600 - Proven and Suspected Early-Onset Bacterial Sepsis: Characteristics, Risk Factors and Outcomes in Extremely Premature Infants

Publication Number: 600.6274

Julide Sisman, University of Texas Southwestern Medical Center, DALLAS, TX, United States; Ahreen Allana, University of Texas Southwestern Medical School, Dallas, TX, United States; Katherine A. Stumpf, UTSW-Dallas, Dallas, TX, United States; L. Steven Brown, Parkland Health, Dallas, TX, United States

Background: Early-onset sepsis (EOS) is rare but potentially fatal among premature infants. Antibiotics are administered empirically to nearly all extremely premature infants often for prolonged periods of time without culture confirmation. However, there has been an increasing concern that early and prolonged use of antibiotics is associated with morbidities and mortality in the preterm population.

Objective: To compare neonatal, maternal and delivery characteristics and outcomes among infants with proven, suspected EOS and the ones with negative EOS evaluation to determine the risk factors for proven and suspected EOS.

Design/Methods: This is a retrospective observational study of 554 infants born at 22 to 29 weeks’ gestation admitted to the neonatal intensive care unit at Parkland Hospital in Dallas, Texas from 2018 to 2023. We defined proven EOS if a blood culture obtained within 72 hours after birth isolated pathogenic bacterial species, and suspected EOS if the duration of antibiotic therapy was ≥ 5 days despite a negative blood culture. Infants who did not meet with these criteria were included as controls. We then compared maternal, neonatal, delivery characteristics and outcomes among the groups. Statistical analyses included Wilcoxon rank sum test and chi-square analysis followed by pair-wise analyses at the < 0.05 level using Bonferroni correction.

Results: Among 554 infants, 12 (2%) had proven EOS, and 90 (16%) had suspected EOS. Infants with proven and suspected EOS were more likely to be born to mothers with clinical chorioamnionitis (P < 0.01), were more likely to have lower gestational age (P < 0.01), lower Apgar scores at 5 minutes (P < 0.01) and Grade III-IV intraventricular hemorrhage (P < 0.01) than the control group [Table 1]. Infants with proven EOS differed than those infants with suspected EOS and controls for the following variables: higher rate of vaginal birth (P < 0.01), and higher rate of mortality (P < 0.01) [Table 1]. Furthermore, those infants with proven EOS were more likely to be born preterm due to preterm labor or induction of labor secondary to pre-labor rupture of membranes with chorioamnionitis (P < 0.01) They were more likely to have acute chorioamnionitis on placental pathology (P=0.03) than those with suspected EOS [Table 2].

Conclusion(s): Delivery characteristics and placental pathology findings were significantly different between the infants with proven and suspected EOS. Vaginal delivery, preterm labor, pre-labor rupture of membranes with chorioamnionitis and acute chorioamnionitis on placental pathology were significantly associated with proven EOS in extremely premature infants.

Table 1.
Table1 for PAS 2025.pdfNeonatal, maternal and delivery characteristics and outcomes of infants with proven EOS, suspected EOS, and controls.

Table.2
Table 2 for PAS 2025.pdfAdditional delivery characteristics for infants with proven and early EOS