Relentless spread and adaptation of non-typeable vanA vancomycin-resistant Enterococcus faecium: a genome-wide investigation

Relentless spread and adaptation of non-typeable vanA vancomycin-resistant Enterococcus faecium:... Abstract Background VRE are prevalent among patients in ICUs. Non-typeable vanA VRE, due to loss of one of the genes used for MLST (pstS), have increased in Australia, suggestive of a new, hospital-acquired lineage. Objectives To understand the significance of this lineage and its transmission using WGS of strains isolated from patients in ICUs across New South Wales, Australia. Methods A total of 240 Enterococcus faecium isolates collected between February and May 2016, and identified by conventional PCR as vanA positive, were sequenced. Isolates originated from 12 ICUs in New South Wales, grouped according to six local health districts, and represented both rectal screening swab (n = 229) and clinical (n = 11) isolates. Results ST analysis revealed the absence of the pstS gene in 84.2% (202 of 240) of vanA isolates. Two different non-typeable STs were present based on different allelic backbone patterns. Loss of the pstS gene appeared to be the result of multiple recombination events across this region. Evidence for pstS-negative lineage spread across all six local health districts was observed suggestive of inter-hospital transmission. In addition, multiple outbreaks were detected, some of which were protracted and lasted for the duration of the study. Conclusions These findings confirmed the evolution, emergence and dissemination of non-typeable vanA E. faecium. This study has highlighted the utility of WGS when attempting to describe accurately the hospital-based pathogen epidemiology, which in turn will continue to inform optimal infection control measures necessary to halt the spread of this important nosocomial organism. Introduction Enterococcus faecium emerged as a significant nosocomial pathogen in the 1990s and, since then, the number of hospital-acquired infections caused by this pathogen has increased worldwide.1 The National Healthcare Safety Network report has ranked E. faecium as the ninth most common pathogen causing healthcare-associated infections.2 In the USA, this equates to ∼3800 E. faecium infections a year, of which >82% were due to VRE. Although less virulent than other multiresistant organisms, VRE still result in greater infection-related morbidity and mortality compared with vancomycin-susceptible enterococci.3 In Australia, VRE colonization is uncommon in healthy individuals (prevalence 0.2%); however, among patients in ICUs, its prevalence is much higher (4.4%).4,5 ICUs are a major reservoir of VRE, with the risk of VRE infections closely associated with prior VRE colonization, varying from 0%–45% (median 4%) in VRE-colonized patients to <2% among non-colonized patients.5 Other risk factors for VRE infection include severe neutropenia, solid organ transplantation and presence of indwelling devices such as urinary catheters.6,7 In Europe and the USA, vancomycin resistance in E. faecium is predominantly due to acquisition of the vanA operon, whereas, until recently, the majority of Australian VRE strains have harboured the vanB operon.8–10 Factors influencing this emergence of vanA VRE across Australia are incompletely defined.11 A non-typeable E. faecium characterized by loss of one of the genes used in the MLST of enterococci, the pstS gene, was identified in Australia in 2015.12 This loss was mediated through a chromosomal inversion event and appears to have occurred repeatedly over time. Since its initial detection, there has been a dramatic increase in the number of non-typeable E. faecium identified across multiple jurisdictions in Australia, suggesting the appearance of a lineage that is both new and hospital-acquired.12 To characterize better the extent and significance of this lineage and its apparent spread to disparate geographical regions, we applied genome-wide analysis to examine vanA-carrying E. faecium strains isolated from patients within ICU wards across the jurisdiction of New South Wales, Australia. Methods Study setting and selection of isolates Between February and May 2016 all E. faecium isolates identified by conventional PCR as harbouring the vanA operon, i.e. vanA positive, from 12 New South Wales ICUs were studied. These ICUs encompass both secondary (n = 4) and tertiary/quaternary (n = 8) referral centres, caring for acute surgical and medical patients. All participating ICUs routinely perform rectal VRE screening on admission and weekly until ICU discharge. Only one isolate per patient was included. Unless positive prior to the study commencement, this was the initial isolate in the majority of cases. A total of 240 isolates were collected and represented isolates from both rectal screening swabs (n = 229) and clinical samples (n = 11). The hospitals were grouped into six major local health districts (LHDs), i.e. Hunter New England Health Service (LHD-1), Sydney Local Health District (LHD-2), South-Eastern Sydney Local Health District (LHD-3), South-Western Sydney Local Health District (LHD-4), Western Sydney Local Health District (LHD-5) and the Illawarra-Shoalhaven Local Health District (LHD-6). Library preparation and genome sequencing All VRE isolates were subcultured on to horse blood agar and incubated overnight at 37°C in 5% CO2 to ensure purity before analysis. Genomic DNA extraction was performed using the Blood and Tissue Mini Kit (Qiagen, Australia) for Gram-positive bacteria as per the manufacturer’s instructions. DNA extracts were treated with 1 U of RNase. Total DNA concentration was quantified using Picogreen (Invitrogen, Australia) and 1 ng/μL DNA was used to prepare DNA libraries employing the Nextera XT Library Preparation Kit (Illumina, Inc., CA, USA). Multiplexed libraries were sequenced using paired-end 150 bp chemistry on the NextSeq 500 (Illumina, Australia). Genome analysis Sequence reads were mapped against the reference isolate Aus0004 (GenBank accession no. GCA_000250945.1, an ST17 isolate containing the pstS gene) using Stampy v1.0.2313 with pre-BWA alignment. Variants were called using FreeBayes v1.1.0-dirty14 and filtered for read depth (minimum 20), read base quality (minimum Phred score 30) and mapping quality (minimum 30). Variation at indels and in the presence of mobile elements was excluded from the mapping-based analysis. Recombination was identified and masked from the phylogeny using Gubbins15 with the maximum likelihood phylogeny generated with RaxML v8.2.1016 and 100 bootstrap replicates. Mapping metadata to the phylogeny was performed in R using the ggtree package.17 The presence or absence of the pstS gene (this gene is one of seven genes used in the MLST scheme) was determined by in silico MLST performed on the de novo assembled contigs using SPAdes v3.10.1.18,19 Exploration of the number of possible pstS loss events was performed using ancestral state reconstruction within the R package ape.20 Early experiments showed the majority of isolates were missing the pstS gene. Subsequent comparative genomics was undertaken following mapping of reads to the closed pstS-negative genome DMG1500801 (GenBank accession no. GCA_900094185.1) and study isolates. To allow for contextual comparisons, the collection was supplemented with 43 (37 Victorian and 6 Australian Capital Territory) previously sequenced, non-typeable Australian E. faecium genomes (Table S1, available as Supplementary data at JAC Online). Although this dataset also contained 22 New South Wales isolates, these isolates were excluded from the analysis as the hospital/LHD origin was not described to allow for appropriate assignments in further analysis. To examine the spread of this clone between New South Wales and Victoria a phylogeographical analysis was performed as implemented through Bayesian Evolutionary Analysis Sampling (BEAST2).21 The optimal clock and tree model was selected based on the presence of convergence and Bayes factor following 100 000 000 iterations. Resultant trees were thinned prior to generating the final maximum clade credibility tree. Additional data including a jitter effect of ±0.01° of latitude or longitude between isolates from the same hospital within each LHD, to assist in visualization, were added using an in-house script prior to running SpreaD3.22 Similarly, BEAST2 was used to determine the molecular clock rate. All raw sequencing reads were deposited in NCBI project number PRJNA415172 (Table S1). Results Two hundred and forty vanA-positive isolates (representing 80% of all VRE isolated across participating LHDs) were obtained during the 4 month study period from the six LHDs (Figure S1). Although the number of isolates from each LHD varied widely (range 13–101), no significant changes were seen over time in the number of positives within any LHD resulting in an average of 60 (range 53–67) isolates obtained per month across the time period. The majority (89.6%, 215 of 240) of these isolates were newly acquired in the ICU, i.e. patients had initial screening swabs that were VRE negative but subsequently either had screening swabs that were positive for VRE or developed VRE infection. Of the 11 isolates from patients with infection, the likely source was the bloodstream (n = 5), abdominal collection (n = 4) or urinary tract (n = 2). The total number of infections represented 4.6% of all isolates detected. Over half of these occurred in one LHD, LHD-3 (n = 6), with the remaining infections occurring in LHD-2 (n = 1) and LHD-4 (n = 4). Of note, 54.5% of all infections (n = 6) occurred in patients who previously screened VRE negative. Determination of the pstS-negative clone Relative to the reference genome, 43 815 SNPs were detected across all isolates. Similar to previous observations,23 a substantial number of recombination events were identified (n = 1507) with masking of these events resulting in a final SNP matrix of 10 531 SNPs. In silico MLST identified the absence of the pstS gene in 84.2% (202 of 240) of all vanA isolates. On further examination, two different non-typeable STs were identified based on different allelic backbone patterns, which were most similar to VRE ST17 (allele profile: atpA-1, ddl-1, gdh-1, purK-1, gyd-1, pstS-1, adk-1) or ST80 (allele profile: atpA-9, ddl-1, gdh-1, purK-1, gyd-12, pstS-1, adk-1), profiles and are forthwith classified as ST17N and ST80N, respectively. ST17N predominated (71.3%; 144 of 202) with ST80N isolates accounting for 28.7% (58 of 202) of all pstS-negative isolates (Figure 1 and Figure S2). Figure 1. View largeDownload slide Maximum likelihood phylogeny based on whole genome sequence following masking of recombination of 240 vanA E. faecium isolates. Isolates represent positive screening or infection (red tips) strains recovered from New South Wales intensive care patients. Metadata are represented by concentric rings around the tree with the inner ring representing the LHD location of the isolate, while the outer ring indicates the in silico multilocus ST of the isolate. 17N* and 80N* isolates correspond with pstS-negative isolates with the remaining alleles most closely representing either ST17 (allele profile: atpA-1, ddl-1, gdh-1, purK-1, gyd-1, pstS-1, adk-1) or ST80 (allele profile: atpA-9, ddl-1, gdh-1, purK-1, gyd-12, pstS-1, adk-1), respectively (see the text for more details). This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 1. View largeDownload slide Maximum likelihood phylogeny based on whole genome sequence following masking of recombination of 240 vanA E. faecium isolates. Isolates represent positive screening or infection (red tips) strains recovered from New South Wales intensive care patients. Metadata are represented by concentric rings around the tree with the inner ring representing the LHD location of the isolate, while the outer ring indicates the in silico multilocus ST of the isolate. 17N* and 80N* isolates correspond with pstS-negative isolates with the remaining alleles most closely representing either ST17 (allele profile: atpA-1, ddl-1, gdh-1, purK-1, gyd-1, pstS-1, adk-1) or ST80 (allele profile: atpA-9, ddl-1, gdh-1, purK-1, gyd-12, pstS-1, adk-1), respectively (see the text for more details). This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Infection (versus colonization) was not associated with the pstS status of the isolate. The disproportionately higher number of infections detected in LHD-3 were evenly spread between pstS-negative and -positive isolates with no genomic feature detected to explain this observation. Furthermore, no significant gene or SNP difference was detected between infective and non-infective isolate genomes. The phylogeny supports previous observations of pstS gene loss occurring independently multiple times (Figure 1), while ancestral state reconstruction demonstrates that although ST17N and ST80N isolates share a common ancestor, ST80N emerged independently several times (Figure S3). Mechanistically, loss of the pstS gene would require either multiple inversion events, as hypothesized by Carter et al.,12 or recombination across this genomic region. The former mechanism appears less probable when ST80N isolates are included in the analysis, as this would hinge on two further independent mutation events in both the atpA and gyd housekeeping genes. Therefore, recombination, as detected across this region (see Figure S4), is the more likely mechanism leading to the emergence of this clone. Spread and dissemination of the pstS-negative clone Since the emergence of the pstS-negative lineage, it has not only disseminated widely across all six LHDs (Figures 2 and 3) but has also spread between Victoria and New South Wales on multiple occasions. The maintenance of this clone, within New South Wales, and spread to other LHDs (inter-hospital transmission) appears centred around LHD-2, which has the only jurisdictional liver transplant centre, liver transplantation being a known patient risk factor for VRE. This suggests that the expanding ‘burden’ of pstS-negative isolates may be partly secondary to local E. faecium adaptation and emergence of new ST80N isolates, within each LHD, following the initial spread of an ancestral clone (i.e. a pstS-positive isolate). Figure 2. View largeDownload slide Phylogeographical history of pstS-negative E. faecium clone spread in Victoria and New South Wales under a discrete diffusion model. Circular polygon area is proportional to the number of isolates maintaining that location. Blue connecting lines show transmission between Victoria and New South Wales. Insert depicts a magnification of the Sydney LHD networks revealing extensive dissemination across the ICUs in different LHDs. See Figure 3 for corresponding maximum likelihood phylogeny. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 2. View largeDownload slide Phylogeographical history of pstS-negative E. faecium clone spread in Victoria and New South Wales under a discrete diffusion model. Circular polygon area is proportional to the number of isolates maintaining that location. Blue connecting lines show transmission between Victoria and New South Wales. Insert depicts a magnification of the Sydney LHD networks revealing extensive dissemination across the ICUs in different LHDs. See Figure 3 for corresponding maximum likelihood phylogeny. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 3. View largeDownload slide Maximum likelihood phylogeny of pstS-negative E. faecium isolates following masking of recombination. Coloured shapes at tree tips represent the origin of each of the isolates. Highlighted nodes depict intra-LHD transmission events consistent with an outbreak. See Table S2 for more details. Several inter-LHD and inter-state transmission events are also seen. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 3. View largeDownload slide Maximum likelihood phylogeny of pstS-negative E. faecium isolates following masking of recombination. Coloured shapes at tree tips represent the origin of each of the isolates. Highlighted nodes depict intra-LHD transmission events consistent with an outbreak. See Table S2 for more details. Several inter-LHD and inter-state transmission events are also seen. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. At an LHD level, all districts had evidence of both intra- and inter-ICU transmission. For intra-ICU events to be considered as evidence of an outbreak, rather than new acquisition events, isolates had to differ by ≤2 SNPs based on the molecular clock estimates of ∼9 SNPs/genome/year (3.0 × 10−6; 95% highest posterior density 7–11 × 10−6 SNPs/site/year), which corresponds with 2.25 SNPs over the 4 month study time frame. At least one outbreak was detected within each LHD (Figure 3, depicted by the highlighted nodes). These outbreaks were protracted, with genomically indistinguishable isolates found >46 and 92 days after the index case (Table S2). Discussion This study provides important insights into the rapid evolution of vanA E. faecium in Australian hospitals following its emergence.12 It further reveals a new development in the evolution of pstS-negative VRE with the identification of MLST non-typeable VRE that are ST80N. Ancestral state reconstruction suggests that both ST17N and ST80N originated from a common ancestor carrying the pstS allele (Figure S3). A previous analysis of ST17N isolates12 similarly found a single common ancestor with three separate deletion events in its evolution. However, the present analysis inclusive of ST80N strains indicates that the most likely explanation for this emergence and spread is multiple recombination events in separate lineages resulting in the loss of this allele. This does not discount the initial event occurring as a result of a large inversion event mediated by flanking ISs.12 The prevalence of vanA VRE has noticeably increased over time from 2.6% of all E. faecium bacteraemias in 2013 to 19.9% in 2015 throughout Australia.24,25 This increase has coincided with the appearance of pstS-negative E. faecium, which was identified as the second most predominant MLST type in 2015.25 The origins of the new clone, based on the current analysis and epidemiological data obtained from sequenced blood culture isolates by the Australian Group on Antimicrobial Resistance, suggest that its emergence and maintenance is within New South Wales.24 Furthermore, it appears that the emergence of vanA VRE and pstS-negative E. faecium is linked. The factors driving these changes deserve further study. Examination of the genomic relationship between isolates revealed local dissemination of pstS-negative vanA VRE within hospitals demonstrated by multiple outbreaks. Although frequent reintroductions into the ICUs cannot entirely be discounted, the reservoir of this clone is more likely each LHD’s ICU. Detected outbreaks were not only numerous but also protracted with >92 days between isolates in one instance. Relaxing the number of mutations defining an outbreak would have resulted in outbreaks being present for the entire study period, despite local hospital infection control measures. Evidence for inter-LHD transmission was also detected. What is responsible for these events is unclear. Unlike patient movement within any LHD, movement between LHDs is infrequent. Therefore inter-LHD transmissions suggest a possible environmental reservoir of this clone. This hypothesis requires investigation, however, as hospital-associated lineages are not generally detected within individuals from the community.26 Nevertheless, the loss of the pstS gene in vanA VRE clearly represents a new development in the adaptation of this pathogen in Australia. Our study, similar to that by Raven et al.,27 highlights the utility of WGS in hospital epidemiology. Not only does WGS allow for greater accuracy of cluster association, it greatly enhances our understanding of transmission chains, which would be entirely undetected using previous typing methods. The importance of studying screening isolates is demonstrated in our study as it pre-dates the trends seen in bloodstream infection surveillance data and unmasks the total genome adaptation leading to new E. faecium clones. Several potential limitations of our study are acknowledged. First, the study was conducted over a relatively short time span (i.e. 4 months) within a single state in Australia. Although limited epidemiological data were obtained for each isolate, overall trends could only be gauged from recently published national bloodstream surveillance data, rather than local surveillance data. Finally, given the high proportion of colonization isolates and the relative weighting of pstS-negative isolates, we were unable to examine possible pathogen factors to account for the success in persistence of this clone. In conclusion, this study has confirmed and quantified the presence of MLST non-typeable ST17N vanA E. faecium in hospital settings but has also documented the emergence of MLST non-typeable ST80N vanA E. faecium in Australia. Understanding the evolution, emergence and dissemination of this clinically significant nosocomial clone will improve control interventions. This study further supports the utility of genome sequencing in understanding and monitoring the hospital epidemiology of VRE. Acknowledgements We wish to thank Bradley Watson, Helen Ziochos, Heather Wren and Chris McIver whose efforts ensured that all isolates were captured, worked up and stored appropriately during the study period. Funding This work was supported by an NSW Health Public Health Pathogen Genomics Partnership grant. Transparency declarations None to declare. Supplementary data Tables S1 and S2 and Figures S1 to S4 are available as Supplementary data at JAC Online. References 1 Cetinkaya Y , Falk P , Mayhall CG. Vancomycin-resistant enterococci . Clin Microbiol Rev 2000 ; 13 : 686 – 707 . Google Scholar CrossRef Search ADS PubMed 2 Weiner LM , Webb AK , Limbago B et al. Antimicrobial-resistant pathogens associated with healthcare-associated infections: summary of data reported to the National Healthcare Safety Network at the Centers for Disease Control and Prevention, 2011-2014 . Infect Control Hosp Epidemiol 2016 ; 37 : 1288 – 301 . Google Scholar CrossRef Search ADS PubMed 3 Bhavnani SM , Drake JA , Forrest A et al. A nationwide, multicenter, case-control study comparing risk factors, treatment, and outcome for vancomycin-resistant and -susceptible enterococcal bacteremia . Diagn Microbiol Infect Dis 2000 ; 36 : 145 – 58 . Google Scholar CrossRef Search ADS PubMed 4 Padiglione AA , Grabsch EA , Olden D et al. Fecal colonization with vancomycin-resistant Enterococci in Australia . Emerg Infect Dis 2000 ; 6 : 534 – 6 . Google Scholar CrossRef Search ADS PubMed 5 Ziakas PD , Thapa R , Rice LB et al. Trends and significance of VRE colonization in the ICU: a meta-analysis of published studies . PLoS One 2013 ; 8 : e75658 . Google Scholar CrossRef Search ADS PubMed 6 National Health and Medical Research Council . Australian Guidelines for the Prevention and Control of Infection in Healthcare . Canberra : Commonwealth of Australia , 2010 . 7 Infection Control Service, Communicable Disease Control Branch . Clinical Guidelines for the Management of Patients with Vancomycin-Resistant Enterococci (VRE): Version 6.2 . Adelaide : Government of South Australia , 2017 . 8 Coombs GW , Pearson JC , Daley DA et al. Molecular epidemiology of enterococcal bacteremia in Australia . J Clin Microbiol 2014 ; 52 : 897 – 905 . Google Scholar CrossRef Search ADS PubMed 9 Top J , Willems R , Bonten M. Emergence of CC17 Enterococcus faecium: from commensal to hospital-adapted pathogen . FEMS Immunol Med Microbiol 2008 ; 52 : 297 – 308 . Google Scholar CrossRef Search ADS PubMed 10 Werner G , Coque TM , Hammerum AM et al. Emergence and spread of vancomycin resistance among enterococci in Europe . Euro Surveill 2008 ; 13 : pii=19046. 11 van Hal SJ , Espedido BA , Coombs GW et al. Polyclonal emergence of vanA vancomycin-resistant Enterococcus faecium in Australia . J Antimicrob Chemother 2017 ; 72 : 998 – 1001 . Google Scholar PubMed 12 Carter GP , Buultjens AH , Ballard SA et al. Emergence of endemic MLST non-typeable vancomycin-resistant Enterococcus faecium . J Antimicrob Chemother 2016 ; 71 : 3367 – 71 . Google Scholar CrossRef Search ADS PubMed 13 Lunter G , Goodson M . Stampy: a statistical algorithm for sensitive and fast mapping of Illumina sequence reads . Genome Res 2011 ; 21 : 936 – 9 . Google Scholar CrossRef Search ADS PubMed 14 Garrison E , Marth G. Haplotype-Based Variant Detection From Short-Read Sequencing. http://arxiv.org/abs/1207.3907. 15 Croucher NJ , Page AJ , Connor TR et al. Rapid phylogenetic analysis of large samples of recombinant bacterial whole genome sequences using Gubbins . Nucleic Acids Res 2015 ; 43 : e15 . Google Scholar CrossRef Search ADS PubMed 16 Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies . Bioinformatics 2014 ; 30 : 1312 – 3 . Google Scholar CrossRef Search ADS PubMed 17 Yu G , Smith DK , Zhu H et al. ggtree: an r package for visualization and annotation of phylogenetic trees with their covariates and other associated data . Methods Ecol Evol 2017 ; 8 : 28 – 36 . Google Scholar CrossRef Search ADS 18 Bankevich A , Nurk S , Antipov D. SPAdes: a new genome assembly algorithm and its application to single-cell sequencing . J Comput Biol 2012 ; 19 : 455 – 77 . Google Scholar CrossRef Search ADS PubMed 19 Seeman T. MLST—Scan Contig Files Against PubMLST Typing Schemes. 2017 . https://github.com/tseemann/mlst. 20 Paradis E , Blomberg S , Bolker B et al. Package ‘Ape’. 2017 . https://cran.r-project.org/web/packages/ape/ape.pdf. 21 Bouckaert R , Heled J , Kühnert D et al. BEAST 2: a software platform for Bayesian evolutionary analysis . PLoS Comput Biol 2014 ; 10 : e1003537 . Google Scholar CrossRef Search ADS PubMed 22 Bielejec F , Baele G , Vrancken B et al. SpreaD3: interactive visualization of spatiotemporal history and trait evolutionary processes . Mol Biol Evol 2016 ; 33 : 2167 – 9 . Google Scholar CrossRef Search ADS PubMed 23 van Hal SJ , Ip CL , Ansari MA. Evolutionary dynamics of Enterococcus faecium reveals complex genomic relationships between isolates with independent emergence of vancomycin resistance . Microb Genom 2016 ; 2 : doi:10.1099/mgen.0.000048. 24 Coombs G , Daley D on behalf of the Australian Group on Antimicrobial Resistance. Australian Enterococcal Sepsis Outcome Program (AESOP) 2016 Final Report. http://agargroup.org.au/wp-content/uploads/2017/08/AESOP-2016-Final-Report-2017.pdf. 25 Coombs GW , Daley DA , Thin Lee Y et al. Australian Group on Antimicrobial Resistance Australian Enterococcal Sepsis Outcome Programme annual report, 2014 . Commun Dis Intell Q Rep 2016 ; 40 : E236 – 43 . Google Scholar PubMed 26 Lebreton F , van Schaik W , McGuire AM. Emergence of epidemic multidrug-resistant Enterococcus faecium from animal and commensal strains . mBio 2013 ; 4 : e00534 – 13 . Google Scholar CrossRef Search ADS PubMed 27 Raven KE , Gouliouris T , Brodrick H et al. Complex routes of nosocomial vancomycin-resistant Enterococcus faecium transmission revealed by genome sequencing . Clin Infect Dis 2017 ; 64 : 886 – 93 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Antimicrobial Chemotherapy Oxford University Press

Relentless spread and adaptation of non-typeable vanA vancomycin-resistant Enterococcus faecium: a genome-wide investigation

Loading next page...
 
/lp/ou_press/relentless-spread-and-adaptation-of-non-typeable-vana-vancomycin-dAd0yGga0Y
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: journals.permissions@oup.com.
ISSN
0305-7453
eISSN
1460-2091
D.O.I.
10.1093/jac/dky074
Publisher site
See Article on Publisher Site

Abstract

Abstract Background VRE are prevalent among patients in ICUs. Non-typeable vanA VRE, due to loss of one of the genes used for MLST (pstS), have increased in Australia, suggestive of a new, hospital-acquired lineage. Objectives To understand the significance of this lineage and its transmission using WGS of strains isolated from patients in ICUs across New South Wales, Australia. Methods A total of 240 Enterococcus faecium isolates collected between February and May 2016, and identified by conventional PCR as vanA positive, were sequenced. Isolates originated from 12 ICUs in New South Wales, grouped according to six local health districts, and represented both rectal screening swab (n = 229) and clinical (n = 11) isolates. Results ST analysis revealed the absence of the pstS gene in 84.2% (202 of 240) of vanA isolates. Two different non-typeable STs were present based on different allelic backbone patterns. Loss of the pstS gene appeared to be the result of multiple recombination events across this region. Evidence for pstS-negative lineage spread across all six local health districts was observed suggestive of inter-hospital transmission. In addition, multiple outbreaks were detected, some of which were protracted and lasted for the duration of the study. Conclusions These findings confirmed the evolution, emergence and dissemination of non-typeable vanA E. faecium. This study has highlighted the utility of WGS when attempting to describe accurately the hospital-based pathogen epidemiology, which in turn will continue to inform optimal infection control measures necessary to halt the spread of this important nosocomial organism. Introduction Enterococcus faecium emerged as a significant nosocomial pathogen in the 1990s and, since then, the number of hospital-acquired infections caused by this pathogen has increased worldwide.1 The National Healthcare Safety Network report has ranked E. faecium as the ninth most common pathogen causing healthcare-associated infections.2 In the USA, this equates to ∼3800 E. faecium infections a year, of which >82% were due to VRE. Although less virulent than other multiresistant organisms, VRE still result in greater infection-related morbidity and mortality compared with vancomycin-susceptible enterococci.3 In Australia, VRE colonization is uncommon in healthy individuals (prevalence 0.2%); however, among patients in ICUs, its prevalence is much higher (4.4%).4,5 ICUs are a major reservoir of VRE, with the risk of VRE infections closely associated with prior VRE colonization, varying from 0%–45% (median 4%) in VRE-colonized patients to <2% among non-colonized patients.5 Other risk factors for VRE infection include severe neutropenia, solid organ transplantation and presence of indwelling devices such as urinary catheters.6,7 In Europe and the USA, vancomycin resistance in E. faecium is predominantly due to acquisition of the vanA operon, whereas, until recently, the majority of Australian VRE strains have harboured the vanB operon.8–10 Factors influencing this emergence of vanA VRE across Australia are incompletely defined.11 A non-typeable E. faecium characterized by loss of one of the genes used in the MLST of enterococci, the pstS gene, was identified in Australia in 2015.12 This loss was mediated through a chromosomal inversion event and appears to have occurred repeatedly over time. Since its initial detection, there has been a dramatic increase in the number of non-typeable E. faecium identified across multiple jurisdictions in Australia, suggesting the appearance of a lineage that is both new and hospital-acquired.12 To characterize better the extent and significance of this lineage and its apparent spread to disparate geographical regions, we applied genome-wide analysis to examine vanA-carrying E. faecium strains isolated from patients within ICU wards across the jurisdiction of New South Wales, Australia. Methods Study setting and selection of isolates Between February and May 2016 all E. faecium isolates identified by conventional PCR as harbouring the vanA operon, i.e. vanA positive, from 12 New South Wales ICUs were studied. These ICUs encompass both secondary (n = 4) and tertiary/quaternary (n = 8) referral centres, caring for acute surgical and medical patients. All participating ICUs routinely perform rectal VRE screening on admission and weekly until ICU discharge. Only one isolate per patient was included. Unless positive prior to the study commencement, this was the initial isolate in the majority of cases. A total of 240 isolates were collected and represented isolates from both rectal screening swabs (n = 229) and clinical samples (n = 11). The hospitals were grouped into six major local health districts (LHDs), i.e. Hunter New England Health Service (LHD-1), Sydney Local Health District (LHD-2), South-Eastern Sydney Local Health District (LHD-3), South-Western Sydney Local Health District (LHD-4), Western Sydney Local Health District (LHD-5) and the Illawarra-Shoalhaven Local Health District (LHD-6). Library preparation and genome sequencing All VRE isolates were subcultured on to horse blood agar and incubated overnight at 37°C in 5% CO2 to ensure purity before analysis. Genomic DNA extraction was performed using the Blood and Tissue Mini Kit (Qiagen, Australia) for Gram-positive bacteria as per the manufacturer’s instructions. DNA extracts were treated with 1 U of RNase. Total DNA concentration was quantified using Picogreen (Invitrogen, Australia) and 1 ng/μL DNA was used to prepare DNA libraries employing the Nextera XT Library Preparation Kit (Illumina, Inc., CA, USA). Multiplexed libraries were sequenced using paired-end 150 bp chemistry on the NextSeq 500 (Illumina, Australia). Genome analysis Sequence reads were mapped against the reference isolate Aus0004 (GenBank accession no. GCA_000250945.1, an ST17 isolate containing the pstS gene) using Stampy v1.0.2313 with pre-BWA alignment. Variants were called using FreeBayes v1.1.0-dirty14 and filtered for read depth (minimum 20), read base quality (minimum Phred score 30) and mapping quality (minimum 30). Variation at indels and in the presence of mobile elements was excluded from the mapping-based analysis. Recombination was identified and masked from the phylogeny using Gubbins15 with the maximum likelihood phylogeny generated with RaxML v8.2.1016 and 100 bootstrap replicates. Mapping metadata to the phylogeny was performed in R using the ggtree package.17 The presence or absence of the pstS gene (this gene is one of seven genes used in the MLST scheme) was determined by in silico MLST performed on the de novo assembled contigs using SPAdes v3.10.1.18,19 Exploration of the number of possible pstS loss events was performed using ancestral state reconstruction within the R package ape.20 Early experiments showed the majority of isolates were missing the pstS gene. Subsequent comparative genomics was undertaken following mapping of reads to the closed pstS-negative genome DMG1500801 (GenBank accession no. GCA_900094185.1) and study isolates. To allow for contextual comparisons, the collection was supplemented with 43 (37 Victorian and 6 Australian Capital Territory) previously sequenced, non-typeable Australian E. faecium genomes (Table S1, available as Supplementary data at JAC Online). Although this dataset also contained 22 New South Wales isolates, these isolates were excluded from the analysis as the hospital/LHD origin was not described to allow for appropriate assignments in further analysis. To examine the spread of this clone between New South Wales and Victoria a phylogeographical analysis was performed as implemented through Bayesian Evolutionary Analysis Sampling (BEAST2).21 The optimal clock and tree model was selected based on the presence of convergence and Bayes factor following 100 000 000 iterations. Resultant trees were thinned prior to generating the final maximum clade credibility tree. Additional data including a jitter effect of ±0.01° of latitude or longitude between isolates from the same hospital within each LHD, to assist in visualization, were added using an in-house script prior to running SpreaD3.22 Similarly, BEAST2 was used to determine the molecular clock rate. All raw sequencing reads were deposited in NCBI project number PRJNA415172 (Table S1). Results Two hundred and forty vanA-positive isolates (representing 80% of all VRE isolated across participating LHDs) were obtained during the 4 month study period from the six LHDs (Figure S1). Although the number of isolates from each LHD varied widely (range 13–101), no significant changes were seen over time in the number of positives within any LHD resulting in an average of 60 (range 53–67) isolates obtained per month across the time period. The majority (89.6%, 215 of 240) of these isolates were newly acquired in the ICU, i.e. patients had initial screening swabs that were VRE negative but subsequently either had screening swabs that were positive for VRE or developed VRE infection. Of the 11 isolates from patients with infection, the likely source was the bloodstream (n = 5), abdominal collection (n = 4) or urinary tract (n = 2). The total number of infections represented 4.6% of all isolates detected. Over half of these occurred in one LHD, LHD-3 (n = 6), with the remaining infections occurring in LHD-2 (n = 1) and LHD-4 (n = 4). Of note, 54.5% of all infections (n = 6) occurred in patients who previously screened VRE negative. Determination of the pstS-negative clone Relative to the reference genome, 43 815 SNPs were detected across all isolates. Similar to previous observations,23 a substantial number of recombination events were identified (n = 1507) with masking of these events resulting in a final SNP matrix of 10 531 SNPs. In silico MLST identified the absence of the pstS gene in 84.2% (202 of 240) of all vanA isolates. On further examination, two different non-typeable STs were identified based on different allelic backbone patterns, which were most similar to VRE ST17 (allele profile: atpA-1, ddl-1, gdh-1, purK-1, gyd-1, pstS-1, adk-1) or ST80 (allele profile: atpA-9, ddl-1, gdh-1, purK-1, gyd-12, pstS-1, adk-1), profiles and are forthwith classified as ST17N and ST80N, respectively. ST17N predominated (71.3%; 144 of 202) with ST80N isolates accounting for 28.7% (58 of 202) of all pstS-negative isolates (Figure 1 and Figure S2). Figure 1. View largeDownload slide Maximum likelihood phylogeny based on whole genome sequence following masking of recombination of 240 vanA E. faecium isolates. Isolates represent positive screening or infection (red tips) strains recovered from New South Wales intensive care patients. Metadata are represented by concentric rings around the tree with the inner ring representing the LHD location of the isolate, while the outer ring indicates the in silico multilocus ST of the isolate. 17N* and 80N* isolates correspond with pstS-negative isolates with the remaining alleles most closely representing either ST17 (allele profile: atpA-1, ddl-1, gdh-1, purK-1, gyd-1, pstS-1, adk-1) or ST80 (allele profile: atpA-9, ddl-1, gdh-1, purK-1, gyd-12, pstS-1, adk-1), respectively (see the text for more details). This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 1. View largeDownload slide Maximum likelihood phylogeny based on whole genome sequence following masking of recombination of 240 vanA E. faecium isolates. Isolates represent positive screening or infection (red tips) strains recovered from New South Wales intensive care patients. Metadata are represented by concentric rings around the tree with the inner ring representing the LHD location of the isolate, while the outer ring indicates the in silico multilocus ST of the isolate. 17N* and 80N* isolates correspond with pstS-negative isolates with the remaining alleles most closely representing either ST17 (allele profile: atpA-1, ddl-1, gdh-1, purK-1, gyd-1, pstS-1, adk-1) or ST80 (allele profile: atpA-9, ddl-1, gdh-1, purK-1, gyd-12, pstS-1, adk-1), respectively (see the text for more details). This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Infection (versus colonization) was not associated with the pstS status of the isolate. The disproportionately higher number of infections detected in LHD-3 were evenly spread between pstS-negative and -positive isolates with no genomic feature detected to explain this observation. Furthermore, no significant gene or SNP difference was detected between infective and non-infective isolate genomes. The phylogeny supports previous observations of pstS gene loss occurring independently multiple times (Figure 1), while ancestral state reconstruction demonstrates that although ST17N and ST80N isolates share a common ancestor, ST80N emerged independently several times (Figure S3). Mechanistically, loss of the pstS gene would require either multiple inversion events, as hypothesized by Carter et al.,12 or recombination across this genomic region. The former mechanism appears less probable when ST80N isolates are included in the analysis, as this would hinge on two further independent mutation events in both the atpA and gyd housekeeping genes. Therefore, recombination, as detected across this region (see Figure S4), is the more likely mechanism leading to the emergence of this clone. Spread and dissemination of the pstS-negative clone Since the emergence of the pstS-negative lineage, it has not only disseminated widely across all six LHDs (Figures 2 and 3) but has also spread between Victoria and New South Wales on multiple occasions. The maintenance of this clone, within New South Wales, and spread to other LHDs (inter-hospital transmission) appears centred around LHD-2, which has the only jurisdictional liver transplant centre, liver transplantation being a known patient risk factor for VRE. This suggests that the expanding ‘burden’ of pstS-negative isolates may be partly secondary to local E. faecium adaptation and emergence of new ST80N isolates, within each LHD, following the initial spread of an ancestral clone (i.e. a pstS-positive isolate). Figure 2. View largeDownload slide Phylogeographical history of pstS-negative E. faecium clone spread in Victoria and New South Wales under a discrete diffusion model. Circular polygon area is proportional to the number of isolates maintaining that location. Blue connecting lines show transmission between Victoria and New South Wales. Insert depicts a magnification of the Sydney LHD networks revealing extensive dissemination across the ICUs in different LHDs. See Figure 3 for corresponding maximum likelihood phylogeny. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 2. View largeDownload slide Phylogeographical history of pstS-negative E. faecium clone spread in Victoria and New South Wales under a discrete diffusion model. Circular polygon area is proportional to the number of isolates maintaining that location. Blue connecting lines show transmission between Victoria and New South Wales. Insert depicts a magnification of the Sydney LHD networks revealing extensive dissemination across the ICUs in different LHDs. See Figure 3 for corresponding maximum likelihood phylogeny. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 3. View largeDownload slide Maximum likelihood phylogeny of pstS-negative E. faecium isolates following masking of recombination. Coloured shapes at tree tips represent the origin of each of the isolates. Highlighted nodes depict intra-LHD transmission events consistent with an outbreak. See Table S2 for more details. Several inter-LHD and inter-state transmission events are also seen. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 3. View largeDownload slide Maximum likelihood phylogeny of pstS-negative E. faecium isolates following masking of recombination. Coloured shapes at tree tips represent the origin of each of the isolates. Highlighted nodes depict intra-LHD transmission events consistent with an outbreak. See Table S2 for more details. Several inter-LHD and inter-state transmission events are also seen. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. At an LHD level, all districts had evidence of both intra- and inter-ICU transmission. For intra-ICU events to be considered as evidence of an outbreak, rather than new acquisition events, isolates had to differ by ≤2 SNPs based on the molecular clock estimates of ∼9 SNPs/genome/year (3.0 × 10−6; 95% highest posterior density 7–11 × 10−6 SNPs/site/year), which corresponds with 2.25 SNPs over the 4 month study time frame. At least one outbreak was detected within each LHD (Figure 3, depicted by the highlighted nodes). These outbreaks were protracted, with genomically indistinguishable isolates found >46 and 92 days after the index case (Table S2). Discussion This study provides important insights into the rapid evolution of vanA E. faecium in Australian hospitals following its emergence.12 It further reveals a new development in the evolution of pstS-negative VRE with the identification of MLST non-typeable VRE that are ST80N. Ancestral state reconstruction suggests that both ST17N and ST80N originated from a common ancestor carrying the pstS allele (Figure S3). A previous analysis of ST17N isolates12 similarly found a single common ancestor with three separate deletion events in its evolution. However, the present analysis inclusive of ST80N strains indicates that the most likely explanation for this emergence and spread is multiple recombination events in separate lineages resulting in the loss of this allele. This does not discount the initial event occurring as a result of a large inversion event mediated by flanking ISs.12 The prevalence of vanA VRE has noticeably increased over time from 2.6% of all E. faecium bacteraemias in 2013 to 19.9% in 2015 throughout Australia.24,25 This increase has coincided with the appearance of pstS-negative E. faecium, which was identified as the second most predominant MLST type in 2015.25 The origins of the new clone, based on the current analysis and epidemiological data obtained from sequenced blood culture isolates by the Australian Group on Antimicrobial Resistance, suggest that its emergence and maintenance is within New South Wales.24 Furthermore, it appears that the emergence of vanA VRE and pstS-negative E. faecium is linked. The factors driving these changes deserve further study. Examination of the genomic relationship between isolates revealed local dissemination of pstS-negative vanA VRE within hospitals demonstrated by multiple outbreaks. Although frequent reintroductions into the ICUs cannot entirely be discounted, the reservoir of this clone is more likely each LHD’s ICU. Detected outbreaks were not only numerous but also protracted with >92 days between isolates in one instance. Relaxing the number of mutations defining an outbreak would have resulted in outbreaks being present for the entire study period, despite local hospital infection control measures. Evidence for inter-LHD transmission was also detected. What is responsible for these events is unclear. Unlike patient movement within any LHD, movement between LHDs is infrequent. Therefore inter-LHD transmissions suggest a possible environmental reservoir of this clone. This hypothesis requires investigation, however, as hospital-associated lineages are not generally detected within individuals from the community.26 Nevertheless, the loss of the pstS gene in vanA VRE clearly represents a new development in the adaptation of this pathogen in Australia. Our study, similar to that by Raven et al.,27 highlights the utility of WGS in hospital epidemiology. Not only does WGS allow for greater accuracy of cluster association, it greatly enhances our understanding of transmission chains, which would be entirely undetected using previous typing methods. The importance of studying screening isolates is demonstrated in our study as it pre-dates the trends seen in bloodstream infection surveillance data and unmasks the total genome adaptation leading to new E. faecium clones. Several potential limitations of our study are acknowledged. First, the study was conducted over a relatively short time span (i.e. 4 months) within a single state in Australia. Although limited epidemiological data were obtained for each isolate, overall trends could only be gauged from recently published national bloodstream surveillance data, rather than local surveillance data. Finally, given the high proportion of colonization isolates and the relative weighting of pstS-negative isolates, we were unable to examine possible pathogen factors to account for the success in persistence of this clone. In conclusion, this study has confirmed and quantified the presence of MLST non-typeable ST17N vanA E. faecium in hospital settings but has also documented the emergence of MLST non-typeable ST80N vanA E. faecium in Australia. Understanding the evolution, emergence and dissemination of this clinically significant nosocomial clone will improve control interventions. This study further supports the utility of genome sequencing in understanding and monitoring the hospital epidemiology of VRE. Acknowledgements We wish to thank Bradley Watson, Helen Ziochos, Heather Wren and Chris McIver whose efforts ensured that all isolates were captured, worked up and stored appropriately during the study period. Funding This work was supported by an NSW Health Public Health Pathogen Genomics Partnership grant. Transparency declarations None to declare. Supplementary data Tables S1 and S2 and Figures S1 to S4 are available as Supplementary data at JAC Online. References 1 Cetinkaya Y , Falk P , Mayhall CG. Vancomycin-resistant enterococci . Clin Microbiol Rev 2000 ; 13 : 686 – 707 . Google Scholar CrossRef Search ADS PubMed 2 Weiner LM , Webb AK , Limbago B et al. Antimicrobial-resistant pathogens associated with healthcare-associated infections: summary of data reported to the National Healthcare Safety Network at the Centers for Disease Control and Prevention, 2011-2014 . Infect Control Hosp Epidemiol 2016 ; 37 : 1288 – 301 . Google Scholar CrossRef Search ADS PubMed 3 Bhavnani SM , Drake JA , Forrest A et al. A nationwide, multicenter, case-control study comparing risk factors, treatment, and outcome for vancomycin-resistant and -susceptible enterococcal bacteremia . Diagn Microbiol Infect Dis 2000 ; 36 : 145 – 58 . Google Scholar CrossRef Search ADS PubMed 4 Padiglione AA , Grabsch EA , Olden D et al. Fecal colonization with vancomycin-resistant Enterococci in Australia . Emerg Infect Dis 2000 ; 6 : 534 – 6 . Google Scholar CrossRef Search ADS PubMed 5 Ziakas PD , Thapa R , Rice LB et al. Trends and significance of VRE colonization in the ICU: a meta-analysis of published studies . PLoS One 2013 ; 8 : e75658 . Google Scholar CrossRef Search ADS PubMed 6 National Health and Medical Research Council . Australian Guidelines for the Prevention and Control of Infection in Healthcare . Canberra : Commonwealth of Australia , 2010 . 7 Infection Control Service, Communicable Disease Control Branch . Clinical Guidelines for the Management of Patients with Vancomycin-Resistant Enterococci (VRE): Version 6.2 . Adelaide : Government of South Australia , 2017 . 8 Coombs GW , Pearson JC , Daley DA et al. Molecular epidemiology of enterococcal bacteremia in Australia . J Clin Microbiol 2014 ; 52 : 897 – 905 . Google Scholar CrossRef Search ADS PubMed 9 Top J , Willems R , Bonten M. Emergence of CC17 Enterococcus faecium: from commensal to hospital-adapted pathogen . FEMS Immunol Med Microbiol 2008 ; 52 : 297 – 308 . Google Scholar CrossRef Search ADS PubMed 10 Werner G , Coque TM , Hammerum AM et al. Emergence and spread of vancomycin resistance among enterococci in Europe . Euro Surveill 2008 ; 13 : pii=19046. 11 van Hal SJ , Espedido BA , Coombs GW et al. Polyclonal emergence of vanA vancomycin-resistant Enterococcus faecium in Australia . J Antimicrob Chemother 2017 ; 72 : 998 – 1001 . Google Scholar PubMed 12 Carter GP , Buultjens AH , Ballard SA et al. Emergence of endemic MLST non-typeable vancomycin-resistant Enterococcus faecium . J Antimicrob Chemother 2016 ; 71 : 3367 – 71 . Google Scholar CrossRef Search ADS PubMed 13 Lunter G , Goodson M . Stampy: a statistical algorithm for sensitive and fast mapping of Illumina sequence reads . Genome Res 2011 ; 21 : 936 – 9 . Google Scholar CrossRef Search ADS PubMed 14 Garrison E , Marth G. Haplotype-Based Variant Detection From Short-Read Sequencing. http://arxiv.org/abs/1207.3907. 15 Croucher NJ , Page AJ , Connor TR et al. Rapid phylogenetic analysis of large samples of recombinant bacterial whole genome sequences using Gubbins . Nucleic Acids Res 2015 ; 43 : e15 . Google Scholar CrossRef Search ADS PubMed 16 Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies . Bioinformatics 2014 ; 30 : 1312 – 3 . Google Scholar CrossRef Search ADS PubMed 17 Yu G , Smith DK , Zhu H et al. ggtree: an r package for visualization and annotation of phylogenetic trees with their covariates and other associated data . Methods Ecol Evol 2017 ; 8 : 28 – 36 . Google Scholar CrossRef Search ADS 18 Bankevich A , Nurk S , Antipov D. SPAdes: a new genome assembly algorithm and its application to single-cell sequencing . J Comput Biol 2012 ; 19 : 455 – 77 . Google Scholar CrossRef Search ADS PubMed 19 Seeman T. MLST—Scan Contig Files Against PubMLST Typing Schemes. 2017 . https://github.com/tseemann/mlst. 20 Paradis E , Blomberg S , Bolker B et al. Package ‘Ape’. 2017 . https://cran.r-project.org/web/packages/ape/ape.pdf. 21 Bouckaert R , Heled J , Kühnert D et al. BEAST 2: a software platform for Bayesian evolutionary analysis . PLoS Comput Biol 2014 ; 10 : e1003537 . Google Scholar CrossRef Search ADS PubMed 22 Bielejec F , Baele G , Vrancken B et al. SpreaD3: interactive visualization of spatiotemporal history and trait evolutionary processes . Mol Biol Evol 2016 ; 33 : 2167 – 9 . Google Scholar CrossRef Search ADS PubMed 23 van Hal SJ , Ip CL , Ansari MA. Evolutionary dynamics of Enterococcus faecium reveals complex genomic relationships between isolates with independent emergence of vancomycin resistance . Microb Genom 2016 ; 2 : doi:10.1099/mgen.0.000048. 24 Coombs G , Daley D on behalf of the Australian Group on Antimicrobial Resistance. Australian Enterococcal Sepsis Outcome Program (AESOP) 2016 Final Report. http://agargroup.org.au/wp-content/uploads/2017/08/AESOP-2016-Final-Report-2017.pdf. 25 Coombs GW , Daley DA , Thin Lee Y et al. Australian Group on Antimicrobial Resistance Australian Enterococcal Sepsis Outcome Programme annual report, 2014 . Commun Dis Intell Q Rep 2016 ; 40 : E236 – 43 . Google Scholar PubMed 26 Lebreton F , van Schaik W , McGuire AM. Emergence of epidemic multidrug-resistant Enterococcus faecium from animal and commensal strains . mBio 2013 ; 4 : e00534 – 13 . Google Scholar CrossRef Search ADS PubMed 27 Raven KE , Gouliouris T , Brodrick H et al. Complex routes of nosocomial vancomycin-resistant Enterococcus faecium transmission revealed by genome sequencing . Clin Infect Dis 2017 ; 64 : 886 – 93 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Journal of Antimicrobial ChemotherapyOxford University Press

Published: Mar 16, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off