De-escalation of antimicrobials in the treatment of bacteraemia due to antibiotic-sensitive pathogens in immunocompetent patients

The aim of this study was to examine the safety and efficacy of de-escalating antimicrobial therapy in immunocompetent patients presenting with bacteraemia due to antibiotic-sensitive pathogens. We screened 1,350 positive blood cultures identified in a single, 1,065-bed university hospital over 5 years, and retained 310 cases of bacteraemia due to antibiotic-sensitive pathogens, including (1) methicillin-sensitive staphylococci, (2) penicillin-sensitive streptococci, (3) beta-lactam-sensitive (a) Escherichia coli, and (b) Klebsiella species. The efficacy of appropriate initial empirical antimicrobial therapy, the performance of de-escalated pathogen-directed therapy, and the safety and efficacy of de-escalated therapy were evaluated. Among 270 appropriately treated patients, 16 (6%) died, versus 6 (15%) among 40 who were inappropriately treated (p = 0.04). While 201 of 270 patients (74%) who received appropriate initial empirical therapy were candidates for de-escalation, the treatment was de-escalated in only 79 (39%). De-escalation was associated with (1) a trend toward a lower (a) death rate (1 vs. 5%) and (b) treatment failure (4 vs. 10%), and (2) (a) a 4-day longer median duration and (b) a $50 higher median cost of antimicrobial therapy (p < 0.001). When the pathogen was sensitive to antimicrobial therapy and the initial empirical treatment was effective, de-escalation of antimicrobial therapy in immunocompetent patients with bacteraemia was safe and associated with acceptable outcomes. The rate of de-escalation of antimicrobial therapy was low.