1054. Validation of a Current Severe Sepsis Antibiotic Treatment Algorithm: Identifying Areas for Further Refinement
Session: Poster Abstract Session: Stewardship: Improving Treatments
Friday, October 4, 2013
Room: The Moscone Center: Poster Hall C
Posters
  • ID Week 2013 Sepsis Algorithm Poster.pdf (1.8 MB)
  • Background: The 2012 Surviving Sepsis Campaign guideline has identified the initiation of active antibiotic therapy within 1 hour of severe sepsis and septic shock (SS/SS) as a key goal of therapy.  Empiric antibiotic selection at our institution is guided by a SS/SS treatment algorithm/orderset based on hospital-wide blood culture data and tailored to suspected source and healthcare exposure.  Study objectives were to 1) determine if our existing algorithm/orderset adequately treats SS/SS and 2) provide guidance on best practices in designing a treatment algorithm for SS/SS.

    Methods: This retrospective, single center study evaluated SS/SS patients with positive microbiologic cultures over a 1 year period.  Study endpoints included the percent of patients who received active therapy against causative pathogens within 1 hour.  Antibiotic recommendations were vetted against the organisms isolated within each arm of the algorithm.

    Results: ­ 248 of 664 SS/SS patients had positive cultures.  105 of 248 (42%) patients received any antibiotic and 21% received active therapy against their causative pathogen(s) within 1 hour. Median time to administration of active therapy was 3.2 hours (range = 0 to 305 hours).  Use of the algorithm orderset occurred in only 84/248 (34%) encounters.  Risk factors for inactive therapy within 1 hour included not using the algorithm orderset (p=0.014), allergy to beta-lactams (p=0.016), and polymicrobial or fungal infections (p<0.001).  Adherence to the orderset would have provided active empiric therapy against 84% of pathogens.  Multi-drug resistant gram negatives, vancomycin-resistant enterococci, and fungi were commonly missed by the algorithm.  Algorithm recommendation mismatches with causative pathogens occurred most often with community/nosocomial abdominal and nosocomial skin/skin structure sources.

    Conclusion: Use of the sepsis algorithm orderset improved the likelihood that patients received active therapy within 1 hour.  Hospital antibiograms may not be adequate to develop antibiotic treatment recommendations for SS/SS for some suspected sources. A population specific evaluation is necessary to develop an antibiotic treatment algorithm tailored to the local SS/SS patient population.

    Elizabeth S. Zhu, Pharm.D., BCPS, Pharmacy Services, University of Cailfornia, Davis Medical Center, Sacramento, CA, Cinda Christensen, Pharm.D., BCPS-ID, Pharmacy Services, University of California, Davis Medical Center, Sacramento, CA and Hien H. Nguyen, MD, MAS, Internal Medicine, University of California, Davis Medical Center, Sacramento, CA

    Disclosures:

    E. S. Zhu, None

    C. Christensen, None

    H. H. Nguyen, None

    Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 2nd with the exception of research findings presented at the IDWeek press conferences.