Early- and Late-Onset Neonatal Sepsis: Pathogens, Maternal Risk Factors, and the Predictive Value of Hematological Biomarkers
DOI:
https://doi.org/10.62838/amsm-2026-0030Keywords:
neonatal sepsis, biomarkers, mean platelet volume, red cell distribution widthAbstract
Background: Neonatal sepsis is a major cause of morbidity and mortality, especially in low- and middle-income countries (LMICs). Preterm neonates are at higher risk due to immune immaturity, comorbidities, and prolonged NICU hospitalization. Early recognition and treatment are crucial, but no ideal biomarker exists. Mean platelet volume (MPV) and red cell distribution width (RDW) have been suggested as potential markers. This study evaluates their significance.
Objectives: To determine the incidence of early- and late-onset sepsis (EOS, LOS), analyze causative pathogens, and assess biochemical parameters (RDW, MPV, platelet count [PLT], and C-reactive protein [CRP]). The study also examines maternal conditions influencing sepsis and neonatal outcomes.
A retrospective analysis was conducted on 139 NICU patients at CCU Sarajevo from January 1, 2021, to December 31, 2023.
Results: Neonatal sepsis incidence was 12.5%, with 25.2% classified as EOS and 74.8% as LOS. EOS occurred mostly in full-term neonates (mean birth weight 2,044g), while LOS was more common in preterms (1,830g). Escherichia coli was the main EOS pathogen, while Acinetobacter baumannii caused 19.4% of LOS cases. Thrombophilia in mothers was significantly associated with EOS. CRP was the most predictive biomarker for EOS, while MPV, CRP, and PLT were significant in LOS. RDW was not relevant. Mortality was 22.3%, higher in LOS.
Conclusion: No single biomarker reliably predicts neonatal sepsis. A combination improves diagnostic accuracy, emphasizing the need for further research.
Downloads
Downloads
Published
How to Cite
License
Copyright (c) 2026 Edina Spahic, Sabina Terzić

This work is licensed under a Creative Commons Attribution 4.0 International License.
Acta Marisiensis Seria Medica provides immediate open access to its content under the Creative Commons BY 4.0 license.






