Changing the paradigm for hospital outbreak detection by leading with genomic surveillance of nosocomial pathogens

Abstract

The current paradigm for hospital outbreak detection and investigation is based on methodology first developed over 150 years ago. Daily surveillance to detect patients positive for pathogens of particular importance for nosocomial infection is supported by epidemiological investigation to determine their relationship in time and place, and to identify any other factor that could link them. The antibiotic resistance pattern is commonly used as a surrogate for bacterial relatedness, although this lacks sensitivity and specificity. Typing may be used to define bacterial relatedness, although routine methods lack sufficient discriminatory power to distinguish relatedness beyond the level of bacterial clones. Ultimately, the identification of an outbreak remains a predominately subjective process reliant on the intuition of experienced infection control professionals. Here, we propose a redesign of hospital outbreak detection and investigation in which bacterial species associated with nosocomial transmission and infection undergo routine prospective whole-genome sequencing. Further investigation is based on the probability that isolates are associated with an outbreak, which is based on the degree of genetic relatedness between isolates. Evidence is provided that supports this model based on studies of MRSA (methicillin-resistant Staphylococcus aureus), together with the benefits of a 'Sequence First' approach. The feasibility of implementation is discussed, together with residual barriers that need to be overcome prior to implementation.

Keywords: bacterial sequencing; nosocomial; outbreak detection; transmission.

Fig. 1

Current and proposed approach to the detection of hospital outbreaks. (a) Current practice for the detection of hospital outbreaks based on surveillance and epidemiology. The pattern of antibiotic resistance is commonly used as a surrogate for bacterial relatedness, and formal bacterial typing may be used during outbreak investigation. (b) A proposed alternative in which outbreak detection is led by routine sequencing of bacterial species that are commonly associated with nosocomial outbreaks and infection.

Fig. 2

Epidemiology and phylogeny of an MRSA outbreak. Left: timeline (in days) of an outbreak that affected infants on a special care baby unit (SCBU), and that went on to affect family clusters. A total of 26 people were affected: infants treated on the SCBU who were known to be MRSA positive during admission (P1–14) or who were not known to be carriers during admission but were detected after discharge (P16-18); mothers who were (P19–22), or were not (P23–24), inpatients on the maternity ward; and partners (P25–26). The length of the boxes shown for infants on SCBU represent duration of hospital stay. A healthcare worker was detected who was also carrying the outbreak strain (denoted by H). Darker vertical blue blocks show times on the SCBU when there were no known carriers of MRSA. Right: phylogenetic tree of MRSA isolated from patients 1–26, together with 20 individual MRSA colonies from a staff member (denoted by H). SNP, single nucleotide polymorphism. Adapted from reference [5].

Fig. 3

Contextualization of outbreak investigations of CC22 MRSA studied at Cambridge University Hospitals. The maximum likelihood tree was based on 22 238 core SNPs for 783 ST22 genomes drawn from the British Society of Antimicrobial Chemotherapy bacteraemia resistance surveillance programme between 2001 and 2010; 7 isolates from an MRSA outbreak on a neonatal intensive care unit (NICU, green) [4]; 15 isolates from an MRSA outbreak that focused on a special care baby unit (SCBU, orange) but extended to other wards and the community [5]; and 42 isolates sequenced as part of an MRSA outbreak investigation on a hepatology ward (nine isolates from four patients with bacteraemia (P1–4; pink filled dots); and the remainder from patients who were MRSA carriers on the same ward during a comparable timeframe (pink open dots) [7]. Reproduced from reference [2].