Pediatric infection control enters new era with children’s hospital network

Planners envision nationwide pediatric infection prevention network

Long overshadowed by infection control efforts targeted at adults, a new era in pediatric infection prevention is opening through a nationwide hospital network that is expected to end the dearth of data in the field and standardize surveillance and intervention efforts.

Currently, some 50 pediatric hospitals have joined in the first national effort to address nosocomial infections, antibiotic resistance, and other infection control issues in hospitalized children. The project is funded by the Centers for Disease Control and Prevention through a cooperative agreement with the National Association of Children’s Hospitals and Related Institutions (NACHRI) in Alexandria, VA.

Baseline data-gathering surveys were completed in 1998. Plans for this year include the first national nosocomial infection prevalence survey in pediatric hospitals. While there has been increasing concern about preventing antibiotic resistance and pediatric nosocomial infections, surveillance systems and intervention approaches traditionally have been targeted at adult patients, notes an ICP at one of the participating hospitals.

ICP role is critical to success of pediatric network

"It is definitely needed," says Pat Medcalf, RN, infection control director at Children’s Medical Center in Dallas. "At the current time, there is not really any organized pediatric network of children’s health care facilities that can provide guidance for the development of policies, surveillance tools, or the evaluation of programs in general. I’m very excited that ICPs have a big role in this whole process."

In addition to well-established sources of nosocomial infections such as invasive devices and health care workers, pediatric patients may be exposed to maternal infections, contaminated infant formula, visitors and siblings, and fomites such as toys shared with other patients on pediatric units, experts in the field advise.¹ Host factors contributing to risk for nosocomial infection include congenital abnormalities and immaturity of the immune system, especially in newborns and premature infants. The rates of nosocomial infections in individual neonatal intensive care units (NICUs) have been reported to be as high as 7% to 25%.

"In the pediatric setting, you have all of the childhood communicable diseases," Medcalf adds. "You have to deal with pertussis, chickenpox, and RSV [respiratory syncytial virus]. You also probably have more visitors because you have whole families involved. Just the way that you interact with children is more intimate, so you are more likely to have exposures. [But] in general, people who work in a pediatric facility feel a little more attached to their patients and might want to do a little more for them. There are definitely still problems with infection control techniques, and we don’t have 100% compliance by any means, but nobody wants to see a kid sick."

While some pediatric infection data are collected as part of CDC’s National Nosocomial Infections Surveillance (NNIS) system, planners envision development of a far-reaching pediatric network that could lead to widespread interventions supported by CDC guidelines. NACHRI is a nonprofit group of more than 150 acute care children’s hospitals, large pediatric units, and related institutions. Overall, there are some 250 pediatric hospitals in the United States, and the network may extend to children’s hospitals in Canada as well.

"We’re trying to develop a pediatric network that would include virtually all children’s hospitals in the United States and Canada," says William Jarvis, MD, one of the project principals and chief of the investigations and prevention branch at the CDC hospital infections program. "One of the complaints about NNIS is that most of the protocols were driven primarily for adult patients. One of the desires of the pediatric infectious disease and infection control community has been for recommendations and guidelines from CDC that are specific to their patient population. We see this as a golden opportunity to do that."

For example, infection control studies may be done at five to 10 of the hospitals. Recommenda tions for practices that demonstrate efficacy can then be extended to the rest of the group, he adds.

As an initial step to gather baseline data, the participating hospitals were surveyed last year regarding the numbers and types of infectious disease clusters and exposures they investigated from Jan. 1, 1996, to March 31, 1998.

"We realized that in pediatric facilities, probably more than in adult acute care facilities, a lot of time is spent on little clusters of varicella, MRSA, etc.," Jarvis says. "We were actually surprised at the number and amount and types of outbreaks that were occurring."

Indeed, some 28 hospitals responding to the survey reported a median of 5.5 infectious disease clusters investigated for the period, though there was substantial variation among reporting institutions (range 1-21).² The most frequently implicated pathogens were varicella, gram-negative bacilli, methicillin-resistant Staphylococcus aureus (MRSA), aspergillus, Bordetella pertussis, Neisseria meningitidis, RSV, rotavirus, and vancomycin-resistant enterococci (VRE).

Focusing on occupational health

Infectious disease clusters occurred most often in NICUs. In some cases, health care workers were the source of varicella or pertussis exposures, while others had signs and symptoms of pertussis after exposures but no laboratory confirmation of infection. The finding underscored the need for improved health care worker immunizations, antimicrobial use, and compliance with infection control practices.

"I don’t know that [pediatric] ICPs have spent a great deal of time on occupational health issues," Jarvis says. "Often that is an occupational health department responsibility. We were surprised to see how many of the outbreaks seemed to be transmission to health care workers, or health care workers transmitting to other workers or patients. The occupational arm of the process is a really important one that more attention needs to be focused on."

Another survey of nosocomial infection rates revealed considerable variability in NICU and pediatric intensive care unit (PICU) rates. Sources of variation include differences in case mix, patient care practices, and data-collection methodology. Of the 32 hospitals responding to the survey, 30 use current or modified CDC infection definitions. Median overall infection rates per 1,000 patient days were 10.6 (3.5-14.7) in NICUs and 11.6 (1.1-29.2) in PICUs. Median nosocomial bloodstream infection rates per 1,000 central venous catheter days were 6.9 (2.5-17.6) in NICUs and 6.6 (0-12.2) in PICUs.²

"What we are finding is that some of the pediatric populations that are at high risk are not totally different from adults," Jarvis says. "NICU, hematology/oncology patients, and those that undergo surgical procedures are the ones at greatest risk. It is very similar to adults in that those that are exposed to [invasive] procedures and devices are at highest risk."

The need for standardized surveillance methods was underscored by the survey finding that 42% of the hospitals could not provide nosocomial infection rate data that was suitable for comparative analysis, adds Beth Stover, RN, CIC, one of the investigators in the project and an ICP at Kosair Children’s Hospital in Louisville, KY. Of particular concern was the variability in the number of months that hospitals were conducting surveillance on high-risk populations, with some reporting only three months of data on pediatric ICUs and intensive care nurseries, she notes.

Surveillance should be consistent

"Pediatric ICU and intensive care nurseries are certainly populations in children’s hospitals where there are infection control problems that need to be addressed," Stover says. "You may identify a colonization or nosocomial infection problem early if you are performing consistent nosocomial infection surveillance in these high-risk populations. When that surveillance is not being done consistently, then there is potential for gaps or problems to occur that may be not be identified as readily or rapidly."

A contributing factor to the finding may be that pediatric ICPs, much like their colleagues in adult settings, are seeing an expansion of job duties without a commensurate increase in program resources.

"That may be a part of why many hospitals do not have the capabilities of performing surveillance in the high-risk units consistently throughout the year," Stover notes.

As the project continues, standardizing surveillance methods will be an immediate priority, says Shirley Girouard, PhD, RN, FAAN, vice president of child health and financing at NACHRI.

"One of the major findings from the surveys, and from the analysis of those rates and cluster data, is the fact that there are really issues around how surveillance is conducted and how people are measuring the outcomes related to clusters and nosocomial infections," she says. "That really is a significant finding. We suspected that, but we did not know. So one of the first efforts that we are going to be undertaking shortly is working with children’s hospitals to systematize or standardize the approach to collecting data."

In addition, project plans for this year call for a point-prevalence survey to assess infection rates and costs, something that has not been done for the pediatric population on a national scale. Researchers also will assess the prevalence of antibiotic-resistant pathogens in intensive care units in children’s hospitals. (See related story, p. 31.)

Looking for the iceberg effect’

"This will be really the first national estimate of infection rates in high-risk pediatric patients in this country," Jarvis says. "Another proposal in the project, which will hopefully be initiated this year, is to do culture surveys of intensive care unit patients looking for multiresistant organisms. This will be the first attempt to look at the VRE iceberg effect’ and [determine] how big the problem is of colonization with VRE, MRSA, and other antimicrobial-resistant organisms."

In that regard, the pediatric network also will survey the facilities for pharmacy databases, ascertaining which have computerized data and what information is being collected.

"It is a first step toward looking at the relationship between antimicrobial use and resistance, and then hopefully developing some interventions to try and reduce the inappropriate antibiotic use," Jarvis says.

As part of that effort, the facilities will be assessed for appropriate vancomycin use as recommended by the CDC’s Hospital Infection Control Practices Advisory Committee to prevent increasing resistance to the last-line drug. CDC data from pediatric facilities in Atlanta suggest that vancomycin is being used inappropriately, and the survey will attempt to determine if that is indicative of a larger problem, Jarvis explains. Examples of inappropriate use include cardiac surgeons and neurosurgeons using vancomycin as routine prophylaxis, as well as hematology oncologists frequently using vancomycin as empiric therapy for rule-out sepsis, he says. In addition, the project eventually may include some observational studies of hand washing and other infection control practices in pediatric hospitals, particularly in NICUs, he adds.

"I think in terms of compliance with hand washing, I’m hoping it is a little better, but it probably won’t be a whole lot better than the 20% to 40% of the time we see it in the adult care setting," Jarvis adds. "We are looking at this [project] as very much a child heath issue. We really could have a major impact on a very large population."

[Editor’s note: Pediatric ICPs or others interested in participating in the program can contact NACHRI via phone: (703) 684-1355; fax: (703) 684-1589; or by writing to 401 Wythe St., Alexandria, VA 22314.]

References

1. Harris JS. Pediatric nosocomial infections: Children are not little adults. Infect Control Hosp Epidemiol 1997; 18:739-742.

2. Seigel J, Goldmann D, Shulman S, et al. Infectious disease cluster and exposures investigated in U.S. pediatric health care facilities from 1/1/96 to 3/31/98. Abstract 81. Presented at the annual meeting of the Infectious Disease Society of America. Denver; Nov. 12-15, 1998.

3. Stover B, Shulman S, Bratcher D, et al. Pediatric and neonatal intensive care unit nosocomial infection rates at U.S. children’s hospitals. Abstract 638 Sa. Presented at the annual meeting of the Infectious Disease Society of America. Denver; Nov. 12-15, 1998.