How do makers of implanted medical devices react when one of their products starts breaking?
One answer can be found in the case of a hernia repair device made by a subsidiary of C. R. Bard Inc. In late 2005, the company sent out a recall, urging doctors to stop using some versions of the product because a plastic component could break and cut through a patient's internal organs and tissue.
At the time, Bard executives said they knew about some serious injuries potentially caused by the device, which is known as the Kugel patch. Since then, the Food and Drug Administration has received reports of more than 80 injuries and other problems possibly related to it, including several fatalities.
Bard officials said recently in a statement that they had not recalled the product sooner because complaints about failures were too few and unrelated to raise any alarms. The company also said that both it and the subsidiary, Davol Inc., had reacted responsibly.
But when F.D.A. officials inspected Davol in early 2006, they apparently found other reasons that company officials and the agency might not have been alerted earlier. For example, inspectors reported "discrepancies" and "inconsistencies" in how Davol tracked and analyze device-related complaints. And, in several instances, Davol also did not accurately report the possible severity of complaints to the agency, a copy of an inspection report obtained by The New York Times under the Freedom of Information Act shows.
Bard officials, who declined to be interviewed, said in a statement that they have since addressed the agency's concerns, and agency officials said they are monitoring the company's progress.
Still, at a time when the use of implanted medical devices is growing sharply, the episode of the hernia patch is an example of what some experts say is a far wider problem: that some major manufacturers, while contending that they carefully monitor product safety, are not as rigorous as they should be.
Harold Pellerite, a consultant to the medical device industry who worked for two decades at the F.D.A., said he was surprised to discover how little attention many companies pay to the issue.
"Firms take the opposite approach of what the F.D.A. expects," said Mr. Pellerite, who works for Quintiles Consulting in Rockville, Md.
Davol is not the only company where agency inspectors have found problems with how a manufacturer tracks and reports possible device failures.
Another case in point: a liquid plastic material called Enteryx once sold by Boston Scientific as a treatment for patients with severe acid reflux. The company learned after a patient's death in mid-2004 that some doctors, trying to inject the product at a specific site in the base of the esophagus, could miss and cause damage to other organs, including the heart.
But when agency inspectors went to Boston Scientific in mid-2005, they could find no evidence that the device maker was adequately tracking reports from doctors about that precise problem, a warning issued last year by the agency shows. Soon after the inspection, the company recalled Enteryx, which may have caused dozens of injuries, and perhaps some deaths.
In a statement, Boston Scientific, which was also warned by the agency in that same letter about systematic failures in its product-safety review, said it had since overhauled its procedures.
"We have greatly improved the methods we use to analyze complaints and complaint trends," the company said.
Under F.D.A. rules, producers are required to have systems in place to collect and analyze any complaints they receive from doctors and hospitals about their devices. They are also required to send to the agency any report indicating that a device's failure may have contributed to a patient's death or injury.
But the agency gives device makers significant leeway about how to set up such systems, including what guidelines to use to help them determine when to issue product alerts to doctors or to issue a recall. In addition, a company can decide not to forward a complaint to the agency if it decides after an internal review that its product was not at fault.
One result, said several experts like Mr. Pellerite, is that companies tend to give themselves the benefit of any doubt.
Not long ago, one group of producers, makers of implanted heart devices, agreed to adopt guidelines about how doctors should be alerted to potential product hazards after a controversy involving the Guidant Corporation. But the broader industry has not followed suit, even though seemingly innocuous devices like the hernia patch can cause as much, if not more, havoc in patients.
Kenneth W. Goodman, a co-director of the ethics program at the University of Miami who also served as an adviser to Guidant, said he believed that companies needed to do so because the potential stakes involving implanted medical devices are so high.
"If a doctor holds a device in their hand that might fail and you know about it, then the doctor needs to know about it," Mr. Goodman said.
Chris Lavanchy, a director at ECRI Inc., a firm that evaluates medical devices, said that companies were concerned that sending up a flare too early might jeopardize sales when there was no reason for concern.
"As a producer, you want to feel justified there is a real problem before undertaking a recall," Mr. Lavanchy said.
In the case of the Kugel patch, both Bard and the F.D.A. have urged doctors not to remove the device unless a patient starts experiencing abdominal pain or a fever, which may be a sign that a fractured device has cut tissue or is causing a peritonitis-like infection.
The recalled large Kugel patches were used frequently to treat abdominal hernias. The small and medium-size versions were not affected. (The Kugel patch is just one of several types of devices used to repair hernias.)
Cynthia J. Wilson, who works for a health care system in Milwaukee, said that she had to undergo emergency bowel surgery in 2004 to repair a hole caused by a dislocated patch.
"I went to the hospital for two weeks and my friends were praying for me," said Ms. Wilson, who has filed a lawsuit against Bard.
The company declined to say how many lawsuits involving the device had been filed or whether it had settled any of them.
The Kugel patch is made of two pieces of mesh that surround a flexible plastic ring. To implant it, a doctor folds a patch, places it at the site of the hernia, which is a spot where an organ pops through internal tissue that has become weak or torn.
The released ring then springs back into its original shape, flattening the patch. The mesh like material then serves as a substrate that allows internal tissue to grow and resolve the hernia.
Davol, which is based in Cranston, R.I., makes the patch in various sizes and shapes. And it was not long after it introduced the largest units in 2002, F.D.A. records show, that company officials began to receive a small but disproportionately high number of complaints about broken rings in those "extra large" versions of the device.
Some of those complaints, agency records indicate, suggested that serious injuries like the one experienced by Ms. Wilson might be occurring.
Bard officials said that the initial numbers were small given the number of patches sold, and the problems cited were too random to form a troubling pattern.
But in mid-2005, Bard said, Davol received 10 complaints about ring breaks over a three-month period starting that June. Six of those reports, including one involving a possible fatality, came from Germany, half of them from the same doctor there.
Bard said that Davol officials initially had concluded, based on their review, tha
t the ring breaks were occurring because doctors were improperly folding the patch while implanting it.
As a result, Davol halted production in August 2005 of the extra-large patches, began to hold training sessions for doctors in Germany and started work on revising its instructions to doctors elsewhere about how to implant the device, Bard said.
But the company kept distributing the patch and doctors kept implanting it. That allowed the problem to continue because Davol's assessment proved wrong.
In early December 2005, tests run by the company on a failed patch "suggested for the first time that the source of the ring break was caused by a failure at the ring weld," according to Bard.
Davol soon recalled the extra-large versions of the patch and, several weeks later, expanded that recall to include large versions of the device. It was after that initial patch recall that F.D.A. officials went to Davol's headquarters in early 2006 to do an inspection.
What they found there, their report indicates, was a range of problems, including flaws in the system used by Davol to track complaints.
For example, while Davol had purchased a new tracking system in 2004, that program was incompatible with the earlier one it had been using, resulting in significant "discrepancies" in how problems were recorded and reviewed, agency inspectors reported.
Davol officials disagreed, that same report shows, saying they cross-checked complaints by hand. But inspectors also reported that company officials had also understated in several reports to the agency the potential severity of device-related injuries, including three reports involving injuries and one report involving a possible device-related death.
In a statement, Bard said that it had not submitted those reports to mislead the agency but said that the company had incorrectly categorized the complaints based on the information it had at the time.
Davol also retained an outside safety consultant around the time of last year's inspection to review its operations and recommend changes, the F.D.A. report shows. The company has since redesigned all the versions of the Kugel patch that it recalled, and those products are back on the market.
Tim Ulatowski, an F.D.A. enforcement official, said the agency was monitoring Davol's response to its concerns, adding that it was satisfied with what it had done.
Mr. Lavanchy, the device expert at ECRI, said he believed that the pace of Davol's reactions to problems with the Kugel patch seemed understandable given that it appeared from the company's account that it was in the dark for a long time about both the scope of the problem with the device and its cause.
But Professor Goodman of the University of Miami was less forgiving about how long it took for the company to take action.
"It doesn't take a surgeon to figure out that if you have a plastic object in an abdomen that breaks, it is going to hurt someone pretty badly," he said.