LAUSR

Introduction/Hypothesis: Septic shock presents a unique challenge to antimicrobial stewardship, with previous data suggesting that empiric antibiotics are continued inappropriately beyond 72 hours. The purpose of this study was to characterize antibiotic use, adverse effects, and outcomes in culture-negative (CNSS) versus culture-positive septic shock (CPSS). Methods: This was a single center retrospective cohort in septic shock patients. Patients were identified via the medical intensive care unit (MICU) sepsis database between April 2014 and July 2018. Exclusion criteria: outside hospital transfer, < 18 years old, positive viral or fungal culture, or death within 72 hours. Cultures within 72 hours of septic shock onset categorized patients as either CNSS or CPSS. At 72 hours, CDC criteria for infection was used to determine presence of infection vs. sepsis mimics. After 72 hours, prolonged empiric antibiotic therapy (PEAT) was defined as any antibiotic for patients without diagnosis of infection or more broad than the microorganism cultured. The primary outcome was rate of PEAT (number of PEAT antibiotics/total number of antibiotics) in CPSS and CNSS; secondary outcomes included antibiotic adverse effects and mortality. Categorical data were analyzed using Chi-square or Fisher’s Exact and continuous data using Mann-Whitney U. Results: A total of 216 patients were included: 108 CPSS and 108 CNSS. Of the 43 patients who did not meet CDC criteria for infection, 40 were culture-negative. The rate of PEAT did not differ between CPSS (30.5%) vs. CNSS (34%)(p=0.799). In CPSS, 20 (16.7%) of PEAT antibiotics were due to excess duration and 100 (83.3%) were due to being more broad than the microorganism identified. There were no significant differences in secondary outcomes. However, there was an observed increase in a composite of superinfections (multidrug resistant organisms, fungemia, and C. difficile) with increasing PEAT duration: no PEAT (17%), PEAT < 5 days (23.8%), PEAT ≥ days (31.8%). Conclusions: The rate of PEAT did not differ between CPSS and CNSS. However, there was an observed increase in a composite of antibiotic harm with increasing PEAT duration. Antimicrobial stewardship strategies and enhanced microbiologic diagnostics should be employed to reduce PEAT and the potential harm associated with PEAT.