Biomarker-guided antibiotic duration strategies using C-reactive protein and/or procalcitonin for neonatal sepsis: a...

Prolonged empirical antibiotic exposure in neonates is associated with adverse outcomes and antimicrobial resistance. Serial C-reactive protein (CRP) and procalcitonin (PCT) measurements may support earlier discontinuation, but neonatal protocols remain heterogeneous. To map evidence on CRP- and/or PCT-guided antibiotic-duration strategies in neonatal sepsis and summarize their effects on antibiotic exposure and safety. Following Joanna Briggs Institute guidance and PRISMA-ScR, PubMed/MEDLINE, Embase, Web of Science and the Cochrane Library were searched from inception to December 2023. Peer-reviewed English-language studies of neonates aged 0-28 days with suspected, probable or culture-proven early- or late-onset sepsis were included when serial CRP and/or PCT formed part of an explicit antibiotic-stopping strategy. Two reviewers independently screened records and charted study characteristics, protocols and outcomes. Fifteen studies involving 4755 neonates were included: five randomized trials, six cohort studies, two prospective observational studies and two systematic reviews with meta-analyses. Seven studies evaluated CRP-guided discontinuation, six evaluated PCT-guided strategies and four used combined approaches. Biomarker-guided care reduced mean antibiotic duration from 7.3 to 4.5 days, a mean reduction of 2.8 days (38.6%). Reductions were directionally consistent and appeared greater in early-onset than late-onset sepsis. Thirteen studies reported treatment failure and 14 reported mortality; neither outcome showed an apparent increase with biomarker-guided discontinuation. Antimicrobial-resistance, ecological and cost outcomes were infrequently reported. CRP- and/or PCT-guided stopping strategies may reduce antibiotic exposure in neonatal sepsis without an apparent increase in treatment failure or mortality. Standardized protocols and further implementation, cost-effectiveness and resistance-focused research are needed, particularly in resource-limited settings and extremely preterm infants.