Moving the Dial on Predicting Early Onset Sepsis in Neonates; Are We There Yet? :

Although the incidence of early-onset sepsis (EOS) among neonates in the United States has decreased significantly over the years, identifying those infants at significant risk and intervening early remains important. In this month’s Hospital Pediatrics (10.1542/hpeds.2014-0126), investigators compared 4 different risk assessment approaches in neonates ≥ 35 weeks gestation in the first 6 hours of life, specifically the 2010 guidelines from the Centers for Disease Control and Prevention (CDC), the neonatal EOS calculator from Kaiser Permanente, and 2 locally developed algorithms, one adapted from the 2010 CDC guidelines and later updated following the 2019 American Academy of Pediatrics EOS clinical report. Investigators aimed to compare recommendations regarding laboratory evaluation, empiric antibiotics, and predictive capability for cases of EOS. Over 8000 neonates delivered from 2014-2018 in a single institution in the first 72 hours of life were included. As expected, more infants assessed by the CDC 2010 guidelines met criteria recommending laboratory evaluation and antibiotic administration compared with the other risk assessment tools. There were 5 cases of EOS (local incidence of 0.62/1000), all of whom were treated empirically with antibiotics prior to 6 hours of life and in 4 of the 5 cases, the mothers were diagnosed with chorioamnionitis. The investigators report that the EOS calculator recommended empiric antibiotics in all but 1 case. In that one case, clinical judgement prompted empiric treatment emphasizing the importance of the clinical assessment in all cases. The authors appropriately highlight the challenges with the subjective nature of identifying clinical illness or neonatal sepsis with a multivariate risk assessment tool such as the neonatal EOS calculator. Overall, what is reassuring in reading this study, is that we have, for the most part, successfully moved towards less unnecessary treatment and laboratory testing of late preterm and term neonates at risk for EOS. We have reasonably accurate risk assessment tools but, importantly, we must continue to retain our clinical judgement when applying risk assessment tools to patients.