The spatial and temporal distribution of SARS-CoV-2 from the built environment of COVID-19 patient rooms: A multicentre prospective study.

ABSTRACT

BackgroundSARS-CoV-2 can be detected from the built environment (e.g., floors), but it is unknown how the viral burden changes over space and time surrounding an infected patient. Characterising these data can help advance our understanding and interpretation of surface swabs from the built environment. MethodsWe conducted a prospective study at two hospitals in Ontario, Canada between January 19, 2022 and February 11, 2022. We performed serial floor sampling for SARS-CoV-2 in rooms of patients newly hospitalized with COVID-19 in the past 48 hours. We sampled the floor twice daily until the occupant moved to another room, was discharged, or 96 hours had elapsed. Floor sampling locations included 1m from the hospital bed, 2m from the hospital bed, and at the rooms threshold to the hallway (typically 3 - 5m from the hospital bed). The samples were analyzed for the presence of SARS-CoV-2 using qPCR. We calculated the sensitivity of detecting SARS-CoV-2 in a patient with COVID-19, and we evaluated how the percentage of positive swabs and the cycle threshold of the swabs changed over time. We also compared the cycle threshold between the two hospitals. ResultsOver the 6-week study period we collected 164 floor swabs from the rooms of 13 patients. The overall percentage of swabs positive for SARS-CoV-2 was 93% and the median cycle threshold (for positive swabs) was 33.7 (IQR 30.9, 37.5). On day 0 of swabbing the percentage of swabs positive for SARS-CoV-2 was 81.1% and the median cycle threshold was 33.7 (IQR 32.1, 38.3) compared to swabs performed on day 2 or later where the percentage of swabs positive for SARS-CoV-2 was 98.1% and the cycle threshold was 33.4 (IQR 30.7, 35.7). We found that viral detection did not change with increasing time (since the first sample collection) over the sampling period, OR 1.65 per day (95% CI 0.68, 4.02; p = 0.27). Similarly, viral detection did not change with increasing distance from the patients bed (1m, 2m, or 3m), OR 0.85 per metre (95% CI 0.38, 1.88; p = 0.69). The cycle threshold was lower (e.g. more virus) in The Ottawa Hospital (median Cq 30.8) where the floors are cleaned once daily rather than the Toronto hospital (median Cq 37.3) where floors were cleaned twice daily. ConclusionsWe were able to detect SARS-CoV-2 on the floors of rooms of patients with COVID-19 and the viral burden did not vary over time or by distance from the bed. These results suggest floor swabbing for the detection of SARS-CoV-2 in a built environment such as a hospital room is both accurate and robust to variation in sampling location and duration of occupancy.