Continuous Surveillance by Whole-Genome Sequencing to Identify and Manage Methicillin-Resistant Staphylococcus aureus Outbreaks

Abstract

Background. Methicillin-resistant Staphylococcus aureus (MRSA) is a common and clinically significant pathogen associated with high mortality. Although epidemi-ologic analyses by infection control agencies are routinely undertaken, the full genetic basis of this pathogen is not routinely monitored or defined with precision greater than multilocus sequence typing (MLST) results, complicating the detection and management of outbreaks of this highly clonal pathogen. We implemented a whole-genome sequencing surveillance program to map the genetic diversity of all MRSA bacteremias identified in an urban hospital center over the course of one year and used it to manage a large outbreak during this time. Methods. Pacbio long-read sequencing was used to assemble high-quality complete genomes for 137 MRSA isolates. Phylogenetic analyses were performed using Mugsy, Gblocks and RAxML to determine relatedness of bacterial strains and identify transmission events, which were correlated with epidemiological temporal-spatial analyses. Additionally, genomes were compared with NUCmer to characterize the evolutionary changes that occurred in endemic strains. Results. WGS revealed a diverse MRSA population that included 36% ST105, 36% ST8, 16% ST5, and 12% other sequence types. Isolates within each ST were genetically distinct by WGS indicating that direct transmissions were generally rare; however , in one case we confirmed a large outbreak of a subpopulation of MRSA ST105 in a neonatal intensive care unit. Related cases were identified in other wards, including a case that pre-dated the epidemiologically determined outbreak cluster by one month. Conclusion. We demonstrate that continuous surveillance of WGS is both feasible and necessary to identify transmissions of MRSA with high accuracy and precision, and to map the full extent of outbreaks. Notably, near-real time sequencing of MRSA isolates permits early detection of transmission events compared to standard epidemiological methods, enabling immediate infection control interventions to pre-emptively manage outbreaks. Disclosures. All authors: No reported disclosures.