Effect of an antibiotic algorithm on the adequacy of empiric antibiotic therapy given by a medical emergency team

Abstract

Introduction: Delayed administration of effective antimicrobial therapy increases mortality in patients with septic shock. Empiric antibiotic selection in this setting can be inaccurate. The objective of this study was to determine whether an antibiotic algorithm (AA) tailored to institutional resistance patterns improves the adequacy of antimicrobial therapy.

Methods: A retrospective review of our rapid response system database was performed. Patients with possible sepsis with positive microbiological culture results were enrolled. Pathogens identified by culture were used to determine adequacy of antibiotic selection before and after implementation of an AA.

Results: A total of 234 patients with septic shock were reviewed (before AA, n = 36; after AA, n = 198). Seventy-two patients had positive cultures and were enrolled (before AA, n = 13; after AA, n = 59). Significantly more patients received adequate coverage after AA implementation (54% vs 86%, P = .02). Before AA, inadequate Gram-negative coverage was the most common reason for failure. Reasons for failure in the after-AA group were nonadherence to the algorithm (n = 5) and multidrug-resistant pathogens (n = 3). The algorithm failed in patients with vancomycin-resistant enterococci (n = 3), multidrug-resistant Klebsiella pneumoniae (n = 1), and Candida albicans (n = 1).

Conclusions: The use of an AA significantly improves the adequacy of empiric antimicrobial therapy.