An evaluation of screening methods for the detection of extended-spectrum beta-lactamase-producing Escherichia coli and Klebsiella pneumoniae in environmental samples from healthcare settings

Aims Extended-spectrum beta-lactamase-producing Enterobacterales (ESBL-E) infections are increasingly problematic, leading to worse treatment outcomes and higher healthcare expenditure. This study evaluated laboratory workflows for the detection of ESBL-E in healthcare environments. We aimed to optimise workflows for organism yield, detection accuracy and practical feasibility to support future surveillance and transmission studies, which are needed to inform targeted interventions to interrupt the spread of drug-resistant organisms. Methods and Results We sampled sinks, toilets, shower drains, shower heads, and high-touch surfaces, within six healthcare facilities in Liverpool, UK. We then assessed recovery of ESBL-producing Escherichia coli (ESBL-Ec) and Klebsiella pneumoniae (ESBL-Kp) using two swab types (foam and polyester), two pre-enrichment broths (Buffered Peptone Water and Tryptic Soy Broth), two incubation times (4 and 18 hours) and four selective agars (CHROMagar ESBL, cefotaxime-supplemented MacConkey, Membrane Lactose Glucuronide Agar and Simmons Citrate Agar with Inositol). Pre-enrichment for 18 hours significantly increased recovery of third-generation cephalosporin resistant Gram-negatives, compared to 4 hours. The choice of selective agar impacted the number of ESBL-Ec and ESBL-Kp detected and the number of samples requiring additional species confirmation. We found CHROMagar ESBL and cefotaxime-supplemented Membrane Lactose Glucuronide Agar performed best overall, demonstrating the highest yield and detection accuracy (i.e. colour of colonies matched their confirmed species) of ESBL-Ec and ESBL-Kp. Conclusions Pre-enrichment for 18 hours in Buffered Peptone Water, followed by plating onto cefotaxime-supplemented Membrane Lactose Glucuronide Agar, was the most effective screening method in our context, having a high detection efficacy, whilst maintaining a scalable and accessible cost.