Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Multicenter Study
. 2024 Aug 1;178(8):792-799.
doi: 10.1001/jamapediatrics.2024.1840.

Hospital-Onset Bacteremia Among Neonatal Intensive Care Unit Patients

Erica C Prochaska  1   2 , Shaoming Xiao  1 , Elizabeth Colantuoni  3 , Reese H Clark  4 , Julia Johnson  5   6 , Sagori Mukhopadhyay  7 , Ibukunoluwa C Kalu  8 , Danielle M Zerr  9 , Patrick J Reich  10 , Jessica Roberts  11 , Dustin D Flannery  7 , Aaron M Milstone  1   2 ; CDC Prevention Epicenters Program
Affiliations
Multicenter Study

Hospital-Onset Bacteremia Among Neonatal Intensive Care Unit Patients

Erica C Prochaska et al. JAMA Pediatr. .

Abstract

Importance: The Centers for Disease Control and Prevention plans to introduce hospital-onset bacteremia (HOB) as a health care-associated infection measure. The epidemiology and clinical characteristics of HOB among infants admitted to the neonatal intensive care unit (NICU) are unknown.

Objective: To estimate the rate of HOB among infants admitted to the NICU, measure the association of HOB risk with birth weight group and postnatal age, and estimate HOB-attributable mortality.

Design, setting, and participants: This retrospective multicenter cohort study and emulated trial from 2016 to 2021 included a convenience sample of 322 NICUs in the United States. Participants were infants admitted to participating NICUs for 4 or more days.

Exposures: The primary exposures were birth weight and postnatal age. Additional exposures included small for gestational age and central line presence.

Main outcomes and measures: The primary study outcomes were HOB and HOB-attributable mortality.

Results: Of 451 443 included infants, 250 763 (55.6%) were male, 200 680 (44.4%) were female, and 62 091 (13.8%) were born 1500 g or less. Of 9015 HOB events that occurred among 8356 infants (2%) during 8 163 432 days at risk (unadjusted incidence rate, 1.1 per 1000 patient-days; 95% CI, 1.0-1.2), 4888 HOB events (54.2%) occurred in the absence of a central line. Within the first 2 weeks after birth, the HOB rate was 14.2 per 1000 patient-days (95% CI, 12.6-16.1) among infants born 750 g or less, to 0.4 events per 1000 patient-days among infants born more than 2500 g (95% CI, 0.4-0.5). Among infants born 750 g or less, the relative HOB risk decreased by 90% after day 42 compared with days 4 to 14 (incidence rate ratio [IRR], 0.10; 95% CI, 0.1-0.1). Conversely, among infants born more than 2500 g, the relative HOB risk increased by 50% after day 42 compared with days 4 to 14 (IRR, 1.5, 95% CI, 1.2-1.9). Compared with otherwise similar infants without HOB, infants with HOB had an absolute difference in attributable mortality of 5.5% (95% CI, 4.7-6.3).

Conclusions and relevance: This study found that HOB events in the NICU are associated with increased mortality. Birth weight is an important risk factor for HOB; however, the relative rate of HOB decreases over postnatal age among low-birth-weight infants and increases among infants born more than 2500 g. Identifying strategies to prevent HOB and programs to decrease HOB risk are urgently needed to reduce infant mortality.

PubMed Disclaimer

Conflict of interest statement

Conflict of Interest Disclosures: Dr Prochaska reported personal fees from the Society for Healthcare Epidemiology of America outside the submitted work. Dr Kalu reported from Bristol Myers Squibb Foundation and personal fees from IPEC Experts outside the submitted work. Dr Zerr reported grants from a Johns Hopkins University subcontract made to Seattle Children’s Hospital during the conduct of the study, grants from a Merck award made to Seattle Children’s Hospital, and consultant fees from Allovir for service on end-point adjudication committees outside the submitted work. Dr Roberts reported funding from Johns Hopkins for data acquisition from her local institution during the conduct of the study and funding from the Centers for Disease Control and Prevention for part of her faculty salary as a contracted subject matter expert outside the submitted work. No other disclosures were reported.

Figures

Figure 1.
Figure 1.. Inclusion and Exclusion Criteria for Infants Included in the Analysis of Hospital-Onset Bacteremia (HOB) and HOB-Attributable Mortality
aInfants were excluded for a length of stay less than 4 days or less than 14 days after an early-onset sepsis event. bInfants were excluded for having missing birth weight, gestational age, or sex or having only culture results left blank or marked as not available. cCultures were excluded for results being blank or marked as not available. dInfants were excluded if they had a positive blood culture in the first 1 to 3 days of admission or days 1 to 3 of life or had group B streptococcus in the first 7 days of life.
Figure 2.
Figure 2.. Unadjusted Hospital-Onset Bacteremia (HOB) Rate Stratified by Birth Weight Groups
The HOB rate was defined as the number of HOB events per 1000 patient-days, stratified by birth weight groups (A) and birth weight groups binned in 2-week postnatal age intervals (B). Error bars represent 95% CIs.
Figure 3.
Figure 3.. Emulated Trial to Estimate the Attributable Mortality From Hospital-Onset Bacteremia (HOB) Among Infants Admitted to the Neonatal Intensive Care Unit
The cumulative 7-day mortality rate attributed to any first-time HOB event in the first 4 to 28 days of life (A), the cumulative 7-day mortality rate among infants without an HOB event, matched for postnatal age, clinical and site-level characteristics (B), and the absolute difference in cumulative 7-day mortality rate between infants with a first-time HOB event and matched infants without an HOB event (C). Error bars represent 95% CIs.
See this image and copyright information in PMC

Comment in

References

    1. Verstraete E, Boelens J, De Coen K, et al. Healthcare-associated bloodstream infections in a neonatal intensive care unit over a 20-year period (1992-2011): trends in incidence, pathogens, and mortality. Infect Control Hosp Epidemiol. 2014;35(5):511-518. doi: 10.1086/675836 - DOI - PubMed
    1. Ely DM, Driscoll AK, Matthews TJ. Infant mortality by age at death in the United States, 2016. NCHS Data Brief. 2018;(326):1-8. - PubMed
    1. Centers for Medicare & Medicaid Services . Hospital-acquired conditions. Accessed March 1, 2024. https://www.cms.gov/medicare/payment/fee-for-service-providers/hospital-...
    1. Advani SD, Murray TS, Murdzek CM, Aniskiewicz MJ, Bizzarro MJ. Shifting focus toward healthcare-associated bloodstream infections: the need for neonatal intensive care unit–specific NHSN definitions. Infect Control Hosp Epidemiol. 2019;41(2):181-186. doi: 10.1017/ice.2019.310 - DOI - PubMed
    1. Sood G, Leekha S. Controversies in healthcare-associated infection surveillance. In: Bearman G, Munoz-Price S, Morgan DJ, Murthy RK, eds. Infection Prevention. 2018:267-276.