RESUMO
BACKGROUND: In 2017, a new Dutch neonatal early-onset sepsis (EOS) guideline was implemented. It is an adaptation from the United Kingdom National Institute for Health and Care Excellence guideline and focuses on maternal and neonatal risk factors. We aim to assess if this guideline performs better at reducing the rate of antibiotic treatment for EOS than the old Dutch categorical EOS guideline, which focused primarily on group B streptococcus (GBS) testing and prophylaxis. METHODS: We performed a single-center retrospective cohort study in the Netherlands. Data were collected from two 12-month epochs (2015 vs. 2019). Neonates were included when treated for suspected EOS or when observed for an elevated EOS risk. RESULTS: The empirical antibiotic rate was 4.6% in both years. Prolonged antibiotic treatment (>48 u) increased from 24% in 2015 to 39% in 2019 ( P = 0.021). Adherence to the guideline decreased from 98% in 2015 to 84% in 2019 ( P < 0.001). Strict adherence in 2019 would have led to more antibiotic treatment (5.1% instead of 4.6%). The EOS incidence rate was comparable, namely 0.6% in 2015 and 0.0% in 2019 ( P = 0.480). The change in the definition of risk factors in 2019 led to less antibiotic treatment in case of a maternal fever during birth, from 48% in 2015 to 26% in 2019 ( P < 0.001). CONCLUSIONS: The new Dutch categorical EOS guideline does not achieve its intended purpose of reducing empiric antibiotic therapy for suspected EOS. We advocate the need for a new screening strategy.
