Methods: We performed a retrospective, quasi-experimental study of adult patients with bacterial BSI between December 1, 2014 to February 28, 2015 (pre-intervention) and December 1, 2015 to February 29, 2016 (intervention). During the intervention, the AS team received a once-daily list of patients with new BSI from Microbiology Laboratory, reviewed available data in the chart at the time of positive blood culture, and made recommendations to the primary service if deemed appropriate. Before the intervention, the AS team was available for consultation upon request by the primary service. Baseline continuous variables were analyzed using student t-test or Wilcoxon sum rank test, while categorical variables were compared using a χ2 test. In-hospital survival was evaluated by the Kaplan-Meier method with log rank statistics.
Results: Of the 264 patients included in the study, 134 were in the pre-intervention group and 130 in the intervention group. The mean age was 60 years old, and 55% of patients were male. There were more patients with chronic kidney disease and on dialysis in the pre-intervention group. The most common organisms included Escherichia coli (26%) and Staphylococcus aureus (23%). Time to optimal therapy was shorter in the intervention group than the pre-intervention group (56 ± 50 hours versus 86 ± 95 hours, p =0.05). The percentage of de-escalation events were similar between both groups (81% versus 73%, p =0.24). No differences in duration of intensive care unit or hospital stay from positive BSI, rate of Clostridium difficile infection, or in-hospital mortality were noted.
Conclusion: Daily review of positive BSI can be considered as an antimicrobial stewardship strategy to shorten time to optimal antimicrobial therapy. Real-time notifications for positive BSI or more frequent reviews of positive BSI may have an even larger impact on antimicrobial prescribing and patient outcomes.
T. P. Trang,
S. B. Doernberg, None
C. Macdougall, None