Hi, everyone. I guess we'll get started. We have a great presentation. I'm going to take a minute and say how completely thrilled I am to be on the stage with Dr. Arnold. In full disclosure, I believe she is the coolest thing about TVT right now. So I'm Joan Michaels. I call myself the queen of TVT. I don't know what we're going to call her after the end of this or she'll never speak to me again, but if you've ever heard me talk before, you know my thing is KCCQ. People make fun of me, say it'll be on my tombstone. And when I present, I've been known to use, in the spirit of enticing folks to really collect and complete KCCQ, perhaps I've used Tony Soprano, Walter White to say do it or die. It has to be done. But I have the real deal up here with me today, and she is, I think, going to be amazing. We've been talking to her as much as we can for the past couple of days about KCCQ and how important it is. So let me just say what I know about Dr. Suzanne Arnold. There's just one bad thing about her. I didn't realize she went to Ohio State, and I hate Ohio State because I'm Notre Dame. Somebody should have told me that. I thought it was bad enough she was from Kansas City because I hate the Kansas City Chiefs. But I'm going to suck it up and power through. She studied at Ohio State, we'll overlook that, and found her way to St. Louis, and then with Kansas City with the St. Luke's Mahi Mid-America Heart Institute, which the ACC and NCDR collaborates with a lot. Perhaps some of you know Dr. John Spertus who also is very supportive and active and really developed the KCCQ. So we're going to talk about KCCQ today. We're going to talk about how important it is. And Dr. Arnold will explain not only for TAVR, but if any of you were with us yesterday talking about the continued journey that we're getting into with Tricuspid that we are in the middle of with Mitral, and how we think KCCQ will continue to be a real pivotal strong point to make sure we do. So in addition to being, unfortunately, a really poor choice of sports teams, Dr. Arnold has authored or been involved in over 200 trials, if that's correct. And I think over 60 of them had something to do with KCCQ. So she knows a little bit about the material. And one other thing in saying how, I mean, I'm starstruck being next to her. I got to tell you this. And I said, wow, Kansas City, wow, you're just as good as being a Swifty here. I really feel like I'm in, you know, and I really think she thinks I'm a little weird. But if you support me, let's welcome Dr. Arnold to the podium. So obviously I've been tasked to talk about the KCCQ, which I am obviously very passionate about and hopefully is not too incredibly dry. But I'm going to talk about kind of what is it? Why is it important? How to interpret it? How to potentially use it clinically? What are some of the challenges we have in analysis of the KCCQ? And then also at the end I will touch a little bit about its role in the TPT composite endpoint and how it fits there. So while I don't have any industry disclosures, I do have funding from the FDA to look at the validity of the KCCQ in patients with tricuspid regurgitation and also some funding from the NIH and NHLBI looking at health status in mitral valve surgeries. So first and foremost, health status is a meaningful outcome. Patients care about it. If you look at mortality, which obviously being alive or dead is really important, these are data from Partner 3 which was TAVR versus surgery in patients low-risk. And you see these are the curves for mortality. But what's important is, what happened to the other 98% of patients? Did they feel well? Were they symptomatic? And this is where the concept of health status can really be useful to understand, are there differences between the treatments? How are the patients actually doing? We use the term health status and quality of life sometimes synonymously, but they are actually a little bit different. A lot of times when we're talking about health status, it is specific to a one-disease process or disease-specific health status, which the KCCQ is one of those measures. So conceptually, you have a disease, whether that's valve disease or heart failure, that disease then causes symptoms. Those symptoms cause functional limitations, whether that's the ability to socialize with friends and family, or being able to walk, or carrying groceries. Those functional limitations then impact a patient's quality of life, which I like to think about as the discrepancy between what a patient thinks they can't, what they want to do, and what they can do. And symptoms, functional limitations, and quality of life are really what encompass health status. Again, patients care about this. I don't think this is surprising. But the degree to which they care about this versus other outcomes I think is a bit enlightening. So this was just one study that looked at this, 51 patients with HF-Ref, mean EF was 64%. And what they asked them to do was to rank a number of hypothetical treatments based on their impact on outcomes. So for example, you would have this particular new treatment. You would not be tired, but you would still be short of breath, depressed, but you'd get a three-month extension or increase in the length of life. And based on all of these kind of new hypothetical treatments, they found that 65% placed more importance on symptoms and quality of life as opposed to prolonging survival. The other reason that health status is important to measure is that it may actually be the only outcome that differs between two treatments. These are data from Partner 2A, which was surgery versus TAVR in intermediate-risk patients. And if you look at death or disabling stroke, there was no difference between the two treatments. But there was an early health status benefit in the TAVR patients. And we all know this. We know the patients with TAVR, they recover quicker. And patients care about that. And if you had not measured health status, you wouldn't really be able to quantify the benefit of this treatment. The FDA has recognized that a treatment may demonstrate effectiveness based on improvement in health status alone in terms of being approved. So point number two, kind of moving to the KCQ specifically, how do we interpret this? So just a little background on the KCQ. It was developed by John Spertus in Kansas City. The original version had 23 questions with five domains, physical limitations, symptoms, social limitations, self-efficacy, which is how does a patient feel that they have the ability to manage their own heart failure? And then quality of life. And the overall summary score, which is what's typically used as the primary outcome in a lot of these trials, is the average of the physical limitations, symptoms, social limitations and quality of life domains. The shortened version, which is what we're most familiar with and used in the TBT Registry is 12 questions. This was developed in 2015 and was originally designed more to use as a clinical tool. So something that could be implemented in clinics, but also is I think very useful in registries where time is limited and we don't want to fatigue our patients with a lot of questions. So this is 12 questions in three domains, physical limitations, symptoms, and quality of life. And the overall summary score is the average of those domains. Scores for all domains and summary scores range from 0-100 and higher is better. So how do we interpret the actual scores? We've used NYHA as kind of a surrogate measure. NYHA is obviously not exactly what the KCQ is measuring. But just as a rough estimate, when we think about scores of the KCQ 45 or below, that's kind of NYHA Class 4. 45-60 is Class 3. 60-75, Class 2. And above 75 is Class 1. To put that into perspective of what do we see in these studies, these are just an average baseline for different TAVR M-Tier and T-Tier studies. This is the Triluminate Single-Arm Study, just to be aware. And we see that obviously in Partner 1, which was the high-risk or actually the inoperable patients, their KCQ at baseline was incredibly low. These were very sick, symptomatic patients. Whereas as we move more into the intermediate or low-risk patients, those go up. But you kind of see a lot of these were starting in that 40-60 range. In terms of how do you interpret changes? So we've generally considered 5 points to be a small clinical change. 10 points is kind of moderate. And 20 is large. Obviously there's space in between. 15 is kind of a moderate to large, that type of thing. From a prognostic perspective, every 5-point increase in the KCQ is associated with a 9% relative decrease in all-cause mortality and an 11% decrease in cardiovascular death or rehospitalization. And just to kind of again put that into perspective of what do we see in some of these clinical trials or in the registries, generally after any of these transcatheter valve interventions we're seeing very large changes by one month, especially in the TBT where it's not controlled against other treatments. So I just want to show a little bit again about what do we expect to see and what do we know about the patterns of KCQ improvement in these transcatheter valve studies. This was the original extreme risk. So you see a large improvement at one month with a little bit of continued improvement at six months and roughly stable between 6 and 12. Moving to the high-risk, those patients generally started a little bit better and had more of their health status improvement within one month. This is just real-world data. Partner 2, and then finally the low-risk where these are patients that generally had a fairly isolated valve issue. So they didn't have a lot of other comorbidities. They didn't have a lot of other heart failure issues. So their KCQ started higher and saw almost all of the health status benefit within one month. If we extend that out, we have data now up to five years in the extreme and high-risk. You see that most of this is relatively stable over time, although there's kind of a small decrement in the quality of life when you get past two years. Looking at the MTR, one of the differences here compared with TAVR is that almost all of the health status benefit is realized within that first month. Limited is still a place where we're learning. These are all single-arm studies. So we don't have a lot, well, there's some randomized data and more upcoming. But what we have seen at least in the single-arm studies is that a lot of the improvement is again within one month and not quite at the level that was seen with MTR or with TAVR, but still moderate to large improvements. Just one note about tricuspid disease in the setting with the KCQ. So the KCQ was originally developed in patients with HFRAF. Those are left-sided disease. They have a lot of pulmonary symptoms, pulmonary edema, dyspnea. And there's been a question about whether or not the KCQ is appropriate to be used in tricuspid patients who have more right-sided symptoms. So more congestive symptoms, edema, fatigue, and things like that. So it has been validated in a number of left-sided disease processes, so HFPAS, aortic stenosis, hypertrophic cardiomyopathy. But it has not yet been validated in tricuspid disease, which is something that we're actually actively working on. I don't want to share too much about that. All right. So just to again talk about how does the KCQ relate to other outcomes. I showed before that it is related to the association — it's associated with increased risk of death. But it's also in the setting of transcatheter valve interventions. These are data from pre-TAVR and death at one year post-TAVR. You can see a nice graded relationship between the KCQ and risk of mortality. It's also associated with increased costs, primarily through hospitalizations or heart failure hospitalizations. When we look at changes, those are again associated with heart failure hospitalization and mortality, both in the setting of HF-REF, but also after both TAVR and MitraClip, I'm sorry, MTR. The one-month change in KCQ after each of these procedures is significantly associated with subsequent risk of heart failure hospitalization and mortality. So the patients that improve are more likely to do better, again not surprisingly. So point number 3, I think this is one thing that comes up is whether or not we can use other measures. Do we really need to use the KCQ or can we just use NYJA? So much easier. I can just say this person has a NYJA Class 3 instead of actually administering a 12-item score. One of the things that we have seen in a few different studies is there's a lot of discrepancy between what physicians assess as a NYJA class and what patients actually report as their own symptoms and limitations. So these are just data from TAVR showing KCQ score on the X-axis and the kind of NYJA disparities. And so you can see that, so for example, in the patients who reported a KCQ score of 75-100, so those would be considered actually NYJA Class 1, 46% of the physicians thought they were NYJA Class 3. So I mean, there's clearly a disconnect between what we think and what patients actually report for themselves, beyond the fact that it is a very coarse measure. There's a lot of other issues with NYJA. This is just a look at the different sites and the amount of ‑‑ there's not just overestimation or underestimation. There's just kind of a lot of inaccuracy in NYJA. If you look at exercise capacity, this is an important measure. But it is very different than what the KCQ is actually looking at. So these are data from COAPT and looking at the KCQ and 6-minute walk. And on the left side, it's the association of those with mortality. And on the right side is heart failure hospitalization. What you see is that heart failure hospitalization is more associated with the KCQ. So KCQ at baseline is going to predict your risk of heart failure hospitalization going forward, whereas exercise capacity predicts mortality. This is looking at changes. So change in exercise capacity versus change in KCQ. And the important thing here is they both, if you improve in either of these, that's a good thing. It's not surprising. But what you see is that the KCQ is much more likely to change after intervention. This is not surprising because the KCQ is disease‑specific. So if you treat a disease, that disease‑specific health status measure is going to show an effect. So if you're really trying to look at, is this treatment affecting my patient's disease and is it actually improving what I'm trying to improve, KCQ is a much better measure. So how can we potentially use this clinically? Again, just kind of summarizing what I showed earlier. But these are the estimated KCQ changes after different transcatheter valve interventions. But one of the things that is important is while most patients improve, not everyone does. And identifying those patients who don't improve and targeting interventions to those patients may actually be useful. So these are data from Partner 1 looking at baseline KCQ on the X axis and 6-month KCGQ on the Y axis. And what you see is that 17% of the patients still reported KCGQ scores less than 45 at 6 months. And 16% were 45-60. So these are essentially people who are reporting NYJA Class 3 or 4 persistent symptoms at 6 months. And the question becomes, you know, can we potentially identify these patients and really try to understand why didn't they improve and can we do something differently either from a valve perspective or from a heart failure perspective. So just a few key relationships with the KCGQ. If you look at the KCGQ before you intervene, if you have a low KCGQ you're more likely to die. If you have a low KCGQ, however, you are also more likely to have an increase. So if you survive, you do better. Not better, but you are more likely to improve. If you look at post-intervention, if you have a low 1-month score, you're more likely to die. But if you improve, that reduces that risk. So these are just some important kind of relationships to think about. One of the things that we have done is kind of put this clinical pathway in as one way that the KCGQ could be used on a clinical basis. So if you look at the KCGQ at baseline, kind of the idea perhaps is that if you had a very low KCGQ, you have a very high risk of mortality. And at least you need to have that conversation with the patient of whether or not it makes sense to go forward with intervention. That kind of 40-75, we kind of said this may be the best patients to target. Because they're really going to improve their quality of life and they're probably going to survive. The KCGQ scores that are high, I think this may be a little bit different now. But you potentially could think about surveillance. Obviously we don't really know quite yet TAVR in asymptomatic patients, but I think we're getting more data that it might actually be useful. But theoretically you could watch those patients. At 30 days, you want to look for how does the KCGQ change? Is it still below 60? Or has it not improved above at least 10 points? And if you see those, then perhaps looking at, you know, is there paravalvular leak that we can address? Is the LV function, do we need to send the patient more to advanced heart failure versus do we need to evaluate other comorbidities like depression or things like that? Okay. The next thing that I want to talk about is, so what are some of the challenges that we have when we interpret the results of the KCGQ or health status results? So the first thing I want to talk about is the placebo effect. This is the issue that the KCGQ is self-reported and is subject to some patient biases and how do we potentially address that? The second thing are mean changes and continuous factors. The KCGQ is a number that goes from 0 to 100. And sometimes we're looking at differences between groups of, you know, 3 points and what does that mean? Missing data due to loss of follow-up, how can we potentially use modeling to account for that? And then finally, something that we've dealt with probably more in the era of MTIER is missing data due to death and how do we potentially, you know, only patients who are alive can fill out these forms and how do we potentially address that? In terms of the placebo effect, we know that when we are looking at a medicine and its impact on symptoms and quality of life, we always have to test it against a placebo. These are data from Merlin, which is an old, old study, but looking at the association or looking at the improvement in Angina, which was a Seattle Angina questionnaire. You can see the placebo arm and the treatment arm, the treatment arm is in red, they both improved substantially from baseline. If you only looked at the treatment arm, you would say, hey, this medicine really, really works. When you start getting into interventions and especially into surgeries, it becomes a little bit more challenging to test that theory. This is one of my favorite studies. It's very old. But there was this procedure called Transmyocardial Laser Revascularization where you would go in and basically drill these little laser holes around an area that was ischemic myocardium with the idea that it would reduce Angina. And it did. It did. But one of the things that they did is they tested it versus a sham. So they took the patients all into the cath lab OR and got access. The patients had no idea whether or not they got the procedure or not. And what they found was that all the patients improved. Their Angina was reduced and there was no difference between the sham and the procedure. And this procedure completely went away. Now you know, there are some issues with this. Obviously they weren't using a disease-specific measure. And when this has been looked at in say Orbita of stents, there is improvement if you're using some disease-specific measures. But this is something that's important to recognize. We say patients feel better, but is it what we've told them or what we've actually done to them that's impacting this? So this is again, these are the data from TBT in TAVR and in M-Tier. And you can see, wow, these patients did a lot better. Now that being said, I actually think these are real results. So please do not hear that I think this is all placebo effect. But how can we show this? What can we do? It's very difficult to do sham procedures. And is it ethical to do sham procedures? And there's a lot of issues with that. So one of the things that we can do is we can say, well, let's correlate these to some physiologic changes. Can we say that if we're reducing B and P levels or reducing MR, and is that associated with our changing in the KCCQs and therefore we then believe the KCCQ? These are data from COAPT. And if you looked at the change in KCCQ from baseline to one month, the M-Tier is on the left and GDMT is on the right. And these are bends of residual MR at one month. And so in this right green bar, these are patients who had residual 3-4 plus MR. So we did this M-Tier on them, but they didn't really have a significant improvement in their mitral regurgitation. They still reported a fairly significant increase in their KCCQ scores. But if you follow them out to two years, that completely went away. And so this is one of the things that we can use to say, hey, perhaps, you know, at least if you follow these people over time, that that, you know, quote, unquote placebo effect is not durable. So if you're seeing benefits in the KCCQ beyond a certain amount of time, perhaps it's more believable. All right. So the next challenge that we have is how do you interpret mean changes? So when we talk about a treatment, we're applying that to a population. And in that population there are some patients who do well, some patients who aren't affected, and some patients who do poorly. But what we report in a lot of these studies is the mean change. And that does not reflect any single person that was actually seen in that study. And how do we interpret that mean change? Just as an example, so these are four studies of transcatheter valve interventions. So partner 1B, S3, partner 3, and then COAPT. And you can see that in the studies where the treatment was compared with medical therapy, there were these large differences. And it's pretty easy to see that this is you don't have to worry about this variability within. But if you look at, say, S3 or partner 3, there was a mean difference of around 2 points in the KCCQ. And I've told you earlier that about 5 points is considered clinically meaningful. So the question becomes, well, is 2 points not clinically meaningful? Well, that's not necessarily the case. Because this is on average. And on average does not represent the variability within that measurement. So one of the things you can do is you can look at the proportion of patients who improved and the proportion who did not. And this was done, this is those data from partner 3. And what you can do, this is the blue line is the, just to orient you, this is the change in KCCQ on the X axis. This is the percentage of patients who had at least that percentage change. So they had at 20 points, that would be patients had at least a 20-point or greater improvement. What you can do is you can create, calculate numbers needed to treat. So what this says is that you would need to treat at least, you would need to treat 17 patients with TAVR to have at least one patient have an additional one patient have a 20-point improvement. And so it just is ability to translate that 2.6-point mean difference into something that's more clinically meaningful. The other thing that we can do is we can say, well, maybe we can identify the target group. Who are the patients that we want to look at? Who are the ones who improved? And this is when you start getting into what we call heterogeneity of treatment benefit or trying to identify the variability within the data. So again, this is the same data that I showed before. And this is the idea with this was where that, if you've heard the term the poor outcome or the acceptable outcome that we use in TBT, I just want to go through kind of what it means and how we came up with it. So this was the figure in which we conceptualized this. What we thought was that, well, if at the end of, this is six months, but the poor outcome is mostly at one year. If at the end of one year you have a KCQ score at least of 60, that's acceptable. If you're below 60, you're NYJ Class 3, 4, that's not really what we're trying to achieve with these treatments. The only issue with this becomes this area in the upper right corner. So these are patients who actually started with a pretty good quality of life. And for those patients, the important thing is we don't want to make them worse. We just want to make sure that they're staying at least the same. And so this is where this little part came up. So the idea of an acceptable outcome at one year, it's a KCQ score at least 60 and no decline of 10 or more points. So that's kind of why that extra piece of the definition is there. So using that definition we then said, okay, can we potentially identify who are those patients who did well and who are those patients who did not do well? Just to show you what are the outcomes that we've been seeing, these are different outcomes at one year after TAVR through the years from Partner 1 to Partner 3, TBT being in the middle. And at the beginning about half of the patients had a poor outcome at a year. So either dead or still had a persistently poor quality of life. That is obviously significantly improved to Partner 3, which is not surprising. We try to identify who are these patients. The first thing we do is we look at individual comorbidities. So patients with severe lung disease, patients on dialysis, patients who have a very poor baseline health status. And we see that we're able to say, okay, about 50% of the people who have severe lung disease have a poor outcome at a year. And that's a lot. But it's also flipping it if half of them have a good outcome or acceptable outcome at a year. So can we do better? And this is where we came up with this Poor Outcome Risk Model. I don't want to go into too much detail of this. But we've developed what we validated it and then we re-estimated it in TBT. And the important thing is the patient characteristics that were associated the most with the highest risk of poor outcomes were being on home oxygen, advanced chronic kidney disease, being wheelchair-bound or having a very, very low KCCQ baseline, ADL dependencies, unintentional weight loss and moderate to severe dementia. In that group, you can see on that right side a substantial proportion of those patients were either dead or had persistently poor quality of life at a year. Now thankfully, we're not seeing these patients nearly as much as we did five or ten years ago. But they still exist. And recognizing who are those patients ahead of time to at least have conversations, important conversations with the patients and their families prior to procedure is important. So Challenge 3, and this is I think something that's really relevant for kind of the TBT registry is, how do we deal with missing data due to loss of follow-up? And how does that potentially impact our analysis? Sometimes we're able to, if we look at multiple measures over time, which in the TBT we're talking about baseline one month and one year, we can kind of do some modeling to kind of account for some of that missing data. But having any missing data, particularly when there's only two measures of follow-up makes this really challenging. It's also important to know that patients who have loss of follow-up, particularly in registries, are often very different than the ones who actually have data. So going after the patients who are perhaps the hardest ones to get can really be helpful to know. Because when we present these data, when we have these data, we really want to show what is it showing to the whole population? If we're only looking at a sliver, we aren't able to get the full picture. So just to kind of show initially what does that mean? So these are those raw data that I showed. Some of the things that we've done in different modeling, there's a lot of different things that we can do. We can do growth models, pattern-mixing models, linear regression, blah, blah, blah. These are ways that we can deal with missing data. But importantly, all of these strategies assume the missingness is random, or it's not in formative. And that's not always the case. So this is just data from one of the studies that we did early on in TBT. This was looking at just looking at the overall quality of life or health status data at a year and what we had to do in terms of missing data. So we look at there's 26,000 patients who underwent TAVR. 5,000 had no baseline KCQ. We also had to exclude a bunch of patients because they came from sites where less than 50% collections. So that was kind of our threshold, which thankfully things have gotten better. And then of those who survived a year, another 2,700 had no follow-up. So we start from 26,000 patients and we end up with 7,800 who had one-year KCQ data. And we're trying to say these are the health status data of all the patients who survived one year. And it's really only a fraction of those who actually underwent TAVR. Now we look at again those data in terms of how does that impact. If we look at those patient characteristics of the included sites versus excluded sites, there's not a lot of difference. Because if we're excluding sites that had less than 50% data collected, a lot of that's administrative things. They're coordinators that aren't really working hard. And they're not selectively choosing one patient or another to get the data. But it's kind of a structural issue. But when we look at those who were the included sites, missing data versus data available, you can see there are some differences. So those who had missing data were more likely to be female. They were higher risk. They had lower baseline KCQ scores. They had more lung disease, more kidney disease. So it obviously impacts or biases the interpretation of the data at one year that are available. There are some ways that we can deal with this from a statistical perspective. But it's not perfect. So the final challenge that's been I think unique to the valve population, maybe not unique, but has come up more in the valve population is missing data due to death. The concept is that health status can only be assessed in surviving patients. It kind of makes sense. We talked earlier that patients who have worse health status are more likely to die. So if you're systematically removing the sickest patients from the analysis, your outcomes are going to look better than they otherwise would had those patients been included. And so that can basically, it biases the results and makes the KCQ scores at follow-up higher. This is really important if you're dealing with a study where there's a difference in mortality between the two groups. Just as an example, this is the first place that we really saw this issue and had to struggle with this. These are data from COAPT. Again COAPT was MTR versus guideline medical therapy in patients who had severe functional MR. And there was a pretty significant improvement in a reduction in mortality in the MTR arm. And again, in the guideline directed medical therapy, if you're systematically removing these patients, the health status at the end is going to look better than it would otherwise. So these were kind of the raw data. So if you look at MTR versus GDMT, just going out to two years, there was about a 13-point mean difference in KCQ scores at two years. And that's significant. So you walk away from this study saying, okay, this does also improve mortality, it reduces heart failure, hospitalization, and it also improves health status. But if you're able to, and we did this kind of fancy statistical thing where we kind of co-modeled the mortality and health status data, so we're sort of accounting for that difference in mortality, you see that the difference is actually about 19 points. Again, had those patients survived, you would have seen a more significant improvement in health status in the MTR arm. All right. So the last thing on my list is how does the KCQ function and what is its role in this TBT 30-day composite? So we have this composite in point for, I think it's only for TAVR I believe, correct? Okay. So it's only for TAVR. And it's supposed to kind of be this overarching measure of quality, one piece of it. And how do we develop it? So one of the things that we did is we looked at a bunch of patients who had undergone TAVR in kind of what we call the modern era, which at that point was 2015 to 2017. And we used a hierarchical 30-day outcome. So death was the worst, then stroke, then AKI, bleeding, PVR, and then none. And we then talk about the site difference. So each patient is paired with a hypothetical patient having the same risk factors and treated in an average-performing reference hospital. The way that we came up with this, and this is a complicated slide, and I don't want to get too much into the weeds on this. But what we did is we looked at the impact of complications at 30 days with one-year mortality and one-year health status. And the ones that mattered most were the ones that ended up higher on that hierarchy. So stroke, for example, that was the first complication that was most important. Had a very strong effect on long-term mortality as well as long-term health status if the patient survived. And you just kind of go down the list. But that's also why, for example, major vascular event and new pacemaker were not included in that. Because while they do impact outcomes to some degree, it's really not as important. All right. So in summary, hopefully I have convinced you that health status is a meaningful outcome. It's important to patients. They care about how they feel. And if you don't measure it, you're never going to be able to assess that as an outcome. You're not going to know how 90% of your patients are doing. But analyzing and interpreting these data have challenges. KCQ can be used clinically for both patient selection as well as risk assessment pre-valve intervention as well as assessment of response after intervention. And integrating survival with health status after valve intervention I think provides a more complete picture of patient outcomes. Because it's not just about whether or not they live or survive, or live or die. And while that's important, it's also about how do they feel. I think that is my last slide. All right. Well, thank you. I appreciate your attention. »» Thank you very much, Dr. Arnold. I think we have time for a couple of questions. And I'm going to mosey on over there to the moderator. But to be honest, show of hands, how many of you in your HART team meetings and your decision-making venues, whatever you call them, before deciding whether or not the patient should have TAVR or not, how often do you bring in KCQ as a talking point? Is it a talking point? Or perhaps after Dr. Arnold's talk, will it be a talking point? Is that something you do discuss pre-procedure? Okay. As we're talking these three days, we're sort of building the conversation and we're trying to get people back into the concept of the HART team concept of shared decision-making and what should we do and should we do it now. This is important. Rose from Penn, are you in the room? So when you're talking, and again, it's kind of that building conversation with people who are doing much better with baseline. We struggle sometimes with 30-day and one-year. And people try to do that. And again, the story, I get to repeat myself because I've been around too long. And it's Penn. So full disclosure, Rose would say that if she was here. But back in the early days they had a very difficult time with their follow-up, follow-up for everything, but definitely with KCQ. And I don't know if this is helpful, but I always like to throw it out there. They changed their practice. And again, those were the days when the folks came back to the site of the index procedure for their follow-up. But right at the point of entry when my name is Joan Michaels, hello, Mrs. Jones, I will be taking care of you for the rest of your life. And that really impressed not only the patient, but you know, as we joke, trying to find a parking place in West Philadelphia is not easy. So the person who took the day off of work to drive mom into the appointment would all of a sudden say, what did she just say? So their completion rate of their 30-day and one-year KCQ follow-up appointments, if Rose was in the room, she would validate this. It just was never a problem again. Again, this was the early days of TAVR when this whole business of 30-day and one-year follow-up was bizarre. But impressing your patient and the family with shared decision-making at the point of contact in the first meet-and-greet really has helped. And I think, you know, we try to have tricks of how do you get these patients to pick up the phone, to answer the mail, to answer the questions. But I'll see what questions we have here. Thank you so much. I also, I'll say one more thing. The vascular complications, we hate that. So if you have anything to do, don't ever include that in the risk model or anything. I'm glad you pointed that out. But that's a show of hands. Is that not a real problem, pain point? And I say, who cares? It's not in there. »» And we showed that actually. We showed it both in TBT as well as the study before that which were data from Partner that vascular complications in the absence of a bleeding issue, it's not that vascular complications aren't important, but if they don't cause a major bleed, they're really not that clinically relevant. So yeah, I'm a fan. Trying to find an easy one, even if you're a Kansas City fan. I'm not sure what maybe this means, but how is KCCQ accurately calculated? How do you calculate it? Well so there's an algorithm for it. The scoring algorithm is, I mean, it's not something that's like an easy calculator that you can do. And that's part of the software part. But you know, it's a Likert scale and each one counts for a certain number of weighting. And then with like the physical limitations, if you say I did not do because of other reasons because it's always asking specific to heart failure, those are excluded, but you have to have a certain number of physical limitations questions answered or else it's considered missing. But the scoring algorithm is not something that is published so that you can just do it by hand. »» The question is, at MAHI what strategies have you found most helpful in obtaining follow-up? Do you have any secrets? »» I mean, you know, one thing in Kansas City is that parking isn't super hard. So no, I'm just kidding. No. You know, we actually are pretty good at keeping track of our patients, especially now for that first year. And part of that's because we don't have a huge area that we're treating anymore because there are a number of different kind of TAVR places through the city. So most of the people we're getting are not, you know, rural, from far away. And so we're able to really get them at 30 days and one year in office. And so, I mean, they can complete the KCQ in the waiting room. From a side part, so not from a clinical part, but we do a lot of studies where we collect where the data collection place for different studies and the KCQ as well as all of these health status measures are really designed to be able to be completed by phone. So we, you know, I would say that's the other thing with this is that it is incredibly valid to do this by phone. It's also built into Epic. There's an app that one of our, well he was a fellow, but now he's faculty, developed to also try to do it, to try to monitor KCQ clinically. So there are structures to do it. But I think the trick would be phone. Because it's very easy to be done, completed that way. »» And as we said, we want it to be part of the conversation and kind of a starting point. But as a last sort of crossover, is it okay to do it in the waiting room before the patient? »» Yes. Absolutely. Unless they've already been sedated and then it's probably not a good idea. But no. And I'm not a huge, it hasn't really, I don't think that it should be done really by proxy That's the other issue. So if a patient's on a ventilator, you know, I don't think it's probably a good idea to have a family member complete it. But the KCQ, I didn't go into this, but it asks over the last two weeks, so it's asking the patient to integrate what are their symptoms over the last two weeks. And most people, you know, have been relatively stable before that two weeks. They're symptomatic, but they're not, you know, it's a little different if you get them in the setting of a hospitalization. But it's perfectly fine to ask them at any time before the procedure. »» Nobody in this room does this. But once upon a time somebody did ask this question. Could I ask the patient after the TAVR, how did you feel to, how did you feel, so the answer to that is no. »» Yeah. »» We asked that to John Rumsfeld and he was not, he kind of gave us a haphazard. »» Not a good idea. Yeah. »» So don't do that. »» No. »» Is there a good source for coordinator or office to use or calculate the KCCU before it is entered into the registry? Is there a good source? »» No. I mean I think again that the scoring algorithm is part of the licensing part. So that's not a publicly available thing. It is, you know, it is built in EPIC. That's the one thing that I could say. So if those are, and that's kind of licensed through EPIC, but they can put those in and then you can get that, you know, like our advanced heart failure clinic has those and it pops up as part of the vital signs, the KCCU does. So there are ways to do it, but the calculator is not publicly available. »» And as a future, do you see this continuing to be an important tracking for both TAVR, aortic and mitral and tricuspid if you were? »» I mean I think it's incredibly important. I mean obviously I come to this from a research perspective because, you know, we've learned so much about how our patients are doing long-term and what are the impact of complications. And we would never have known those things had we not been tracking these data. But I also think that from a clinical perspective, tracking these patients over time and trying to identify who are the patients who are not responding the way that we would expect them to, you know, say somebody has a KCCU improvement of, you know, 8 points after the procedure, they may say, oh, I feel better. But they're not feeling as much better as they should. And so I think that those are some trigger points that might I think be helpful. »» And that was a trick question, not for you, absolutely not. But we get that, well, with low risk we don't have to do that anymore. And so, yes, we do. And quality of life will be important at high, intermediate or low risk. And again, I don't know if you want to show hands, but some folks who have started to do low risk felt that the coverage decision didn't count anymore, the 5-meter walk didn't count anymore. And certainly you didn't have to do KCCU anymore because they're now low-risk patients. And that's not a true statement. »» To me that's like counter, I mean that's counter. Because if you look at the mortality rates and the hospitalization rates in these patients, you're not moving the needle on those. So that's really the one thing that you're improving is their quality of life. »» And I said this to Dr. Arnold. And again, if you've ever heard me say this, I'll say this again. And Dr. Brindis is in the audience, if he's still there, he could back me up. We meet with our stakeholders, which includes our CMS and our FDA stakeholders and a lot of other folks a couple times a year. And if CMS over the years has been clear about one answer that we've asked them is that they've always been clear with everything that's in the coverage decision, it's a little bit cloudy sometimes. However, it does say patient-reported outcome. They're very clear about saying they are willing to continue to pay for this procedure as long as you could continue to prove that you've improved the quality of life of their constituents of their patients. So I think that's a heavy endorsement for quality of life and they value it and KCCU will not go away. »» And you might have answered this, does it matter if it's phone? I know you said phone's okay, but digital paper, is there any best practice or best way? »» No, I think all of them are acceptable. It's what the patient's willing, other than proxy. I think digital, phone, paper, whatever the patient is most acceptable of is fine. »» Do you ever see a place for the full KCCU score versus the clinical model? What situations? »» Yeah, I think that if there are a few, I think from a research perspective in clinical trials and when you start looking at the individual domains, you can kind of understand a little bit in a deeper dive how different things are impacted. So for example, the quality of life domain in most of these studies starts out the lowest and improves the most. I think that that is those kind of individual domains. While you can get some sense of that with the Cases U12, I think from a clinical perspective there's no, I don't see much of a role of it at all. And I think from a registry perspective, we use the overall summary score almost all the time. If we're not using that, we use the clinical summary score. The FDA sometimes likes that better because it excludes the quality of life domain. But yeah, I would not, I can't think from a clinical perspective what you would really gain from the 23 item. »» Okay. That's good to know. Question is, are there any other tools that are on the horizon that we should be aware of? First question. And well, I'll let you maybe address that one if there's anything out there. »» Yeah. So I mean, we're not actively creating any new health status measures. I think that, you know, we have good ones for angina, for PAD, for heart failure. I think the one thing that we're actively working on is how does the KCQ do in tricuspid patients? sort of supplement to the KCQ in the tricuspid patients? And that's the stuff that we're working on with the FDA right now. »» But it's not a spoiler alert. We've talked about this in our previous session. The one thing that we're hanging on to with our tricuspid patients will be improving their quality of life. So please, please, please, if you leave with anything, that KCQ is here to stay. And work that into your practice with all three eval procedures. Because that's what has been shown so far with whatever's out there with tricuspid, that it's the quality of life that's moving the needle. I think we got most of the answers. Again, I hope you're not mad at me for making fun of the Cleveland or Kansas City and Ohio State. »» I can definitely hold my own with the Buckeyes. The Chiefs I care about, but you know, my Buckeyes are my soul. »» You are now officially welcomed into Hashtag Tavern Nation as our very own Swifty. So thank you. »» Awesome.

Kansas City Cardiomyopathy Questionnaire

KCCQ

health status

quality of life

heart failure

transcatheter valve interventions

clinical outcomes

physical limitations

self-efficacy