Mortality risks associated with empirical antibiotic activity in E. coli bacteraemia: an analysis of electronic health records

Background Reported bacteraemia outcomes following inactive empirical antibiotics (as judged by in vitro testing) are conflicting, potentially reflecting heterogeneous effects of species, minimum inhibitory concentration (MIC) breakpoints defining resistance/susceptibility, and times to rescue therapy. Methods We investigated adult inpatients with Escherichia coli bacteraemia at Oxford University Hospitals, UK, from 04-February-2014 to 30-June-2021 receiving empirical amoxicillin-clavulanate with/without other antibiotics. We analysed 30-day all-cause mortality from index blood culture using Cox models by in vitro amoxicillin-clavulanate susceptibility (activity) using the EUCAST resistance breakpoint (>8/2mg/L), categorical MIC, and a higher resistance breakpoint (>32/2mg/L), adjusting for other antibiotic activity and multiple confounders including comorbidities, vital signs, and blood tests. Results 1720 E. coli bacteraemias (1626 patients) were treated with empirical amoxicillin-clavulanate. 30-day mortality was 193/1400 (14%) [any active baseline therapy] and 52/320 (16%) [inactive baseline therapy] (p=0.17). With EUCAST breakpoints, there was no evidence that mortality differed for inactive vs. active amoxicillin-clavulanate (adjusted HR, aHR=1.27 [95%CI 0.83-1.93;p=0.28]), nor of an association with other antibiotic activity (p>0.18). Considering categorical amoxicillin-clavulanate MIC, MICs>32/2 were associated with mortality (aHR=1.85 vs. MIC=2/2 [0.99-3.73;p=0.054]). Using the higher resistance breakpoint, MICs>32/2 were independently associated with higher mortality (aHR=1.82 [1.07-3.10;p=0.027]), as were MICs>32/2 with active baseline aminoglycoside (aHR=2.34 [1.40-3.89;p=0.001), but not MICs>32/2 with active baseline non-aminoglycoside antibiotic(s) (aHR=0.87 [0.40-1.89;p=0.72). Conclusions EUCAST-defined amoxicillin-clavulanate resistance was not associated with increased mortality, but a higher resistance breakpoint was. Additional active baseline non-aminoglycoside antibiotics prevented amoxicillin-clavulanate resistance-associated mortality, but active baseline aminoglycosides did not. Granular phenotyping and comparison with clinical outcomes may improve AMR breakpoints. Summary In patients with E. coli bacteraemia, high-level resistance to baseline empirical amoxicillin-clavulanate (minimum inhibitory concentration >32/2 mg/L) was associated with increased 30-day mortality, which was not compensated for by single-dose aminoglycosides.