FDA lax on monitoring medical implants
This story was originally published by ProPublica.
ProPublica is a Pulitzer Prize-winning investigative newsroom. Sign up for The Big Story newsletter to receive stories like this one in your inbox.
Each prescription drug you take has a unique code that the government can use to track problems. But artificial hips and pacemakers? They are implanted without identification, along with many other medical devices. In fact, the FDA doesn't know how many devices are implanted into patients each year – it simply doesn't track that data.
The past decade has seen numerous high profile cases of malfunctioning medical devices, which have led to injury or even death. Critics say the FDA's minimal monitoring of devices contributes to these problems.
"If you're lucky, you might find a sticker on the operating room note that was left over from the product," said Richard Platt, who runs the Harvard Pilgrim Health Care Institute. Otherwise, there is little way of knowing what device was used.
Right now, the FDA depends mostly on voluntary reports from doctors, patients, manufacturers and hospitals to notify them of problems with devices already on the market. The agency does have some power to require manufacturers to conduct further studies or track a particular device once it is sold. But many devices don't get that level of surveillance.
"It's much like a patchwork of streams of information getting to the FDA," said cardiologist Frederic Resnic of Brigham and Women's Hospital, who has worked with the FDA on medical device safety monitoring. "The FDA is relying on anecdotal and very variable information about the safety of medical devices."
If manufacturers get word from a doctor or hospital about a death or injury that occurred as a result of their product, they are legally obligated to investigate the event and report it to the FDA. But the process isn't straightforward, as has become clear in the recent controversy over the malfunctioning St. Jude's Riata defibrillator leads (wires that connect a defibrillator to the heart). The FDA said an individual doctor's report helped alert them to the problem, but it was months before the device was recalled.
According to attorney William Vodra, a regulatory law expert and member of the Institute of Medicine panel that published a report on medical device safety last year, the number of doctors who actually contact manufacturers is small.
And after being notified of patient harm, manufacturers can minimize their own responsibility if they point the blame elsewhere, said health policy expert Diana Zuckerman, president of the National Research Center for Women & Families.
For example, if someone dies from complications in a surgery to remove an implant, the manufacturer may argue that it was the surgery – not the implant – that killed the patient.
"You have a system that is not rigorous, the standards are not always understood, and they are interpreted differently by different people," Zuckerman said.
The FDA responds to the criticism by pointing out that while every medical device carries a potential risk, the vast majority of devices perform well and improve patient health. An FDA spokeswoman emphasized that the agency must evaluate thousands of medical devices each year, and is constantly looking for ways to better and more quickly identify problems.
While the FDA makes the adverse event reports publically available in a searchable database, it doesn't have a standardized system for reviewing reports once they are sent in, said Vodra, the attorney. A disclaimer on the site specifically states that the data is "not intended to be used either to evaluate rates of adverse events or to compare adverse event occurrence rates across devices."
"What you would normally consider the simplest kind of data analysis is not done," said Zuckerman. Often, doctors catch a malfunctioning device before the FDA ever notices.
In one case, a group of Pennsylvania doctors noticed that several patients were showing severe complications a few years after getting an IVC filter – a device designed to capture blood clots. Bits of the filter were breaking off, causing chest pain and a dangerous build-up of fluid and pressure around the heart. In 2010 the doctors conducted their own study and found that the filter broke in a quarter of all patients who used it.
On the day that study was published, the FDA issued a warning saying it had received over 900 reports of problems with IVC filters since 2006, and that the device was meant to be removed after a few months, not left in permanently.
There have been numerous attempts at reform. Five years ago Congress ordered the FDA to set up a post-market surveillance system to track the safety of all medical projects, but a system hasn't yet been set up for medical devices.
A year later the FDA announced the Sentinel Initiative, which would combine existing data from electronic health records and medical claims to track drugs, vaccines, and devices. Some groups of hospitals or other organizations have voluntarily set up registries to collect information about the make and model of devices.
While the FDA has made significant progress on tracking drugs, it's not yet in a position to do the same thing for devices, according to Harvard's Platt, who is the principal investigator of Mini-Sentinel, the FDA's pilot program for the national system. The data isn't there.
The FDA has long acknowledged the need for a unique device identifier system, and got permission from Congress to set one up five years ago. No such system of ID-tags exists yet, but after several recent high profile medical device failures, the issue getting some attention from Congress. A proposed Senate bill, which cleared the Health, Education, Labor and Pensions Committee last week, sets a timeframe for implementing a unique identification system, among other reforms.
"If UDI's were used in a consistent way, we could use the same kinds of techniques we've developed for drugs for devices," said Platt. "It would be a huge breakthrough."