Hi, everybody. Thanks for being here. Sorry about the glitch here. We have a little video show. We can show it later. First of all, we're really appreciative to have you all here. We thought we'd share something that a journey that we've been on. Typically, Edwards Lifesciences, we get very focused on those patients who are treating the critically ill or those people going through a big procedure, and we try and be expert to make sure that these patients have a wonderful experience during that one or two where we're really involved. Over time, we've gotten quite an education. We've always learned in the past from the physicians, and their impression is that all the patients that need the procedures show up at their office door and that they get treated.
More as the time has gone on, we've come to understand that many of these patients indeed never come to the office door of the treater. We've been on a journey to learn more about that, and we're going to invest a little bit of time with you to share part of our journey. We've asked Don Bobo to assemble a team of some experts and share our journey with you. Don, welcome to the stage. Thanks.
Thanks, Mike. Welcome, everybody, and let me go ahead and start by asking Dr. Matt Brennan and Maddie Petrescu to come join me on stage. I'll take a few minutes and try to frame the journey we've been on in the past year, understanding what patients really experience and encounter as they're on a journey to being diagnosed with aortic stenosis. I'll try to keep mine moving, then Dr. Brennan will talk about a number of the learnings that we had from this work. Then Dr. Petrescu, who's the head of the echo lab at Swedish in Seattle, will actually talk about one of our initiatives to try to improve the quality and close treatment gaps around echo. With that, let's go ahead and see if the slides work. Yes.
This is an oversimplified attempt to express the process that patients go through when they get diagnosed with severe aortic stenosis. There's typically an echo exam that's done where they have to measure parameters. You see behind me three columns that get filled out, and depending on how severe the jet velocity is or the gradient, determines whether the physician goes to the next step, and that is attempts to evaluate symptoms. It's the combination of signs and symptoms that actually drive the diagnosis of aortic stenosis. We actually had a chance in the last year to get a very large cohort of data. This is a group of 80 million patients that have been collected and the data curated over the last 10 years.
From this, we said, "Look, what if we could take the EHR data, which is comprehensive, physician notes, tests, echo results, and the claims data, and actually create a cohort of patients where we have all of the information on their journey over 3 years through a diagnosis of aortic stenosis. Wonder what we could learn. Wonder how that understanding would change our view of this disease." We ended up with a longitudinal time series march of these patients across time, and we were able to understand insights by provider, by system, by who connected the signs and symptoms, and what ultimately happened to these patients. Let me spend a brief moment taking you on this journey. At the very beginning, this is a group of patients 2 years before the cohort that were treated.
You had some of these that had information in the health record that said they were moderate AS. You had some where there was information saying that they were severe AS, and you had some that were undiagnosed. There was no particular mention of echo findings or symptoms. This is the starting point on this journey for these patients. As we go through this is going to reflect largely a group of patients that get diagnosed. As we're able to play through this 3-year tape, at some point, you reach a stage where there are signs and symptoms that align up with the guideline classification for severe aortic stenosis. At this particular point in time, a majority are being seen by a general cardiologist. A minority are being seen by a PCP.
Now, it's important to understand that we were able to connect signs and symptoms. It doesn't mean that the physician actually caught this guideline diagnosis. At some point, a significant minority got sent to a treater, and you can see the PCP actually picked up some of these patients and sent them to a general cardiologist. The red group went to a treater, the gray group unreferred. The next step in the march is a group of these were treated. In this particular illustration, you see for the first time the emergence of mortality as these patients were or weren't sent through the PCP or general cardiologist to a treater.
If you look at this group a year after the cohort that were treated, you had about a third that were treated, you had about a third that died, and you had the rest that were left untreated. They're in the system. There are clear signs and symptoms. This understanding really drove a couple of learnings for us. First of all, if I were to take you back in other conferences where we talked and when Larry talks about his business, we used to say, "Half these patients are asymptomatic." In this particular data set, they're almost all symptomatic. Physicians may not catch that, but there are clear signs and symptoms that say this is guideline disease. It's not about symptoms. The other thing we realized is as we looked across the stakeholders, patients, physicians, or systems, there are barriers to diagnosis and referrals across every stakeholder.
Patients often don't understand the symptoms. They don't appreciate the burden of the disease. They don't understand the mortality risk that they face. Physicians oftentimes can't connect signs and symptoms over time. They're busy. Oftentimes, these symptoms get conflated with old age. Systems where we do echoes and there are referrals, a lot of learnings around what goes on when a patient encounters a system. Tonight we'll have the opportunity to have Dr. Brennan bring to life some of the learnings around physicians and patients, and then Dr. Petrescu will talk specifically about an initiative we have. The other learning we experienced is if I take you back over a couple of three years, we would look at the 1.2 million aortic stenosis patients and say probably 18%-20% are being treated. This is a prevalence.
As we take ourself through this data set, it's probably one in 10. The magnitude of undertreatment ended up being bigger than we thought. While highly effective treatments are essential, we say there's probably more we should be doing. There is a systematic approach that we're looking at to actually increase disease and therapy awareness to try to get at some of these treatment gaps and barriers. One of the initiatives that some of you may have had a chance to see back around the corner is a service partnership with centers to actually get into the echo lab and significantly improve the way echoes are collected and the feedback the echo labs are able to respond to in referral. Dr. Petrescu will take you through that, but here's a little bit of a teaser.
On my left, there's a chart that think of every column as 100 patients. Where there's gray, the three important echo variables were measured. Where it's white, patients went for an echo, and those parameters were never measured. When echo labs respond to this or say, "Look, at least if a patient goes for an echo, we should measure it," systems that have implemented CardioCare have seen this kind of improvement in the quality of the echo that's being collected. While this journey and these learnings, there's no one silver bullet. We really look at this and say a broad effort going at disease awareness with patients and providers, therapy awareness with physicians, and some targeted program like CardioCare are going to be an important part of the growth of this particular therapy segment.
With that as a setup, let me go ahead and invite Dr. Matt Brennan, he's a physician at Duke. He's been partnering with us on this work over the last year or two to come and bring to life some of the additional learnings we had as we did this work. Matt?
Thanks, Don. Appreciate it. Don and his team have asked me to share with you a little bit about the journey of an AS patient so that you can understand from our perspective what we see as clinicians. I'm an interventional cardiologist. I work at Duke. I've been on faculty for the past, I guess 8 years now. Did my training at Duke as well. Run a clinic that's basically valvular and coronary heart disease. This is what I deal with all day long. I'm going to try and help you to understand the journey of the patient. Hopefully, if I do my job, at the end of this talk, you're going to be able to look and understand what are those deficiencies in our care and where is it that we're losing patients.
You'll understand the scope of the problems so that you can reflect back to the numbers that Don has given you and try and understand why this drives us to do better and where the areas for improvement lie. I'm going to start with a very deeply personal story. I'm going to call this patient Mr. G. He was a patient of mine for a couple of years and became a good friend. He and his wife came to every appointment together. A terrific guy. 52 years old when I met him. He was a debilitated carpenter. He had had a fall at work and had fractured his back and had significant leg weakness and pain, which he dealt with absolute grace.
He also happened to have a bicuspid aortic valve, and when I met him, his aortic valve gradient was over 60, which Don just told you over 40 is severe. Over 60 is quite severe. Despite that, his ejection fraction was normal. He was completely asymptomatic, although arguably he couldn't get out and run a marathon. Could you really assess that? We engaged on a program of checking his what's called a proBNP. It's a marker of cardiac strain every 6 months to evaluate whether or not his ventricle was feeling the effects. We were going to use that in his symptoms in addition to check his symptoms to tell when he needed to go for a valve. Something came up.
Mr. G's leg weakness progressed, the pain progressed, he came in the clinic, and he needed a back surgery. His back surgeon was pretty insistent on it. He was pretty insistent that he couldn't go on anymore. We talked about the possibilities. Mr. G could have gone for his surgery first. At the time, TAVR was not an option for him. He was not high enough risk by the standards at the time. We talked it through, and reluctantly, I gave him clearance for his surgery after talking to him about the devastating hemodynamic collapse that he could experience with his induction to his surgery, et cetera. Remarkably, he went through his back surgery, had no problems. He flew through it, was observed overnight. In the morning, was looking terrific. He and his wife went home.
He collapsed and died that afternoon. His wife called me to tell me thanks for the care that I had provided. She said her husband would have wanted her to do that, and wanted me to know how much they appreciated the opportunity to get through the surgery, so on and so forth. I don't tell you about his case because I enjoy sharing the ghosts. We all have ghosts that come with us in our care as clinicians. It's not because I want to burden you with that, but more because I want you to understand that this is a very fickle disease. It's a disease that can turn on a coin, and we have to be prepared to meet that challenge. These can be some of the most difficult patients in our practice as cardiologists. They tend to be older patients.
They tend to be more comorbid. I'll give you another example. This is a common scenario, but this gentleman I met, we'll call him Mr. W. I met in the ICU, the intensive care unit. I was a consultant. He was in cardiogenic shock. He was, by all means, on death's door bed. He was still conversant, late 60s, early 70s. The team, the ICU team, called me in to see was there anything else that we could offer him. Why in the world was his ventricle so sick? His heart, why was it not squeezing the way that it was supposed to? I went back and looked at the echo, and in fact, it had been interpreted incorrectly. He appeared to have what's called low-flow, low-gradient aortic valve stenosis. I suggested that he needed an aortic valve replacement.
I found a surgeon, a younger surgeon, who was willing to take on his care. The surgery was rocky, but he got through, and his ventricle flew. When you relieve this obstruction, people love it. He now comes in annually. I see him. His ventricle's back to completely normal. He's having a great life, sees his grandkids. A much more uplifting story. On a different day, with a different consultant, with a different surgeon, this patient may get a different treatment. This is one of many stories that all of us carry around, hopefully it helps you to understand why this disease process in our country is so undertreated. Although we've been improving over time, if you look in the right set of figures here, pre-TAVR, early TAVR, contemporary practice. Our treatment rates, and now these are for diagnosed patients, so objectively diagnosed patients.
Our treatment rates are increasing, we still leave 65% of patients by the Optum data set, and I think this is probably true based on practice, untreated. That's a problem. It's partly built because of historic paradigms. We've come up, all of us, and the systems were built in an era where many of our patients couldn't be treated because surgery, open surgery, was the only possibility. Outdated systems, outdated ways of thinking, have led to this place where a lot of patients are no longer being treated, this can be fixed. This is a problem that can be fixed, and it needs to be fixed for our patients' good. We know that when patients with severe symptomatic aortic valve stenosis are treated, their survival improves, their quality of life improves, and it really doesn't matter who they are. It doesn't matter who they are.
African American, women, older patients, doesn't matter who they are. Their symptoms will improve. Again, data from the Optum data set, but it's been replicated before. We know that we've got a treatment that works. We know that we have a problem. Why not just fix the system, right? To understand this problem, why are patients being undertreated, you really have to understand the road that these patients follow, the appropriate road to get to treatment, because this is not a disease process where you want to take the road less traveled. That's not here, right? I'm going to ask you to, for a second, imagine a patient. This patient has calcific degenerative aortic valve disease, and I take you back in time. This patient's 40s, 50s, no signs of any problems.
They've got normal physical exams all the way along, suddenly their valve starts to harden. Ideally, the primary care physician puts a stethoscope on the patient's chest and says appropriately, "Look, you've got hardening of your valve. I'm going to send you for an echocardiogram." That patient goes for an echocardiogram. They get their diagnosis. They're referred to a cardiologist. The cardiologist then follows them as that disease progresses, eventually they become symptomatic. The patient's immediately referred for treatment, within a span of a couple of weeks, the patient gets treated. They go on with the rest of their life. No problems. The problem is this road has hazards. The murmur is missed. The echo's misinterpreted. The symptoms are misappropriated to COPD or something else. The referral is delayed. The patient's misinformed about what's going on. Ultimately, the treatment's delayed and lives are lost.
Clearing these hazards out of the way, though, will save lives, that's what we're about here. That's what we're trying to do with these initiatives. Although 80% of the U.S. population has annual contact with the medical system, if you look in the right-hand panel, very few of our providers are actually equipped with the skills that it takes to diagnose aortic valve stenosis. If you look the second line over, set of bars over, you'll see aortic valve stenosis diagnostic accuracy of a graduating medical resident, less than 10%. This was a study that was done in 1997, published in JAMA, that's been reproduced on multiple fronts. Those of us who practice, probably a little bit better than this, but I think this gives you a sense of the scope.
80% of patients will see their doc every year. Just because you see a doc doesn't mean that that doc knows what they're listening for when they pull out that stethoscope. This shows you the problem here. Patients missing their visits, they have an opportunity. The doc puts the stethoscope on but doesn't recognize what the problem is. These are opportunities missed. To understand why these compound into undertreated patients, you need to understand that we were all raised in a system that said that aortic valve stenosis was an indolent disease. Like prostate cancer, for example, you have aortic sclerosis, or you have mild aortic valve disease, sir or ma'am, you will probably never have to deal with this in your life. That's what we were all taught. That's far from the truth.
The truth is that seven years from an aortic sclerosis diagnosis, most patients will develop severe aortic stenosis. Four years from a mild to moderate diagnosis, they'll have severe aortic valve stenosis. That brings up a problem when you have people who are used to sitting back and saying, "Hey, this is not a problem." I would say that it's actually a bigger deal in my mind, bigger than that patients are not getting echoes, is that patients are getting those echoes. The provider is getting a note back that says, "Hey, you've got mild. Your patient has mild disease." Now that patient and the provider sit back and they say, "Oh, this is no big deal. I don't have to think about it again" until the patient shows up like that Mr. W, who's in the intensive care unit on death's bed.
That's his initial incident heart failure episode, and we're just looking back through his chart, and five years earlier, he got a diagnosis of mild aortic valve stenosis. That's a major problem, but it's one, again, that's correctable. Let's say that your doc does the right thing, that you're diagnosed appropriately, that you get referred on to a cardiologist. These, again, are data from the Optum dataset that hopefully show you that it depends your likelihood of actually getting treatment. You go to a cardiologist, you're going to get standard care across the board. That's what we offer. Except not really. It depends on which cardiologist you go see as to whether or not you actually get treated. There's an over 200% chance of a different treatment. You'll get valve replacement or not, depending on which cardiologist you go to in the United States.
Randomly, somebody picks up the phone, picks up a consultation, 200% difference between whether or not you're treated, and that makes a difference. The bottom third of referring providers will have a 23% increase in the risk of death for their patients. It makes a huge difference. This is a problem that can be fixed. Unfortunately, in our society, this does not impact everyone the same. The first sign of this, you look at the U.S. Census, minority patients, minorities, individuals, are over 20% of people in this country. They're 5%-10% of the treated people, patients in this country. 20% of the population, 5%-10% of those treated. We looked in the Optum database to see, did we see the same problem there? The answer is yes. African Americans, grossly undertreated in this country.
Women, grossly undertreated in this country, and the elderly. This is a problem that affects a large group of patients. Thinking of it as opportunities, this is an opportunity for us to reach out to a population and really make a difference. This is something that lives that can be saved by just correcting a few of our biases, intrinsic biases. Let's say that this patient, the right thing happens. This patient gets referred on for treatment. Is that patient going to get treated? Well, the first thing to know is that delays are inevitable. In our system, these are data from Northwestern, prominent medical system. Data that Chris Malaisrie presented in 2014. One in four patients waited five weeks from their referral to treatment until they actually got the treatment. Five weeks, one in four.
That five-week wait corresponds with an 8% increased mortality rate for those patients. It matters when you get treated as well. Finally, 17% of patients who get referred for treatment, 17% who get referred, they show up at the office to get their treatment, don't get treated. Why? One in three of those patients who don't get treated will elect not to be treated themselves. Why in the world would you elect for that? It's essentially a day procedure now. It's a couple of day procedure. You're out of the hospital with a TAVR procedure within a couple of days. Why in the world would you take on that risk of mortality, the quality of life hit? The same one in three say that they're uncertain about the choice that they've made. Why?
If you look at the other characteristics, if you really dig into these patients, what they tell you is that they weren't adequately prepared for the decision. They just weren't educated. This is education. Education's easy. We can do this. We can do that. That's easy. Can we move the needle? The answer is yes. We're already moving the needle. The question is, do any of you think that 30% treatment rates are adequate movement of that needle? I think I would certainly, as a doc, say the answer is no. We're leaving 65% of people with a deadly disease untreated. We can move this needle way farther along, and we have to for our patients' good. With the advent of TAVR, the bottom lines, if you look here. You've got two sets of lines. You have pre-TAVR and post-TAVR.
With the advent of TAVR, what you see is that in the elderly, we've made great strides. We've made great strides. Those are the bottom lines here. We've moved from a place where the patients over 80 were grossly untreated to a place where they're just kind of untreated. They're still quite untreated, though. We still have room to go in women. We still have room to go in racial minorities as well. This is an area where we can make strides, and it's an area where we need to. Education is key for our providers. Education is key for our patients and this community of patients. It's also key that we put in place systems that keep up with the technology, and systems that will recognize patients, diagnose them appropriately, and refer them quickly to therapy.
Dr. Petrescu is going to tell you a little bit about one of those systems, CardioCare, and the implementation at her facility, and help you to understand one of the attempts, and the success model that it brings of trying to get these patients treated. Hopefully I've left you with an understanding of what the problem is here. We have patients who are being under-diagnosed. We have patients who are being under-treated. These are not high hurdles. We're at 35% treatment rates, and these are the folks who are diagnosed. Think about all the people who the PCP that wasn't one of those 10% and knew what they were listening for, they missed it. Or the echo was missed, or the symptoms were chalked up to smoking.
There's a huge group of patients here who are untreated, and our goal over the next several years, part of my career, is going to be dedicated to trying to reach those patients and get them in so that we can get them treated, we can save their lives. Thank you for your time.
Okay. That was really great. It was really great to see your presentations, it's really an honor for me to be here with all of you guys, to share with you guys my story of a cardiologist, to take you on a journey into the heart of what happens in the echo lab, what happens when we take care of our patients, how does all that relate to all the data that you've seen tonight, to the population health data, to patient outcomes, to quality standards, and things like that. A little bit about me. I've been a cardiologist at Swedish Medical Center in Seattle for more than 10 years. Swedish is part of the Providence system that includes about 50 hospitals. It's a center of excellence. It's a coronary center. It's a really great place to work at.
A lot of great state-of-the-art programs, it's a very collaborative environment. I personally wear a lot of hats that I do, but amongst my roles, the one that is most meaningful to me is my role as a director of the echo lab. This is something I'm very passionate about, I'm happy to say that at this point in time, we have a state-of-the-art echo lab, we have a great team, a phenomenal team, that we really are aligned in our values of excellence and quality. We're always trying to raise the bar on quality because we know that best quality translates to best patient outcomes. We're going to talk about that.
I want to take you on a journey a little bit first on what is that echo lab experience like, when the patient is coming in, as you saw, the patient experience and the patient journey, what does it look like? Give you a little bit of insight. When the patient is coming in and getting evaluated, getting diagnosed, then getting to the proper therapy, there's a process, the very first thing that happens to these patients, they have to get an echo. The echo is the very first thing that needs to happen. The echo is done, from the echo, there's all this information that is being used to make a diagnosis of what is the structural pathology for that patient and what is the severity.
To be honest, as a cardiologist and as a part of the echo lab, I have to say that echo is a lot of times taken for granted. Oh, it's an echo. Everybody focuses on the procedures and things like that, really, the echo is a cornerstone of what's going to happen to these patients because the information, the data that we use from the echo is going to determine, is this patient going to have a surgery or not? Is this patient going to receive a transcatheter procedure or not? Or is the disease advanced that it needs therapy or not? We're making critical patient life decisions based on the echo. It only makes sense that we expect that the information from this echo, we expect that it should be of the highest possible quality and accuracy, consistent, reliable, because we're dealing with patient life.
We're making decisions based on this echo data. In the ideal world, we would like to believe that, yes, all echoes are going to have this high quality and accuracy. In the real world, unfortunately, that's not always the case. In the real world, there are actually a variation of quality that can exist at multiple levels in the process that can actually significantly impact the accuracy of the echo, and hence, potentially impact misdiagnosis or under-diagnosis. You already saw what happens when we under-diagnose. We could be missing patients. Patients could be dying as a result of that. Let me take you on a journey of how that would actually happen for a patient. When the patient comes in to get evaluated and they get an echo, there's a sonographer that comes in to perform the echo.
The sonographer comes in and gets a series of images and performs measurements, which are then used to come up with a diagnosis. Well, each sonographer is supposed to follow a specific echo protocol. The echo protocol basically has all the requirements of what they have to measure. The echo protocol is the first variant that determines the quality of that echo. Because not all echo protocols are created equal. Some echo protocols are going to be very thorough and complete and standardized, whereas other echo protocols are going to be more minimal, basic, and even lacking. It's going to vary from one institution to the other. They're not all created equal. The ones that are comprehensive are going to be better in diagnosis. The ones that are lacking could potentially miss things. That's the first variant. The second variant is at the level of the sonographer.
Sonographers are not all going to be the same. There are some sonographers that are going to put a lot of effort and really do a great job on the imaging, making sure that they're getting accurate measurements, making sure that they're always following the protocol as they're required to do, and then there are some sonographers that may not always be compliant to the protocol. They may not always be trying as hard, and so forth. That's the second variation that can impact the quality and hence the accuracy of that echo. The third is at the level of the interpreting cardiologist.
Once the sonographer has done their share, then it goes to the cardiologist, like Matt and I, and then we get the echo, and then it is the cardiologist's responsibility to review all the data, to settle any discrepancies in the data, and then to come up with a final diagnosis. There's a variation to the cardiologist. There's a variation to the experience, the expertise. Some are board-certified in echo, and others aren't. That can also impact this. These are some of the things that are going on that, a lot of times, there's not as much awareness, and I can share with you a story that really drives home this important point of quality of the echoes and how that can impact patients' lives. We were talking about this not too long ago in our conference.
40-year-old male, no prior medical history, developed sudden onset of dyspnea and presents to an outside institution, not Swedish. When he arrives there, he's found to be in congestive heart failure. He gets an echo. The echo that is done at that institution reports mild mitral regurgitation and normal ejection fraction. Nothing is really in the way of the heart. In the meantime, the patient continues to develop progressive heart failure, now is in cardiogenic shock on multiple pressors. Within 2 days of that presentation, is sent to Swedish, to us, to see if we can help the patient out. Upon arrival to the ICU, there's an echo that we get right away, which discovers the presence of a flail mitral leaflet related to a chordal rupture with torrential wide open MR, but it was in eccentric jets.
Sometimes you have these irregular mitral regurgitation jets, and it was clearly there, and it was severe, and that was the cause for this patient's presentation. The patient goes on to get surgery while in cardiogenic shock, and we know that when patients have cardiogenic shock, that's going to impact the outcome. It's not going to be as favorable. The patient thankfully made it, but barely made it. We said, "Okay, let's take a look at these echoes side by side in our meeting." We took a look at the echo side by side, the one at Swedish and the one 2 days at the outside facility, and sure enough, it was there, the pathology, but it was missed. It was completely missed. Why is that? We figured out that, number 1, the protocol at their facility was quite, let's just say, limited.
It was not as comprehensive. That was the 1st thing. 2nd thing, the sonographer performance also was a bit suboptimal. Sonographers have a responsibility to really chase the jets when they're eccentric, and the quality of that was not as good. Also, the interpreting cardiologist missed the diagnosis that was there. The problem is that all 3 quality checks were missed. This really shows you, really illustrates how clearly patients' lives are dependent upon the quality standards of these echo studies. It's stories like these that motivate our team at Swedish to really, really always raise the bar on quality, not settle for anything lower than that, because we're talking about patients' lives here, linked to quality. This is a big deal we're talking about.
That's where we're at, I can share with you that although we have a state-of-the-art echo program now, it wasn't always the case. When I arrived many years back, it was actually kind of a ghost town. There was not much in the way of leadership there. The protocols were quite minimal. There was not much quantitative assessment. It took days or even weeks to get an echo read, believe it or not. There was only one person who was echo board-certified, and that was myself. That's how the status was. By the way, that is quite typical of many other institutions across the country. When I got there and when I saw that, I was like, "Oh my goodness, we have lots to do here." I really was motivated to get the quality standards and to get everything working.
I have to say, that it was not easy. I met a lot of resistance. I was not popular because I was trying to get people to do more and work more, the sonographers and the cardiologists, nobody likes to work more and get out of their comfort zone. Anyway, the bottom line is, over the years, I realized that what worked, what got us there, was I had to get a good team. I couldn't do it alone. I got a good team of myself and we had a great team leader for the non-invasive lab, we also got role models. We got new sonographers to serve as role models. That was really important, getting a good team.
The second thing was that it was hard to get people to just do things because people, human nature, you're like, "Why would you want to do more?" It was really, we realized more about inspiring and motivating them so that they want to do more by helping them to realize that what they are doing, they're not doing it for me, but they're doing it to actually because they're part of a team and they're impacting patients' lives. That's kind of how we got there, you can see why quality is so important to us. When we heard about CardioCare, we saw the opportunity because we saw that it allowed us to have visibility of our echo data and how this transparency would be very important in uncovering what it is that we need to work on.
This is something that we were very excited about, some people were a little objective to that. They're like, "Well, why do you need to do anything? Because your lab is doing so great." The answer to that is, well, how do you know what you don't know until you know? I mean, how do you know until you actually look and have this visibility? Because when we talk about quality is not something that we have achieved at the ceiling, then we're done, and we can all go home. Quality is something that we can always improve upon. It's a dynamic and evolving process that we can always work and improve upon. I'm going to show you guys some of the data that we got with partnering with CardioCare.
When we sat down and we uncovered some really incredible insights by looking at our data. The first insight was we realized how important it is to standardize sonographer performance. You've seen the heat map. We're looking here at echo lab protocol compliance, which means how good were the sonographers at being compliant to the protocol, which requires that all three of these measurements are made. As you can see at the top, there's specific measurements that you see at the top, velocity, mean pressure gradient, and AVA are required to be assessed in each and every single echo. If you are getting those measurements, then you're compliant, and if you're not, if there's any white spaces anywhere, that means compliance is not there. You can see we're 91% compliant.
Remember the heat discussed. To clarify, each column is 100 patients, and then each bar over there represents one echo, one patient. If you have a white space, it means that a measurement wasn't done. If it's gray, it means that it's normal or mild, orange means moderate severity, and red means severe. Okay. We looked at our data and we said, "Wow, we're 91." First thing we said is, "Hey, we're pretty good." Because if you look at other places around the country, that number is like in the 40s. We said, "Okay, relatively speaking, we're doing pretty good." We said, "Wow, there's a lot of white spaces there." Our eyes were drawn to the white spaces because those white spaces, what it represents is that those patients did not have the opportunity to have a comprehensive echo.
That means they were denied the opportunity to have the highest level of accuracy and hence potentially be at risk for underdiagnosis or misdiagnosis. You might say, well, it's only 9%, but if you look at population data, we're talking like 20,000 patients. That represents about 1,800 patients. What we did is we said, "Okay, let's take a look at the compliance, linking it to the sonographer level. We know exactly who's doing what. We can quantify that, and we know who needs to do the work." We were able to do that, provided that it's feedback. We repeated the evaluation, and you can see now the compliance improved to 93%. First of all, we already started out to the 91%, so it was already pretty high.
If you have other places where they start lower, like in the 40s, that interval net profit will be much greater. Even for us, 91 to 93, I know it may seem like it's only 2%, but 2% for, again, 20,000 large population health data, that represents 400 patient lives. To be honest, one patient, to be honest, is enough to matter. I mean, that one white space could represent the patient I talked about in the cardiogenic shock that was missed because they didn't have a comprehensive echo. That was the first insight. The next insight that we uncovered was that we identified significant variability across locations. You can see here the Site A and Site B, looking at the compliance. They're both Swedish, but they're different campuses. Site A represents the main campus, and then Site B is about 30 minutes away.
You can see that there's quite a significant difference in variation there in the same Swedish name, but different campuses. This is really, it was a jaw-dropper for us when we saw this because we had known Site B had some issues, but we didn't realize it was this bad. All these white spaces could be patients, again, being denied opportunity to have a proper diagnosis. Then hence be at risk for being undertreated as a result of underdiagnosis. Imagine if geographically the patient goes to Site A, they have the better opportunity, whereas if they go to Site B because of geographic reasons, well, that basically denied the opportunity to a proper comprehensive evaluation. This is a big deal. For patients, they don't know. From a patient perspective, they're trusting the name Swedish. They're not realizing there's a discrepancy.
It's not really fair to the patient. Anyway, we looked at this and we said, "Wow, this is really important." We realized how critical it is that we want to make sure that quality standards are really consistent across all campuses under the one name, Swedish, under the same system, Swedish. This is really critical data that really motivated us to start talking with leadership at the higher-ups to really start the conversations of getting the Site B up to speed. Without this kind of data, those conversations could have been quite uncomfortable. Now with this fact-based data, we were able to really start those conversations and really get things going. The last thing with the insight was the importance of care pathways. We heard all about the underdiagnosis. Right. We heard about this and how patient mortality.
I mean, that's just incredible data. How can we do better? How can we actually make a difference in patients' lives and save patients' lives and get the diagnosis and the treatment up? These are care pathways that you see here. The first red bar, 100%, represents patients that met one or more of those parameters. That met severe criteria. Out of them, 64% were diagnosed with severe AS, 25% with moderate. You can see how many went on to get referred and how many went on to get therapy completed. When you look at this data, we realize how critical, because this is the data that's so important for us as cardiologists because we're actually needing the data for the sake, for the survival of our patients.
We can consider how important it is to develop care pathways for the severe and the moderate patients so that for the severe patients, we would have care pathways in place that would ensure that once somebody's diagnosed with severe AS, that there's prompt recognition, communication, then referral to the cardiology team to make sure these patients are getting timely treatment. The same thing for the moderates. For the moderates, there should be care pathways in place to make sure that these patients with moderates are getting yearly echo evaluations so that once they hit severe, they are promptly recognized so they can be treated. A clinical story I can share with you guys, we were just discussing this not too long ago in our TAVR conference. We sit around the table, specialists, surgeons, cardiologists, imaging, and we all discuss patients because they're very complex.
You need a team to make a decision. We had a patient that presented, 90-year-old patient with multiple comorbidities, presented with shock state, an echo that shows severe aortic stenosis and ejection fraction of 15%-20%. Patient is in heart failure, also has pneumonia, has a whole bunch of stuff. Now we're trying to figure out what do we do with this patient. We said, "Well, let's take a look. When was this patient's last echo?" When we looked, we saw that the last echo the patient had was four years ago when the patient was diagnosed with moderate aortic stenosis. From four years. Within this four-year period of time, there was no other echo done.
Had there been in place a care pathway that assured that there were yearly evaluation with echo, then once this patient would have hit the severe, then this patient could have been recognized and treated before the low EF, before the heart failure, before the cardiogenic shock, and perhaps this patient's life could have been saved. Unfortunately, this patient did not make it because he was just too sick. Anyway, stories like these that really show us how these care pathways really serve as opportunities to save patients' lives. The next slide that you can see, this is just recent. We just got a hold of this days ago. That you can see when we take a look at what therapy is being implemented on our patients, more than 50% is medical management. I mean, there's no medical management of aortic stenosis. Medical management equals no management.
This is a big deal. We said, "Well, who is really taking care of these patients?" You can see here in the bar that two-thirds of these patients that are being referred, the care stops with the general cardiologist. They don't get referred further, unfortunately. This is new data that we are really now having a lot of opportunities that we can work upon. I will end by saying that as a cardiologist, one of the most frustrating things, I'm sure Matt will agree, is that one of the most frustrating things for us is when we get to patients and it's too late, or when things got missed, like the cases I showed, because the quality is substandard. This is really heartbreaking for us because then we can't really help anyone.
The way that we see cardiac care is by partnering up with institutions across the country, that this can really serve as a platform over which we could really powerfully transform the way we take care of patients by shifting to population health care. Which allows us to have the visibility of data, this transparency, which allows us then to identify these opportunities as you saw, which can be used then to bridge the diagnostic and the therapeutic gap and hopefully save patients' lives because all patients really deserve the highest possible quality. Thank you so much for your attention.
We got about 10 or 15 minutes to do some Q&A. There are microphones that'll be around. Go ahead and raise your hand and we'll give you a microphone.
Thanks. Hi. Thanks. Isaac Rowe from Goldman Sachs. Doctor, I really enjoyed the first part of your presentation. I thought you made an interesting comment around access to the technology. It's something that's come up I think a little bit in the recent MEDCAC conversations as it relates to giving equal and fair access to the marketplace for TAVR. There's a little debate about whether or not that should happen. I'd just like you to weigh in on how you think that's going to play out just speaking from your own region. Maybe both of you can comment on how you think that's going to play out, because it seems like a big part of tapping into the market is just giving the technology broader access.
I think it was for you.
He said Doctor. I mean, I can give you my perspective. There are clearly two sides to every argument. My perspective is that increased access leads to better outcomes. That's my perspective for patients. I think that that's played out previously with other technologies. It's played out with PCI, for example, in the U.S. C-PORT, which was a trial, big trial to look at should we be able to have STEMI care and PCI. This was PCI care in smaller hospitals. The answer was yes. Patients do better, do just as well at those institutions. I think the technology for TAVR has gotten to the same place.
The closer you keep a treatment to a patient's care, I think the more likely you are that patients are going to say yes to that treatment and get the care that they need. From my perspective, I think being closer to the patients is the right thing, which means dispersion of technology. You have to do it responsibly. Proctoring is important. Education's important. Quality checks are absolutely important. For TAVR and for SAVR both, there are registries. For a time, I ran both of those analytic groups at Duke. They are solid registries that track outcomes. My opinion is that the right way to judge programs is not based on their volume but rather on their outcomes. If you have a small volume center that's doing excellent outcomes, why would you shut that center down?
If you have a high volume center that's doing poorly, you can do as many TAVRs as you want. If you're killing people, you ought to be re-educated. I think we ought to be using prior experience to dictate who stays open. To answer your specific question about the MEDCAC, I watched it on video. I didn't go. I was on call that day. I definitely weighed in with public comments. I know the CMS folks who are trying to wrestle with, grapple with this issue very well. I've had multiple conversations with them. I've been very clear about my opinion, which is that we need open dispersion of the technology. For my money, I think that industry has done the right thing with making sure that that's a responsible rollout.
I don't see any reason to believe that it's going to be any different moving forward. Thank you for the question.
Maddie, do you have a different view?
I think you answered it perfectly. I very much agree with everything you said.
Perfect. Yes.
Hi, Glenn Novarro with RBC. This is more for you, Don. In Dr. Brennan's presentation, 65% of patients that are coming through the system are still not getting treatment. What is Edwards doing to get that number higher and to educate the cardiologists in the system? Thanks.
Thanks for the question. Some of this, Larry, tomorrow in his talk, he'll get after, but I had one slide that kind of illustrated a little bit of our own journey. As we've looked at this, we think there's real opportunities for education to patients, and so that's a theme. I think there's a lot of education interaction with physicians and general cardiologists that don't always or can't always put signs and symptoms together in a way that would motivate the kind of action that Matt and Maddie talk about. We've increased our investment in those areas where we can't do it on our own. Partnerships with physicians and groups to try to get that out, we think is going to be key.
I would expect we're going to continue to learn what works and what's effective to get the hazards, I guess, as Matt called them, off the road. Simply, it's incremental investment that we're making.
Thanks. A question for Dr. Brennan. I was just wondering, curious that the slide that you put up that 10% of generalists are able to diagnose aortic stenosis through auscultation was shocking. If you could just let us in on what percentage of patients do you think you examined with severe aortic stenosis you could diagnose just by auscultation? Should we be assuming that generalists are not even laying the stethoscope on the patient's chest? Lastly, some of us was in the backroom with the Eko technology. Is there any drive that elderly patients or every patient should have a non-physician kind of automated technology to auscultate during their annual physicals? Thanks a lot.
I'd say yes. Be done. No, absolutely to your last question, that's the first time that I've seen that technology. That's exciting to me, and I think it's exciting beyond rolling it out to primary care docs. I think it potentially has impact if you roll it out to a CVS clinic that has a blood pressure cuff. I mean, every Yes. It should be. That's a phonocardiograph. We used to do those back in the day. I would say we really got away from them. Absolutely, that would help with diagnosis. Personally, I would say that I'm reasonably accurate at diagnosing severe AS. I'm occasionally tripped up by a bicuspid, seem to have this S2 that we listen for, and sometimes I get tripped up by those.
I would say the way that you teach yourself is you listen to the chest, you go back and look at the echo report. Remember, I did an additional five years of training beyond my internal medicine residency, right, to become an interventional cardiologist. Paid a special interest and a special attention to diagnosis of valve disease through physical exams. I would say that I'm pretty good, but nowhere close to perfect. We hear something, ideally, I think if a primary care doc could just, or the device, the Eko device could simply say there's a murmur. Give the patient an alert that says, "You've got a murmur, get an echo." If we could take it that far, we would expand tremendously these numbers.
The difference between the 30% treatment rates that I showed and the 10% that Don showed is the difference between diagnosed and undiagnosed, right? I think we can make a huge difference in our treatment rates if we were to do that. Yeah, I see a role for that device, an exciting role for that device. It takes time. When I came out of medicine residency, no, I couldn't do this. I couldn't do this.
Let's go this side.
Thank you very much. Jason Mills, Canaccord Genuity. For both doctors, what do you think the single most impactful thing that can be done to
Let's say double the number of TAVRs that go into patients in any given year by any constituency, whether it be manufacturers improving the devices, imaging technologies, improving the care pathways. I know I'm putting you on the spot, and there are probably many things that could be done collectively and maybe a couple of things together. If you had to pick the single most impactful thing that you think that would impact the number of patients that get these life-saving devices, what would it be?
Maddie, you want to go first?
I'm going to say, because actually we were looking at this data, we have our structural team, we got together, this is exactly what we were looking. If I were to choose a top thing, because we noticed, and you saw the data, that it really depends who you refer to. It really depends on that, because if you refer to valve specialists, the chances are you're going to probably get a much better therapy. Valve specialists are really trained into teasing these things out. Are you symptomatic? Well, if you're not having symptoms, what are you like, really getting to the bottom of, are you truly symptomatic or not? There's all these tests that can be done, like stress tests, to really uncover. There's also indications for surgery even though you're asymptomatic. There's a lot more awareness with valve specialists.
Whereas in general, when we looked at our data, the cardiologists that were not in the structural arena, they were not referring because they were calling the patients asymptomatic. When we looked at our doctors, well, they were asymptomatic, but their Vmax was in the critical, which means, yeah, that does meet criteria. They were asymptomatic, but they were not really asking the questions. They were not really, can that patient do a stress test? They were not following the patients properly. It really does matter who you refer to. I think there's a lot of room for education. What we did in our facility, when we saw the data that we can see that medical management is higher in the general cardiologist versus a valve, we actually then started educating the cardiologist. Just educating and educating, showing data, showing why you've got to do more.
Ultimately, my personal feeling is that patients would be better served if they go to the valve specialist, the valve team, because they have the higher level of knowledge and updated information and the guidelines, and really take the expertise in this area.
Matt?
I think that would absolutely move the bar. I think so. There's a lot of low-hanging fruit here, a lot, and it's because we designed the system with old technology in mind. There's a lot of low-hanging fruit. If I had the dollars to invest, I'd invest them in direct-to-patient communication, personally. That would be the importance of your physical exam if you have been told that you have a murmur, the importance of an echo, and critically, don't leave it at a one-time echo. Because what I see more than any other issue is that the patient had a mild aortic stenosis diagnosis. They were told that they were fine. Five years later, they're in the ICU. Don't leave it at that one-time echo.
It would be a direct-to-patient education campaign if I had to put. There's so much low-hanging fruit here, that's if I had to put one.
Yeah. One final question. Bruce?
I guess in the last slide, it showed that there are like 60% of the suspected patients did in fact have severe stenosis. It was confirmed, 20% were ultimately treated, 24% were medically managed. Why were those 24% medically managed? That seems like the high-yield thing where you don't have to change anything because they're already in the bucket.
When we looked at this data, and we saw these patients weren't actually getting referred to the cardiologist. The cardiologist, they were having to make decisions, do these patients need to get treatment or not? The problem is that the general practitioner, the general cardiologist were making the call on that. I think the problem with that is that they don't, number one, they're not asking the right questions about the patients. Number two, they don't have the expertise and the knowledge bases that, for instance, there are criteria that even though they're asymptomatic, you actually can still have surgery. How do you know you're truly asymptomatic? There's tests that can be done.
I think the problem with our data is, what we found when we actually looked, we really looked in great detail, is that cardiologists are just not as expertise in really teasing out the symptoms and as aware of the guidelines. This is what's going on around the country. Patients are being under-managed.
Those 24%, to be clear, met guidelines for intervention.
Those patients were severe aortic stenosis. Whether or not they met guidelines depends on, you have to go back to each patient to find out were they symptomatic? What are the specific cutoffs? We don't have the information if they met the guidelines, but it's severe aortic stenosis.
What we found from the Optum data is that 60% or 67% of patients who meet those guideline definitions, severe and symptomatic, aren't being treated. The 67% of patients who meet those guideline definitions, severe and symptomatic, aren't being treated. It's a large group of patients. I just finished up this past year a big grant with PCORI, which is the Patient-Centered Outcomes Research Institute to create a shared decision-making tool online for patients to help patients and families understand the disease better. I agree that educating our docs. Cardiologists tend not to be stupid people. They tend to be some of the higher achievers in internal medicine residencies to get selected for that field. They can be taught. I think for a general cardiologist, I do think that we absolutely need to direct, again, education at those general cardiologists. Also, again, back to the patients themselves.
I think if patients are calling for treatment or if they're asking the right questions, they're ultimately going to get treated. That's been my experience.
Okay. I think what we're going to do is to end the formal program at this point. Thanks very much, Dr. Brennan, Dr. Petrescu, Don, for your informative session. They're going to hang around for a while, and you are also welcome to do that. There's some coffee and dessert, and we're happy to mingle with you for a little while longer. Take care.