Abstract
Vancomycin-resistant Enterococcus faecium (VREfm) has become a significant nosocomial pathogen due to its potential to cause outbreaks. Whole-genome sequencing (WGS) is considered the reference method for determining genomic relatedness among outbreak strains, but its routine use in clinical microbiology laboratories remains challenging. Consequently, faster and simpler typing methods are needed. Fourier transform infrared spectroscopy (FTIR) captures the unique infrared fingerprint of each isolate, enabling the comparison of spectral profiles to infer genomic relatedness. In this study, we evaluated the performance of FTIR for identifying genomic clusters of VREfm in a tertiary hospital, in comparison with three WGS-based methods: core-genome multilocus sequence typing (cgMLST), core-genome single nucleotide polymorphism analysis (cgSNP), and split k-mer analysis (SKA). A total of 87 VREfm isolates, collected between April 2020 and October 2023, were typed using both FTIR and WGS. Among these, 56 were associated with three outbreaks in the surgery, nephrology, and oncohematology units, according to conventional epidemiology. Concordance between typing methods was assessed using the Adjusted Rand index (AR) and Adjusted Wallace coefficient (AW). All three WGS-based methods yielded similar clustering results and revealed one monoclonal and two polyclonal outbreaks. Using cgMLST as the reference, an optimal FTIR cutoff range of 0.210-0.227 was determined. FTIR clustering results showed strong concordance with WGS-based methods; however, concordance with SKA was slightly lower. These findings suggest that FTIR provides clustering information comparable to WGS-based methods, providing a rapid and practical alternative to support timely infection control measures during VREfm outbreaks.