In the wake of the FDA’s approval of TAVR in the United States, patients who underwent the novel procedure at hospitals that did not participate in clinical trials had outcomes similar to those of patients treated at trial centers, according to analysis of Medicare data published in the September 2015 issue of Circulation: Cardiovascular Interventions.

“The unique policy and regulatory environment governing TAVR adoption by hospitals may have contributed to better outcomes during the technology diffusion process,” suggest Nimesh D. Desai, MD, PhD, of the University of Pennsylvania (Philadelphia, PA), and colleagues.

Drawing on Medicare fee-for-service claims data, the investigators compared outcomes for 3,617 patients who underwent TAVR at 68 hospitals that had participated in preapproval trials with those of 1,392 patients who had procedures at 140 centers that did not. Cases were performed between January 2011 and November 2012, a period that spanned FDA approval of TAVR on November 2, 2011. All patients had at least 30-day follow-up, but 90- and 180-day analyses were limited to patients who underwent TAVR before October 1, 2012 (n = 4,138).

Trial institutions were defined as those taking part in a Sapien (Edwards Lifesciences) or CoreValve (Medtronic) preapproval clinical trial or otherwise performing TAVR prior to approval.

Patients treated at both categories of hospitals were similar in age (mean 84 years), sex (51.5% men), and comorbidities, although patients at trial centers had more congestive heart failure and less PAD than those at nontrial hospitals (both P < .01). Trial hospitals had higher overall surgical aortic valve replacement volume (P < .001) and performed more transapical procedures than did nontrial centers (P < .01).

After FDA approval, use of the transapical approach, patient age, and the prevalence of congestive heart failure declined at trial centers, but most procedural and patient characteristics remained unchanged at both trial and nontrial centers.   

Both the number of patients undergoing TAVR and the number of institutions offering the procedure increased steadily over the study period. The gap in patient volume between trial and nontrial hospitals progressively narrowed, reaching near parity by the end of 2012.

Outcomes Largely Unchanged as TAVR Expands

The mean annual rate of TAVR cases per institution, adjusted for the date of the first case performed at each center, was 33.3 for trial hospitals and 26.8 for nontrial hospitals. The rate of TAVR cases per institution during each quarter was relatively stable, except for an increase over the first 3 quarters at nontrial centers. Overall, the patterns suggest that nontrial centers represented an expansion of the population receiving TAVR rather than a redistribution of cases, the authors say.

Aside from a longer postprocedural stay and a higher rate of permanent pacemaker implantation at trial hospitals than at nontrial hospitals, there were no differences in outcomes between the center types (table 1).

After multivariable adjustment, mortality risk was similar for trial vs nontrial hospitals at both 30 days (OR 0.93; 95% CI 0.70-1.23) and 180 days (OR 1.06; 95% CI 0.74-1.52). Older age (≥ 90 years), congestive heart failure, use of the transapical approach, chronic kidney disease, liver disease, and weight loss predicted higher risk-adjusted mortality at 30 days.

Analysis comparing the first 20 and the first 30 cases with subsequent cases showed no impact on 30-day mortality or 180-day survival. Furthermore, comparison of any given case with the subsequent one showed early mortality rates to be stable, suggesting no adverse “learning curve” effect.

Stakeholder Collaboration Key

“TAVR technology dissemination involved new regulatory, proctoring, and multidisciplinary clinical collaboration,” the authors say. The fact that outcomes were preserved at nontrial hospitals “is a particularly notable finding in the setting of a new and complicated procedure in hospitals without any previous experience,” they add. The US experience stands in contrast to the early adoption of TAVR in Europe, where large French and German registries showed 30-day mortality rates between 9.7% and 12.4%, the investigators observe.

“The diffusion of TAVR in the United States followed a plan of rational dispersion resulting from close collaboration between the major professional societies representing cardiologists and cardiac surgeons, specific mandatory criteria to qualify as a TAVR hospital, and active participation from industry to provide proctoring and education,” Dr. Desai and colleagues say.

This program guiding the transition from pivotal trial to commercial use of TAVR differs sharply from the diffusion of other cardiovascular procedures and technologies, including carotid intervention, endovascular abdominal aortic aneurysm repair, and implantable cardioverter-defibrillators, they observe.

Aside from the inherent limitations of a retrospective analysis, the authors acknowledge that the Medicare dataset’s lack of information on LVEF, valve type and size, and STS score may have resulted in confounding.

Multisociety Guidance Helped Ensure Medicare Coverage

“The study highlights that there wasn’t a penalty for adoption of these very complicated technologies outside the trial framework,” Ted Feldman, MD, of NorthShore University HealthSystem (Evanston, Illinois), told TCTMD in a telephone interview.

“The lesSon was apparent right from the start that a multisociety approach was critical to help [the Centers for Medicare & Medicaid Services (CMS)] and the community with the rollout of these technologies,” he added, noting that the professional societies were determined to “make the dispersion of the therapy rational and safe and at the same time available. It was necessary to provide some guidance for site qualification as part of the National Coverage Determination process. [And CMS] was looking to the medical societies for help on how to define operator and institutional requirements for the therapy.”

Also critical to successful dissemination of TAVR was the role of Edwards and, later, Medtronic in “making proctors readily available and providing a lot of field support for getting sites going,” Dr. Feldman said.

The key requirement of a heart team was an evolution of the practice instituted in the SYNTAX trial, in which a surgeon and an interventional cardiologist had to agree that a patient was a candidate for either surgery or PCI, he noted.

“So the combination of some rational baseline requirements for sites and a really careful job by industry to roll out the technology are why these results look the way they do,” he said.

“We have learned more and more about how to roll out a new technology,” Dr. Feldman said. “I think the multisociety approach has worked very well,” he added, noting that a similar strategy has already been adopted for the MitraClip and transcatheter mitral valve replacement.

---

Source: 
McCarthy FH, Groeneveld PW, Kobrin D, et al. Effect of clinical trial experience on transcatheter aortic valve replacement outcomes. Circ Cardiovasc Interv. 2015;8:e002234.

Related Stories:

• Medical Societies Issue Position Paper in Anticipation of Transcatheter Valve Availability
• Societies Recommend TAVR Program Requirements
• Registry Study Summarizes Initial US Experience with First Approved TAVR Device