Due to a lack of rapid, accurate diagnostic tools for early-onset neonatal sepsis (EOS) at the initial suspicion, infants are often unnecessarily given antibiotics directly after birth. We aimed to determine the diagnostic accuracy of presepsin for EOS before antibiotic initiation and to investigate whether presepsin can be used to guide clinicians' decisions on whether to start antibiotics. In this multicenter prospective observational cohort study, all infants who started on antibiotics for EOS suspicion were consecutively included. Presepsin concentrations were determined in blood samples collected at the initial EOS suspicion (t = 0). In addition to this, samples were collected at 3, 6, 12 and 24 h after the initial EOS suspicion and from the umbilical cord directly after birth. The diagnostic accuracy of presepsin was calculated. A total of 333 infants were included, of whom 169 were born preterm. We included 65 term and 15 preterm EOS cases. At the initial EOS suspicion, the area under the curve (AUC) was 0.60 (95% confidence interval (CI) 0.50-0.70) in the term-born infants compared to 0.84 (95% CI 0.73-0.95) in the preterm infants. A cut-off value of 645 pg/mL resulted in a sensitivity of 100% and a specificity of 54% in the preterm infants. The presepsin concentrations in cord blood and at other time points did not differ significantly from the concentrations at the initial EOS suspicion. Presepsin is a biomarker with an acceptable diagnostic accuracy for EOS (culture-proven and clinical EOS) in preterm infants and might be of value in reducing antibiotic exposure after birth when appended to current EOS guidelines. However, the small number of EOS cases prevents us from drawing firm conclusions. Further research should be performed to evaluate whether appending a presepsin-guided step to current EOS guidelines leads to a safe decrease in antibiotic overtreatment and antibiotic-related morbidity.
Read full abstract