microorganism, or when WGS showed that the discharge isolate was not identical to the admission isolate. Results of the perianal sample were not communicated to medical staff or patients, were not registered in the electronic health records (EHR), and hence, no infection prevention measures were taken based on the results, as stated in the protocol approved by the medical ethical research committee of the Erasmus MC (MEC-2017-1011). Microbiological methods Perianal samples were taken with flocked swabs and transported in the accompanying 1mL Amies medium (e-Swabs (Copan Italia, Brescia, Italy)). Perianal samples collected from January 1, 2018, until January 19, 2019, were stored in a -80°C freezer before being processed. To prevent freezing/defrosting damage, 0.2 mL 99% glycerol was added to the samples before freezing (12). Samples taken after January 19, 2019 were processed directly. All samples, regardless of being frozen, were processed following the same protocol. Samples were vortexed for 10 s before 250µL of the sample was inoculated in a tryptic soy broth with vancomycin (50mg/L) and incubated overnight at 35°C. A BrillianceTM ESBL Agar (Oxoid, Basingstoke, UK) was inoculated from the broth with a 10 µl loop and incubated twice overnight at 35°C. Colonies were identified to species level using Matrix-Assisted Laser Desorption/Ionization Time-Of-Flight mass spectrometry (MALDI-TOF [Bruker Daltonics, Bremen, Germany]) and antibiotic susceptibility was tested with the VITEK®2 (bioMérieux, Marcy l’Etoile, France). Antibiotic susceptibility results were interpreted according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines (13). All ESBL-E isolates were stored in a -80°C freezer. Whole genome sequencing WGS was performed for all identified ESBL-E isolates. Total genomic DNA was extracted using the MagNA Pure 96 platform (Roche Applied Science, Mannheim, Germany). Genomic DNA was fragmented by shearing to a size of ~350 bp. Libraries were prepared using the NEBNext® DNA Library Prep kit (New England Biolabs, Ipswich, MA, USA) and subjected to 150 bp paired-end sequencing creating >100x coverage using Illumina technology (Novogene, HongKong, China). De novo genomic assemblies were generated using CLC Genomics Workbench v21 (Qiagen, Hilden Germany) using default parameters. Antimicrobial resistance (AMR) genes were detected and identified using the web-based Comprehensive Antibiotic Resistance Database (CARD) interface (https://card.mcmaster.ca/) restricted to perfect and strict hits (14). Conventional multi locus sequence types (MLST) and core-genome MLST cluster types were determined using each species’ corresponding scheme (https://cgmlst.org/ncs) in SeqSphere+ v5 software (Ridom, Munster, Germany). The identity of all strains was verified by analyzing the genomic assemblies using the online TYGS platform (https://tygs.dsmz.de/) (15). Data collection Patient characteristics were collected from the EHR, including the demographic variables age at admission and sex. For the hospitalization period, data on admission specialization, all antibiotic usage, surgical procedures, ICU admission, length of hospital stay, and number of intra-hospital patient transfers were collected. Intra-hospital transfers were defined as being transferred to another patient room for ≥4 h, and did not include transfers to e.g. the 2 29 Effect of single-occupancy rooms on ESBL-E acquisition
RkJQdWJsaXNoZXIy MTk4NDMw