Time to Positivity of Blood Cultures Supports Antibiotic De-escalation at 48 Hours

Background: Appropriate de-escalation of empirical antimicrobial therapy is a fundamental component of antimicrobial stewardship. Concern for the late detection of bloodstream pathogens may undermine early streamlining efforts and subject patients to protracted courses of nonessential therapy. Objective: To quantify the prevalence of bacterial bloodstream infection (BSI) detection after more than 48 hours of culture incubation. We also assessed the impact of antimicrobial therapy delivered prior to blood sample collection. Methods: We retrospectively evaluated time to blood culture positivity (TTP) in adult patients at an academic tertiary care hospital. Microbiology reports were reviewed to identify the TTP for the first positive blood culture bottle for each episode of BSI occurring from February 1, 2011, to July 31, 2011. Isolates were classified as true pathogens or contaminants. Blood culture results after 48 hours of incubation were compared with results after 120 hours of incubation. Results: The median UP of 416 monomicrobial BSIs and 210 contamination episodes was 13.7 and 24.4 hours, respectively (P < .001). The median TTPs in those who received and did not receive prior antibiotics were 17.0 and 12.8 hours, respectively (P < .001). By 48 hours, 98% of aerobic Gram-positive and Gram-negative BSIs were detected. Culture results at 48 hours were 97% sensitive and had a negative predictive value of 99.8%. Conclusion: Few true BSIs are detected after more than 48 hours of culture incubation. Clinicians may adjust empirical antibiotic coverage at this time with little risk for subsequent bacterial pathogen detection.