Data Raises Questions On FDA Oversight


Flawed Statistics Paint Murky Picture of Rings’ Safety Record

When the medical device trade association AdvaMed petitioned the FDA to reclassify annuloplasty rings in 1997, it argued that data on adverse events involving the rings showed few problems, pointing out that just 465 events had been reported to the FDA between 1991 and 1995.

Based in part on that analysis, the FDA moved the rings from Class III to Class II in 2001, which allowed manufacturers to market new models without having to conduct clinical trials.

Today, there have been about 3,400 adverse events involving annuloplasty rings during the most recent five-year period, and the rings have had more deaths associated with them than any other device in their class. They have been linked to just 56 fewer deaths than replacement heart valves, a Class III device that requires clinical data to gain approval.

When the FDA reclassified annuloplasty rings, the devices weren’t in widespread use. Since then, use of the rings has grown steadily and the consensus among physicians is that they usually provide better outcomes than artificial valves.

Because of flaws in the agency’s Manufacturer and User Facility Device Experience (MAUDE) data, it’s not possible for patients to draw definitive conclusions from the numbers. But they do raise questions about whether the agency was justified in reclassifying the rings.

“This exceptionally large number of reported deaths and serious problems should be a red flag to the FDA that the devices require the closer scrutiny of a Class III device, such as clinical trials and inspections,” said Diana Zuckerman, president of the National Research Center for Women & Families.

Among the problems with the MAUDE data is that there’s no way to figure out the number of adverse events for a device relative to how many have been sold. Manufacturers collect information on how many devices are sold, but the industry considers that information proprietary.

Nor is there a way to be certain whether devices were directly responsible for injuries, illnesses and deaths listed in the database.

That lack of information highlights how little the FDA knows about the effectiveness of devices once they’ve gone on the market.

“You don’t have an adequate postmarket surveillance in place,” said Dr. Robert Hauser, senior consulting cardiologist at Minneapolis Heart Institute Foundation, Abbott Northwestern Hospital. “If they had an adequate postmarket surveillance in place, you can pick up low frequency events and know when problems occur.”

Reporting adverse events is voluntary for hospitals and doctors, but manufacturers are required to report deaths and injuries. About 98 percent of the adverse events on annuloplasty rings involve those made by Edwards Lifesciences, but the FDA can’t say whether the company has done a better job reporting than other manufacturers or whether there’s a problem with its devices. Edwards declined to comment on the Tribune’s findings.

When the FDA inspected Edwards Lifesciences in September 2009, the agency found that the company had failed to report numerous adverse events.