Objective: to inform antibiotic stewardship strategies in a level 3 neonatal intensive care unit by surveillance and assessment of antibiotic use during a 14-month period. To identify scenarios where antibiotic use can be reduced, and implementing interventions while monitoring safety.
Design: pre-post intervention quasi-experimental study. Surveillance and Correction of Unnecessary Antibiotic Therapy (SCOUT) study.
Setting: level 3 Neonatal Intensive Care Unit, Parkland Hospital, Dallas, Texas, United States.
Study sample: 2502 infants admitted to the NICU, 1607 during the baseline period (October 3, 2011 to May 31, 2013) and 895 during the intervention period (October 1, 2013 to June 30, 2014).
Intervention: during the intervention period, all antibiotic treatments were reevaluated systematically at 48 hours, at which point courses of empirical antibiotherapy were discontinued. The duration of therapy in cases of pneumonia and clinical sepsis with negative cultures was limited to five days.
Outcome measurement: the primary outcome was the change in total antibiotic days of therapy per 1000 patient-days between the baseline and intervention periods. The safety outcomes under study were the number of infants that had antibiotic therapy reinstituted within 14 days of completion of a course. The authors also documented the number of cases of late-onset sepsis, necrotising enterocolitis or death in infants ≤ 32 weeks’ gestational age, the prevalence of colonisation by multidrug-resistant organisms and the length of stay.
Main results: antibiotic use decreased from 343.2 to 252.2 of therapy per 1000 patient-days (P < .001), which amounts to an overall reduction of 27%. No differences were observed in the safety outcomes between the intervention and baseline periods, although there was a significant increase in length of stay from seven to eight days.
Conclusion: a thorough assessment of antibiotic consumption in neonatal intensive care units and the implementation of strategies tailored to each facility can help improve interventions to optimise antibiotic use.
Conflicts of interest: the authors declared no competing interests.
Funding source: the study was supported by a Gerber Novice Researcher Award (grant number 5200762201).