2297. Epidemiology of Antibiotic-resistant Pathogens and Empiric Treatment Patterns in Community-Onset Sepsis

BACKGROUND: Guidelines recommend immediate empiric broad-spectrum antibiotics for all patients with suspected sepsis. Understanding the epidemiology of antibiotic-resistant pathogens and empiric treatment patterns in sepsis could inform improvements in antibiotic utilization and outcomes. METHODS: We identified adults admitted during 2009–2015 to 104 US hospitals in the Cerner HealthFacts dataset who met CDC Adult Sepsis Event criteria and had positive clinical cultures within 2 days of admission. We characterized prevalence and empiric treatment rates for methicillin-resistant S. aureus (MRSA), vancomycin-resistant enterococcus (VRE), ceftriaxone-resistant Gram-negative organisms (CRO) (including P. aeruginosa), and extended-spectrum β-lactamase Gram-negative organisms (ESBL). We evaluated associations between in-hospital mortality and either inappropriate empiric therapy (antibiotics inactive against any isolated pathogen) or excessively broad therapy (empiric MRSA or VRE coverage, extended spectrum β-lactam, or carbapenem therapy when targeted organisms were absent), adjusting for baseline characteristics and severity-of-illness. RESULTS: The cohort included 17,962 patients with culture-positive sepsis; 2,965 (16.5%) died in-hospital. The most common culture-positive sites were urine (51.2%), blood (41.8%), and respiratory (16.5%). The most common pathogens were E. coli (33.0%), S. aureus (20.9%), and Streptococcus (13.2%) (Figure 1). Most (81.6%) patients received empiric antibiotics active against all isolated pathogens. Empiric therapy was directed at resistant organisms in 67.5% of cases (primarily vancomycin and extended spectrum β-lactams, Figure 2), but resistant organisms were isolated in only 25.2% (MRSA 11.5%, CRO 12.9%, VRE 2.0%, ESBL 0.8%). Both inappropriate empiric therapy and excessively broad empiric therapy were associated with higher mortality on multivariate analysis (OR 1.30, 95% CI 1.14–1.48 and OR 1.20, 95% CI 1.05–1.38, respectively). CONCLUSION: Most patients with community-onset sepsis do not have resistant pathogens, yet empiric broad-spectrum antibiotics are frequently prescribed. Both inappropriate empiric therapy and excessively broad therapy are associated with worse outcomes. [Image: see text] [Image: see text] DISCLOSURES: All authors: No reported disclosures.