Early-onset sepsis: a cohort study comparing serial clinical observation with laboratory testing.

OBJECTIVE Early onset sepsis (EOS) remains a serious and potentially fatal illness. We aimed to demonstrate that serial clinical observation (SCO) is a feasible strategy associated with fewer laboratory evaluations and unnecessary antibiotic use. STUDY DESIGN We compared the admissions and antibiotic therapy in neonates ≥35 weeks' gestation at risk for EOS in a prospective cohort after the implementation of a new protocol based on SCO (n = 381) with a historical cohort which received laboratory testing (n = 417). RESULTS There was a significant reduction in admissions for suspected sepsis (7.2% vs 2.9%, p = 0.006) and the use of antibiotics (6.1% vs 0.7%, p = 0.000) in the cohort based on SCO. There was no delay in diagnosis. CONCLUSIONS SCO in neonates ≥35 weeks' gestation at risk for EOS, including chorioamnionitis-exposed infants, is a feasible measure that reduces laboratory evaluations and the overuse of antibiotics respecting the bonding mother-infant.