Early-Onset Neonatal Sepsis 2015 to 2017, the Rise of Escherichia coli, and the Need for Novel Prevention Strategies

JAMA Pediatr. 2020 Jul 1;174(7):e200593. doi: 10.1001/jamapediatrics.2020.0593. Epub 2020 Jul 6.


Importance: Early-onset sepsis (EOS) remains a potentially fatal newborn condition. Ongoing surveillance is critical to optimize prevention and treatment strategies.

Objective: To describe the current incidence, microbiology, morbidity, and mortality of EOS among a cohort of term and preterm infants.

Design, setting, and participants: This prospective surveillance study included a cohort of infants born at a gestational age (GA) of at least 22 weeks and birth weight of greater than 400 g from 18 centers of the Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network from April 1, 2015, to March 31, 2017. Data were analyzed from June 14, 2019, to January 28, 2020.

Main outcomes and measures: Early-onset sepsis defined by isolation of pathogenic species from blood or cerebrospinal fluid culture within 72 hours of birth and antibiotic treatment for at least 5 days or until death.

Results: A total of 235 EOS cases (127 male [54.0%]) were identified among 217 480 newborns (1.08 [95% CI, 0.95-1.23] cases per 1000 live births). Incidence varied significantly by GA and was highest among infants with a GA of 22 to 28 weeks (18.47 [95% CI, 14.57-23.38] cases per 1000). No significant differences in EOS incidence were observed by sex, race, or ethnicity. The most frequent pathogens were Escherichia coli (86 [36.6%]) and group B streptococcus (GBS; 71 [30.2%]). E coli disease primarily occurred among preterm infants (68 of 131 [51.9%]); GBS disease primarily occurred among term infants (54 of 104 [51.9%]), with 24 of 45 GBS cases (53.3%) seen in infants born to mothers with negative GBS screening test results. Intrapartum antibiotics were administered to 162 mothers (68.9%; 110 of 131 [84.0%] preterm and 52 of 104 [50.0%] term), most commonly for suspected chorioamnionitis. Neonatal empirical antibiotic treatment most frequently included ampicillin and gentamicin. All GBS isolates were tested, but only 18 of 81 (22.2%) E coli isolates tested were susceptible to ampicillin; 6 of 77 E coli isolates (7.8%) were resistant to both ampicillin and gentamicin. Nearly all newborns with EOS (220 of 235 [93.6%]) displayed signs of illness within 72 hours of birth. Death occurred in 38 of 131 infected infants with GA of less than 37 weeks (29.0%); no term infants died. Compared with earlier surveillance (2006-2009), the rate of E coli infection increased among very low-birth-weight (401-1500 g) infants (8.68 [95% CI, 6.50-11.60] vs 5.07 [95% CI, 3.93-6.53] per 1000 live births; P = .008).

Conclusions and relevance: In this study, EOS incidence and associated mortality disproportionately occurred in preterm infants. Contemporary cases have demonstrated the limitations of current GBS prevention strategies. The increase in E coli infections among very low-birth-weight infants warrants continued study. Ampicillin and gentamicin remained effective antibiotics in most cases, but ongoing surveillance should monitor antibiotic susceptibilities of EOS pathogens.

Publication types

  • Multicenter Study
  • Research Support, N.I.H., Extramural
  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Escherichia coli / isolation & purification*
  • Escherichia coli Infections / epidemiology
  • Escherichia coli Infections / microbiology
  • Escherichia coli Infections / prevention & control*
  • Female
  • Gestational Age
  • Humans
  • Incidence
  • Infant, Newborn
  • Infant, Premature*
  • Male
  • Neonatal Sepsis / epidemiology
  • Neonatal Sepsis / microbiology
  • Neonatal Sepsis / prevention & control*
  • Practice Guidelines as Topic*
  • Prospective Studies
  • United States / epidemiology