Recall brawl: FDA plans no changes to system after bronchoscope controversy

Recall brawl: FDA plans no changes to system after bronchoscope controversy

But CDC is developing electronic warning system

Despite calls for action from the medical community, a Food and Drug Administration (FDA) official says there are no plans to change the current device recall system after the deaths of two patients were linked to previously recalled bronchoscopes.

"Hospitals need to take measures to see that the risk manager, or whoever the appropriate party is, receives the recall notice," says Harold Pellerite, assistant to director of the FDA office of compliance. "[But recall system changes] based on this case? No. Communication is always the most critical part, and that is one of our biggest concerns. In the case of a recall, we want to make sure that people who are affected by it are properly notified. The Olympus America case appeared to be what we expect [from a manufacturer]."

In light of that stance, infection control professionals should make sure their health care facility has some type of fail-safe system to ensure a medical product recall notice doesn’t fly in under the radar. That appears to be what happened at Johns Hopkins University Hospital in Baltimore, where clinicians continued to use the bronchoscopes well after a Nov. 30, 2001, recall letter was sent out by Olympus America Inc. in Melville, NY. The recall notice went to a hospital loading dock area and remained unread by clinicians, says Gary Stephenson, Johns Hopkins spokesman. A factor may have been a general mail backup at the hospital due to the anthrax scares, he adds.

"Obviously, the notification process should have been more aggressive and comprehensive," he says. "There should have been a follow-up, as is customarily done. I tend to think that is more of an anomaly than anything else. It’s just that everything that could have gone wrong in this one case went wrong."

Some 410 Johns Hopkins patients underwent bronchoalveolar lavage (BAL) between June 1, 2001, and Feb. 4, 2002. Bacterial contamination with Pseudomonas aeruginosa was confirmed in three of seven bronchoscopes used for examining the patients. A statement released by the hospital said, "infections related to the contaminated scopes may have contributed to the deaths from pneumonia of two already critically ill patients. However, determining cause and effect is complex in this group of extremely sick patients among whom some deaths are to be expected due to the severity of their illness."

Responding to a request for comment from Hospital Infection Control, Olympus America states: "The recall letters sent to Johns Hopkins, as with every customer, were sent to the address the customer provided to us when they purchased their instruments. The recall letters were sent via certified return-receipt mail, addressed to the attention of the Bronchoscopy Lab Manager/ Risk Manager with "URGENT RECALL" printed in large capital letters on the envelope and atop the first page of the letter."

But both the company and the FDA have been criticized for not aggressively publicizing the recall to ensure all providers were aware.

"This does not look like the product of a very assertive regulatory agency that has patient safety first and foremost in mind," says William Schaffner, MD, chairman of the department of preventive medicine at Vanderbilt University in Nashville, TN, and a senior investigator in a 2001 Tennessee case that originally led to the recall. "Should this not be the occasion to re-look at this whole process of the FDA relationship with manufacturers, the issuance of recall notices, and how the recall is actually going to be carried out?"

Olympus America told HIC it would support the development of a working group composed of industry, FDA, and health care organizations to develop a rapid, efficient system for recall communication. In that regard, the Centers for Disease Control and Prevention (CDC) already is in the process of developing a warning system that may help prevent such occurrences in the future.

"The [CDC] division of healthcare quality promotion is developing a health care safety network, which is an electronic system for monitoring complications associated with invasive devices," says Julie Gerberding, MD, MPH, director of the division. "So this is the kind of thing that a reporting network [like this] could help with. And also, perhaps [it would be] a better method for getting the word out when there is a recall."

Loose cap lets bugs in

The original recall was based on an investigation at Skyline Medical Center in Nashville in September-October 2001, after clinicians began noticing an unusual increase in P. aeruginosa cultures from bronchoscopy specimens.

"We cultured everything imaginable — all the solutions, the washer, and the lines," says Gaye Mayernick, RN, CIC, director of infection control at Skyline. "We were just not able to detect it. We were getting ready to close our [bronchoscopy] unit."

State health officials and CDC investigators were called in, and eventually the problem was identified.

"During the course of the investigation, it became apparent that this problem occurred after the hospital received two new bronchoscopes," Schaffner says. Comparing the scopes against patient records, the investigators found that two older-model bronchoscopes had no positive cultures. "All the positive cultures were associated with the two new bronchoscopes," he says.

The biopsy port on the new scopes had a cap that was designed to remain affixed but was found to become loose over time. The cap apparently became loosened during cleaning, but all of the environmental cultures related to the reprocessor were negative. However, the implicated P. aeruginosa strain was found in the drain of a sink where the scopes were washed off before being sterilized. Investigators did not recover the pathogen from the faucet but hypothesize that it was in the water and multiplied once it stowed away within the scope port beneath the loosened cap.

"If you give a few organisms a chance to get into a protected place, as they did once the cap was loose, they will actually multiply in a little biofilm in that biopsy port housing," Schaffner says. Investigators were able to demonstrate that once contaminated, the bronchoscope could be sent through the reprocessing procedure three times in succession and still not sterilize the housing.

"You could not clean it," Mayernick says. "We did three pre-cleaning processes and sterilized it three times to make sure we had three sets of negative cultures before we used it again. Knowing where the problem was, we thought that it would be easy. We paid particular attention to pre-cleaning that area. All of the cultures after the first two pre-cleanings were still positive for Pseudomonas aeruginosa, and after the third step, some of them were still positive." Thus, because there were bacteria in the biopsy port when BAL specimens were taken, the specimens could become contaminated. That resulted in false-positive microbiology results and some unnecessary antibiotic treatment of patients.

"It became apparent, fortunately in our experience only modestly, that the contamination could actually get down into the patient and cause illness," Schaffner says. "We had one such Pseudomonas pneumonia, which was properly diagnosed and treated."

Investigators alerted the company, the FDA, and the CDC of their findings. "We thought that from our point of view locally that the problem had been solved and the ball was in the park of the company and the FDA," Schaffner says. "The company issued a voluntary recall. In my opinion, the voluntary recall was not conducted aggressively."

The FDA has the authority to mandate a recall of medical devices, but that action has a high threshold, including that the device poses "a significant risk to public health," Pellerite says. Therefore the company undertook a "voluntary" recall with communication and oversight by FDA.

"In this particular case, this problem did not rise to the level of what would fall within the category of us being able to mandate the recall," Pellerite says. "For all intents and purposes, 99% of recalls are voluntary actions. We just monitor to make sure they are effective and their methods are proper. Firms have to report actions that they take to remove products. There is an obligation to report to the agency. The action, however, is a voluntary action."

Still, Schaffner questions why the FDA is not more assertive in such cases. "For some time, many of us have thought that the unit within the FDA that deals with medical devices is probably underfunded," he says. "The procedures they have in working with industry don’t have the same authoritativeness and aggressiveness that the drug unit within the FDA has."

Beyond the FDA issues, Schaffner questions the wording of the Nov. 30, 2001, Olympus recall notice, which states in part: "We are initiating this action after receiving a customer complaint regarding a loosened bronchoscope port and microbial contamination of the port. "Nothing is mentioned regarding false-positive cultures and inappropriate antibiotic therapy," he notes. "Nothing is mentioned about infections. I think that is a very substantial omission."

Of course, the Tennessee case only had one infection, but there is ample evidence in the literature of transmission of pathogens by bronchoscopes, he adds. "The infection hazard is right there," Schaffner says. "This is not complicated stuff. There clearly is a hazard of infection and contaminated cultures and inappropriate antibiotic therapy. There is no doubt about that." Schaffner particularly objected to a sentence in the recall notice that read, "An additional factor in this instance was the failure by the user [the hospital] to maintain the automated endoscope reprocessor to the manufacturers recommended instructions."

While it is technically true that the hospital officials didn’t change the water filter as frequently as they should have, the reprocessor cultures were all negative and the problem was found only in the new scopes, Schaffner says. The outbreak pathogen was not traced back to the reprocessor, but a sink drain.

Observers wonder whether the wording may have blunted the immediacy of the recall by at least implying that part of the problem may have been local. "I think that hurt them," Mayernick says. "That had nothing to do with the persistence of Pseudomonas in that area in the biopsy port. It had nothing to do with it."

It’s a little like a car company recalling a vehicle because a wheel is falling off, but noting that the owners didn’t change their oil, Schaffner says. "This is a sentence that deflects a sense of responsibility from the manufacturer of the device to the hospital. It has nothing to do with the creation of this hazard as far as we could tell," he says.

Olympus tells HIC that the information was included because "our investigation indicated that the maintenance of the [automated endoscope reprocessor] could have been a factor, and we believe the responsible and prudent action was to advise our customers of that information."

The scopes were returned and retrofitted by the company, and the problem appears to have been eliminated. "We actually had our retrofitted scopes back before Johns Hopkins knew about the recall," Mayernick says. "So far, we have had no problem. We did culture them, of course, when they came in to be sure that they were negative before we put them into service. I look at every culture that is done, and haven’t noted any more contamination.

The patient who developed pneumonia was treated and released at Skyline. "She had been admitted and improved," she says. "She is still OK. That is the good news here. Johns Hopkins wasn’t quite that fortunate."