First-generation fluoroquinolone use and subsequent emergence of multiple drug-resistant bacteria in the intensive care unit.

OBJECTIVE: The objective of this study was to determine the relationship between fluoroquinolone (FQ) use and subsequent emergence of multiple drug-resistant bacteria (MRB) in the intensive care unit (ICU).

DESIGN: The authors conducted a prospective observational cohort study and a case control study.

SETTING: The study was conducted in a 30-bed ICU.

METHODS: All immunocompetent patients hospitalized for >48 hrs who did not receive antibiotics before ICU admission were eligible during a 15-month period. Routine MRB screening was performed at ICU admission and weekly thereafter. This screening included tracheal aspirate and nasal, anal, and axilla swabs. Univariate and multivariate analyses were used to determine risk factors for MRB emergence in the ICU. In addition, a case control study was performed to determine whether FQ use is associated with subsequent emergence of MRB.

RESULTS: Two hundred thirty-nine patients were included; 108 ICU-acquired MRB were isolated in 77 patients. FQ use and longer duration of antibiotic treatment were identified as independent risk factors for MRB occurrence (odds ratio [95% confidence interval [CI] = 3.3 [1.7-6.5], 1.1 [1.0-1.2]; p < .001; respectively). One hundred thirty-five (56%) patients received FQ; matching was successful for 72 (53%) of them. Number of MRB (40 vs. 15 per 1,000 ICU days; p = .019) and percentage of patients with MRB (40% vs. 22%; OR [95% CI] = 1.5 [1.0-2.4]; p = .028) were significantly higher in cases than in controls. Although methicillin-resistant Staphylococcus aureus (26% vs. 12%; OR [95% CI] = 1.6 [.6-2.9]; p = .028) and extending-spectrum beta-lactamase-producing Gram-negative bacilli (11% vs. 1%; OR [95% CI] = 4.7 [0.7-30.2]; p = .017) rates were higher in cases than in controls, ceftazidime or imipenem-resistant Pseudomonas aeruginosa (15% vs. 8%), Acinetobacter baumannii (1% vs. 5%), and Stenotrophomonas maltophilia (2% vs. 1%) rates were similar (p > .05) in case and control patients.

CONCLUSION: FQ use and longer duration of antibiotic treatment are independently associated with MRB emergence. Reducing antimicrobial treatment duration and restricting FQ use could be suggested to control MRB spread in the ICU.