July 31, 2012—Analysts from a Belgian health care group have published a review of transcatheter aortic valve implantation (TAVI) online ahead of print in the British Medical Journal (BMJ). Describing the procedure as “risky and costly,” the group of authors sought to examine why, in their estimation, TAVI practice seems to have gone beyond the evidence.

The authors call for better regulation and transparency regarding the use of these devices, claiming that many of the TAVI procedures performed worldwide since the introduction of the procedure cannot be justified on medical or cost-effectiveness grounds. Other clinicians have expressed strong disagreement with the conclusions of the article, as well as the means by which they were reached.

The article's authors are cardiologist Hans Van Brabandt, MD; economist Mattias Neyt, PhD; and Frank Hulstaert, MD, who focuses on health technology assessments. They are with the Belgian Health Care Knowledge Centre in Brussels. Dr. Van Brabandt is also affiliated with the Belgian Centre for Evidence-Based Medicine and Branch of the Dutch Cochrane Centre in Leuven.

In the article, the authors raise concern over the path to market that devices such as transcatheter aortic valves take in Europe, comparing it unfavorably to the more rigorous process required by the US Food and Drug Administration (FDA). They state that the valves are classed as medical devices, but that this distinction means the devices need only obtain a quality certificate (CE Mark approval) to gain access to the European market, following a similar path as household appliances. As a result, TAVI was in use in Europe 4 years before it was commercially available the United States, where more rigorous approval processes are required.

However, the analysts are also “far from convinced” as to the adequacy of the evidence required by the FDA in its evaluation and approval of the Sapien device (Edwards Lifesciences, Irvine, CA), which was gathered in the randomized, controlled PARTNER trial of more than 1,000 patients and published in the New England Journal of Medicine. They further expressed doubt regarding “publication bias, lack of data transparency, unbalanced patient characteristics, and incompletely declared conflicts of interest.” Most concerning to the group was their belief that the use of transapical access in European TAVI procedures exceeds what they believe is justified by clinical evidence.

Cardiac Interventions Today discussed the BMJ article with a recognized expert on valvular repair, Martyn Thomas, MBBS, FRCP. Dr. Thomas is Clinical Director of Cardiovascular Services, St. Thomas' Hospital in London, United Kingdom. He also serves as Co-Principal Investigator of the ongoing SOURCE XT registry, which is a multiregion outcomes study of the Edwards Sapien XT device. Dr. Thomas has previously disclosed receipt of research funding from Edwards Lifesciences.

Dr. Thomas disagrees with the conclusions reached by the BMJ analysis authors, beginning by noting that it is an opinion piece and not a study, and he raises concerns regarding the manner in which disparate data are presented. He believes the analysts put too much weight on flawed trials while focusing insufficiently on the “very-well-performed” randomized PARTNER trial.

Cohort B of the PARTNER trial randomized 358 inoperable patients to transfemoral TAVI or standard medical therapy and was the basis of the FDA approval for the Edwards Sapien transcatheter valve for that patient population in November 2011. Cohort A of the PARTNER trial randomized 699 patients at high risk for surgery and compared outcomes with either surgical valve replacement or the Sapien valve via transfemoral or transapical delivery. The 1-year cohort A results were presented in April 2011 at the American College of Cardiology's annual scientific sessions and were published in the New England Journal of Medicine in June 2011.

On June 13, 2012, the FDA's Circulatory System Devices Panel of the Medical Devices Advisory Committee voted unanimously in favor of recommending approval of the Edwards Sapien transcatheter heart valve via transfemoral and transapical delivery for the treatment of high-risk patients with severe, symptomatic aortic stenosis.

However, the authors of the BMJ analysis focus instead on two European registries that describe transapical use in 20% (UK TAVI) and 25% (FRANCE II) of patients and mortality rates of more than 20% in each, leading to the authors' concerns over this approach. It is noted in the article that the registries did not include randomization to either medical management or surgery; the operative risk associated with the patients in the registry populations is not specified. Citations for these data in the BMJ article do not indicate that they have been peer-reviewed or published, although they have been presented.

The BMJ analysts also refer to data from STACCATO, a Danish trial that enrolled 34 transapical TAVI patients and 36 surgical patients before being halted, at which point, major adverse events had been reported in five transapical TAVI patients versus one surgical patient. Although the BMJ article cites a Heartwire article in which Dr. Michael Mack of the University of Texas at Dallas opines that STACCATO included misinformation based on an invalid study design, it does not provide further analysis regarding that possibility, focusing instead on Dr. Mack's belief that it was “likely to hurt the field.”

“The emphasis on STACCATO is misplaced,” commented Dr. Thomas. “STACCATO was strongly criticized when it was first presented at the Transcatheter Therapeutics annual meeting in 2011. The trial was performed in Scandinavia before the learning curve had been overcome, and it specifically randomized transapical TAVI versus low-risk surgical patients. It was widely believed that this trial was misconceived, and the results have no relevance to current TAVI practice.”

The BMJ article's authors state that they agree with guidance from the UK National Institute for Health and Clinical Excellence (NICE), which indicates that the evidence for TAVI in patients who are suitable for conventional surgery is inadequate. Dr. Thomas believes that quoting the NICE guidance in this context is misguided.

“This guidance was published before the latest 2-year PARTNER results, and it also ignored the PARTNER cohort A cost-effectiveness data,” he said. “The cost-effectiveness data for TAVI versus surgical aortic valve replacement indicated that TAVI was dominant. TAVI was cheaper and better than surgical aortic valve replacement in the American health care system, with a comparative cost per quality-adjusted life years of –$2,000.”

Another point of contention raised in the BMJ analysis involved what it described as a 90-patient follow-up study requested by the FDA. The analysts sought further access to the data from this population, which they characterize as including randomization of 41 inoperable patients to TAVI and 49 to standard therapy and favoring standard therapy; the data were presented at an FDA panel meeting in July 2011. The analysts opined that their not being granted further access to this data was “ethically and scientifically” unacceptable.

Dr. Thomas disagreed with the BMJ analysts' description of the 90-patient segment, commenting, “The authors say that a further 90 patients were randomized in a follow-up study asked for by the FDA. This is not true. The FDA asked for cohort B (inoperable) patients to continue to be randomized while the cohort A (high-risk surgery) group finished its trial randomization. This was to prevent an imbalance in decision-making that would be created [in cohort A] if there were no other treatment option avail- able to inoperable patients. When this group was brought up at the FDA panel—which I attended—the FDA clearly stated that this group was not considered to be part of the PARTNER trial and was too small to be analyzed separately.”

Responding to the BMJ authors' belief that PARTNER Principal Investigator Martin Leon, MD, had financial interests that were not fully disclosed at the time the trial's data were published, Dr. Thomas believes it is difficult to see how this could alter the results of a trial like PARTNER, which was conducted in a meticulous fashion that included safety and event committees. The BMJ analysis mentions the New England Journal of Medicine article's disclosure of Dr. Leon's having received payment for equity holdings when Edwards bought Percutaneous Valve Technologies but not three payments that Dr. Leon was to later receive for milestones including European and US approval of the device. However, the BMJ analysis does not specify any impact this may have had on patient care, the conduct of trials, or regulatory review.

The analysts noted that after rigorous analysis of all the available data, combined with a study of what they consider real-world TAVI practice in Europe, “the arguments supporting the widespread use of TAVI do not stand up to scrutiny.” Going forward, they believe that Europe's regulatory system “should require high-quality randomized trials to show clinical efficacy and safety before granting marketing approval to innovative, high-risk medical devices.”

Dr. Thomas contends that a CE Mark is known to indicate that a device is safe. Although he believes there is clearly a need for randomized trials and additional registries in the development of a device such as a heart valve, he does not think randomized trials like those required before many FDA approvals should be part of the CE Mark process.

“The FDA process is extremely defensive, and rather than protecting United States citizens, it often delays the introduction of new technology,” said Dr. Thomas. “This leads to Europe having access to new technologies 2 to 3 years before the United States, resulting in the fact that this year TAVI will be performed with 14-F devices in Europe and 22- to 24-F devices in the United States.”

In a statement issued to Cardiac Interventions Today regarding the article published in BMJ, Edwards Lifesciences provided the following comment:

“The BMJ editorial contains no new data but is simply a repackaging of old, non-peer-reviewed arguments against TAVI that have been fully and publicly answered. Globally, there exists a remarkable body of clinical, quality-of-life, and economic evidence on Sapien collected over 10 years. These data have been published in more than 350 peer-reviewed publications, extensively presented at public meetings, and reviewed by regulators and payors worldwide.”