A trio of Belgian researchers has expressed grave concern regarding the rapid growth in the use of transcathether aortic valve replacement (TAVR) to treat patients with severe aortic stenosis. Publishing their analysis online July 31, 2012, ahead of print in BMJ, they raise “serious unanswered questions” about the strength of the clinical evidence supporting the procedure as well as the regulatory process enabling it to be widely adopted, especially in Europe.

Dramatic Increase in TAVR in Europe

Economist Mattias Neyt, PhD, of the Belgian Health Care Knowledge Centre (Brussels, Belgium), and colleagues write that an estimated 40,000 TAVR procedures had been performed by the end of 2011, with the number tripling in Europe between 2009 and 2011. Adoption of the novel procedure has been driven in part by need (before TAVR, about one-third of elderly patients with aortic stenosis went untreated because they were considered unsuitable for surgery) and in part by patient demand for the less invasive alternative to surgery, the researchers observe.

In the European Union, medical devices need only a quality certificate to qualify for use in patients, and in fact the Edwards Sapien valve (Edwards Lifesciences, Irvine, CA) and CoreValve (Medtronic, Minneapolis, MN) both received CE mark approval in 2007 before substantial supporting clinical trial evidence was available, Dr. Neyt and colleagues write.

But a health technology assessment commissioned by the Belgian government concluded that only about 10% of patients currently considered for the therapy—those deemed inoperable for technical reasons—should have their procedures reimbursed. In addition, the United Kingdom’s National Institute for Health and Clinical Excellence guidelines indicate that evidence for TAVR use in high-risk patients is “inadequate.”

In the United States, the Sapien valve was approved for inoperable patients by the US Food and Drug Administration (FDA) in late 2011 based on randomized data from the PARTNER trial. “But while the evidence demanded by the FDA exceeded that required in Europe, we remain far from convinced that is it adequate,” the authors assert.

Questions About Follow-up, Methodology

In PARTNER Cohort B, which focused on inoperable patients, TAVR reduced 1-year mortality by 20% compared with standard care alone, suggesting that its use could be justified in this population, the authors say. But, they add, that conclusion is “thrown into doubt” by a follow-up study that included 90 ‘continued access’ patients randomized to TAVR (n = 41) or standard therapy (n = 49). Although the study remains unpublished, Dr. Neyt and colleagues say data presented at an FDA meeting in July 2011 showed that the TAVR group fared worse in terms of 1-year mortality (34.3% vs. 21.6%).

Dr. Neyt and colleagues also cite what they consider a methodological failing of PARTNER Cohort B. “Treatment and control groups are unbalanced in a way that would favor [TAVR],” they say, noting that standard therapy patients had more comorbidities and more prior MI, and were more often frail. “An analysis that adjusted for prognosis at baseline would have produced a more realistic estimate of the effect size,” they claim.

Transapical Approach Problematic

Most concerning, the authors say, is European physicians’ tendency to overuse the transapical approach, accounting for about 25% of TAVR patients in a UK registry and 20% in the FRANCE-2 registry. This “far exceeds what is justified by the clinical evidence,” they say. In fact, a red flag was raised by the Danish STACCATO trial, which was halted when TAVR patients treated via the transapical route showed notably worse outcomes than surgical patients, they add.

“In contrast to the current situation in Europe, we recommend that marketing approval for a high risk device should be granted for specific indications only,” and each indication “should be supported by clinical evidence from high quality randomized trials,” Dr. Neyt and colleagues write.

The authors state that it is difficult to justify reimbursing TAVR for operable patients when the stroke risk is twice as high as for surgery and the transcatheter procedure is more costly, at least based on an analysis of Belgian data.

A ‘Hatchet Job’

Jeffrey W. Moses, MD, of Columbia University Medical Center/Weill Cornell Medical Center (New York, NY), reacted strongly to the paper. “This is not an analysis, this is a hatchet job,” he told TCTMD in a telephone interview.

In his view, the divergent data from the PARTNER follow-up study is basically a nonissue. “Those data didn’t change a thing—the FDA knows that and the New England Journal [which published the initial PARTNER study] knows that,” he asserted.

Dr. Moses also dismissed the significance of the unbalanced characteristics in the 2 arms of PARTNER Cohort B. “No trial is perfect; and when you get into small numbers, imbalances can occur,” he observed. “But there’s nothing in these minor imbalances that would lead to this type of difference in mortality [between TAVR and standard therapy].”

As for possible overuse of the transapical approach, Dr. Moses said “there are some European sites where the surgical groups use the transapical approach because they like it.” Clearly, improvement is warranted in the nonfemoral strategy, he added, ”but in the meantime, you can see in the continued access portion of PARTNER that transapical results improved—strokes went down and patient selection got better.”

Moreover, to portray STACCATO as anything more than a cautionary tale is inappropriate, Dr. Moses said. “STACCATO shows that you shouldn’t do transapical TAVR in low-risk patients, and it shouldn’t be done by inexperienced groups,” he explained. “But for [the authors] to dismiss the transapical approach out of hand is astonishing. I don’t consider transapical the preferred approach, but it’s an option for people who have no other choice.

“All these objections are nitpicking that has been discussed before,” Dr. Moses concluded. “If [TAVR evidence] passed scrutiny in the New England Journal of Medicine, I think we’ve passed a pretty rigorous test.”

Analysis Not Relevant to the United States

In a telephone interview with TCTMD, PARTNER investigator David J. Cohen, MD, of Saint Luke’s Mid America Heart Institute (Kansas City, MO), said the article “raises some valid criticism buried within some not-so-valid criticisms.” But even the former are relevant only to Europe, he added.

“The authors’ biggest concern seems to be the way practice [in Europe] has outstripped the evidence in terms of treating low- and moderate-risk patients, and frankly I agree,” he said. “In the United States, rollout of the technology has been much more careful. TAVR use is constrained by FDA regulation and more importantly by reimbursement policies and the process of the heart team.”

According to Dr. Cohen, it is not clear that TAVR offers any major advantages in lower-risk patients and there may be disadvantages. “That’s why we’re doing the PARTNER II trial and in Europe they’re doing the SURTAVI trial [with CoreValve],” he said. “But by the time SURTAVI has any meaningful results, many, many [European] patients will already have been implanted.”

Dr. Cohen added that the authors’ concerns about the cost-effectiveness of TAVR are “probably realistic.” He noted that in the United States for inoperable patients “TAVR doesn’t save money, it saves lives. The cost-effectiveness ratio is around $50,000 per year of life gained, which is acceptable but not a bargain. In many countries with a lower health care budget, that may be more than they wish to spend.” Also, Dr. Cohen added, in the United States, costs for transfemoral TAVR vs. surgery in operable patients are “largely neutral or slightly lower.”

Related Stories:

• NEJM Letters Take on PARTNER Trial
• PARTNER: TAVI Cost Effective Compared With Standard Care
• PARTNER: TAVR Cost-Effectiveness Evaluated in Inoperable Patients