749 - When and why It happens - risk factors, temporal trends and clinical outcomes of sepsis in micropremature Infants: a multicenter cohort study

Publication Number: 2730.749

Joseph Ting, University of Alberta, Edmonton, AB, Canada; Shikha Gupta-Bhatnagar, McMaster Children's Hospital, Hamilton, ON, Canada; Eugene Ng, University of Toronto Temerty Faculty of Medicine, Toronto, ON, Canada; Brigitte Lemyre, University of Ottawa Faculty of Medicine, Ottawa, ON, Canada; Jo-Anna Hudson BJ. Jo-Anna Hudson, Memorial University of Newfoundland Faculty of Medicine, St. John's, NF, Canada; Marc Beltempo, McGill University, Montreal, PQ, Canada; Yi-Chen Su, mount sinai hospital, Toronto, ON, Canada

Background: Advances in perinatal care have significantly improved survival rates among infants born before 25 weeks' gestational age (GA). However, these "micropremature" infants remain highly vulnerable to infections due to their immature immune systems, underdeveloped skin barrier, and frequent exposure to invasive procedures. A comprehensive understanding of the incidence, timing, and outcomes of sepsis in this population is essential for guiding preventive strategies.

Objective: To examine the risk factors, timing, and clinical outcomes associated with sepsis in infants born at < 25 weeks' GA

Design/Methods: This retrospective cohort study included infants born at < 25 weeks' GA who survived at least two days and were admitted between 2015 and 2024 to one of 32 tertiary-level neonatal intensive care units participating in the Canadian Neonatal Network. Sepsis was defined as a culture-confirmed bacterial or fungal infection in blood and/or cerebrospinal fluid and categorized as early-onset (EOS; ≤2 days of life) or late-onset (LOS; >2 days). Multivariable logistic regression was used to examine the association between sepsis and a composite outcome of mortality, moderate-to-severe bronchopulmonary dysplasia (BPD), periventricular leukomalacia, or severe retinopathy of prematurity (ROP - Stage III or higher), adjusting for potential confounders.

Results: Of the 3,137 eligible infants, 1,390 (44.3%) experienced at least one episode of sepsis, including 125 (9.0%) with EOS. Sepsis was more prevalent among outborn infants, those delivered vaginally, and those exposed to clinical chorioamnionitis. Compared to infants with LOS, those with EOS were more likely to have experienced prolonged rupture of membranes (24 hours to 1 week), higher illness severity (Score for Neonatal Acute Physiology-II >20), and lower 5-minute Apgar scores (Table 1). Cumulatively, approximately 10%, 20% and 30% of sepsis episodes occurred by days 6, 9 and 15 of life, respectively (Figure). Sepsis-related mortality within 7 days of onset was 15% (217/1,390). The composite adverse outcome occurred in 87% (1,217/1,390) of infants with sepsis, compared to 82% (1,427/1,747) of those without sepsis (adjusted OR 1.49, 95% CI 1.28-1.73). Sepsis remained significantly associated with increased risks of adverse outcomes, including death or moderate-to-severe BPD, and severe ROP. (Table 2).

Conclusion(s): Sepsis is a leading cause of morbidity and mortality in "micopremature infants", with one third of cases occurring within the first two weeks of life, highlighting the potential role of perinatal factors and immature skin barriers in its pathogenesis.

Maternal and delivery characteristics of infants
#p-value: between infants with sepsis (EOS and/or LOS) and infants without sepsis;
EOS, early-onset sepsis; LOS, late-onset sepsis; GA, gestational age; BW birth weight; IQR: inter-quartile range; SNAP: score for neonatal acute physiology; ROM: rupture of membrane


Table 2: Adjusted analyses of neonatal outcomes
Infants without sepsis as reference*;
OR: odds ratio; AOR: adjusted odds ratio; BPD: bronchopulmonary dysplasia; PVL: periventricular leucomalacia; ROP: Retinopathy of prematurity; GA: gestational age, SGA: small-for-gestational age; SNAP: score for neonatal acute physiology;
- Model 1: variables associated with p < 0.20 in the univariate analysis (GA, SGA, sex, Apgar Score 5-min, Delivery mode, Inborn and Presence of any of the following: (histological chorioamnionitis, maternal fever within 24 hours before birth, maternal leukocytosis, purulent discharge from cervix)
- Model 2: variables chosen a priori which are clinically significant: GA, SGA, sex, SNAP II>20


- Kaplan-Meier curve- Time to develop first sepsis among all infants
Y-axis: probability of remaining free from first sepsis over time.