An outbreak of imipenem-resistant Acinetobacter baumannii in critically ill surgical patients

OBJECTIVE: To describe an outbreak of imipenem-resistant Acinetobacter baumannii (IR-Ab) and the measures for its control, and to investigate risk factors for IR-Ab acquisition. DESIGN: An observational and a case-control study. SETTING: A surgical intensive care unit (ICU) in a university tertiary care hospital. METHODS: After admission to the ICU of an IR-Ab-positive patient, patients were prospectively screened for IR-AI, carriage upon admission and then once a week. Environmental cleaning and barrier safety measures were used for IR-Ab carriers. A case-control study was performed to identify factors associated with hi-Ah acquisition. Cases were patients who acquired IR-Ab. Controls were patients who were hospitalized in the ICU at the same time as cases and were exposed to IR-Ab for a similar duration as cases. The following variables were investigated as potential risk factors: baseline characteristics, scores for severity of illness and therapeutic intervention, presence and duration of invasive procedures, and antimicrobial administration. RESULTS: Beginning in May 1996, the outbreak involved 17 patients over 9 months, of whom 12 acquired TR-Ab (cases), 4 had IR-Ab isolates on admission to the ICU, and 1 could not be classified. Genotypic analysis identified two different IR-Ab isolates, responsible for three clusters. Ten of the 12 nosocomial cases developed infection. Control measures included reinforcement of barrier safety measures, limitation of the number of admissions, and thorough environmental cleaning. No new case was identified after January 1997. Eleven of the 12 cases could be compared to 19 controls. After adjustment for severity of illness, a high individual therapeutic intervention score appeared to be a risk factor for IR-Ab acquisition. CONCLUSION: The outbreak ended after strict application of control measures. Our results suggest that high work load contributes to IR-Ab acquisition (Infect Control Hasp Epidemiol 2001;22:35-40).