High-Quality TAVR: Does a Minimum Volume Requirement Still Make Sense?

CHICAGO, IL—Interventional cardiologists are grappling with how to best assess the quality of TAVR programs that have cropped up in the United States, with many arguing the focus on volume alone is an outdated metric that misses the bigger picture when it comes to clinical outcomes. 

The debate over the volume-outcome relationship is particularly timely given that the US Centers for Medicare & Medicaid Services (CMS) will convene a panel of the Medicare Evidence Development and Coverage Advisory Committee (MEDCAC) in just a few weeks to examine data related to the procedural volume requirements of a structural heart program.

For some, there is a need to maintain minimum standards and to ensure that those currently performing TAVR have adequate experience—with volume standing in as a surrogate for quality—to handle straightforward procedures as well as complications that may arise.

Others, however, contend that if new volume requirements are instituted, some hospitals will be forced to close, which will lead to a disruption in care and limited patient access to treatment. Currently, there are 580 sites performing TAVR in the United States, but if new minimum volume requirements are adopted, the number of centers performing the procedure could decline by nearly 40%, according to one estimate.

Speaking during a policy forum session at the 2018 Structural Heart Disease Summit, Martin Leon, MD (NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY), called the proposed volume requirements “serious stuff,” noting that if the current volume recommendations are increased, “it certainly runs the risk of changing the dynamics of how and where patients will be treated.”

For Leon, too many patients with aortic stenosis are already neglected, a problem that will be compounded if procedural volume thresholds are increased.

“In the US, there is not a single state where the treatment of diagnosed, severe symptomatic aortic stenosis rises above 40% in the first year,” said Leon. “The average is 24%. So we’re saying that three out of four patients with severe, symptomatic aortic stenosis who may have as high as 50% mortality at 1 year are not being treated with any aortic valve replacement procedure. That is reality.”

Reimbursement decisions, such as those considered by MEDCAC, and tying those to decisions to hospital volume requirements “will absolutely affect the ability to provide care for patients,” said Leon.

Consensus Statement on TAVR Operator and Institutional Requirements   

Last summer, a group of experts from the American Association for Thoracic Surgery (AATS), American College of Cardiology Foundation (ACCF), Society for Cardiovascular Angiography and Interventions (SCAI), and Society of Thoracic Surgeons (STS) published a draft consensus statement suggesting only centers performing a minimum of 50 TAVR procedures per year (or 100 within the last 2 years) and a minimum surgical aortic valve replacement volume of 40 cases per year (or 80 within the last 2 years) be allowed to perform transcatheter valve replacement. 

Carl Tommaso, MD (NorthShore University Health System, Chicago, IL), who co-chaired the AATS/ACCF/SCAI/STS writing committee along with Joseph Bavaria, MD (Hospital of the University of Pennsylvania, Philadelphia), said the finalized recommendations will be published very shortly, hopefully within the month, but would not reveal the finalized details.

To TCTMD, Tommaso said that when they published their first multisociety expert consensus statement on operator and institutional requirements in 2012, “TAVR was just a toddler.” The previous consensus statement suggested hospitals perform a minimum of 20 TAVR procedures (or 40 procedures every 2 years) and 25 surgical aortic valve replacements per year.  

In the US, there are training programs currently turning out experienced operators, said Tommaso, so he suspects there will be no shortage of experienced TAVR physicians. In the future, Tommaso said they are hoping to move beyond institutional volume requirements to outcome-based metrics. “This is going to take some time because we need to develop risk-stratification models in order to get accurate measures,” he said. “TAVR volume will continue to be used as a surrogate for quality, particularly in relatively new sites climbing the learning curve ladder.”

That said, in the years to come, volume will be less important and the focus will be on clinical outcomes as the benchmark, said Tommaso. In other, more established fields, such as hip replacement surgeries, quality is not measured by hospital volume but by clinical outcomes, he noted.

Megan Coylewright, MD (Dartmouth-Hitchcock Medical Center, Lebanon, NH), said she hopes that when MEDCAC reviews the national coverage decision for TAVR they focus not on institutional procedural volume but instead on value and best practices, including shared decision-making. “That’s more important to ensure quality of care than a volume cutoff,” she told TCTMD. “The volume requirement may get in the way of the dissemination and communication of best practices and may limit access.”     

‘Bad Things Can Happen’

Several studies presented at the Structural Heart Summit focused specifically on factors, including volume, that impact clinical outcomes. John Carroll, MD (University of Colorado School of Medicine, Denver), presented data from STS/ACC Transcatheter Valve Therapy (TVT) Registry showing that 70 US sites in 2016-2017 performed fewer than 25 TAVR procedures annually. These low-volume centers, on average, had a higher mortality rate than centers performing more cases.

To TCTMD, Carroll said that despite the introduction of second- and third-generation TAVR devices, expanded indications into lower-risk patients, and increased operator and institutional experience, there remains an association between volume and clinical outcomes. “I would be worried about TAVR getting a black eye if we start seeing the opening of more and more low-volume centers,” he said. “It could change the field, and it’s something that should be of concern to all the stakeholders. Nobody wants to see that.”

While he agrees that direct measurements of quality are important and should drive quality-improvement programs in all sites performing TAVR, Carroll also stressed that minimum volume standards are needed. “I think it would be wrong to do away with standards and say, ‘Oh, it’s such a simple procedure, it’s low risk now,’” he said. “That’s simply not true. Yes, outcomes have improved, but it’s still a major procedure and bad things can happen.”

In another presentation, however, Mark Russo, MD (Barnabas Heart Hospital at Newark Beth Israel Medical Center, Newark, NJ), presented data exploring the volume-outcome relationship with newer devices, specifically the Sapien 3 (Edwards Lifesciences) valve. Also using data from the TVT Registry, Russo reported that the 30-day mortality rate with Sapien 3 was not significantly different among centers performing up to 25 TAVRs per year compared with hospitals doing more than 100 procedures. Similarly, there was no difference in 30-day stroke rates between the groups, he said.

“After maturation of the TAVR procedure and introduction of an enhanced new technology, there was no meaningful relationship between volume and clinical outcomes, even among sites that hadn’t previously done TAVR,” said Russo. “This is an important point. Even new sites can expect to have excellent clinical outcomes early in their experience. It’s not OK to have inferior outcomes just because you’re new to the procedure.”

Like others, Russo argued high volume does not always translate into excellent outcomes, with some low-volume centers achieving solid results while some high-volume centers struggle with national benchmarks. “I think it’s important we use metrics other than volume to measure quality, or shouldn’t use volume as a surrogate for quality, because if we do that we might restrict TAVR availability in lower-volume, high-quality centers as well as underserved populations,” he commented.

Aaron Horne, Jr, MD (Methodist Dallas Medical Center, TX), presented data on 54 TAVRs performed at his institution over a 2-year period. With outcomes and complication rates equal to the national average, Horne noted that nearly one-third of the patients treated at his hospital were African-American. By contrast, just 3.8% of all US TAVR patients are African-American. If judged solely on arbitrary volume metrics, Horne said these quotas will harm patients, particularly racial/ethnic minorities already undertreated.

Nobody Is Closing Clinical Centers

In another presentation, A. Peter Kappetein, MD (Medtronic; formerly Erasmus University Medical Center, Rotterdam, the Netherlands), estimated that approximately 210 US centers wouldn’t meet the proposed AATS/ACCF/SCAI/STS volume requirements, which would translate into 39% fewer centers. Some states would be hit particularly hard, he said, noting that only 12 of 30 centers in Texas would meet the minimum requirements and that much of the western part of the state would not have a working TAVR program. In Kentucky, just two of nine centers would meet the minimum annual volume requirements.

Wyoming ranchers drive a hundred miles to go to Walmart. Carl Tommaso

In the US, Wyoming is the only state without a TAVR program. However, Tommaso noted there are TAVR centers in neighboring states, including Idaho, North Dakota, South Dakota, Montana, and Colorado. “And as John Carroll noted the other day, Wyoming ranchers drive a hundred miles to go to Walmart,” said Tommaso. “That’s not a problem for them as it would be for someone in Evanston. They’re not driving a hundred miles to do anything.” 

Carroll, a member of the AATS/ACCF/SCAI/STS writing committee, said there is no suggestion centers should be closed if they do not meet the annual volume requirements. “Instead, these centers should be allowed to grow,” he told TCTMD. “What would be the major impediment to growth of a low-volume center? That’s allowing more and more centers around them to open so that we start diluting the experience. It also contributes to the escalation of healthcare costs because of the redundancy of services.”

To TCTMD, Tommaso also stated that the purpose of their consensus statement is not to close clinical centers given the need to ensure access to care. “If the center in Anchorage, Alaska, is a low-volume center but they have good outcomes, we don’t care,” said Tommaso. For low-volume/poor-outcome centers, there will be a need for an external review.

“Part of the problem may be that a low-volume center might decide to increase their volume by doing very high-risk patients,” he explained. “We don’t want that. If they’re high-risk patients, send them down the street.”