NASHVILLE, TENN—Each year, about half a million bronchoscopies are performed in the United States. Although most of these procedures are performed without sequelae, nosocomial infection is an occasional complication. Recently, however, two separate hospitals noted a significant increase in the rate of Pseudomonas aeruginosa infections/isolates associated with bronchoscopy.

The investigation into the cause of these outbreaks found a manufacturing defect—loose biopsy port caps—to be the culprit. It also uncovered defects in the product recall system, which allowed many defective devices to remain in use for months after the original recall.

Detailed analyses of these outbreaks were published recently in the New England Journal of Medicine.[1,2] These analyses include important lessons for how future outbreaks should be handled.

OUTBREAKS INVESTIGATED

Following reports of a threefold increase in the number of P aeruginosa specimens detected in bronchoalveolar lavage (BAL) fluid (from 10.4% to 31%), Arjun Srinivasan, MD, and colleagues from Johns Hopkins Hospital in Baltimore obtained culture specimens from the hospital’s bronchoscopes and gastrointestinal endoscopes.[1] Patients at risk were identified from the hospital’s database, and their medical records were reviewed for colonization or infection possibly related to bronchoscopy.

During the outbreak, which began in June 2001 and continued for one year, 414 patients had undergone 665 bronchoscopies. BAL cultures from 97 of these patients grew P aeruginosa. In only 21 patients, P aeruginosa had been isolated from the respiratory tract before bronchoscopy was performed. In 35 patients, the organism could be cultured from non-BAL respiratory tract specimens obtained after the procedure.

Three of the 10 bronchoscopes used in the endoscopy suite tested positive for P aeruginosa. Yet, no breaches in cleaning or disinfecting procedures were found.

BAL culture results were available for 48 patients. Twenty-six of the isolates were genetically related to a P aeruginosa strain that was recovered from the bronchoscopes.

“Endoscopes of all kinds (including bronchoscopes) are very complicated and sensitive instruments,” noted William Schaffner, MD, Chief of Preventive Medicine at Vanderbilt University, Nashville, who took part in an investigation into a similar and concurrent outbreak. “The industry has performed a lot of research to ensure that they are cleaned, disinfected, and dried properly—it’s a multistage process.” The process is not foolproof, though, as both the Johns Hopkins researchers and Dr. Schaffner’s group found out.

The second outbreak occurred at a community hospital in Tennessee.[2] In September 2001, a hospital staff member reported an increase in the number of P aeruginosa and Serratia marcescens isolates among patients who had undergone bronchoscopy.

Laboratory reports for all cases of P aeruginosa and S marcescens occurring in the hospital in 2001 were reviewed. The medical records of all patients who underwent bronchoscopy were also reviewed to see if any had been readmitted to the hospital for P aeruginosa or S marcescens infections within one month of the procedure.

Between July 1 and October 31, 2001, 66 bronchoscopies had been performed in 60 patients. Eight bronchoscopes had been used for these procedures. Of the 43 specimens obtained during the bronchoscopies, 20 were positive for P aeruginosa, and six of these were also positive for S marcescens.

BROCHOSCOPES THE CULPRIT

Between July 1 and September 17, 2001, four of the eight bronchoscopes had been used, but only two tested positive for P aeruginosa. Both were new bronchoscopes from the same manufacturer, Olympus America. These two bronchoscopes (designated A and B) were used for the first time on July 13 and 24, respectively. Of 16 specimens collected with bronchoscope A, 75% were positive for P aeruginosa, as were 62% of specimens collected with bronchoscope B.

Both bronchoscopes were removed from use on September 18, and no further P aeruginosa isolates were found. Likewise, at Johns Hopkins, after removal of the same model of contaminated bronchoscope, the incidence of P aeruginosa in BAL specimens returned to baseline levels.

David Kirschke, MD, a member of the Tennessee team, found that the biopsy port caps on the suspect bronchoscopes were “loose enough to be unscrewed with two fingers, but not so loose that the problem was apparent to staff.”[2] When the threads of the biopsy ports and the insides of the caps were swabbed, nine of 12 specimens tested positive for both P aeruginosa and S marcescens, while the remaining three were positive for P aeruginosa only. One specimen tested positive after the device had been cleaned and disinfected three times using the hospital’s automated endoscope reprocessor. Similarly, in the Johns Hopkins study, one bronchoscope remained contaminated even after ethylene oxide sterilization.

On October 11, the Tennessee hospital received two new bronchoscopes from the manufacturer to replace the contaminated pair. Both bronchoscopes were found to have the same defect (loose biopsy port caps), and one tested positive for P aeruginosa after less than one month of use.

HOW DID THIS HAPPEN?

In 1997, the manufacturer had changed the design of its bronchoscopes—including the devices’ biopsy port caps. The company later reported that because of a failure to conform to production standards, bronchoscopes that did not meet manufacturing specifications had been distributed. Only the newer models, not the older bronchoscopes without the design changes, were associated with culture positivity for P aeruginosa.

How the bronchoscopes became contaminated is unknown, but it could have occurred during procedures in patients with P aeruginosa infections. Furthermore, the loose biopsy port caps interfered with cleaning procedures: When bacteria entered the biopsy port caps, they were protected from any subsequent exposure to disinfectant chemicals.

After the Tennessee group reported their findings to the manufacturer and the FDA, the manufacturer issued a recall of 15 bronchoscope models on November 30, 2001. The recall received national media attention after the Johns Hopkins Hospital team also reported their outbreak of P aeruginosa. Originally involving 4,700 bronchoscopes in the United States, the recall was later expanded to include about 14,000 bronchoscopes worldwide.

PROCEDURES INADEQUATE

“Our general impression is that the recall did conform to the FDA guidelines as they currently exist, and procedures were correctly followed,” observed Dr. Schaffner, who is also a consultant with the Tennessee Department of Health. He added, however, that “these procedures were neither efficient nor effective.”

For example, said Dr. Schaffner, “The recall notice text described ‘a potential structural problem’ without mentioning patient-related consequences.” Among these consequences, he noted, were “the risk of false-positive culture results, which could lead to some patients being treated with unnecessary and potentially toxic antibiotics, [and] the risk of infection.”

Another problem arose from the way in which the recall notices were delivered. The manufacturer first sent letters to the facilities using the bronchoscopes. When this proved inadequate, letters were sent to individual physicians.

“The recall notices often did not reach their intended targets,” Dr. Schaffner said. The recall could have been conducted more rapidly, he suggested, had the manufacturer used its sales force to contact physicians directly.

In addition, neither the manufacturer nor the FDA posted information on their Web sites until the recall was well under way.

The manufacturer subsequently repaired the defect and returned all of the bronchoscopes, Dr. Schaffner acknowledged. However, he continued, “there was no information given about how they were fixed, and there was no information about microbiological testing to prove that they were fixed and that the problem had been solved. Again, the manufacturer could have gone to the users of the bronchoscopes and presented the data: This was the problem. This is how we fixed it. These are the tests we performed. It’s safe.”

The bottom line, according to Dr. Schaffner, is that the two outbreaks could be used as a “sentinel event” to precipitate a review of how the FDA and the medical device industry proceed when a defect is discovered—“especially,” he stressed, “one that has a direct impact on patient health.”

—Gale Jurasek

References
1. Srinivasan A, Wolfenden LL, Song X, et al. An outbreak of Pseudomonas aeruginosa infections associated with flexible bronchoscopes. N Engl J Med. 2003;348:221-227.
2. Kirschke DL, Jones TF, Craig AS, et al. Pseudomonas aeruginosa and Serratia marcescens contamination associated with a manufacturing defect in bronchoscopes. N Engl J Med. 2003;348:214-220.

Return to table of contents