Antibiotic De-Escalation Practices in the Intensive Care Unit: A Multicenter Observational Study

Abstract

Background: There is little known about antibiotic de-escalation (ADE) practices in the intensive care unit (ICU). Objective: The objective was to determine the proportion of patients who received ADE within 24 hours of actionable cultures and identify predictors of timely ADE. Methods: Multicenter cohort study in ICUs of 15 hospitals in Australia and New Zealand. Adult patients were included if they were started on broad-spectrum antibiotics within 24 hours of ICU admission. The ADE was defined as switching from a broad-spectrum agent to a narrower-spectrum agent or antibiotic cessation. The primary outcome was ADE within 24 hours of an actionable culture, where ADE was possible. Results: The 446 patients included in the study had a mean age of 63 ± 16 years, 60% were male, 32% were mechanically ventilated, and 19% were immunocompromised. Of these, 161 (36.1%) were not eligible for ADE and 37 (8.3%) for whom ADE within 24 hours of actionable culture could not be determined. In the remaining 248 patients, ADE occurred ≤24 hours in 60.5% (n = 150/248) after actionable cultures. In the multivariable logistic regression analysis, ADE was less likely to occur within 24 hours for patients with negative cultures (odds ratio [OR] = 0.48, 95% confidence interval [CI] = 0.25-0.92, P = 0.03). Conclusion and Relevance: Timely ADE may not occur in 40% of patients in the ICU and is less likely to occur in patients with negative cultures. Timely ADE can be improved, and patients with negative cultures should be targeted as part of antimicrobial stewardship efforts.