The effect of antibiotic rotation on colonization with antibiotic-resistant bacilli in a neonatal intensive care unit.

This study was designed to test whether rotation of antibiotics can reduce colonization with resistant Gram-negative bacilli in a neonatal intensive care unit (NICU). A monthly rotation of gentamicin, piperacillin-tazobactam, and ceftazidime was compared with unrestricted antibiotic use in side-by-side NICU populations (rotation team vs control team). Pharyngeal and rectal samples were obtained 3 times a week and tested for Gram-negative bacilli resistant to each of the rotation antibiotics. Pulsed-field gel electrophoresis analysis determined the numbers of genetically discordant resistant organisms on each team. The association between colonization with a resistant bacillus (the primary outcome) and team assignment was tested. A total of 1062 infants were studied during a 1-year period. A total of 10.7% infants on the rotation team versus 7.7% on the control team were colonized with a resistant bacillus. No interteam differences were distinguishable when the numbers of genetically discordant resistant organisms were normalized to the total number of team admissions. The incidence of nosocomial infection and mortality also were similar across teams. These data indicate that rotation of parenteral antibiotics according to the applied protocol has no detectable effect in decreasing the reservoir of resistant Gram-negative bacilli in a tertiary-care NICU.