Vancomycin-resistant Enterococcus faecium: a national wide genomic snapshot in Switzerland

Abstract

Purpose

Vancomycin-resistant Enterococcus faecium (VREfm) remains an antibiotic challenge to healthcare systems. In Switzerland, VREfm remains rare but appears to be on the rise. We aimed to conduct a national genomic study of VREfm, identify major circulating clones, and explore evidence of clonal dissemination using genomic epidemiology.

Methods

All Swiss diagnostic laboratories were invited to submit VREfm isolates collected during February–March 2024 since strains with reduced susceptibility to daptomycin had been identified. Isolates underwent species confirmation, antimicrobial susceptibility testing, and whole genome sequencing. Sequence types (STs), core genome multilocus sequence typing (cgMLST), and resistance gene profiles were determined. Clusters were defined based on allele distances.

Results

We analysed 78 VREfm isolates from 17 cantons. Most carried the vanA gene (94%) and exhibited high-level resistance to vancomycin and teicoplanin; the remaining isolates carried vanB. Two clones dominated: ST80 (n = 43) and ST612 (n = 14), both within clonal complex 17. Eight cgMLST-defined clusters were identified. The ST80-Q cluster was confined to a single canton and linked to an outbreak. In contrast, ST612-C was found in 7 cantons, suggesting cryptic inter-cantonal dissemination. All ST612 isolates harboured mutations in liaR and liaS associated with reduced daptomycin susceptibility, although phenotypic resistance was not observed.

Conclusion

This first nationwide genomic survey reveals the emergence and dissemination of dominant VREfm clones in Switzerland, particularly ST80 and ST612. The results highlight the need for national genomic surveillance and coordinated infection prevention strategies to detect and contain high-risk VREfm clones.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10096-026-05505-5.

Introduction

Vancomycin-resistant Enterococcus faecium (VREfm) has emerged as a significant public health threat, particularly within healthcare settings across Europe [1]. This pathogen, known for its intrinsic resistance to many antibiotics and its acquired resistance to vancomycin, poses a serious challenge in the management of healthcare-associated infections. The epidemiology and prevalence of VREfm have seen a troubling rise, particularly in European countries where healthcare systems are under increasing stress to control its spread [1–3].

VREfm is primarily associated with nosocomial infections, such as bloodstream infections, urinary tract infections, and wound infections, particularly in immunocompromised patients [3]. The high adaptability of E. faecium in the hospital environment, combined with its ability to acquire and disseminate resistance genes, has facilitated its persistence and spread [4]. The prevalence of VREfm in Europe has shown a concerning upward trend over the past decades, with significant variations among different countries [1]. While some nations have managed to maintain low levels of VREfm through stringent infection control measures, others have experienced endemic levels, leading to increased morbidity, mortality, and healthcare costs.

Understanding the epidemiological patterns and prevalence of VREfm across European hospitals is crucial for developing effective infection prevention and control (IPC) strategies and antibiotic stewardship programs.

In Switzerland, the epidemiological situation regarding VREfm has been relatively stable but concerning [5]. While the prevalence of VREfm in Switzerland has historically been lower compared to its European neighbours, there has been a notable increase in cases over the past decade [6, 7].

Sporadic outbreaks of VREfm in healthcare settings have been observed, particularly in large hospitals and university medical centres [8]. These outbreaks are often associated with nosocomial transmission, especially in intensive care units (ICUs) and among patients with prolonged hospital stays.

As elsewhere, the factors contributing to the spread of VREfm in Switzerland include the movement of patients between healthcare facilities, the use of broad-spectrum antibiotics, and lapses in IPC practices. Additionally, the proximity to neighbouring countries with higher VREfm prevalence influences cross-border transmission [9].

Switzerland has implemented a national surveillance program to monitor the incidence of VREfm and other multi-drug-resistant organisms. Since 2004, the Swiss Centre for Antibiotic Resistance (www.anresis.ch) plays a crucial role in tracking resistance patterns and guiding public health interventions. In addition, IPC measures in Swiss hospitals are robust, with an emphasis on early detection, isolation of colonized or infected patients, and strict hygiene practices to prevent transmission.

Despite these efforts, challenges remain in controlling VREfm, and increasing prevalence of this pathogen places a burden on healthcare systems, necessitating continued vigilance and adaptation of IPC strategies. Importantly, as there is no national genomic surveillance, the spread of the different genomic clones between health care institutions and between cantons remains obscure.

In early 2024, a national alert concerning VREfm with reduced daptomycin susceptibility prompted coordinated sampling across diagnostic laboratories [10]. This provided a unique opportunity to assemble a countrywide collection of isolates and to characterise them using whole-genome sequencing (WGS), which offers high-resolution typing, detection of resistance determinants, and delineation of putative transmission links beyond the resolution of conventional phenotyping. The objectives of this study were to generate a nationwide genomic snapshot of VREfm in Switzerland; to identify the predominant sequence types and their resistance determinants; and to explore potential inter-hospital and inter-cantonal transmission using WGS data.

Material and methods

In response to a national alert regarding the spread of the ST612 clone with decreased susceptibility to daptomycin, the Federal Office of Public Health requested that all medical laboratories submit all consecutive VREfm isolates identified over a two-month period (February and March 2024) to the National Reference Centre for Emerging Antibiotic Resistance (NARA). In cases of duplicate isolates from a single patient, only the first isolate was retained for further analysis. Species identification was confirmed using MALDI-TOF. Antimicrobial susceptibility testing was performed using broth microdilution with the EUVENC Sensititre plate (Thermo Fisher, Switzerland), according to the manufacturer’s instructions. Minimum inhibitory concentrations (MICs) were determined for vancomycin, teicoplanin, quinupristin/dalfopristin, tetracycline, daptomycin, ciprofloxacin, erythromycin, tigecycline, gentamicin, ampicillin, and chloramphenicol. The reference strain Enterococcus faecalis ATCC 29212 was used as internal quality control. Breakpoint interpretation was performed according to EUCAST guidelines (version 14). As no EUCAST criteria are available for daptomycin in E. faecium, CLSI breakpoints were applied, using the susceptible dose-dependent (SDD) category (≤ 4 mg/L). Isolates with daptomycin MIC values at the upper end of the CLSI SDD range (i.e. ≥ 4 mg/L) or above were considered as showing reduced susceptibility for descriptive purposes. Interpretation of daptomycin MICs should be approached with caution due to known reproducibility limitations in this species.

Whole genome sequencing was performed on the isolates using Illumina platforms at the Institute of Microbiology, Lausanne University Hospital, and the Laboratory of Bacteriology, University Hospital of Basel. Read quality filtering, genome assembly and core genome multi-locus sequence typing (cgMLST) were performed using SeqSphere (version 10.0.6, Ridom) with default parameters. The pubMLST scheme was used to define MLST sequence types (https://pubmlst.org/organisms/enterococcus-faecium, last access 25.03.2026). The cgMLST scheme used for E. faecium was developed by Ridom (https://www.cgmlst.org/ncs/schema/Efaecium/, last access 25.03.2026) which includes 1423 loci. Within the cgMLST scheme, very closely related genomes are 'lumped' together in a Complex Type (CT). Based on our own experience of VREfm investigation, cgMLST clusters were defined as all isolates differing by 5 or less loci.

The analysis of the resistome was performed with AMRFinderPlus version 4.0.3 [11] using the genomes assembled with SeqSphere. All sequences are available at the European Nucleotide Archive (ENA) under the accession project numbers PRJEB97053 and PRJEB80474 (Table S). Genomes were annotated with bakta version 1.12.0 (https://pubmed.ncbi.nlm.nih.gov/34739369/). Alignments of the van transposon were performed with blastn version 2.14.0 (parameters: task = dc-megabplast, perc_identity = 30) (https://pubmed.ncbi.nlm.nih.gov/20003500/), including a maximum of 5 kb upstream and downstream of vanA. The syntheny plot was generated with pyGenomeViz (https://github.com/moshi4/pyGenomeViz).

Results

Between February and March 2024, we collected 72 VREfm isolates and received six additional genome sequences, representing a total of 78 patients hospitalized across 17 different cantons (Table S). The majority of samples (n = 55) were obtained through screening procedures (rectal or perianal swabs, or stool samples), while others originated from urine (n = 11) or various clinical specimens (n = 7). For five cases, the source of the sample was unknown.

Antimicrobial susceptibility and resistance determinants

Antimicrobial susceptibility testing confirmed the multidrug-resistant profile typical of hospital-associated E. faecium Table 1. All 72 tested vanA positive isolates exhibited high-level resistance to vancomycin (MIC > 32 mg/L, with the majority > 128 mg/L) and teicoplanin (MIC ≥ 64 mg/L in most cases), as expected from the genotype. All isolates were resistant to ampicillin and almost all to ciprofloxacin, in line with the presence of pbp5 variants and fluoroquinolone target mutations. In contrast, resistance to last-line agents remained uncommon: only four isolates (5.6%) were resistant to linezolid, and two (2.8%) to tigecycline.

Table 1.

MIC distribution of Swiss isolates of vancomycin resistant E. faecium. Susceptibility breakpoints according to EUCAST v14.0 guidelines are indicated in colours (green, susceptible; orange resistant; yellow, susceptible with increased dose). For daptomycin, a susceptible dose-dependent (SDD) breakpoint of ≤ 4 μg/mL was established by CLSI

Daptomycin MIC values ranged from 1 to 8 mg/L. Although EUCAST does not provide breakpoints for E. faecium, according to CLSI criteria the susceptible dose-dependent (SDD) breakpoint is ≤ 4 mg/L; therefore, most isolates, including one with a MIC of 8 mg/L, fall at or just above the upper limit of the SDD range.

Quinupristin/dalfopristin MICs showed a wide dispersion around the clinical breakpoint, with most isolates clustering at 1–2 mg/L, reflecting the reduced but highly variable streptogramin susceptibility typically observed in E. faecium.

Resistome and chromosomal mutations

Analysis of the resistome confirmed a characteristic hospital-associated clade A1 profile (Table S). All 78 genomes carried vancomycin resistance determinants, predominantly vanA (N = 72) with a minority vanB (N = 6), mirroring the phenotypic glycopeptide resistance pattern. Universally or near-universally present were genes conferring resistance to aminoglycosides (aac(6')-I), macrolides and streptogramins (msr(C)), and β-lactams (pbp5 variants), while erm(B) and dfrG were detected in the majority of isolates. High-level fluoroquinolone resistance was supported by frequent parC_S80I and gyrA_S83Y/S83I substitutions. Mutations in the cell envelope stress response system (liaR and liaS) associated with reduced daptomycin susceptibility were common, particularly among ST612, although phenotypic resistance to daptomycin was not observed. The four linezolid-resistant isolates all carried cfr and clustered within a single ST80 sublineage, highlighting clone-specific acquisition of clinically relevant resistance.

Sequence types, genomic clusters and van transposon structure

MLST analysis of the 78 isolates revealed that the majority belonged to ST80 (N = 43) and ST612 (N = 14). Additionally, twelve other MLST sequence types (STs) were identified, including ST117 (N = 7), ST203 (N = 4), ST2664 and ST168 (N = 2 each), as well as ST17, ST18, ST280, ST761, ST913, ST1299, ST1336, and ST2698, each represented by a single isolate. All MLST ST identified belonged to the hospital-associated Clade A1 (Table S).

cgMLST analysis revealed significant genetic diversity among the isolates, identifying eight distinct genomic clusters, each involving between 2 to 17 patients (Fig. 1 and Table S1). The ST80-Q cluster (17 patients) was detected in two major healthcare institutions within the same canton and showed 0 to 8 loci differences., indicating a high degree of genetic relatedness. Fourteen isolates ST612-C isolates were detected across 8 hospitals and 2 private practices in 7 cantons. These isolates showed minimal differences, with variations of only 0 to 3 loci. The ST80-S cluster (6 patients) was mainly identified in a single tertiary-care hospital; one additional isolate was found in another institution within the same canton. Isolates in ST80-S differed by 0–1 locus.

Fig. 1.

Minimum spanning tree based on cgMLST analysis of 78 Swiss VRE isolates. Each node represents one or more isolates, with the number inside the node indicating the MLST sequence type. The distance between nodes corresponds to the number of allelic differences. Isolates with high genetic similarity (≤ 5 allelic differences) were grouped into clusters (see Materials and Methods), which are highlighted by distinct colours. The grey background delineates the cluster boundaries

An alignment of the van transposon is presented in Figure S1. These results show that within the ST80 and ST117 lineages, distinct vanA transposon structures are observed, suggesting independent acquisitions of the van plasmid. In contrast, isolates belonging to the same genomic cluster exhibited identical transposon structures. The only exception was the ST612-C genomic cluster, in which two isolates displayed a slightly different structure compared with the others. Notably, these two isolates originated from the same canton in Switzerland, which may suggest local evolution of the transposon. However, these two strains also showed a shorter alignment due to a gap upstream of vanA; therefore, the observed structural difference may represent an assembly artifact rather than a true biological variation.

Discussion

This study provides the first nationwide genomic overview VREfm in Switzerland, revealing dominance of hospital-associated clade A1 lineages, particularly ST80 and ST612. Although collected over a short period, the isolates were assembled in the context of a national alert on VREfm with reduced daptomycin susceptibility [12] and therefore represent a comprehensive overview of circulating clones during this coordinated investigation. The predominance of these STs aligns with recent observations from Europe and underscore the persistent threat posed by VREfm to IPC and antimicrobial stewardship efforts [1, 13–15].

The detection of ST612 across seven cantons showed limited allelic divergence (0–3 loci) indicates recent clonal expansion and possible inter-institutional spread during the surveillance window. In parallel, ST80 formed multiple subclusters, including the ST80-Q outbreak restricted to one canton (0–8 loci), and the ST80-S cluster that also contained four linezolid-resistant cfr-positive isolates alongside two linezolid-susceptible cfr-negative isolates (MIC = 2 mg/L). These patterns illustrate how WGS delineates relatedness at high resolution and identifies clusters that may warrant targeted investigations by infection prevention and control (IPC) teams.

Resistome analysis confirmed the characteristic multidrug-resistant profile of clade A1, with near-universal presence of pbp5 variants, aac(6’)-I, msr(C), erm(B) and dfrG, and widespread mutations in parC and gyrA conferring fluoroquinolone resistance, consistent with international reports on hospital-adapted E. faecium [2, 3, 16]. The predominance of the vanA gene and a corresponding phenotype of high-level resistance to both vancomycin and teicoplanin is consistent with European epidemiological data. The minority of vanB-positive isolates in our cohort exhibited the expected teicoplanin susceptibility pattern and aligns with reports showing that vanB remains relatively uncommon in most of Europe, although it has emerged as dominant in Norway [17], Australia [18], and to some degree in Germany [15].

Of particular importance, all ST612 isolates carried liaR and liaS mutations that have associated with reduced daptomycin susceptibility [19]. However, these mutations alone are not sufficient to confer high-level phenotypic resistance and are generally considered markers of an early or intermediate stage in the evolution of resistance. Interpretation of daptomycin susceptibility in E. faecium nonetheless requires caution. EUCAST does not provide clinical breakpoints for this species, and reproducibility of MIC determination is known to be limited. CLSI defines a susceptible dose-dependent (SDD) breakpoint of ≤ 4 mg/L for E. faecium, and several isolates in our collection exhibited MICs at the upper end of this range, including one isolate with a MIC of 8 mg/L. Such borderline or elevated MICs, particularly when combined with resistance-associated mutations, may reflect reduced susceptibility and a potential for rapid emergence of resistance under selective pressure. These findings suggest that genomic markers such as liaFSR mutations may provide earlier indication of evolving daptomycin resistance than phenotypic testing alone. These findings illustrate how genomic data refine interpretation of AMR phenotypes, provide early warning of emerging resistance threats, and distinguish between clonal spread and independent acquisition of resistance determinants.

Comparative analysis of the van transposon structure further refines interpretation of the genomic diversity observed within ST80 and ST612. Within the broader ST80 lineage, multiple distinct vanA transposon configurations were identified, consistent with heterogeneous acquisition of vanA-encoding elements among genetically distant ST80 sub-lineages. This finding supports the interpretation that the cgMLST diversity observed between these sub-lineages is not solely the result of recent diversification within Switzerland but may reflect either multiple introductions of distinct ST80 backgrounds or independent acquisition events within pre-existing ST80 populations. In contrast, within closely related genomic clusters defined by cgMLST, including ST80-Q and ST612-C, the vanA transposon structure was largely conserved, reinforcing the hypothesis of clonal dissemination within these sub-lineages. The minor structural variation observed in two ST612-C isolates from the same canton likely reflects local transposon evolution rather than independent introduction. Nevertheless, given the well-documented structural plasticity of vanA-carrying plasmids and transposons, definitive conclusions regarding the number of introductions cannot be drawn without comprehensive plasmid-level reconstruction using long-read sequencing.

This study has several limitations. The short two-month collection window provides only a snapshot of VREfm epidemiology and limits our ability to assess temporal trends. It also precludes detailed evaluation of long-term transmission dynamics and persistence of specific lineages across institutions. In addition, plasmid profiling and detailed analysis of mobile genetic elements were not performed, limiting our ability to further investigate transmission dynamics at the plasmid level. The lack of patient-level epidemiological data (e.g., hospitalizations and transfers between healthcare institutions, previous hospitalizations or travel abroad, underlying conditions, preceding antibiotic treatment, infections) restricted our ability to correlate genomic findings with transmission pathways or clinical outcomes. Lastly, no longitudinal sampling was performed to distinguish between transient carriage and persistent colonization.

In conclusion, this national survey reveals the emergence and spread of ST80 and ST612 VREfm clones across Switzerland, with both localized and widespread dissemination patterns. These results call for urgent investment in national genomic surveillance, inter-cantonal coordination, and real-time data integration to prevent further spread of high-risk clones. Proactive genomic monitoring, combined with robust epidemiological tracking, will be essential to contain VREfm and preserve the effectiveness of remaining therapeutic options.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

All authors contributed to the conception, design, and execution of this study.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

All sequences are available at the European Nucleotide Archive (ENA) under the accession project numbers PRJEB97053 and PRJEB80474

Declarations

Competing interests

The authors declare no competing interests.

Ethical approval

This work was conducted as part of a routine outbreak investigation led by the National Nosocomial Outbreak Investigation Centre under the mandate of the Federal Office of Public Health of Switzerland and an ethical approval was not needed.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work the authors used ChatGPT (OpenAI) in order to improve the readability and language of the manuscript. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article.