Lack of publicly available scientific evidence on the safety and effectiveness of implanted medical devices

By Diana Zuckerman, PhD; Paul Brown, BA; Aditi Das, PhD

The US Food and Drug Administration (FDA) approves about 400 implanted medical devices each year through an abbreviated process called the 510(k) process, which only rarely requires clinical trials (studies of patients).  These implants include potentially high-risk devices such as heart valves, spinal implants, and hip and knee joint replacements.  In contrast, fewer than 20 implants each year are approved by FDA through the more rigorous Premarket Approval (PMA) process that requires that the impact on the patients’ health be studied in clinical trials

The law requires that all medical implants sold in the U.S. provide a reasonable assurance of safety and effectiveness based on scientific evidence, even if the company doesn’t provide data from studies of patients.  The scientific evidence could include bioengineering studies, for example, to determine if the new implant is expected to be similar to older implants on the market.  This groundbreaking study examined the scientific evidence provided to the FDA and the public for 1155 medical implants.

The study focuses on 50 representative medical implants newly cleared by the FDA 510(k) process from 2008-2012.  Since those implants were required to be substantially equivalent to implants already on the market, called “predicates,” the study also included the 1105 “predicates” for those 50 implants.  The implants included hip implants, surgical mesh, cardiac implants, spinal implants, and dental implants, among others.

By law, every 510(k) medical device application since March 15, 1995 must provide a summary that includes “sufficient evidence” to prove that the new device is substantially equivalent to a device already on the market.  This summary must be available online, linked to a letter on the FDA website granting the company clearance to sell their device in the U.S.  If a company doesn’t provide that summary, the law requires it to provide all the same information within 30 days of a written request from anyone.  This is especially important for implants, since they can cause permanent damage or even be life-threatening if they fail while inside the human body

RESULTS:  Just 3% of the 1155 implants in the study provided scientific evidence that the implant was substantially equivalent to its predicate or that it was safe and effective.   Even when predicates had been permanently recalled because of fatalities or serious injuries, there was no evidence of how the newly cleared implant was similar or different from those predicates.  Since 2009, the FDA leadership has repeatedly focused on the need for transparency at the FDA, but even in 2012, the 3rd year of new leadership at the FDA, only 20% of the implants had any kind of publicly available scientific data.

CONCLUSIONS: For the 1155 implants studied, “sufficient detail to provide an understanding of the basis for a determination of substantial equivalence,” as required by law, was unavailable to physicians, patients, providers, and others who care about patient safety or public health. Numerous newly cleared implants had predicates that had been recalled because of serious risks, calling into question the safety of the newer “substantially equivalent” implants. Regardless of how much effort physicians and providers would make to find out about the safety or effectiveness of new medical implants, they do not have access to such information for approximately 84% of medical implants that have been cleared in recent years, and more than 95% of those cleared in the last 2 decades.

IMPLICATIONS:  Previous criticisms of the FDA medical device approval process by the Institute of Medicine and public health researchers have focused on the lack of clinical trial data proving safety and effectiveness.  The FDA has repeatedly claimed that the agency has other rigorous scientific criteria for approval, such as bioengineering data proving that the new devices are so similar to the ones already on the market that they will be just as safe and effective.  This is the first study to examine the “other scientific data” that the FDA explains that it relies on.  The results indicate that either the companies are not providing solid scientific data proving that the new devices are substantially equivalent to the predicates, or that the companies are not making a sufficiently detailed summary of their scientific evidence publicly available, as required by law.  In cases where an earlier “predicate” was recalled because of serious or possibly fatal risks, the researchers found no publicly available scientific evidence to inform physicians or patients whether the newer implant would have the same potentially fatal flaw.  To safeguard the health of patients, the FDA should enforce the law.

From JAMA Internal Medicine, published online September 29, 2014.  A link to the official abstract of the article is available here. A podcast with Drs. Zuckerman and Sharfstein is available on the JAMA Internal Medicine website and can be found here.