New TAVI Centers Clustered Together, and Outcomes Suffered: TVT Registry

In the first several years after the introduction of TAVI, centers performing the procedure were not distributed evenly across the United States but instead clustered regionally, according to data from the Society of Thoracic Surgeons/American College of Cardiology TVT Registry. In terms of performance, increased density of TAVI sites was associated with lower procedural risks yet also linked to worse clinical outcomes,.

The findings, published in the August 24, 2021, issue of the Journal of the American College of Cardiology, raise questions about whether the goal of balancing the need for access to TAVI with maintenance of quality and safety was achieved as the procedure rolled out across the country.

“Any policy decision is going to have unintended consequences and keeping an open eye to what those downstream consequences might be and what they might mean for TAVR centers, for access to care, and for quality of the procedure is going to be critically important,” lead author Javier Valle, MD (Michigan Heart and Vascular Institute, Ann Arbor), told TCTMD.

As it turns out, they really followed more-economic principles than true access or needs. Javier Valle

The framework set up by professional societies and government agencies, like the US Centers for Medicare & Medicaid Services (CMS), to guide the dissemination of TAVI was well-intentioned, he added. However, “this analysis suggests that minimum standards, minimum numbers, aren’t the only way to judge quality and that there needs to be some more consideration and reevaluation of the ongoing evaluative process for TAVR centers for both opening and maintenance of care,” said Valle.

D. Craig Miller, MD (Stanford University Medical School, CA), took a starker view of the findings, calling them “explosive” and “highly sensitive dynamite.” He argued that this analysis supports shutting down centers that aren’t up to snuff by withholding CMS reimbursement. In an accompanying editorial, he and David Holmes Jr, MD (Mayo Clinic, Rochester, MN), assert that TAVI dispersion in the US has been “irrational,” with the clustering of sites in population centers that only dilute procedural volumes and lead to poorer outcomes.

“We shouldn’t have so many doing so little so close together,” Miller told TCTMD, adding that shuttering poorly performing centers “is absolutely mandatory and it’s essential, but the key question is: will it happen?”

In comments to TCTMD, however, Martin Leon, MD (Columbia University Irving Medical Center/NewYork-Presbyterian Hospital, New York, NY), would not take it that far.

“I view this as being fascinating data which gives us a good understanding as to what the dispersion of TAVR was,” said Leon. “It was less than perfectly rational. I don’t believe that I would go as far as to say it was irrational, but it was suboptimal. And I think that as we get quality metrics, we’re going to have a better understanding of how to both select sites and maintain sites based upon not just the absolute volume of a center, but based upon outcomes and quality indicators that will justify centers being supported or being adjusted.”

Dispersion Has ‘Definitely Not Been Equitable’

Early consensus documents from professional societies with an interest in TAVI called for a “rational dispersion” of the technology, Valle et al note. “This expansion of services has occurred under recommendations from medical societies and requirements from payers to establish procedural minimums to maintain operator and institutional proficiency.”

We shouldn’t have so many doing so little so close together. D. Craig Miller

To examine whether the appropriate balance between access and quality was achieved, the investigators identified TAVI centers participating in the TVT Registry between November 9, 2011, and December 31, 2017. During that time, the number of sites performing the procedure increased from 198 to 556, with a corresponding increase in the density of centers from five to 12 per million Medicare beneficiaries. The median time it took to drive from an existing TAVI site to a new one fell dramatically from 403 minutes to just 26 minutes (P < 0.001) over that span, indicating that new centers tended to pop up near others rather than in areas lacking those services.

“As it turns out, they really followed more-economic principles than true access or needs,” Valle said. Dispersion of the technology, he added, has “definitely not been equitable.”

The researchers then dug deeper into the interplay between case volumes, site density, and patient risk profiles and outcomes.

Consistent with prior research evaluating the volume-outcome relationship in TAVI, higher case volumes—though associated with increased estimates of patient risk—correlated with lower risks of 30-day and 1-year mortality and a composite of adverse clinical outcomes. Similarly, increases in case volume over time were linked to better outcomes.

The story was different when looking at the density of TAVI sites, however. Despite being tied to lower estimates of patient risk, higher site density was associated with a greater risk of 30-day mortality. Increasing site density over time correlated with greater mortality at both 30 days and 1 year.

As the density of TAVI programs increases, case volumes at individual centers decline, Valle noted, saying that “it would make sense that you’d start to see some concerning signs about procedural quality.” He added that this is “something that I think needs to be evaluated more as we continue to see TAVR expand.”

Rethinking Requirements for TAVI Programs

Valle said the findings of the study are not very surprising since the dissemination of PCI programs played out similarly, with centers clustering in major metropolitan areas.

As for how the results play into the discussions in recent years about procedural volume requirements for new and existing TAVI programs, Valle said “what these findings really suggest is it’s not just a numbers game. There’s more to quality than just case volumes.”

Moreover, the study shows “that if the intent was to provide access, that’s not what happened,” Valle said. “So I think there needs to be a reconfiguration or reconsideration of what requirements are there to open up a TAVR program and to maintain quality care. . . . This is a new technology. Saying these are the requirements and then sort of setting it and forgetting it is not what we should be doing. It should be a continuing evaluation and a continuing evaluative iterative process on how we evaluate quality, and that’s what [the TVT Registry] is set up for.”

Indeed, investigators involved with the TVT Registry are working on integrating new quality metrics beyond volume, which will help guide further dispersion of TAVI, Leon said. He added that there are low-volume centers with excellent results as well as high-volume centers with poor results. “When you bring quality into the mix, then I think you can also create the remedial measures to be able to improve care in those centers that are underperforming,” said Leon.

Whatever you say about TAVR has to be equally applied to the alternative treatments like surgery. Martin Leon

Miller was skeptical about the possibility of remediation for poor-performing centers, however. “It sounds good on paper, it’s altruistic, and it’s the American way, it’s democratic. But I don’t think it’s going to make a difference. There’s just too much selfish self-interest here. And it’s all money,” Miller said, adding that there are already many centers performing TAVI—and SAVR for that matter—without meeting minimum requirements. “We’re dealing with greed and avarice and money, and that makes it so stinky.”

When additional data on the proportion of patients meeting the “alive and well” outcome at 1 year within the TVT Registry are available, it will show that many centers are falling short and should not receive further CMS reimbursement for TAVI, Miller predicted. “Will that ever happen? Most people say no, it’ll never happen. It’s just too much lobbying, too much financial pressure, too much at stake for the hospitals and the docs and the industry. It shows you how corrupt US medicine is.”

Valle didn’t advocate for closing centers failing to meet certain quality benchmarks, instead saying “there needs to be some further evaluation as far as quality and making sure that TAVI centers in high-density areas really look at their outcomes and make some tough decisions about what they need to do to maintain reasonable quality standards.”

For Leon, remediation of centers, rather than closing their doors if they fall short of certain standards, “is the appropriate thing to do,” and mechanisms to do so are currently being discussed while the field awaits the new quality metrics in the TVT Registry. But if one were to advocate for shutting down TAVI centers, the same rules would have to apply to SAVR centers, Leon stressed.

“The reality is, whatever you say about TAVR has to be equally applied to the alternative treatments like surgery,” he said. “There may be an indication for low-volume underperforming sites that you cannot improve with remediation to be either shut down or not reimbursed, but I think that should be for TAVR and for surgery. It should not apply just to TAVR. It should apply to any therapy for aortic stenosis that involves a procedure.”

Valle agreed that the issues highlighted in this study are “probably true in a lot of other medical devices, a lot of medical procedures that have been disseminated. . . . I don’t think this is unique to TAVR.”

What Are the Alternatives?

In any discussion like this, Leon said, one has to take a step back and consider the treatment alternatives.

“One of the biggest problems still is that we don’t identify quickly enough and treat promptly enough patients with severe symptomatic aortic stenosis, and that’s by either surgery or TAVR because medical treatment is not an option in patients that have the opportunity for valve replacement,” Leon said. “And even though the outcomes weren’t as good as we had hoped in some of these centers, I would imagine if these centers didn’t exist and fewer patients were treated, that the overall patient outcomes would be even worse. So you just have to have that kind of balanced perspective when you make some of these broad conclusions.”

The population of patients with treatable aortic stenosis in the United States is expected to double in the next 20 years, Leon noted. “So the availability of having a less-invasive approach to allow these patients to be treated and to have access to treatment is nontrivial.”

Still, he acknowledged, there is an opportunity to improve how TAVI is further disseminated.

“I think there was some overcompensation, and I think that we can do it better in the future,” he said. “Now that we have these data, we’ll learn how to do it better in the future, and I think once we have quality metrics we’ll learn even more about what are the best ways to be able to regulate the distribution of both new and old technologies, including TAVR and surgery.”

Note: Leon is the founder and chairman emeritus of the Cardiovascular Research Foundation, the publisher of TCTMD.