New multilocus sequence typing scheme for Enterococcus faecium reveals sequential outbreaks of vancomycin-resistant E. faecium ST1162 and ST610 in a Japanese tertiary medical center

Abstract

Vancomycin-resistant Enterococcus faecium (VREfm) is a major nosocomial pathogen, and molecular epidemiological tools are crucial for controlling its spread. Pulsed-field gel electrophoresis (PFGE) is still used in clinical laboratories despite the increased accessibility of whole-genome sequencing (WGS). As PFGE equipment is no longer commercially available, clinical laboratories need alternative tools. Highly standardized multilocus sequence typing (MLST) is one option. However, the original MLST scheme for E. faecium, designed in 2002, showed inconsistencies with WGS-based typing. Therefore, the new Bezdíček MLST scheme, which offers more accurate genetic similarity based on genome-wide data, has recently been proposed. To clarify its clinical utility in analyzing nosocomial VREfm transmission, we compared both MLST schemes with PFGE using 68 VREfm isolates collected during an outbreak at a Japanese tertiary medical center in 2019. PFGE analysis identified nine clusters among the 68 strains, including two predominant clusters. The original scheme identified five sequence types (ST_Os), of which 82.4% (56/68) were ST_O192. The Bezdíček scheme identified nine sequence types (ST_Bs), subdividing the original ST_O192 into ST_B1162 (30/56), ST_B610 (25/56), and ST_B895 (1/56). Simpson's index of diversity values were 0.635, 0.317, and 0.648 for PFGE, the original scheme, and the Bezdíček scheme, respectively. Combining the Bezdíček scheme with admission records provided clearer outbreak visualization, indicating that two distinct ST_Bs independently caused sequential outbreaks. With high discriminatory power comparable with PFGE and global availability, the Bezdíček scheme is a practical and valuable tool for controlling nosocomial VREfm infections in clinical laboratories.IMPORTANCEIn areas where vancomycin-resistant Enterococcus faecium is common, hospital-acquired infections pose a considerable threat to patients' lives owing to treatment difficulties. Although whole-genome sequencing-based typing has logically become the new reference standard and its accessibility is growing, many clinical laboratories still lack the fundamental resources to exploit its full potential. Limited availability of the traditional pulsed-field gel electrophoresis test in clinical settings has necessitated the use of alternative tools such as Bezdíček multilocus sequence typing. This study tested the clinical utility of the Bezdíček scheme by comparing it with pulsed-field gel electrophoresis. Designed using Czech isolates, this scheme showed comparable discriminatory powers with the traditional method for geographically distinct Japanese isolates and clearly visualized outbreaks. These findings suggest that the Bezdíček scheme is a potential alternative to pulsed-field gel electrophoresis for identifying hospital transmission of vancomycin-resistant Enterococcus faecium in clinical laboratories.

Keywords: Enterococcus faecium; multilocus sequence typing; outbreak; pulsed-field gel electrophoresis; vancomycin-resistant.

Fig 1

Comparison of ST distribution between the original and Bezdíček MLST schemes based on PFGE clustering analysis of the studied strains. (a) The dendrogram of the unweighted pair group method with arithmetic mean (UPGMA) was derived from the similarity of PFGE band patterns. The red line indicates 85% similarity. (b) Cluster analysis is based on 85% similarity. *: singleton. (c) The date of the first VREfm isolate in 2019 is presented as "day X". (d) The circles, diamonds, and triangles represent the use of proton pump inhibitors (PPIs), H2 blockers, and potassium-competitive acid blockers (P-CABs), respectively. The duration of use prior to the detection of VRE is indicated by colors: black for ≥3 days and white for <3 days. (e) Indicates that PPIs and H2 blockers were used. (f) Indicates that PPIs and P-CABs were used. (g) Indicates that vancomycin was used before the isolation of VRE.

Fig 2

Phylogenetic tree and goeBURST full MST analysis of STs detected using the original and Bezdíček MLST schemes. (A, C) UPGMA tree for the concatenated allelic sequences from (A) the original scheme and (C) the Bezdíček scheme. The evolutionary distances were calculated using the maximum composite likelihood method, measured as the number of base substitutions per site. (B, D) Genetic relationships between allele profiles from (B) the original and (D) Bezdíček schemes, overlaid on PFGE cluster distribution and analyzed using the goeBURST Full MST algorithm at the SLV level. Colors outside ST nodes are shown as group founders in light green and common nodes in light blue. Colors inside ST nodes represent PFGE clusters A to I, as indicated in the figure legend. The size of each node represents the number of isolates on a log-scale, whereas the numbers in parentheses denote the number of isolates in each ST node. The black link represents “without recourse to tiebreak rules.” ST_O and ST_B indicate the sequence type according to the original MLST and Bezdíček MLST schemes, respectively.

Fig 3

Visualization of the outbreak using the Bezdíček MLST scheme and tracing carriers. (A) Distribution of detection dates by ST_Bs from the Bezdíček scheme. ST_B indicates the sequence type according to the Bezdíček MLST scheme. (B) The number in the outline represents the strain number. The number in each well represents the number of days spent by each carrier in the same ward before positive confirmation. The orange, blue, and green color coding represents contact risks between ST_B1162, ST_B610, and ST_B1164, respectively. The duration in days is represented by four color shades: 1–3 days, 4–7 days, 8–14 days, and 15 or more days. The red-boxed areas indicate the period during which isolation was presumably concentrated owing to horizontal spread within the ward. The red arrows indicate movements that may have triggered transmission in the ward after contact with the carriers.