Antibiotic De-escalation in the Intensive Care Unit

Abstract

Successful treatment of severe infections in the intensive care unit (ICU) often requires broad-spectrum empiric therapy, while attempting to control the source of infection. However, this liberal antibiotic strategy may be associated with adverse effects on the patients as well as on the overall microbial ecology of the unit. This “antibiotic dilemma” may be solved by early de-escalation of antibiotic therapy, which allows reducing the overall antibiotic exposure of ICU patients by shortening the duration of therapy (including early stop when infection is not confirmed), switching from combined to single therapy, and/or substituting broad-spectrum agent with narrower-spectrum regimen. The opportunity for de-escalation varies across series from 20% to 50%, depending on the empiric antibiotic policy and the epidemiological context. Adapting the antibiotic regimen, possible as early as 24 h after obtaining the first results from adequate samples, is mandatory at 48–72 h, once full microbiological results are obtained. Subsequently, the intensivist must reassess daily the continued need for antibiotics, just like sedation is reassessed daily in mechanically ventilated patients. Several studies have confirmed that early deescalation is safe, and recent evidence suggests that it may even be associated with improved outcome of patients.