difference in prevalence of HRMO. Most studies have been conducted in the United States of America, where the prevalence of HRMO carriage among patients is higher than in the Netherlands, with consequently higher environmental contamination rates (26-28). Secondly, an explanation for the low contamination rates could be the chosen sample method. Based on our selection of sampled surfaces, we decided to sample with premoistened cotton swabs. While this method has some disadvantages, such as difficulty to standardize, they also come with several important advantages (29). Cotton swabs have high recovery rates on wet surfaces, similar or better recovery rates compared to other sampling methods, and they can be used on all surfaces, including surfaces that are more difficult to sample such as door handles (29-31). Additionally, since the swabs were directly placed in a selective broth, we were able to identify HRMO in low concentrations. A third explanation could be that, while other studies focused mostly on “high-touch” surfaces (e.g. bed rails, call buttons) we sampled built-in surfaces, with the exception of the nightstand (7, 23-25). These locations might be less frequently contaminated, but since these surfaces are used by all room occupants, they are potentially a better indicator of differences between multiple-occupancy and single-occupancy rooms. Interestingly, no sink or shower drains were sampled in the other studies, while we identified almost all HRMO on these surfaces, and not on “dry” surfaces (i.e. nightstands, tables). Notwithstanding, the contamination rates observed in our study are low, even after considering the low prevalence of HRMO in the Netherlands and our chosen sample methods. Thus, it is likely that other factors, such as our cleaning protocol, have contributed to these low rates. There are several explanations for the fluctuations over the three year follow-up period in CFU counts per cm2. As expected, the CFU counts in single-occupancy rooms and bathrooms were significantly lower before transferring patients to the new hospital building. However, this was not observed for the ICU rooms or rooms with an anteroom. One explanation for this is the fact that, while the construction of the single-occupancy rooms was mostly finished during the sampling moments, construction of the ICU rooms and rooms with anterooms was still ongoing. Consequently, more construction workers were present in these rooms, leading to relatively higher contamination levels. The fluctuations in CFU counts during the three years most likely reflected the use of the rooms. CFU counts were compared with the CFU counts determined in the old hospital building one month before relocating patients, since we believed that this was more representative for the contamination than the values determined one week before relocating patients. One week before relocating, the number of admissions to the hospital was lower, to prepare for the transfer of patients, and thus locations were used less frequently. We did not correct for use or nonuse of the bathroom by the patient. It is unclear why the CFU counts nine months after opening were higher in single-occupancy rooms. There were no changes in sampling or lab protocol that could explain the increase, and on later sampling moments, this increase in CFU counts was not observed again. A possible explanation is that there were changes in indoor temperature, or in humidity, which can impact the bacterial load (32). However, since we did not measure this, we cannot be sure about this. The final two sampling moments took place during the COVID-19 pandemic. The lower CFU count could be explained by enhanced cleaning and increased disinfection rates with 1000 ppm chlorine. Only four of the included single-occupancy rooms were dedicated for suspected COVID-19 patients, and two of the included isolation rooms were dedicated for COVID-19-care. 162 Chapter 3.2
RkJQdWJsaXNoZXIy MTk4NDMw