Impact of empiric antibiotic use on development of infections caused by extended-spectrum β-lactamase bacteria in a neonatal intensive care unit

Background: The neonatal intensive care unit at Miller Children's Hospital changed from empiric use of cefotaxime and vancomycin (CEF) to tobramycin and vancomycin (TOB) for hospital-acquired infections in November 1999 because of an increase in infections caused by extended-spectrum beta-lactamase (ESBL)-producing bacteria. The objectives of this study were to evaluate the incidence and impact of this change on the development of ESBL infections. Methods: We retrospectively reviewed medical records of infants who received CEF or TOB between January 1998 and December 2002. A standardized form was used to collect demographic data, information on antibiotic use, and culture results. Results: The mean gestational age and birth weight of the 250 infants were 28.8 +/- 4.0 weeks and 1213.1 +/- 662 g, respectively. There were no differences between infants who received CEF (N = 130) or TOB (N = 120) in terms of gestational age, birth weight, device use, invasive procedures, or prior antibiotic use. There were I I ESBL infections. Infants in the CEF group were more likely than those in the TOB group to develop ESBL infection (7.8% versus 0.8%, P = 0.008). There were I I deaths, with none attributed to ESBL infection. In a multivariate analysis, duration of prior ampicillin and gentamicin use and exposure to CEF were associated with ESBL infection [odds ratio (OR): 3.1, 95% confidence interval (Cl): 1.28-7.49, P = 0.012; and OR: 33.7; 95% CI: 1.02-1136, P = 0.05, respectively]. Conclusions: The change from empiric use of CEF to TOB was associated with a significant decrease in the incidence of ESBL infections.