115 - Influence of Antimicrobial Stewardship and Molecular Rapid Diagnostic Test on Antimicrobial Prescribing for Extended-spectrum Beta-lactamase and Carbapenemase-producing Bacteria in Bloodstream Infections

Background: Molecular rapid diagnostic tests (mRDT) may help expedite the time to optimal antimicrobial therapy (TTOT) for extended-spectrum beta-lactamase (ESBL)- and carbapenemase-producing bacteria in bloodstream infections (BSI). The greatest impact of mRDT appears to occur when combined with antimicrobial stewardship program (ASP) intervention. The purpose of this study was to evaluate if mRDT + ASP influences the TTOT for patients with ESBL- and carbapenemase-producing E. coli and K. pneumoniae in BSI compared to conventional microbiological methods with ASP (CONV + ASP).

Methods: Multicenter, retrospective, cohort study evaluating five years of patients that had a positive E. coli or K. pneumoniae blood culture determined to be ESBL- or carbapenemase-producing by mRDT and/or CONV. Patients were excluded if they had polymicrobial BSI, transferred–in with previously identified positive blood cultures, were immunosuppressed, or died before culture results.
Primary outcome was TTOT defined as time from blood culture draw to start of carbapenem therapy for ESBL-producing BSI and ceftazidime-avibactam, meropenem-vaborbactam, or at least one drug active in-vitro with the most-narrow spectrum for carbapenemase-producing BSI. Secondary outcomes were time to microbial clearance (TTMC) defined as the time from index blood culture draw to the time of first negative blood culture or hospital discharge, all-cause hospital mortality, 30-, 60- and 90-day readmission rates, and Clostridioides difficile rates.

Results: A total of 378 patients were included for analysis. Baseline characteristics were balanced between mRDT + ASP (n=164) and CONV + ASP (n=214). Infectious diseases consults were significantly greater for CONV + ASP compared to mRDT + ASP (82.2% vs 34.8%; p< 0.001). The mRDT + ASP demonstrated a statistically significant decrease in TTOT (20.5 hrs [(IQR 17.0-42.2 hrs)] vs 50.1 hrs [(IQR 27.6-77.9 hrs)]; p< 0.001) and TTMC (71.9 hrs [(IQR 54.1-108.5 hrs)] vs 91.2 hrs [(IQR 64.6-134.3 hrs)]; p=0.007). Other secondary endpoints were similar between groups.Table 1. Comparison of baseline characteristics for the mRDT+ASP and CONV+ASP groups
Graph 1. Kaplan Meier time to optimal antimicrobial therapy
Graph 2. Kaplan Meier time to microbial clearance


Conclusion: Our study supports the additional benefit of mRDT to ASP on shortening the TTOT and TTMC in patients with ESBL- or carbapenemase-producing E. coli and K. pneumoniae in BSI compared to CONV + ASP.