VA Medical Centers’ Quality Measures for TAVR Pay Off

Thanks to oversight, reporting, and other stringent standards, veterans’ outcomes “compare favorably” to elsewhere.

VA Medical Centers’ Quality Measures for TAVR Pay Off

US veterans undergoing TAVR at Veterans Affairs (VA) medical centers have outcomes that “compare favorably” to those achieved elsewhere, according to 6 years of data on nearly 1,000 individuals treated in the healthcare system.

The VA healthcare system first began offering TAVR at one center in 2011. Now, the minimally invasive alternative to surgery is offered at eight VA centers out of the approximately 40 to 45 that perform SAVR, Philip S. Hall, MD (University of North Carolina Rex Healthcare, Raleigh), and colleagues report in their paper published online August 28, 2019, ahead of print in JACC: Cardiovascular Interventions.

TAVR’s introduction within the VA system is unique, they point out. Here, Centers for Medicare & Medicaid Services (CMS) reimbursement is not a concern, but VA leadership has adopted VA-specific standards on top of key CMS requirements; these include multidisciplinary site inspections, stringent hybrid-OR standards, as well as oversight and outcomes reporting via the VA Clinical Assessment, Reporting, and Tracking (CART) Program and the National VA Surgical Quality Improvement Program.

“This process was accompanied by the establishment of a national VA quality committee of peer experts to optimize quality improvement opportunities through a real-time alert system and periodic conference calls to monitor and collectively discuss every major adverse event at any VA,” the researchers explain.

Morton Kern, MD (University of California, Irvine, and VA Long Beach Healthcare System), commenting on the study for TCTMD, said: “I think the VA’s peer review process is much more organized and stringent than many of our communities’ peer-review processes, [which are] run by individual hospital systems run by a team of doctors that work at that hospital. But the VA has a really, I would say, vigorous and continuous way to both record and report.”

The question, then, is whether this approach bears fruit for US veterans, a population that tends to be older men with numerous comorbidities. As the investigators note in their paper, “veterans are known to have a different cardiovascular risk profile from nonveterans, and their experience and outcomes after TAVR in a pure veteran population, performed in dedicated intramural TAVR programs, may be quite different from results in the general population.”

For Kern, and for editorialist John C. Messenger, MD (University of Colorado School of Medicine, Aurora), the answer is that much can be learned from how TAVR was rolled out within the VA healthcare system.

The VA CART Program

Hall et al based their analysis on data from the VA CART Program. For the 959 TAVR patients treated between 2012 and 2017, access was transfemoral in 90%, transapical in 5%, transaxillary in 3.8%, and transaortic in 0.3%. Roughly two-thirds received a balloon-expandable Edwards Lifesciences valve, one-third received a self-expanding Medtronic valve, and 2% a mechanically expandable valve from Boston Scientific. Men made up 98% of the cohort, and the mean age was 78.1 years.

Median length of stay was 5 days. By 30 days, 2.9% of patients had died, 14.7% had been rehospitalized, and 17.9% required a pacemaker. By 1 year, the mortality rate was 14.0%.

Predictors of longer hospital stays included nonfemoral access  (OR 1.74; 95% CI 1.10-2.74), heart failure (OR 2.51, 95% CI 1.83-3.44), and atrial fibrillation (OR 1.40; 95% CI 1.01-1.95). A-fib also was linked to higher likelihood of being rehospitalized (HR 1.79; 95% CI 1.22-2.63).

Data from the Society of Thoracic Surgeons/American College of Cardiology TVT Registry, which the VA healthcare system does not participate in due to data-security concerns, were used as the comparison group. In the TVT Registry, TAVR patients treated between 2012 and 2015 had a higher median age (83.0 years), and nearly half were women. Mortality rates were 5.7% at 30 days and 21.0% to 26.0% at 1 year, Hall and colleagues note, saying this may relate to a greater use of nonfemoral access in the TVT Registry. Veterans, on the other hand, were more likely to experience vascular complications and require pacemaker placement.

“Direct comparison and adjustment of risk between registries remains a controversial challenge, and the definition of risk in multidisciplinary teams varies,” the authors caution, adding, “The results reported here are somewhat heterogeneous, as a reflection of the rapid evolution of TAVR over the course of this experience” in terms of best practices, experience levels, device iterations, and patient risk levels.

More TAVR Oversight?

Messenger, in his editorial, also emphasizes that the study is comparing unadjusted outcomes over slightly different periods, during which TAVR was a shifting target.

Still, the quality controls set in place within the VA system appear to be working and might be worthy of imitation, he suggests. “Approaches to consider include more active central oversight implemented via the STS/ACC TVT Registry to allow the identification and public reporting of poorly performing sites using risk-adjusted approaches. Alternatively, the introduction of heart valve center of excellence accreditation may be the most acceptable.”

Messenger goes on to say: “Certainly further oversight needs to be considered before hundreds more low-volume sites are able to perform TAVR in the United States, potentially releasing ‘the horses out of the barn’ without any CART attached to keep them in check and ensuring the right balance between access and quality.”

Kern is chief of medicine and an interventional cardiologist at the Long Beach VA, which is not among the eight VA medical centers offering TAVR. He agreed that something like CART “could be emulated across the country” as a means to collect and examine TAVR data in its totality. One thing to note, Kern said, is that the VA system recognizes the need to only take on what it can and focus its resources on hospitals with the infrastructure for a TAVR program.

Asked by TCTMD whether TAVR operators would be game for additional oversight, Kern replied that “any time you increase monitoring you have to have some resources to do it—somebody has to enter the data, somebody has to follow up—but they probably already have some type of a [reporting] system if they have a TAVR program, so I think the transition to an addition of a program or replacement of a program probably would not be so great.”

One interesting path for future research, Kern said, would be to track the experiences of US veterans who undergo TAVR outside the VA due to there being a limited number of sites in that system. Most VA medical centers have a “university partner,” he explained. “The next part of this study might be to see who in the VA with aortic stenosis is being referred to [these centers] for their procedure. My guess is it would be a fair number, because we know the technology works, we know the technique is safe, it provides great benefit, it’s a natural. And the VA, if it can’t provide the service, contracts for the service” and pays the university partner instead.

  • Hall reports receiving funding from the American College of Cardiology Foundation/Merck Research Award 2016.
  • Messenger reports receiving institutional grant support to the University of Colorado School of Medicine from Philips Medical Systems.
  • Kern reports no relevant conflicts of interest.