ABSTRACT
BACKGROUND: Bacterial whole-genome sequencing (WGS) and determination of genetic relatedness is an important tool for investigation of epidemiologically suspected outbreaks. AIM: This prospective cohort study evaluated a comprehensive, prospective bacterial WGS-based surveillance programme for early detection of transmission of most bacterial pathogens among patients at a paediatric oncology hospital. METHODS: Cultured bacterial isolates from clinical diagnostic specimens collected prospectively from both inpatient and outpatient encounters between January 2019 and December 2021 underwent routine WGS and core genome multi-locus sequence typing to determine isolates' relatedness. Previously collected isolates from January to December 2018 were retrospectively analysed for identification of prior or ongoing transmission. Multi-patient clusters were investigated to identify potential transmission events based on temporal and spatial epidemiological links and interventions were introduced. FINDINGS: A total of 1497 bacterial isolates from 1025 patients underwent WGS. A total of 259 genetically related clusters were detected, of which 18 (6.9%) multi-patient clusters involving 38 (3.7%) patients were identified. Sixteen clusters involved two patients each, and two clusters involved three patients. Following investigation, epidemiologically plausible transmission links were identified in five (27.8%) multi-patient clusters. None of the multi-patient clusters were suspected by conventional epidemiological surveillance. CONCLUSION: Bacterial WGS-based surveillance for early detection of hospital transmission detected several limited multi-patient clusters that were unrecognized by conventional epidemiological methods. Genomic surveillance helped efficiently focus interventions while reducing unnecessary investigations.
