Risk factors for neonatal sepsis

Obstet Gynecol. 1996 Feb;87(2):188-94. doi: 10.1016/0029-7844(95)00402-5.


Objective: To determine the associations between maternal characteristics, intrapartum events, and neonatal sepsis by multivariate analysis.

Methods: We enrolled 823 women from a high-risk population and analyzed maternal and neonatal demographic and outcome variables with univariate analysis and multivariate logistic modeling.

Results: Two-hundred sixteen women (26%) were colonized with group B streptococci, 82 (10%) developed chorioamnionitis, and 141 (17%) delivered prematurely. Culture-proven neonatal sepsis or meningitis was found in 15 of 833 (1.8%) neonates, and 101 of the remaining 818 (12.3%) infants were suspected to have sepsis or pneumonia. Multivariate analysis of risk factors for proven neonatal sepsis demonstrated a statistically significant association with decreasing gestational age, duration of internal monitoring for more than 12 hours (odds ratio [OR] 7.2, 95% confidence interval [CI] 1.6-32.2), maternal group B streptococcal infection (OR 4.2, 95% CI 1.4-13.1), chorioamnionitis (OR 4.4, 95% CI 1.2-16.1), and endometritis (OR 6.4, 95% CI 1.2-34.2).

Conclusion: Through the use of multivariate modeling, we determined that chorioamnionitis or endometritis, preterm delivery, group B streptococcal colonization, and a prolonged duration of internal monitoring are independent risk factors for neonatal sepsis. We postulate that the presence of a foreign body that traverses the birth canal may facilitate ascending peripartal infection.

MeSH terms

  • Confidence Intervals
  • Female
  • Humans
  • Infant, Newborn
  • Logistic Models
  • Multivariate Analysis
  • Odds Ratio
  • Prospective Studies
  • Risk Factors
  • Sepsis / epidemiology*
  • Sepsis / therapy