70. Impact of the Accelerate Pheno™ System on Clinical and Antimicrobial Outcomes among Inpatients with Gram-Negative Bacteremia at a 528-bed Community Teaching Hospital

BACKGROUND: Traditional methods in blood culture analysis require 24-72 hours to yield identification (ID) and antimicrobial susceptibility testing (AST) results, which may contribute to the use of empiric broad-spectrum antibiotic therapy. Hence, the primary objective of this study was to determine the impact of rapid blood culture analysis with the Accelerate Pheno™ system (AXDX) on time to antibiotic de-escalation. METHODS: This was a single center, case-control analysis of adult inpatients with E. coli or Klebsiella spp. bacteremia. Cases were prospectively identified by the antimicrobial stewardship team between August and October 2020 after the implementation of AXDX in July 2020. Subjects were matched to historical controls (July 2018-July 2020) based on age (± 3 years), gender, source of infection, and identified organism. The primary outcome was time to antibiotic de-escalation and time to oral antibiotic therapy from the time of positive blood cultures. Secondary outcomes included hospital length of stay, 30-day mortality, 30-day readmission, and 60-day C. difficile infection. Outcomes were compared using descriptive and inferential statistics. RESULTS: Of 33 cases identified, 30 (91%) were matched with historical controls. E. coli bloodstream infection was identified in 24 (80%) subjects while Klebsiella spp. was identified in 6 (20%) subjects. The average age was 66 years (SD ± 19) and there was an even distribution of males and females in both groups. Other demographics were similar between groups. The median time to species identification [14 hours (IQR 13 – 18) vs 34 hours (29 – 39), p< 0.001) and AST [20 hours (19 – 37) vs 45 hours (38 – 51), p< 0.001] from laboratory registration was significantly shorter in cases. The average time to antibiotic de-escalation was 1.7 (±1.2) days for cases compared to 2 (±1.3) days for controls (p=0.460). Median time to oral antibiotic therapy from positive blood cultures was 2.9 (1.8 – 4.7) days for cases and 3.4 (2.5 – 5.1) days for controls (p=0.166). There were no significant differences in the secondary outcomes. CONCLUSION: AXDX did not appear to have a significant impact on time to antibiotic de-escalation and time to oral antibiotic therapy. However, time to organism ID and AST results were significantly shorter in the AXDX cohort. DISCLOSURES: All Authors: No reported disclosures