Diagnostic value of maternal, cord blood and neonatal biomarkers for early-onset sepsis: a systematic review and meta-analysis.

Background: An accurate diagnosis of early-onset sepsis (EOS) is challenging because of subtle symptoms and the lack of a good diagnostic tool, resulting in considerable antibiotic overtreatment. A biomarker, discriminating between infected and non-infected newborns at an early stage of the disease, could improve EOS prediction. Numerous biomarkers have been tested, but have never been compared directly.

Objectives: We aimed to provide a comprehensive overview of early biomarkers and their diagnostic value in maternal samples, umbilical cord blood, and neonatal serum.

Data Sources: PubMed-Medline, EMBASE, The Cochrane Library, and Web of Science were searched up to 1 March 2023, without restrictions on publication date, population, or language.

Study Eligibility Criteria: Articles describing the diagnostic value of at least one biomarker in the detection of EOS in neonates, independent of gestational age, were included.

Assessment Of Risk Of Bias: The QUADAS-2 tool was used to assess study quality.

Methods Of Data Synthesis: Three independent researchers assessed the articles using Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Meta-analysis was performed with all manuscripts describing diagnostic accuracy using a random-effects model.

Results: Of 2296 identified articles, 171 reports were included in the systematic review and 69 in the meta-analysis. Literature showed mixed and inconsistent evidence for most biomarkers and sample types, because of a lack of a uniform EOS case definition, small sample sizes, and large heterogeneity between studies. Interesting markers were procalcitonin (pooled sensitivity 79%, 95% CI 71-84%; specificity 91%, 95% CI 83-96%, n = 11) and interleukin (IL)-6 (pooled sensitivity 83%, 95% CI 71-90%; specificity 87%, 95% CI 78-93%, n = 8) in umbilical cord blood and presepsin (pooled sensitivity 82%, 95% CI 62-93%; specificity 86%, 95% CI 73-93%, n = 3) and serum amyloid A (pooled sensitivity 92%, 95% CI 75-98%; specificity 96%, 95% CI 78-99%, n = 4) in neonatal serum. Studies on the combination of biomarkers were scarce.

Conclusions: A biomarker stand-alone test is currently not reliable for direct antibiotic stewardship in newborns, although several biomarkers show promising initial results. Further research into biomarker combinations could lead to an improved EOS diagnosis, reduce antibiotic overtreatment, and prevent associated health-related problems.