Background: There are wide variations in antibiotic use in neonatal intensive care units (NICUs). Limited data are available on antimicrobial stewardship (AS) programs and long-term maintenance of AS interventions in preterm very-low-birth-weight (VLBW) infants.
Methods: We extended a single-centre observational study carried out in an Italian NICU. Three periods were compared: I. "baseline" (2011-2012), II. "intervention" (2016-2017), and III. "maintenance" (2020-2021). Intensive training of medical and nursing staff on AS occurred between periods I and II. AS protocols and algorithms were maintained and implemented between periods II and III.
Results: There were 111, 119, and 100 VLBW infants in periods I, II, and III, respectively. In the "intervention period", there was a reduction in antibiotic use, reported as days of antibiotic therapy per 1000 patient days (215 vs. 302, p < 0.01). In the "maintenance period", the number of culture-proven sepsis increased. Nevertheless, antibiotic exposure of uninfected VLBW infants was lower, while no sepsis-related deaths occurred. Our restriction was mostly directed at shortening antibiotic regimens with a policy of 48 h rule-out sepsis (median days of early empiric antibiotics: 6 vs. 3 vs. 2 in periods I, II, and III, respectively, p < 0.001). Moreover, antibiotics administered for so-called culture-negative sepsis were reduced (22% vs. 11% vs. 6%, p = 0.002), especially in infants with a birth weight between 1000 and 1499 g.
Conclusions: AS is feasible in preterm VLBW infants, and antibiotic use can be safely reduced. AS interventions, namely, the shortening of antibiotic courses in uninfected infants, can be sustained over time with periodic clinical audits and daily discussion of antimicrobial therapies among staff members.
Keywords: antibiotic use; antimicrobial stewardship; clinical audit; early onset sepsis; extremely low birth weight; late-onset sepsis; neonatal intensive care unit; newborn; very low birth weight.