Surveillance of vancomycin-resistant enterococci reveals shift in dominating clones and national spread of a vancomycin-variable vanA Enterococcus faecium ST1421-CT1134 clone, Denmark, 2015 to March 2019

We describe clonal shifts in vanA Enterococcus faecium isolates from clinical samples obtained from patients in Denmark from 2015 to the first quarter (Q1) of 2019. During Q1 2019, the vancomycin-variable enterococci (VVE) ST1421-CT1134 vanA E. faecium became the most dominant vanA E. faecium clone and has spread to all five regions in Denmark. Among 174 E. faecium isolates with vanA, vanB or vanA/vanB genes in Q1 2019, 44% belonged to this type.

We describe clonal shifts in vanA Enterococcus faecium isolates from clinical samples obtained from patients in Denmark from 2015 to the first quarter (Q1) of 2019. During Q1 2019, the vancomycin-variable enterococci (VVE) ST1421-CT1134 vanA E. faecium became the most dominant vanA E. faecium clone and has spread to all five regions in Denmark. Among 174 E. faecium isolates with vanA, vanB or vanA/vanB genes in Q1 2019, 44% belonged to this type.
We describe the clonal shift for vanA Enterococcus faecium during the last 4 years and the national spread of a vancomycin-variable vanA E. faecium ST1421-CT1134 clone in Denmark. The aim is to highlight the importance of using molecular methods for detecting vancomycin-variable enterococci (VVE), and to alert other countries about this emerging nosocomial clone.

Vancomycin-variable enterococci
Vancomycin-variable enterococci (VVE) are E. faecium harboring the vanA gene complex, but being phenotypically vancomycin susceptible [1,2]. VVE can only be detected by molecular methods and cannot be cultured on selective vancomycin-containing media. Different clones of VVE have caused nosocomial outbreaks and development of vancomycin-resistant revertant mutants in vitro and in vivo has been described [1,[3][4][5]. This makes the detection of VVE highly important in clinical samples in order to assure relevant antibiotic treatment and in screening samples to avoid nosocomial spread. In 2015 and 2016, sporadic VVE with different genetic background were detected in the Capital Region of Denmark, in connection with vancomycin-resistant enterococci (VRE) outbreaks (data not shown). In 2016, a VVE clone belonging to ST1421-CT1134, which displays variable vancomycin susceptibility (minimum inhibitory concentration (MIC) 1 to ≥ 256 mg/ml) was detected in screening samples from a hospital in the Capital Region [5]. One strain, Efm-V1511, belonging to this clone was characterised by Hansen et al. [5]. Efm-V1511 had a 49.6 Kp plasmid, which carried the Tn1546 (vanA transposon). Tn1546 was truncated in vanX by a 252 bp 3' deletion explaining the vancomycin susceptibility of Efm-V1511. In ST1421-CT1134 isolates resistant to vancomycin, resistance could be attributed to changes in ddl disrupting gene function sometimes accompanied by changes in vanS, increased pHVH-V1511 copy number or the existence of an additional vanA-containing plasmid encoding a functional vanX [5].

National surveillance of vancomycinresistant and vancomycin-variable enterococci
We have previously described the surveillance of vancomycin-resistant enterococci (VRE) in clinical isolates in Denmark from 2005 to 2015 [6]. In the present study, we follow up and describe the data from isolates obtained from 2016 through the first quarter (Q1) of 2019. Since 2005, VRE isolates from clinical samples, e.g. urine, blood and tissue, as opposed to screening (faecal) isolates have been voluntarily submitted to Statens Serum Institut (SSI) from Danish Departments of Clinical Microbiology (DCM) for species identification, genotyping and surveillance ( Figure 1) [7]. Only one isolate per patient per 12 months was included. All VRE isolates (699 E. faecium and 30 E. faecalis) were tested for the presence of vancomycin resistance genes vanA and vanB by PCR from 2005 through 2014. From 2015 through Q1 2019, all clinical VRE/VVE isolates (n = 1,935) underwent whole-genome sequencing (WGS) as previously described [6]. From the WGS data, multilocus sequence type (MLST), and van genes were extracted in silico. The isolates were further subtyped in SeqSphere + (Ridom GmbH, Münster, Germany (http://www.ridom.de/seqsphere/)) using the cgMLST scheme by de Been et al. [8] (Figure 2).
The E. faecium isolates belonged to 29 sequence types (STs). ST80 (22%), ST203 (65%) and ST1421 (9%) were most prevalent. Typing by cgMLST revealed 156 different complex types (CTs).  Modified from DANMAP 2017 [7]. was only detected from clinical samples from the Capital Region. In 2018, 34% of the E. faecium isolates belonged to ST1421-CT1134 and were detected in the Capital Region, the Region Zealand and from one DCM in the Region of Southern Denmark (Table 1, Table 2). During Q1 2019, ST1421-CT1134 vanA E. faecium was the most prevalent type (44%) ( Table 1). It was detected in all five regions of Denmark, 50 isolates from the Capital Region, one isolate from Region Zealand, 23 isolates from the Region of Southern Denmark, two isolates from Central Denmark Region and one isolate from the North Denmark Region (Table 2). Furthermore, ST1421-CT1134 vanA E. faecium spread to the Faroe Islands during 2018 and 2019 (data not shown).

Discussion and conclusion
During 2005 to Q1 2019, most of the Danish clinical VRE isolates have been vanA E. faecium isolates. This study shows that predominating clones shifted over time and, importantly, the emergence of a vancomycinvariable clone, ST1421-CT1134 vanA E. faecium, that has spread to all the five Danish regions in 2019.
ST80-CT14 vanA E. faecium was highly prevalent in the Capital Region during 2012 to 2015 [9]. The vanA E. faeciumconstituting Group2_ST80 in the paper by Pinholt et al. [9] belonged to ST80-CT14 (data not shown). On a national level, the numbers of ST80-CT14 vanA E. faecium decreased during 2016 to 2018, and this clone was not detected during Q1 2019.
Because of differences in diagnostic algorithms, there is a detection bias of VVE. It seems very likely that ST1421-CT1134 vanA E. faecium have been underreported in some regions at least during some periods. Thus, the rising incidence could partly be explained by  increasing molecular testing of vancomycin susceptible isolates. However, a sharply increasing incidence has also been seen in DCM with extensive testing for VVE.
The origin of ST80-CT14 vanA E. faecium and ST203-CT859 vanA E. faecium are still unknown. vanA E. faeciumisolates belonging to ST1421-CT1134 have also been reported from Australia, but these isolates have not been VVE [10]. Why these three clones were so successful is unknown.
The spread of the VVE clone, ST1421-CT1134 vanA E. faecium, in Denmark is of concern, especially since VVE diagnostic is challenging. Because of this, the clone is likely to be underdiagnosed, which facilitates further spread. Since cross-border spread has been described for VRE, countries with patients transferred from Denmark should be aware of the vancomycin-variable ST1421-CT1134 vanA E. faecium clone.