CMS Advisory Panel Splits on TAVR Volume Requirements

Confidence in whether the requirements are necessary hinged on how panelists interpreted the data on volume-outcome relationships.

CMS Advisory Panel Splits on TAVR Volume Requirements

If the Centers for Medicare & Medicaid Services (CMS) was looking for a clear consensus on the need for procedural volume requirements in its national coverage determination (NCD) for TAVR, it didn’t get it from yesterday’s meeting of its advisory committee.

Members of the Medicare Evidence Development and Coverage Advisory Committee (MEDCAC) heard conflicting interpretations of the evidence on whether volumes of PCI, surgical aortic valve replacement (SAVR), and TAVR do in fact influence outcomes in patients undergoing TAVR. That divide left its mark when the time came to vote on various questions dealing with the importance of volume requirements for TAVR programs.

Panelists were asked to weigh whether there was sufficient evidence to support procedural volume thresholds for hospitals to start or maintain TAVR programs, for operators to begin programs, and for heart teams to maintain such programs. They were also tasked with determining whether the benefits of having such requirements would outweigh any harms in terms of limited access to TAVR. The panel consisted of 12 members who participated in the discussion, nine of whom had votes that counted toward the official tally.

The average of most votes fell between 3 and 4 on a scale of 1 (low confidence) to 5 (high confidence). The weakest support was seen for the question asking whether the benefits of meeting procedural volume requirements to begin a TAVR program outweigh the harms (average 3.11).

On the flip side, panelists indicated strong support for a question asking whether there is sufficient evidence that a certain number of SAVR and TAVR procedures should be required for the principle cardiovascular surgeon on the heart team to begin a new TAVR program (average 4.33) and a similar question about volume of structural procedures for the principle interventional cardiologist (average 4.22).

Still, the wide range of responses for each question point to some uncertainty about the data regarding the importance of procedural volume as it relates to TAVR outcomes, which was hashed over at CMS headquarters in Baltimore, MD, yesterday.

CMS convened the MEDCAC meeting to help in its decision-making as it takes another look at the NCD for TAVR, which was released in 2012 and established requirements for coverage that included minimum volumes of SAVR, catheterizations, and PCIs that hospitals and heart teams had to meet in order to begin and maintain TAVR programs. Those requirements were consistent with recommendations found in a 2012 consensus statement from the American Association for Thoracic Surgery (AATS), American College of Cardiology (ACC), Society for Cardiovascular Angiography and Interventions (SCAI), and Society of Thoracic Surgeons (STS).

The goal of the MEDCAC meeting was to assess whether the evidence supports such requirements. The same societies that produced the 2012 consensus statement have already weighed in, recently releasing revised recommendations that maintain volume thresholds for TAVR programs while also shifting the focus to more direct measures of quality.

Panelists heard from several experts, some of whose presentations concluded that there is not a strong relationship between procedural volumes and TAVR outcomes. Others, however, concluded that there is indeed a clinically meaningful association between the two.

Volume Is Not a Useful Surrogate for Quality

Peter Pelikan, MD (Providence Saint John’s Health Center, Santa Monica, CA), one of the authors of the letter that spurred CMS to reexamine this issue, argued that because TAVR has become more streamlined and outcomes have improved since 2012, the NCD should be revised. He presented data disputing a volume-outcome relationship.

“In the age of electronic records where we can actually measure quality and no longer need to use volume as a surrogate for quality, I propose that the time is now to change the NCD and use quality not procedural volume as a requirement for TAVR,” he said.

Continuing to adhere to volume requirements, Pelikan argued, may induce centers to perform unnecessary PCIs or SAVRs to maintain their TAVR programs. In addition, he said, volume criteria create barriers to access by limiting the number of hospitals that are able to perform TAVR.

If such requirements are removed, Pelikan said, it would improve access to care by allowing new TAVR programs to open, enhance quality of care by allowing patients to receive treatment from their regular healthcare team, and boost outcomes by emphasizing quality across centers.

Volume Is Still Important

Leading the defense of volume requirements were Carl Tommaso, MD (NorthShore University HealthSystem, Skokie, IL), and Joseph Bavaria, MD (Hospital of the University of Pennsylvania, Philadelphia), the co-chairs of the updated multisociety consensus statement. They argued that there is indeed evidence supporting volume requirements, even as the field moves toward greater reliance on direct measures of quality.

“There’s a well-known and robust body of literature showing a volume-outcome relationship in almost every complex medical procedure,” Bavaria said. “This is true for TAVR as well.”

He added that the steady overall improvement in TAVR outcomes over the last 5 years has hidden the concerning signals of worse outcomes among low-volume centers. “Reducing volume standards would sacrifice quality for expansion of access, without any scientific evidence that 584—as we speak today—[US TAVR] centers is inadequate,” he said.

Tommaso and others said minimum procedural volumes are needed to ensure that quality can be reliably measured, and at least one panelist agreed. “I agree quality trumps quantity every day of the week,” Gregory Dehmer, MD (Carilion Clinic, Roanoke, VA), said. “I don’t think anybody would argue about that, but how can you measure quality with no volume?”

A Compromise?

Representing AdvaMed, a medical technology lobbying group, at the meeting, Martin Leon, MD (NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY), argued that contemporary data do not support a strong link between volumes of PCI, SAVR, and TAVR and outcomes, calling for a greater focus on direct quality measures.

He indicated that he agrees with much of the recently published consensus statement, but disagrees with the timing of the transition to predominantly quality-based—rather than volume-based—assessments of TAVR programs.

“The narrative from the consensus document makes good sense with clear goals to rely on quality metrics rather than crude site volume thresholds to determine TAVR and surgery performance and site readiness as a new or existing TAVR center,” he said. “The main difference in opinion is the need for acceleration in timing to the quality metric platform, without a burdensome and arbitrary increased volume transition period of 7 years, which will limit patient access.”

Direct measure of quality of care should begin immediately using data that are already available, he said.

Leon offered a compromise for AVR volume recommendations, saying that quality metrics should supplant volume thresholds; surgery volume requirements should be eliminated as a determining factor for new and existing TAVR centers and be replaced by a quality metric; PCI volume requirements should be kept but lowered; and TAVR volume requirements should be maintained at 2012 levels rather than increased as recommended in the new consensus statement.

“TAVR quality metrics should be integrated in the proposed new NCD to rapidly replace the need for volume requirements and to more closely monitor the clinical outcomes of all TAVR centers, especially the low-volume centers, with corrective measures for poor performance installed as needed,” Leon said.

That suggestion for a balance between volume requirements and more direct quality assessments seemed to resonate with several of the panelists, including Daniel Ollendorf, PhD (Institute for Clinical and Economic Review, Boston, MA). He said that after reviewing evidence for complex procedures within and outside of cardiology leading up to the meeting, “I felt like I needed to be convinced that there was not a volume-outcome relationship. I don’t feel convinced that there is not a volume-outcome relationship.

“All that being said, [and] this is not what we’re voting on today,” Ollendorf continued, “volume as the only surrogate for quality also makes no sense to me.”

Volume Requirements and Access to Care

One recurring topic of discussion throughout the day was whether procedural volume requirements for TAVR programs are harming access to care, particularly among underserved groups. There was no consensus on this issue.

During one presentation, Aaron Horne, Jr, MD (Cardiac and Vascular Interventional Group, Dallas, TX), board member of the Association of Black Cardiologists, argued that volume requirements do create barriers to access based on a number of factors, including geography, sex, ethnicity, race, socioeconomic status, provider preference, and hospital setting.

Leon, too, cited potential access issues, especially in the context of the anticipated growth in TAVR case volume as indications expand and the population ages.

Others disagreed. Serving as a representative of SCAI, Ted Feldman, MD (Evanston Hospital, Chicago, IL), said: “The society is confident that procedural volume requirements for TAVR programs outweigh the harms of limiting access to TAVR to only hospitals that meet these volume requirements.”

And Tommaso noted that the density of TAVR centers is greater in the United States than in high-performing European countries, and touted the “broad geographical access, with rare exceptions,” provided by US centers.

Panelists were generally in agreement that volume requirements might contribute to access issues, but that numerous other factors that affect access to cardiovascular care in general also play a role.

Disclosures
  • Pelikan reports no relevant conflicts of interest.
  • Leon reports that AdvaMed supported his travel and accommodations for the meeting. He also reports receiving research support from Abbott, Boston Scientific, Edwards Lifesciences, and Medtronic.

We Recommend

Comments