Rapid antimicrobial susceptibility testing in patients with bacteraemia due to Enterobacterales: an implementation study.

AIMS OF THE STUDY: The goal of this descriptive study was to assess the performance as well as the extent of the clinical impact of rapid automated antimicrobial susceptibility testing in patients with bacteraemia due to Enterobacterales. We also aimed to analyse how rapid automated antimicrobial susceptibility testing influences clinical decision-making.

METHODS: This single-centre study conducted at the University Hospital of Zurich included data from all consecutive patients with Enterobacterales bacteraemia from November 2019 to October 2020. There was no control group. The primary outcome was the effect of rapid automated antimicrobial susceptibility testing on antibiotic therapy (no adjustment, escalation to a broader-spectrum antibiotic or de-escalation to a narrower-spectrum antibiotic). Rapid automated antimicrobial susceptibility testing results were further compared to susceptibility tests using European Committee on Antimicrobial Susceptibility Testing (EUCAST) standard methods and erroneous results were noted. Additionally, we investigated turnaround times for rapid automated antimicrobial susceptibility testing and routine diagnostic testing.

RESULTS: We analysed 106 patients with 116 episodes of bacteraemia due to Enterobacterales, with Escherichia coli and Klebsiella pneumoniae being the most frequent isolates. Almost 8% of pathogens were multidrug resistant. Rapid automated antimicrobial susceptibility testing showed category agreement in 98.4% of all interpretable cases. A significant reduction of more than 20 h in turnaround times could be achieved with rapid automated antimicrobial susceptibility testing compared to the routine diagnostic workflow. In the majority of cases, rapid automated antimicrobial susceptibility testing had no effect, given that the empirical therapy was already correct or circumstances did not allow for de-escalation. In 38.8% of cases, antimicrobial therapy was adjusted, whereas eight cases were de-escalated based on rapid automated antimicrobial susceptibility testing alone.

CONCLUSIONS: Rapid automated antimicrobial susceptibility testing may be a valuable and safe way to accelerate diagnosis. In particular, time to suitable therapy can be shortened in cases of incorrect therapy. However, physicians are reluctant to de-escalate antibiotic therapy based on rapid automated antimicrobial susceptibility testing alone, limiting its impact in everyday clinics. To further explore the potential of rapid automated antimicrobial susceptibility testing, a stringent/compulsory antibiotic stewardship programme would be a valuable next step.