Antibiotic use in serious infections

PURPOSE OF REVIEW: To concisely review the selection of empiric antibiotics, optimization of antibiotic dosing based on pharmacokinetic principles, and antibiotic de-escalation in critically ill patients. RECENT FINDINGS: Empiric therapy decisions should be based on patient risk for multidrug-resistant organisms. Administration of single-agent broad spectrum therapy is generally sufficient and agents with anaerobic activity should be avoided except in patients with specific sources of infection. Prolonged infusions of beta-lactams, therapeutic drug-monitoring, and refinement of historical dosing recommendations should be implemented in most patients to optimize pharmacokinetics in critically ill patients. Antibiotic de-escalation is essential component of antimicrobial stewardship and should be facilitated by consideration of rapid diagnostic test results coupled with evidence of clinical improvement. The duration of antibiotic therapy in most cases should be 1 week or less in patients who have had the antibiotic regimen de-escalated. SUMMARY: Given the high prevalence of antibiotic administration, it is imperative to implement evidence-based practices regarding empiric antibiotic selection, dose optimization, and a steadfast commitment to antibiotic de-escalation to enhance patient outcomes given the inevitability of continuing infections due to multidrug-resistant organisms in critically ill patients.