No abstract

Whole genome sequencing (WGS) of healthcare-associated pathogens is recognised as the gold standard for isolate typing and the recognition of transmission networks and outbreaks. However, it remains reasonably expensive to process small numbers of isolates in real-time, and frequently requires specific expertise to enable both sequencing and the analysis of sequencing outputs, limiting its generalisability and turnaround. Spectrometry has revolutionised species identification in clinical laboratory workflows, and has more recently been applied to strain-level identification to facilitate low-cost, routine strain typing in clinical laboratories. However, studies to date of its clinical performance for strain-level typing are conflicting, and limited evaluation has been undertaken on environmental healthcare-associated isolates. We therefore compared its performance with WGS for Enterobacter cloacae, Citrobacter freundii and Klebsiella pneumoniae isolated from sink drains across nine hospitals and investigated whether it could be used as a screening tool prior to WGS. We found its sensitivity and specificity to cluster isolates when compared with WGS were generally poor and highly variable dependent on species and the single nucleotide polymorphism (SNP) distance threshold used to cluster isolates.