PEDIATRICS Vol. 110 No. 3 September 2002, pp. 481-485
Rate, Risk Factors, and Outcomes of Nosocomial Primary Bloodstream Infection in Pediatric Intensive Care Unit Patients
* Division of Infectious Diseases, Departments of Internal Medicine
Pediatrics, Washington University School of Medicine, St Louis, Missouri
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective. The objective of this study was to determine the rate, risk factors, and outcomes of nosocomial primary bloodstream infection in pediatric intensive care unit (PICU) patients.
Design. Prospective cohort study.
Settings. This study was performed at St Louis Children’s Hospital, a 235-bed academic tertiary care center with a combined 22-bed medical and surgical PICU.
Patients. Subjects for this study were patients admitted to the PICU between September 1, 1999, and May 31, 2000.
Interventions. None.
Outcome Measures. Patients were monitored for the development of nosocomial bloodstream infections from the day of PICU admission until 48 hours after PICU discharge.
Results. Of 911 patients, 526 (58%) were male and 674 (74%) were white. Congenital heart disease (29%), lung disease (25%), and genetic syndrome (18%) were common. There were 65 episodes of primary bloodstream infection in 57 patients; 5 were polymicrobial and 7 patients had multiple bloodstream infections. Coagulase-negative Staphylococcus was the leading cause of bloodstream infection (n = 28), followed by Enterobacter cloacae (n = 8). The rate of bloodstream infection was 13.8 per 1000 central venous catheter days. In multiple logistic regression analysis, patients with bloodstream infection were more likely to have multiple central venous catheters (adjusted odds ratio [aOR]: 5.7; 95% confidence interval [CI]: 2.9–10.9), arterial catheters (aOR: 5.5; 95% CI: 1.8–16.3), invasive procedures performed in the PICU (aOR: 4.0; 95%CI: 2.0–7.8), and be transported out of the PICU (aOR: 3.4; 95% CI: 1.8–6.7) to the radiology or operating room suites. Severity of illness as measured by admission Pediatric Risk of Mortality score, underlying illnesses, and medications were not associated with increased risk of nosocomial bloodstream infection.
Conclusions This study identified a high rate of bloodstream infection among St Louis Children’s Hospital PICU patients. Risk factors for bloodstream infection were related more to process of care than to severity of illness. Additional research is needed to develop interventions to reduce nosocomial bloodstream infections in children.
Key Words: pediatric intensive care unit • nosocomial • bloodstream infection
Abbreviations: PICU, pediatric intensive care unit • NNIS, National Nosocomial Infections Surveillance • SLCH, St Louis Children’s Hospital • PRISM III, Pediatric Risk of Mortality • aOR, adjusted odds ratio • CI, confidence interval
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bloodstream infections represent a major cause of hospital-acquired infections in pediatric intensive care unit (PICU) patients.1,2 According to the 1999 National Nosocomial Infections Surveillance (NNIS) report,2 bloodstream infections (28%) are the most frequent nosocomial infection reported in PICUs, followed by ventilator-associated pneumonias (21%). PICUs have one of the highest central venous catheter-associated bloodstream infection rates—7.7 infections per 1000 central venous catheter days—with only burn units and neonatal intensive care units surpassing these rates of bloodstream infection.5 Coagulase-negative staphylococci (38%) are the most common pathogen, although Gram-negative rods are isolated in 25% of PICU patients with bloodstream infections.2
The majority of literature concerning bloodstream infection is generated from studies of critically ill adult patients. In adults, independent risk factors for developing bloodstream infections include prolonged hospitalization, severity of illness at admission, comorbidities, exposure to invasive procedures, inappropriate antimicrobial therapy, immunosuppressive therapy, use of steroids, parenteral nutrition, and histamine type-2 receptor blockers.1,7–11 Critically ill adult patients with bloodstream infections have a longer hospital length-of-stay, higher treatment costs, and greater risks for mortality.3 In a cohort of critically ill adult surgical patients, the excess hospital length-of-stay resulting from bloodstream infection is 24 days.3 The mean attributable cost of bloodstream infection is $40 000 per survivor, and the attributable mortality rate is 35%.3 Unfortunately, the substantial amount of literature on the epidemiology of nosocomial infections in adults cannot be directly extrapolated to children because of the significant differences between adults and children which include age, underlying medical conditions, process of care, and type and distribution of pathogens.2,4,5
Most PICUs care for a heterogeneous population of children of varying age, diagnosis, and underlying illness, ranging from infants with congenital anomalies to adolescents with multiple traumas. Consequently, even patients admitted to the same PICU may differ substantially in their risk for nosocomial infections. Additional epidemiologic studies in PICU patients in nonoutbreak settings are necessary to better understand the scope of the problem to develop interventions. We performed a prospective cohort study to determine the rate, risk factors, and outcomes of bloodstream infection in PICU patients.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Setting
This study was performed at St Louis Children’s Hospital (SLCH), a 235-bed teaching hospital affiliated with Washington University School of Medicine. SLCH is located in the metropolitan St Louis area and has a 300-mile radius referral base in southeastern Missouri and southwestern Illinois. SLCH has a combined medical and surgical PICU with 22 beds. There are
1400 admissions per year. In April 2000, a new facility for the PICU opened with an increase to a 28-bed capacity. In the new facility, patient rooms are separated by walls rather than curtains, and each room has a sliding glass door and a sink placed next to the door. In the old facility there were 12 sinks and 22 beds. The agents for handwashing are Triclosan soap and Alcare alcohol foam, both by Steris (Mentor, OH).
Patients
Since September 1999, there has been an ongoing prospective cohort study of nosocomial infections in the PICU. Participants for this study were patients admitted to the PICU between September 1, 1999, and May 31, 2000, excluding patients who were older than 18 years of age, died within 24 hours of admission, or were neonatal intensive care unit patients on extracorporeal membrane oxygenation occupying PICU bed space. Approval was obtained from the Washington University School of Medicine Institutional Review Board; a waiver of informed consent was requested and granted.
Data Collection
All eligible study patients were monitored for the development of nosocomial infections from the day of PICU admission until 48 hours after PICU discharge. Data on demographics, underlying medical conditions, surgeries and procedures performed, use of antibiotics, steroids, stomach acid suppressants and immunosuppressants, hospital length of stay, PICU length of stay, PICU discharge disposition, central venous catheter days, and ventilator days were collected from the medical records and daily flow sheets by 1 of the investigators (A.M.E.). The variable "procedures performed in the PICU" was a composite variable, describing the location in which procedures such as intubation, arterial and central venous catheter placement, mediastinal exploration, and initiation of extracorporeal membrane oxygenation occurred. Admission severity of illness was calculated using the Pediatric Risk of Mortality (PRISM III) score, which is a weighted score, estimated using 17 physiologic parameters from the first 24 hours of PICU admission.12 Information on potential risk factors was collected on both cases and noncases throughout the duration of PICU stay. Antibiotic-coated central venous catheters are not used in this PICU. Patients with >1 central venous catheter placed were defined as having "multiple central venous catheters." "Transport out of the PICU" was defined as the physical movement of the patient out of the intensive care unit to the radiology department, operating room, or cardiac catheterization laboratory; specific destinations were not evaluated separately in the multivariate analysis. There were no clusters of infection or outbreaks during the study period.
Definitions
Centers for Disease Control and Prevention (CDC) definitions were used to diagnose nosocomial primary bloodstream infections.13 Primary bloodstream infection cases were diagnosed with bacteria or fungi in their blood, which were not present or incubating before hospital admission and without evidence of other localized infections. Microorganisms causing bloodstream infection were isolated using routine blood cultures and their antibiotic susceptibility profiles were obtained from laboratory reports. Coagulase-negative Staphylococcus species was reported as a cause of bloodstream infection only if 2 or more blood cultures drawn on different occasions were positive for coagulase-negative Staphylococcus sp. Staphylococcal strains were not speciated. In children with >1 episode of nosocomial bloodstream infection, only the first episode of bloodstream infection was analyzed.
Data Analysis
SPSS Version 10 (SPSS Inc, Chicago, IL) was used in the statistical analysis. Differences between means were analyzed by Wilcoxon rank sum for nonnormally distributed variables and differences in distribution of categorical variables were analyzed using the 
2 test. All statistical tests were 2-tailed, and a P value of <.01 was considered significant. Clinically relevant statistically significant variables from univariate analysis were used to create a logistic regression model to assess the differences between patients with and without bloodstream infection. Variables affecting <10% of patients were not considered for entry into the logistic regression model. Regression diagnostics were used to identify outliers and colinearities. Outliers were checked and corrected for mistakes. When colinearity was detected between variables, only the variable with the highest effect size was kept in the regression model and others were recoded. The fit of the model was tested using the Hosmer and Lemeshow test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
A total of 911 patients were admitted to the PICU between September 1, 1999, and May 31, 2000. Patient demographics are reported in Table 1. There was a white predominance (74%). Slightly over half of the patients were male (58%). Approximately half of the patient population was under 3 years of age. Congenital heart disease (29%), lung disease (25%), and genetic syndrome (18%) were common. Less than 5% of the patient population were admitted because of immunodeficiency, diabetes, renal disease, multiple trauma, or burn. Fifty percent of the patients had an arterial catheter, and 12% had multiple central venous catheters. Seventy percent were mechanically ventilated, and 30% had 1 or more transfusions during their stay in the PICU.
|
Fifty-seven children developed 65 episodes of primary bloodstream infections during their PICU stay. Five episodes of bloodstream infection were polymicrobial, and 7 patients had multiple bloodstream infections. The rate of bloodstream infection in this population was 13.8 per 1000 central venous catheter days. The most common causative microorganisms are reported in Table 2. Gram-positive bacteria were responsible for most of these infections (55%). Coagulase negative Staphylococcus (n = 28) was the leading cause of bloodstream infection, followed by Enterobacter cloacae (n = 8). Candida species were responsible for all 5 episodes of fungemia. There were insufficient numbers of cases caused by any 1 organism to determine relationships between organism type and patient characteristics.
|
The mean time to bloodstream infection was 11.7 days, median 10 days (range: 2–33 days).
Univariate analysis of differences in characteristics of patients with and without bloodstream infections is summarized in Table 1. Only factors occurring in >10% of patients are reported in this table. Age, gender, and ethnicity were not significantly different between patients with and without bloodstream infections. Severity of illness at admission (PRISM III score) was higher in patients with bloodstream infections. A higher proportion of patients with bloodstream infection received transplants, immunosuppressive agents, histamine type-2 receptor blockers, total parenteral nutrition, antimicrobial agents, and steroids. Patients with underlying medical conditions, invasive devices, and those who underwent surgeries had an increased risk for developing bloodstream infections in the univariate analysis.
Most central venous catheters were placed in the internal jugular vein. It was not possible to assess the relationship between site of catheterization and risk of bloodstream infection, as there were fewer patients with femoral and nontunneled subclavian central venous catheters, and often these sites were used to place subsequent central venous catheters.
Logistic regression analysis was used to identify independent risk factors for bloodstream infections (Table 3). Patients who had multiple central venous catheters (adjusted odds ratio [aOR]: 5.7; 95% confidence interval [CI]: 2.9–10.9) and arterial catheters (aOR: 5.5; 95% CI: 1.8–16.3) were nearly 6 times more likely to develop bloodstream infections than patients who were not exposed to these factors. Patients who had invasive procedures performed in the PICU (aOR: 4.0; 95% CI: 2.0–7.8) or those who were transported out of the PICU (aOR: 3.4; 95% CI: 1.8–6.7) were also at increased risk for acquiring bloodstream infections. Severity of illness at admission, underlying medical conditions, length of stay before infection, and medications used before infection were not independent risk factors for bloodstream infection in the logistic regression analysis.
|
Univariate analysis of the outcomes of bloodstream infection revealed that patients with bloodstream infections had longer PICU (24.4 ± 26.5 vs 4.1 ± 11.5 days) and hospital (41.2 ± 38.4 vs 13.0 ± 24.9 days) length of stay. The crude mortality rate associated with bloodstream infection was 19.2%.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
We performed a prospective cohort study to determine the rate, risk factors, and outcomes associated with bloodstream infection in PICU patients. The incidence of bloodstream infection is 13.8/1000 central venous catheter days, which is within the 75th to 90th percentile reported by NNIS.2 The distribution of causative microorganisms for bloodstream is similar to what has been reported in the NNIS data.2 These infections are predominantly caused by Gram-positive organisms, mostly coagulase-negative Staphylococcus species. Enterobacter sp was the most common Gram-negative species reported in PICU patients with bloodstream infections. In multivariate analysis, risk factors related to process of care, specifically multiple central venous catheters, arterial catheters, procedures performed in the PICU, and transport out of the PICU were associated with bloodstream infection. Underlying medical conditions, severity of illness at admission, and length of PICU or hospital stay were not associated with bloodstream infection in logistic regression analysis.
The presence of an arterial catheter or multiple central venous catheters also increased the risk for developing bloodstream infections. Arterial catheters are frequently used in the management of critically ill patients for continuous blood pressure monitoring, and to obtain blood gas measurements. Frequent sampling through stopcocks may lead to increased opportunity for introduction of microorganisms into these catheters. Contamination of central venous catheters at the time of insertion or when accessing the catheter may lead to subsequent development of catheter-related infections.14 Measures to reduce catheter-related bloodstream infection that have proven effective in adults include adherence to maximal sterile and barrier precautions during catheter placement, meticulous attention to CDC guidelines for catheter care, adherence to proper hand-washing practices, minimization of the number of people manipulating the catheters, and limitation of multiple uses for central venous catheters.14–16 The SLCH PICU has a written policy for central venous catheter insertion and care that closely adheres to the CDC guidelines, recommending maximal sterile barrier precautions for insertion; use of betadine or alcohol before accessing central venous catheters; keeping catheters clamped, capped, or connected to running fluids at all times; and minimizing the number of times the catheter is accessed when possible. No specific policies or procedures exist for the management of arterial catheter stopcocks.
Patients who had procedures performed in the PICU and those who were transported out of PICU to the radiology or operating room suites were also at higher risk for bloodstream infection. Breach of sterile technique during procedures and patient transport may predispose these patients to the risk of infections. Adhering to strict aseptic techniques, minimizing the number of staff handling a patient, and increasing access to disinfectants and sterile gloves during procedures and patient transport may reduce bloodstream infections.
There are a few limitations to this study. This study was performed at a single academic tertiary care center, so results from this study may not directly apply to other institutions with different patient populations and medical practices. Because the data were obtained over the course of only 9 months between September and May, seasonal variations in frequency of bloodstream infection and causative microorganism could not be assessed. Severity of illness (PRISM III score) was calculated only at admission, and this score probably does not reflect severity of illness at the time of infection, especially as the median time to bloodstream infection was 10 days. The prospective cohort study design limited our ability to determine the attributable excess mortality and length of stay resulting from bloodstream infection. A more appropriate study design to determine the attributable morbidity and mortality of bloodstream infection is a nested case-control study matching patients on severity of illness at the time of infection, which is planned.
The risk factors identified in this study are different from those described for adults and are more related to the processes of care. This study identified several areas where implementation of preventive strategies could reduce a large proportion of nosocomial infections. Additional analysis is needed to estimate the attributable morbidity, mortality, and economic impact of bloodstream infections in the PICU.
|
ACKNOWLEDGMENTS |
|---|
This work was supported by grants from Zeneca through the Pediatric Infectious Diseases Society (Dr Elward), and in part by Centers for Disease Control and Prevention cooperative agreement UR8/CCU715087 (Dr Fraser).
|
FOOTNOTES |
|---|
Received for publication Dec 17, 2001; Accepted Apr 3, 2002.
Reprint requests to (V.J.F.) Infectious Diseases Division, Washington University School of Medicine, Campus Box 8051, 660 S Euclid Ave, St Louis, MO 63110. Email: vfraser{at}im.wustl.edu
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
- Milliken J, Tait GA, Ford-Jones EL, et al. Nosocomial infections in a pediatric intensive care unit. Crit Care Med.1988; 16 :233 –237[Web of Science][Medline]
- Richards MJ, Edwards JR, Culver DH, Gaynes RP, and the National Nosocomial Infections Surveillance System. Nosocomial infections in the pediatric intensive care units in the United States. Pediatrics.1999; 103(4) . Available at: http://www.pediatrics.org/cgi/content/full/103/4/e39
- Pittet D, Terara D, Wenzel RP. Nosocomial bloodstream infection in critically ill patients: Excess length of stay, extra costs and attributable mortality.
JAMA.1994; 271
:1598
–1601
[Abstract/Free Full Text] - Richards MJ, Edwards JR, Culver DH, Gaynes RP, and the National Nosocomial Infections Surveillance System. Nosocomial infections in medical intensive care units in the United States. Crit Care Med.1999; 27 :887 –892[CrossRef][Web of Science][Medline]
- Centers for Disease Control and Prevention. Monitoring hospital-acquired infections to promote patient safety-United States, 1990–1999. MMWR Morb Mortal Wkly Rep.2000; 49 :149 –153[Medline]
- Haley RW, Culver DH, White JW, et al. The efficacy of infection surveillance and control programs in preventing nosocomial infections in US hospitals. Am J Epidemiol.1985; 12 :182 –205
- Pittet D, Harbarth S, Ruef C, et al. Prevalence and risk factors for nosocomial infections in four university hospitals in Switzerland. Infect Control Hosp Epidemiol.1999; 20 :37 –42[CrossRef][Web of Science][Medline]
- Singh-Naz N, Sprague BM, Patel KM, Pollack MM. Risk factors for nosocomial infection in critically ill children: a prospective cohort study. Crit Care Med.1996; 24 :875 –878[CrossRef][Web of Science][Medline]
- Tokars JI, Cookson ST, McArthur MA, Boyer CL, McGeer AJ, Jarvis WR. Prospective evaluation of risk factors for bloodstream infection in patients receiving home infusion therapy.
Ann Intern Med.1999; 131
:340
–347
[Abstract/Free Full Text] - Velasco E, Thuler L, Martin L, Dias V, Goncalves V. Risk factor for bloodstream infections at a cancer center. Eur J Clin Microbiol Infect Dis.1998; 17 :587 –590[CrossRef][Web of Science][Medline]
- Wey SB, Mori M, Pfaller MA, Woolson RF, Wenzel PR. Risk factors for hospital-acquired candidemia: a matched case-control study.
Arch Intern Med.1989; 149
:2349
–2353
[Abstract/Free Full Text] - Pollack MM, Patel KM, Ruttimann UE. PRISM III: an updated pediatric risk of mortality score. Crit Care Med.1996; 24 :743 –752[CrossRef][Web of Science][Medline]
- Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definition for nosocomial infections. In: Olmsted RN, ed. Infection Control and Applied Epidemiology: Principle and Practice. St Louis, MO: Mosby; 1996:A1–A20
- Raad II, Hohn DC, Gilbreath BJ, et al. Prevention of central venous catheter-related infections by using maximal sterile barrier precautions during insertion. Infect Control Hosp Epidemiol.1994; 15 :231 –238[Web of Science][Medline]
- Pearson ML, for the Hospital Infection Control Advisory Committee. Guidelines for prevention of intravascular device-related infections. Am J Infect Control.1996; 24 :262 –293[CrossRef][Web of Science][Medline]
- Nystrom B. Impact of handwashing on mortality in intensive care: examination of the evidence. Infect Control Hosp Epidemiol.1994 :15 :435 –436[Web of Science][Medline]
PEDIATRICS (ISSN 1098-4275). ©2002 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  C. Hatler, L. Buckwald, Z. Salas-Allison, and C. Murphy-Taylor Evaluating Central Venous Catheter Care in a Pediatric Intensive Care Unit Am. J. Crit. Care., November 1, 2009; 18(6): 514 - 520. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Costello, D. F. Morrow, D. A. Graham, G. Potter-Bynoe, T. J. Sandora, and P. C. Laussen Systematic Intervention to Reduce Central Line-Associated Bloodstream Infection Rates in a Pediatric Cardiac Intensive Care Unit Pediatrics, May 1, 2008; 121(5): 915 - 923. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. S. Lakshmi, M. Jayashree, S. Singhi, and P. Ray Study of Nosocomial Primary Bloodstream Infections in a Pediatric Intensive Care Unit J Trop Pediatr, April 1, 2007; 53(2): 87 - 92. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Bhutta, C. Gilliam, M. Honeycutt, S. Schexnayder, J. Green, M. Moss, and K J S Anand Reduction of bloodstream infections associated with catheters in paediatric intensive care unit: stepwise approach BMJ, February 17, 2007; 334(7589): 362 - 365. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Larsen Vancomycin-Heparin Lock Solution Decreases the Rate of Sepsis in Neonates AAP Grand Rounds, January 1, 2006; 15(1): 5 - 6. [Full Text] [PDF]
|
|
|
|
|
|
A. M. Elward, C. S. Hollenbeak, D. K. Warren, and V. J. Fraser Attributable Cost of Nosocomial Primary Bloodstream Infection in Pediatric Intensive Care Unit Patients Pediatrics, April 1, 2005; 115(4): 868 - 872. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 OTHER ARTICLES NOTED (Nov 01 to 18 Oct 02) Evid. Based Nurs., January 1, 2003; 6(1): e1 - 1. [Full Text] [PDF]
|
|
|
|
 |
|
 Risks for Nosocomial Bloodstream Infections in a Pediatric ICU Journal Watch Infectious Diseases, October 25, 2002; 2002(1025): 8 - 8. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
