Abstract
OBJECTIVE: Discrepancies or inaccuracies between testing methods can create confusion or lead to clinical harm if antibiotics are inappropriately chosen. We report our clinical experience using the Accelerate Pheno™ followed by routine automated susceptibilities by the Vitek®2 for positive blood cultures with gram-negative rods. DESIGN: This was a retrospective review of positive gram-negative blood cultures, including comparison of susceptibility testing results and impact on clinical care. SETTING: Academic teaching hospital. PATIENTS PARTICIPANTS: All patients admitted to the hospital with gram-negative bacteremia from January 2020 to December 2022. METHODS: Microbiology was reviewed for discrepancies as defined by very major errors (VMEs), major errors (MEs), and minor errors (mEs). Clinical charts were reviewed for antibiotic therapy. RESULTS: Positive blood cultures with gram-negative rods were included (n = 262). Between these two methods, overall essential agreement was 93.7% (2162/2304) and categorical agreement 93.5% (2159/2306). There were 147 discrepancies noted, including 6 VMEs, 25 MEs, and 116 mEs accounting for 96 patients. Antibiotic choice was changed in 8 patients due to perceived suboptimal empiric therapy based on the rapid susceptibility results. CONCLUSIONS: The Accelerate Pheno tended to over-call resistance compared to the Vitek®2. Few patients (8) received the incorrect antibiotic based on the Pheno result. Stewardship programs may choose to optimize their rapid antibiotic susceptibility testing reporting to help minimize confusion and guide appropriate antibiotic selection.