Vast Numbers of Severe Symptomatic AS Patients Untreated? A Plea and Some Pushback
By one estimate, 65% don’t get referred for aortic valve replacement, posing questions for who and how to track patients with early AS.
CHICAGO, IL—A fresh approach to managing the vast majority of patients with aortic stenosis (AS) who are never referred for treatment, or are referred when it’s too late to intervene, could radically reduce population-level mortality among people with this progressive disease, according to J. Matthew Brennan, MD (Duke University, Durham, NC), who made a passionate plea for change at TVT 2019.
“There have been some amazing presentations here,” he said. “We’ve been talking about incremental benefit, changes in our technique that could save stroke risk and affect our patients in the 1-2% range. I’m about to talk to you about something that should affect our patients on a population level by a 70% mortality reduction. So, it’s a little different scale.”
Over the past decade, TAVR has made inroads into the large number of patients with aortic stenosis who never get treatment in the United States, but even today, 65% of severe symptomatic AS patients remain untreated, Brennan noted. That’s despite the fact that 60% of untreated patients are actually followed by cardiologists.
Pulling data from a range of different studies, Brennan showed figures indicating that rates of referral to aortic valve replacement (AVR) vary markedly by cardiologist, with some referring almost all of their patients and others referring none. “This is such a big deal that one of the biggest determinants of whether a patient gets treated is which cardiologist’s office they walk into,” he said. “That’s as big of a determinant [of treatment] as their age, for example. That seems like a little problem with a terminal illness—leaving it untreated.”
Other work, he continued, has shown that patients who don’t get treated or are managed by a “less aggressive” cardiologist face a 20% increased risk of mortality in the first year after their diagnosis. Race, age, and gender all play a role as well, with non-Hispanic black and female patients both being 20% less likely to be referred for AVR, while elderly patients are 40% less likely.
“We’re just seeing the tip of the iceberg,” Brennan told TCTMD, noting that the numbers he cited are among patients who actually get a diagnosis. The reasons for this are complex and are partly due to referrals, partly due to cultural views, and physician bias can’t be ruled out. “We don’t know,” he said simply. “[Physician bias] has been shown in other fields so it would make sense that it would play out here, but we don’t have those data to demonstrate that. But in the community it’s clearly an issue.”
The murmur is missed, the echo is misinterpreted, the symptoms are misappropriated, the referral is delayed, the patient is misinformed, the treatment is delayed, and the lives are lost. J. Matthew Brennan
As for why patients don’t get referred for valve replacement, he said the reasons are multifactorial: “The murmur is missed, the echo is misinterpreted, the symptoms are misappropriated, the referral is delayed, the patient is misinformed, the treatment is delayed, and the lives are lost. And clearing those hazards is going to save lives.”
Brennan noted that many cardiologists and trainees—not to mention primary care physicians—are no longer comfortable using stethoscopes. “So we fall back on the echocardiogram, because the echocardiogram is always right. Only, the echo is only as good as the people interpreting and collecting the data.” Here, he showed numbers from 18 hospitals in the CardioCare network indicating that 48% of patients who met the criteria for a diagnosis of severe AS are not accurately diagnosed on echo. Of those correctly diagnosed (52%), fewer still (43%) are referred for specialized assessment and just 29% are ultimately treated.
There’s also the outdated view that aortic stenosis progresses slowly. “This is not the indolent disease that we were taught about in fellowship,” he said. On the contrary, aortic sclerosis typically progresses to severe aortic stenosis in 7 years and to mild-to-moderate stenosis within 4 years. Delays to treatment cost lives, he continued. In a study looking at the one out of every four patients who waited 5 weeks from referral to intervention, that delay translated into an 8% increase in mortality.
A Multipronged Approach
The problems are “fixable,” Brennan argued, but the solution must be multifactorial. The most critical component is implementing programs and technology to systematically follow-up on patients, such that someone sent for an echo and diagnosed with mild aortic stenosis isn’t simply discharged back to their primary care physician. Rather, a diagnosis of “mild” or “moderate” should be accompanied by a reminder to the referring physician that echo follow-up is warranted within a specific period of time.
Brennan also called for more use of “objective treatment criteria” rather than subjective symptoms.
“When the tobacco farmer from Chatham County tells me, ‘I don’t feel I’m short of breath’ because he’s a stoic old guy, we can’t use that as one of our diagnostic criteria any more,” he argued. “I understand it’s part of the guidelines, but it doesn’t make sense. Why do we wait for the disease to progress and the ventricle to fibrose before we treat these patients?”
Better patient education is also urgently needed, said Brennan, pointing to analyses showing that one in three medically managed patients chose “no treatment” after being referred for aortic valve replacement. The most common reason cited was that they felt “the final decision about my AS treatment was not the right one for me” and only 15% of patients surveyed said they felt their heart valve doctors had involved them in the decisions about their care. This, said Brennan, suggests that patients “are not feeling informed and they are feeling listened to.”
What’s needed are better tools for educating patients and training providers to deliver it, he continued, along with refining technology and techniques so that patients are not denied care based on anatomic considerations alone.
We’re drowning in the clinic, and we just don’t have the capacity to follow these patients. We can send them back with a specific recommendation but don’t follow them longitudinally. James Hermiller Jr
Speaking with TCTMD, Brennan stressed that the problem is twofold.
“The first issue here is that we need to recognize that these patients are out there, in each of our systems. We need to go looking for them and put in place systems to try to counter the problem. Part of that includes routine follow-up, and part of that includes making it easier to get to treatment—whether that’s expanded capacity of the system or whether that’s diagnostic criteria. The other part is coming up with new techniques and technologies, and informing our patients so they can be involved in those shared decision-making conversations,” he explained.
In his practice, Brennan has started putting a specific recommendation in the echo reports, alerting referring doctors to the need to send their patient back for follow-up echo within a specific period of time. He’s also involved in a pilot project involving the Society for Cardiovascular Angiography and Interventions and the electronic health record (EHR) company Avalere.
“Whether it is an EHR, or whether it's a coordinator, in clinic, who has a list of people who are mild to moderate [and] who we know are supposed to be coming back, we can start to call them," he said. “This is going to increase our echo volumes, so good for us. But it’s also going to increase the diagnostic capabilities of the clinic, so good for our patients.”
To let patients fall through the cracks should not be an option. "We should be tracking these patients," he insisted. "It's a terminal illness, it's very treatable, so why aren't we doing this? Would we do this with cancer? No. We'd be tracking these patients and they'd be treated."
To TCTMD, Brennan noted that there's also an economic argument for tracking and treating patients. "This has been shown over and over to be cost-saving. First of all, the cheapest patient is the dead patient, but aside from that reality these patients end up using huge amounts of healthcare resources because they’re hospitalized at really high rates, they’re in their doctors' offices frequently, they’re using the emergency room, and then ultimately they’re in hospice care, and you don’t need that with these patients."
Coming Up for Air
Panelists following Brennan’s presentation applauded his bold appeal, but said they had reservations about how to implement the kinds of changes he proposed, particularly if the burden for follow-up falls on already overwhelmed TAVR physicians.
“We’re really just keeping our head above water,” said session co-moderator John Webb, MD (St Paul’s Hospital, Vancouver, Canada). “We’re getting patients referred with aortic stenosis, and we review the chart and say: well, it’s not truly severe yet so why would I get involved?” Then he added, prompting chuckles: “I’m hoping in the future our surgeons will have more time.”
“We’re like you, John,” said James Hermiller Jr, MD (St. Vincent Heart Center, Indianapolis, IN). “We’re drowning in the clinic, and we just don’t have the capacity to follow these patients. We can send them back with a specific recommendation but don’t follow them longitudinally.”
This kind of problem marks a sea change for TAVR, Mark Russo, MD (Robert Wood Johnson University Hospital, New Brunswick, NJ), noted. The focus used to be on just getting patients through procedures alive. “Now we’ve gotten to the point where, on large scales, we can do these procedures safely at lots of different centers. But the procedure is the easy part; now it’s actually getting the patients on the table that’s become the challenge,” he said.
Whether the population of untreated severe symptomatic AS patients was as large as Brennan had estimated from claims data is also unclear. Speaking with TCTMD, David Cohen, MD (University of Missouri-Kansas City School of Medicine), estimated that 65% is more likely the “upper bound” of this hypothetical group.
Valve disease comes in patients of all shapes and sizes with all different levels of complexity and the valve isn’t always their only problem. So it is still very important to have the general cardiologist engaged. David Cohen
“We have a very active program,” Cohen said. “We get referred a lot of patients, and a lot we send out without treatment because they are not severe symptomatic AS, they are severe lung disease with mild AS, or they are heart failure with moderate AS. So these decisions remain nuanced and getting that out of claims data is not so easy.”
He also balked at the idea of structural heart interventionalists and surgeons taking on the burden of patient follow-up. “I think we have to rely on our general cardiology colleagues to play a part of the role here, if we give them a plan to follow,” he said. “I recognize it would be nice for us to completely own valve disease, but valve disease comes in patients of all shapes and sizes with all different levels of complexity and the valve isn’t always their only problem. So it is still very important to have the general cardiologist engaged.”
Hinted at in Brennan’s talk is the idea of intervening earlier in TAVR patients, something actively being studied in trials like EARLY TAVR. At TVT, Cohen reminded TCTMD that the community needs more data, and more time. “I want to know what we’re getting with treating them earlier, because one thing we know for sure that we get with treating them earlier is we start the clock for needing to treat them again,” he said. Currently valve care means intervening at the “golden moment. That’s the moment just before something bad is going to happen,” Cohen explained. “Because once you replace the valve they don’t just have aortic stenosis at that moment, they have a different disease: they have a bioprosthetic valve.”
He continued: “The EARLY TAVR trial will help and will tell us something, but what we’re beginning to understand from the whole dialogue at this meeting is that low-risk patients change so many parts of the equation with respect to reintervention and longevity. How do we deal with the inevitable first failure, the second failure? And we don’t have that well worked out.”
Brennan JM. Under-treatment of aortic stenosis in the United States: a coordinated path forward. Presented at: TVT 2019. June 14, 2019. Chicago, IL.
- Brennan reports being a consultant for Atricure and Edwards.