probably we'll discuss on this. Let's see, indeed, we try to do our best to minimize as much as we could, all these potential confounders, so we did it in a different way, indeed.
The first way to do it was to adjust all the potential differences among group but what was very also interesting to remind is that, when you look at the 25 clinical and imaging variables and creating the aortic annulus diameter that was incorporated in the matching, that actually 21 of the 25 variables were already there. We were balanced between the two groups, existing that indeed most of the case, the operators, we are not so much directing or at least if there was selecting it was not captured but all of these valuables because again, out of 25, the correction needed to do the matching was only affecting 4 variables, mainly. Those variables were already pretty well-matched between the two populations.
The other way we did it was to look at what is called falsification endpoint, that it is endpoints that are supposed to be unrelated to the devices to verify that indeed, we have not selected a population that will have issues that are not related to the device itself. We look at, let's say, mortality by infection, mortality by cancer, to verify that, indeed, this kind of event where it did well balanced between the two groups suggesting that the mortality effects that we observed was not related to this kind of unbalance related to something else that was not captured by analysis.
Dr Amit Khera: Yeah, I think that was quite an important observation you mentioned. The first that these two groups are generally well-balanced to begin with, even before with all the matching parameters and then certainly the falsification endpoint helped to add validity to the findings.
Dharam, I'm going to turn it over to you. Maybe you can show this from an associate editor's perspective. What are some of the observations you found interesting about this study and what are some of the considerations we had in some of the discussions about it?
Dr Dharam Kumbhani: I'll just remind our listeners that this was also a late breaker at AHA last year in 2019, so this is really a very important finding. As Eric briefly pointed out, there haven't really been head to head comparisons between, the two dominant valves in the market even though TAVR has pretty much become the dominant strategy for treatment of aortic stenosis.
At the end of the day, it's an observational analysis. We have to take the findings with that in mind. At a minimum, it's first a lot of debate and discussion about the need to have randomized trials and our belief that, perhaps, TAVR is a class effect may not always be true. I think that hypothesis would certainly need to be tested and that's what this paper really sparks as far as discussion going forward.
Dr Amit Khera: Maybe I'll ask both of you. One of the challenges of any type of observational research is time period. First there was a hint towards even maybe a greater effect in the more recent time period than the distant time period. Also, there's always commonly changes in technology, especially interventional field where a study comes out and it's already obsolete because there's some new technology. There are some newer generations of valves that have come out. Do you think that it would affect these findings in any way? Maybe we'll start with you, Eric.
Dr Eric Van Belle: That's always an issue. Again, because it's a very rapidly evolving field and if you want to have strong data, you need to have really long-term follow-up. You need to have mortality data. There is some kind of contradiction between both that the field is evolving very quickly but then to have solid data, you need to have some time.
What we could say, indeed, as a study period was 2013 and 2015, but the device that we are using at that time were already really well matured and also the devices that were used at that time was usually the ones that were used for the comparison with the surgical techniques. Again, these devices are not so much obsolete since they were accepted and used again, when you need this one device study to compare with surgery.
Of course these devices have still had some evolution and change, and it is for the good of the patient, but again, as mentioned there, I'm seeing what is very, very important is that this finding is, in my view, intriguing enough to say, okay, even if it's difficult to conduct this kind of randomized study, it has to be done now because we need to really know. Let's say 80% of the patients could indeed be treated with any of the two device in this large margin of patients. Do we have to choose one or the other one to start with? This has already been well answered in a larger randomized trial.
Dr Amit Khera: Dharam, maybe I'll ask you, do you think this large randomized trial, are you optimistic that that would happen? Certainly it sounds like it's something that would be very helpful for the field. What are your thoughts on whether that's actually going to occur?
Dr Dharam Kumbhani: I know that there are some head to head trials ongoing. I don't know if they will have the sample size to really drill down, as far as hard endpoints, mortality, for example. I think the field clearly needs it. The question is, who's going to sponsor a trial like that? There's probably not much incentive for industry to sponsor something like that. Really it would fall down to whether there's a way for government agencies to partner with industry or other ways to run this. I do agree with Eric that that's really very important and hopefully we'll see that in the field going forward.
I did want to comment on the next iteration of devices as far as what we may see now. The mortality signal, I know we've talked about it. It's an observation study. It's hard to know if there's confounding, and even with all the sophisticated statistical analyses that the team did, there's always a possibility that somehow there was sicker patients that received self-expanding valves.
The signal for paravalvular regurgitation is not just in this study. We've observed it in many other studies and for other self-expanding platforms as well. Both the SCOPE trial and the St. Jude trial last year, both came around the same time. They were self-expanding platforms and both of them showed a higher paravalvular regurgitation rate compared to the balloon expandable rate. That may be a real thing, and I don't know if that is an inherent design flaw in the self-expanding platform or if there are ways that that could be mitigated going forward. Again, I think the trials, for it to be meaningful, it would be obviously important to collect and have short term and imaging markers. Really, what the field needs is long-term evaluation of these two strategies.
Dr Amit Khera: I want to take both Dharam Kumbhani and Dr Eric Van Belle l from Lille University Hospital. Thank you both for joining today.
Dr Greg Hundley: This program is copyright, the American Heart Association 2020.