Other studies have suggested a cutoff value for the number of CFU for hand contact surfaces in the healthcare environment. Dancer et al. suggested 5 CFU/cm2, however, due to our cutoff value of 100 CFU per RODAC, which translates to a maximum of 3.95 CFU/cm2, we were unable to determine if this criteria was exceeded (33). Griffith et al. suggested <2.5 CFU/cm2 as a cutoff value, a value that they found was practicable for all sites after disinfection (34, 35). Nonetheless, CFU counts are not helpful to determine if a source is contaminated with HRMO. While we did not determine the correlation between CFU counts and HRMO presence, other studies have not shown a correlation between CFU counts and HRMO presence (36, 37). WGS was performed on all identified isolates. No persistent contamination over time was identified in the new hospital building. Remarkably, in isolates that were considered to be genetically closely related, variation in the presence of AMR genes was detected. We believe this to be the result of plasmid gain/loss in strains of otherwise identical genetic background. Plasmid gain/loss as possible explanation for these observations fell beyond the scope of this study. Another interesting result is that one K. pneumoniae strain was of ST16 (Supplement 4). This strain is an important emergent lineage of K. pneumoniae, has caused multiple outbreaks within European hospitals, and is known to carry multiple carbapenem resistance genes (38, 39). However, the strain identified in the old hospital did not carry any gene encoding carbapenem resistance. Another interesting finding is that seven E. hormaechei strains, both from the old and the new hospital building, were of ST78 (Supplement 4). ST78 isolates are successful One Health clones that are considered high risk and are of global interest (40). Additionally, nosocomial infections with this ST, both in Europe and Asia, are increasingly reported (41, 42). The ST78 isolates we identified from the hospital environment were CPE and carried blaOXA-48. As far as we know, these strains have not yet lead to nosocomial infection in our patients, but it is important to monitor presence of this strain. Strengths and limitations The main strength of this study is that we sampled the old and the new hospital building, with identical sampling methods and sampling locations. A second strength of our study is the follow-up period of three years in the new building. This follow-up period not only provided us with the opportunity to look at a situation where environmental contamination had developed, but also provided time for that contamination to build up further. Thirdly, we did not focus on environmental contamination with one type of HRMO, but looked at the presence of MRSA, ESBL-E, CPE, CP-PA, CR-AB, and VRE. Finally, we sampled a large number of rooms, on different wards, including isolation-, hematology-, and ICU rooms. A limitation of our study is that we were not able to sample every room at every sample moment. When a patient was cared for in isolation, in a non-isolation room, we did not sample this room, but we sampled a nearby room instead. During the next sampling moment, the original room was sampled again. Secondly, it is likely that our study shows an underestimation of the environmental contamination. This could be due to our chosen sampling method or the selected sample sites. On the other hand, every sample method or selection of sampled surfaces will inherently introduce bias, and hence, it is unlikely that 3 163 Environmental contamination with MDRO in single-occupancy rooms
RkJQdWJsaXNoZXIy MTk4NDMw