Prolonged empirical antibiotic therapy despite negative blood cultures in neonates: a retrospective stewardship-oriented analysis

BACKGROUND: Empirical antibiotic therapy is routinely initiated in neonatal units for suspected sepsis; however, continuation of treatment despite negative blood cultures represents a major antimicrobial stewardship challenge. Prolonged exposure in early life may increase the risk of adverse outcomes and antimicrobial resistance. OBJECTIVES: To evaluate the prevalence and independent predictors of prolonged empirical antibiotic therapy in neonates with negative blood cultures over a three-year period. METHODS: This retrospective cohort study included 550 neonates treated with empirical antibiotics between January 2023 and January 2026. Clinical data, laboratory parameters (C-reactive protein, white blood cell count, procalcitonin), gestational age, birth weight, indication for therapy, blood culture results, and antibiotic regimens classified according to the WHO AWaRe framework were analyzed. Prolonged therapy was defined as antibiotic treatment exceeding five days despite negative cultures. Days of therapy (DOT) and DOT per 1000 patient-days were calculated. Multivariable logistic regression identified independent predictors. RESULTS: Blood cultures were negative in 82.4% of neonates. Among these, 41.8% received antibiotics for more than five days. The mean duration of therapy was 6.7 ± 2.4 days, and DOT reached 724 per 1000 patient-days. Prematurity (< 37 weeks) was independently associated with prolonged therapy (OR 2.36; 95% CI 1.58-3.52; p < 0.001), particularly < 32 weeks (OR 3.41; 95% CI 2.01-5.79; p < 0.001). Birth weight < 1500 g (OR 2.87; 95% CI 1.74-4.72; p < 0.001), elevated CRP (> 20 mg/L) (OR 1.94; 95% CI 1.29-2.91; p = 0.002), and mechanical ventilation (OR 2.11; 95% CI 1.32-3.37; p = 0.001) were significant predictors. "Watch" antibiotics accounted for 38.6% of initial regimens. CONCLUSIONS: Prolonged empirical antibiotic therapy despite negative blood cultures remains frequent in neonatal practice and is independently associated with prematurity, very low birth weight, inflammatory markers, and clinical severity. Targeted antimicrobial stewardship interventions are warranted to optimize neonatal antibiotic use.