Deadly Pseudomonas strain kills 4 NICU infants
Strain found on HCW’s hands, NICU surfaces
A recent deadly outbreak in a neonatal intensive care unit at Children’s Hospital in Boston was caused by a strain of Pseudomonas aeruginosa that was antibiotic-susceptible but strikingly virulent, reports Bela Matyas, MD, medical director of the epidemiology program at the Massachusetts State Health Department in Boston.
The hospital temporarily closed the NICU after four infants died of bloodstream infections in a six-week period in July and August. Heightened virulence of the outbreak strain was reflected both in terms of attack rate and subsequent mortality, says Matyas. All five infants colonized with the pathogen developed symptomatic bloodstream infections and four (80%) died, he notes. The route of transmission and several other aspects of the outbreak remain under investigation, but the epidemic strain was recovered from the hands of one health care worker and two surfaces in the NICU environment, he adds.
"It is speculative as to the direction of acquisition," he says. "It is not clear at this point how the five kids got it."
At least two of the fatal bloodstream infections appear to have been preceded by pneumonia, Matyas adds.
"They were all intubated children and the most likely portal of entry was the respiratory system," he says. "It’s very likely that all four of them had pre-existing pneumonias prior to the bloodstream infection forming, but the cause of death was bacteremia."
P. aeruginosa causes about 10% of the 2 million nosocomial infections reported annually, but the rapid progression of the infections in the outbreak was uncommon.
"Pseudomonas is not, generally speaking, anywhere near this aggressive an organism," Matyas tells Hospital Infection Control.
From colonization to death in 12 hours
In one case, for example, an infant lapsed into a coma and died of bloodstream infection 12 hours after a sputum sample showed only colonization with the pathogen, he says.
"That is extraordinarily fast for an organism like Pseudomonas, which would typically take days to get to that point," Matyas says. Appropriate antibiotic treatment delivered orally and via inhalation was initiated after the colonization was confirmed, and clinicians would have likely implemented intravenous administration to treat symptomatic infection, he notes.
"So the issue with respect to treatment wasn’t that the bug was resistant," he says. "But rather, treatment was difficult to initiate in time because by the time symptoms were clear it was already too late."
Despite the indicators of atypical virulence, Matyas discounted initial speculation that some sort of "superbug" Pseudomonas caused the nosocomial outbreak.
"We don’t know enough about all the different strains that are out there to say that this is unique," he says. "There may be dozens of strains that can do this, but they don’t attack very often, so we don’t see this happening very often."
Hospital officials declined to comment for this article, but distributed a question-and-answer handout for patients and the public and held a Sept. 15, 1997, press conference in Boston. (See related story, p. 167. ) A statement made at the press conference by Edward J. O’Rourke, MD, medical director of infection control, included the following remarks:
"Pseudomonas bacteria are very common in our general environment, as they grow in moist areas or in water. These bacteria do not normally cause disease in healthy infants, but can cause illness when a patient is severely ill already or when medical interventions, such as the tubes used to attach the ventilator to an infant’s lungs, are used. Because most of the infants in the NICU are on ventilators, this group of patients has a higher risk of infection with these bacteria. . . . It has been reported that this may be a unique strain of bacteria. From a medical and scientific point of view, it is speculative to say that this is an unusual or special strain. It is correct that we have had four deaths due to this infection, but in the context of an infection in a critically ill infant, it is not possible to say there is anything unusual about this organism."
In addition to the one health care worker in the NICU who had a positive hand culture for the epidemic strain, three other NICU workers cultured positive for different Pseudomonas strains, Matyas says. However, transient carriage on dry hands would be very unusual as the source of transmission in an outbreak of Pseudomonas, which typically causes problems by contaminating feeding solutions or other moist sources in the environment, he adds. The two positive environmental samples were a flow meter on a ventilator and a faucet handle in an NICU sink, he adds. However, the environmental sources were not considered significant reservoirs to fuel such an the outbreak, he added.
"Most nosocomial outbreaks of Pseudomonas involve some major component of environmental transmission," he says. "That is not evident here."
For example, another Pseudomonas outbreak in a newborn ICU reported earlier this year by clinicians at Temple University in Philadelphia was tracked to feeding solution contaminated during preparation. Failure to properly disassemble and disinfect a blender used to mix baby formula ultimately was determined to be the cause of the outbreak, which led to the deaths of one infant in the NICU. In light of the outbreak, investigators reported other hospitals may not be following optimal formula preparation and monitoring policies. (See guidelines, p. 168.) Tests revealed the formula blender housing and blades were contaminated with the strain that caused the outbreak, which was reported in New Orleans at the 1997 annual conference of the Association for Professionals in Infection Control and Epidemiology (APIC) by Keith St. John, MS, CIC, infection control professional at Temple University Hospital.
"Hospital policies concerning infant formula preparation and monitoring are often lacking," he said at APIC. "I must say the dietary department cleaned up their act. In fact, we have banned the blender. The blender is no more in our institution as a result of this."
The use of powdered formula was adopted as a cost-saving measure, but now the hospital has returned to the practice of buying prepared formula, he noted. Any additives to the prepared formula are done under pharmacy supervision using laminar flow hoods and aseptic conditions.
"I just caution everyone, take a look at your facility, because we are all trying to save a dollar," he told APIC attendees.
Another NICU outbreak reported in a poster session at APIC involved three low-birthweight infants who became infected with Candida lusitaniae at the Medical University of South Carolina in Charleston. Transmission could have occurred via the hands of health care workers, contaminated hospital equipment, or a common vehicle to which all three babies were subjected. One of the infants died.
Though hand cultures were not done, investigators concluded that because the organism was never isolated from equipment or environmental furnishings, it was most likely that hand carriage was implicated in the horizontal spread of the organism from one infant to another. Interventions recommend by a multidisciplinary infection control task force included improved hand washing practices, reported Beth Rhoton, RN, MS, CIC, infection control practitioner at the facility. Educational inservices were conducted, and waterless alcohol hand washing stations were added to ease and encourage compliance. In addition, a practice monitoring sheet was developed so health care workers could observe each other as part of learning a standardized approach to infection control in the unit. (See monitoring sheet, p. 166.)
"We put together a basic infection control practices list hand washing is one, of course and there are some related to isolation precautions and the handling of equipment," Rhoton tells Hospital Infection Control. "Eventually we hope to have everybody in the unit trained so that everybody is working with the same basic infection control ideas. It is not to be punitive. Part of the educational process will be how to communicate to somebody that they are not doing something right, so that we don’t have hurt feelings."
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content