Penumbra’s JET 7 Recall: A Kick Start to Change?

A deep dive into the rocky history of the JET 7 reperfusion catheter with Xtra Flex Technology, published this week in JAMA Internal Medicine, is aiming to renew discussions on medical device safety.

Its manufacturer, Penumbra, voluntarily recalled the device in December 2020 due to the potential for distal tip damage during use, a move that the US Food and Drug Administration categorized as Class I, its most serious type. The agency by that point had received over 200 medical device reports associated with the product. Among them were 14 deaths and a range of serious injuries like vessel damage, hemorrhage, and cerebral infarction.

Kushal T. Kadakia, MSc (Harvard Medical School, Boston, MA), the paper’s lead author, told TCTMD that the saga caught his group’s attention in part because of how widely JET 7 with Xtra Flex, cleared under the FDA’s 510(k) approval pathway, is used—the recall covered over 22,000 devices across the United States.

“It’s not really about a specific device or a specific company,” but rather a “case study” that captures the big picture, he stressed. Falling on the 10-year anniversary of the Institute of Medicine’s look into the 510(k) program, “it seemed like a timely opportunity to see where we are in the field and where we still need to go.”

Under the 510(k) pathway, companies seek approval from the FDA to market a novel device considered to be as safe and effective as, or “substantially equivalent” to, an existing product. The agency’s decision can rest on either clinical or preclinical data.

But time and time again, concerns have been raised that speed can come at a price.

B. Hadley Wilson, MD (Sanger Heart & Vascular Institute/Atrium Health, Charlotte, NC), vice present-elect of the American College of Cardiology (ACC), said that the paper offers a useful look at one example of how things can go wrong.

“Obviously for any occasion where there is harm from medical devices, I think we always need to reexamine our processes to see if this could’ve been avoided not only with that particular case but also [others] like this in the future,” he commented to TCTMD, describing the article as “very sophisticated and detailed.”

It Starts With the 510(k)

One aspect of the 510(k) program that may be unfamiliar to cardiologists is just how far the links can stretch in the evidence base, said Wilson. “Some devices are approved after several iterations of the [process] and may not even resemble the original device for which the 510(k) was approved.”

A novel product in the 510(k) pathway is viewed in light of a “predicate device” that is used as a comparator. “In the case of Penumbra, the original 2007 device was cleared under 510(k) using the Merci catheter as its predicate device, which itself was authorized based on evidence generated from nonrandomized, single-arm trials,” Kadakia and colleagues note.

“The Merci device itself was cleared using the percutaneous catheter from Concentric (the Modified Concentric Retriever) as its predicate device, even though the latter device had a different indication, with use intended for the removal of foreign bodies (eg, a coil) as opposed to thrombi in patients who had stroke,” they write, pointing out that Penumbra’s pivotal trial was a single-arm study with 125 participants.

You shouldn’t have to rely on a stockbroker to tell you there’s a problem. Kushal T. Kadakia

The first known death linked to JET 7 with Xtra Flex was reported to the FDA Manufacturer and User Facility Device Experience (MAUDE) database in October 2019, 4 months after 510(k) clearance, and other complications followed. By July 2020, Penumbra had alerted clinicians to issues with the device’s distal tip. Thereafter, the researchers say, the company reported record-high sales in its October 2020 earnings call.

Only after activist investors drew attention to the safety concerns did the topic grab news headlines. “To me as a scientist, as a future clinician, that felt a bit strange that the alarm bells for a device failure are being rung on Wall Street before they’re being rung on Main Street,” said Kadakia. “That really caught our eye—irrespective of the specific problems of a device, that speaks to a larger failure of our system. You shouldn’t have to rely on a stockbroker to tell you there’s a problem.”

What Role for Physicians?

Reports to MAUDE are, as of now, voluntary and the onus rests on those reporting to ensure that they include pertinent details. Here is where tools like unique device identifiers (UDIs) could ease the burden. “MAUDE relies on people to report [adverse events] themselves, and oftentimes those forms don’t have all the information,” Kadakia noted. “If reports are delayed by months, or missing key information about malfunctions or adverse events, then it constrains our ability to understand patients’ real-world experiences with these products.”

As to what individual cardiologists can do, Hadley said it would be great if things went in the opposite direction, with the FDA and others reaching out to say: “We have an easy, safe, secure, confidential way of now reporting on new devices [to see postmarket] if there are any early warning signs.”

Barriers to reporting should be lowered, he urged.

Professional societies like the ACC can also be involved. Hadley pointed to the National Cardiovascular Data Registry as an example. “It certainly would be nice, like the airlines, to be able to have a safe reporting system for ‘near misses,’” that could draw attention to particular areas, he said.

‘Issues Across the Life Cycle’

JET 7 is just one of thousands of devices the FDA clears annually under the 510(k) program. Within the past 9 years, approximately 13% have been subject to any recall and 1% subject to a Class I recall, the paper notes.

Its authors offer several critiques of that pathway, among them the potential for “creep” as multiple products piggyback on existing devices; the fact that “determination of substantial equivalence is biased in favor of manufacturers, who can selectively choose comparators, and performance thresholds”; and how broadly “risk” is defined when classifying devices, a factor that informs their requirements for approval.

“JET 7, like 43% of all devices and 89% of 510(k)-cleared devices, is classified as Class II, even though the invasive nature of the stroke procedures in which the device is used is far from moderate risk,” the authors say. “Indeed, it seems incommensurate for JET 7 to be classified in the same risk category as surgical face masks and powered wheelchairs.”

The approval process involves trade-offs, Wilson explained. “Every cardiologist, I would hope, would really love to see more clinical data and randomized trials and so forth, but I believe we all also appreciate the cost and expense and time required for those types of trials. It’s really a very delicate balance between trying to get medical devices to help our patients in a timely manner and hopefully less-costly manner versus safety, which is paramount.”

It’s really a very delicate balance between trying to get medical devices to help our patients in a timely manner and hopefully less-costly manner versus safety, which is paramount. B. Hadley Wilson

And after a device has reached the market, there are other obstacles when it comes to tracking outcomes. Solutions suggested by the paper include increased update of UDIs, national postmarket registries (as in Iceland), and active surveillance programs (eg, the National Institutes of Health-funded POST SCD study in San Francisco).

Another positive sign, Kadakia et al say, is the FDA’s National Evaluation System for health Technology (NEST).

“Although digital tools such as UDIs and initiatives such as NEST that collate data from diverse sources (eg, claims, registries, electronic health records) can enhance the efficiency of surveillance, the FDA requires additional resources to move beyond individual demonstrations and toward improvements in everyday practice,” they stress, suggesting that Medical Device User Fee Amendments (MDUFA) could help increase funding. These payments were established under the Medical Device User Fee and Modernization Act of 2002, enacted by the US Congress, and are currently in the process of being reauthorized.

Indeed, Kadakia concluded, there are “issues across the life cycle that we have an opportunity to improve.”

An editorial by Jessica Holtzman, MD, Christina Lalani, MD, and JAMA Internal Medicine Editor Rita F. Redberg, MD (all University of California, San Francisco), also calls for improvement.

The passive surveillance that’s currently the norm in the United States “is estimated to capture 15%, at best, of all actual adverse events,” the editorialists note.

At the FDA, the process of tracking these events can be lengthy and tedious, they add. “According to the most recent Office of the Inspector General Report, from 2010, less than one-half of reports that were reviewed by analysts at the FDA were reviewed within 60 days of receipt. An updated Office of the Inspector General report is long overdue and sorely needed.” They, too, call on Congress to provide the agency with money to pursue these goals.

For Hadley, the pending appointment of cardiologist Robert Califf, MD, as head of the FDA also bodes well. He “is quite experienced and savvy in this area,” said Hadley.