Ebola Virus Stability on Surfaces and in Fluids in Simulated Outbreak Environments

We evaluated the stability of Ebola virus on surfaces and in fluids under simulated environmental conditions for the climate of West Africa and for climate-controlled hospitals. This virus remains viable for a longer duration on surfaces in hospital conditions than in African conditions and in liquid than in dried blood.


The Study
We tested the stability of EBOV on 3 materials commonly found in an Ebola treatment unit (ETU) in West Africa: 1) utility-grade (308) stainless steel washers (McMaster-Carr, Atlanta, GA, USA); 2) plastic (Teflon [polytetrafluoroethylene]; McMaster-Carr); and 3) Tyvek (from the front of a coverall). For each time point, 3 disks (4-cm diameter) of each material were placed individually into wells of a 6-well plate. Five samples (10 μL/sample) containing a total dose of 10 6 50% tissue culture infectious doses (TCID 50 s) of EBOV in cell-free medium were evenly distributed on the disks. The plates were divided into groups, and each group was placed into a plastic HEPA-filtered box and placed at 21°C, 40% RH, or 27°C, 80% RH. The samples were dried naturally, and virus titers were determined over a 14-day period.
In the surface and fluid stability experiments, all samples were stored at −80°C until titration (1 freeze-thaw cycle of EBOV samples that did not change virus titer). Titrations were performed on Vero E6 cells as described (10,11). The TCID 50 per milliliter for each sample at each time point was calculated by using the Spearman-Karber method (11).
Because viral decay rates often exhibit first-order kinetics (12), we log 10 transformed our TCID 50 calculations to represent virus titer and used a linear regression analysis (Prism version 6.05; GraphPad, San Diego, CA, USA) to determine the log 10 reduction rate of EBOV on each surface at both environmental conditions ( Figure 1; Table 1). We also determined whether linear regression models were significantly different from each other at the p<0.05 level by using an analysis of covariance equivalent test in Prism. Overall, virus remained viable longest in hospital conditions and on Tyvek. Viable EBOV was detectable for 3 days on Tyvek at tropical conditions.
The stability of EBOV in water was assessed by diluting 150 µL virus stock in 2.85 mL of Ambion diethylpyrocarbonate-treated water (Thermo Fisher Scientific, Pittsburgh, PA, USA) and removing residual protein and medium with 1 initial and 2 rinse spins on Amicon Ultra Centrifugal Filters 100K MWCO (Merck, Darmstadt, Germany). EBOV was more stable in water at 21°C and had an ≈1 log 10 reduction/day in water at 27°C (Table 1; Figure 2, panel A).
The stability of EBOV in human blood was assessed by spiking blood samples from a healthy human volunteer EBOV maintained viability for a longer duration in liquid than in drying blood regardless of initial titer or environmental condition.

Conclusions
We found that EBOV can persist on surfaces common in an ETU, highlighting the need for adherence to thorough disinfection and doffing protocols when exiting the ETUs and careful handling of medical waste. In addition, EBOV maintains viability for a longer duration in liquid than in dried blood. EBOV in blood of experimentally infected NHPs persists for a similar duration as EBOV in spiked human blood. A recent study showed that blood in the body cavity of an NHP contained viable EBOV for up to 7 days after death (13). We detected viable EBOV in drying blood for up to 5 days at both environmental conditions in human and NHP blood. Therefore, dried and liquid blood from an infected person in their home or ETU should be treated as potentially infectious. The finding that EBOV remains viable in water for as long as 3 (27°C) or 6 (21°C) days at the experimental concentration warrants further investigation into the persistence of the virus in aqueous environments, such as in wastewater or sewage canals. Viable EBOV has been isolated from urine (14) but not from human stool (8). Therefore, the potential for dissemination of EBOV through wastewater remains unknown. This study is subject to several limitations. First, because standard volumes for samples were used, different volumes or matrices could influence the stability of EBOV under the tested conditions. Second, blood samples from the NHPs might have different immunologic or biochemical conditions, which can potentially influence virus stability. Third, the experimental conditions in the laboratory are sterile, but in disease-endemic areas and ETUs, bacteria or chemicals could influence EBOV viability.
Overall, we found that different environmental conditions, fluids, and surfaces influence the persistence of EBOV. These findings demonstrate that such factors are crucial in understanding transmission and improving safety practices.  conditions. A) EBOV stability in water at 2 environmental temperatures. Virus concentration was reduced at a significantly faster rate in 27°C water than in 21°C water (p = 0.0001). B) Stability in drying or liquid EBOV-spiked human blood samples at 2 environmental conditions. Virus concentration was reduced at a significantly faster rate by drying than in liquid blood at both conditions (p<0.0001 for each condition). No significant difference between reduction rates in virus titer in drying human blood at both conditions was found (p = 0.92). Triplicate samples were taken at each time point. Error bars indicate mean ± SEM virus titer. Dashed line indicates the limit of detection for the assay. An analysis of covariance equivalent test was used to compare linear regression models and determine differences in virus reduction rates. TCID 50 , 50% tissue culture infectious dose. EBOV maintained viability for a longer duration in liquid than in drying blood regardless of initial titer or environmental condition.

Conclusions
We found that EBOV can persist on surfaces common in an ETU, highlighting the need for adherence to thorough disinfection and doffing protocols when exiting the ETUs and careful handling of medical waste. In addition, EBOV maintains viability for a longer duration in liquid than in dried blood. EBOV in blood of experimentally infected NHPs persists for a similar duration as EBOV in spiked human blood. A recent study showed that blood in the body cavity of an NHP contained viable EBOV for up to 7 days after death (13). We detected viable EBOV in drying blood for up to 5 days at both environmental conditions in human and NHP blood. Therefore, dried and liquid blood from an infected person in their home or ETU should be treated as potentially infectious. The finding that EBOV remains viable in water for as long as 3 (27°C) or 6 (21°C) days at the experimental concentration warrants further investigation into the persistence of the virus in aqueous environments, such as in wastewater or sewage canals. Viable EBOV has been isolated from urine (14) but not from human stool (8). Therefore, the potential for dissemination of EBOV through wastewater remains unknown. This study is subject to several limitations. First, because standard volumes for samples were used, different volumes or matrices could influence the stability of EBOV under the tested conditions. Second, blood samples from the NHPs might have different immunologic or biochemical conditions, which can potentially influence virus stability. Third, the experimental conditions in the laboratory are sterile, but in disease-endemic areas and ETUs, bacteria or chemicals could influence EBOV viability.
Overall, we found that different environmental conditions, fluids, and surfaces influence the persistence of EBOV. These findings demonstrate that such factors are crucial in understanding transmission and improving safety practices. conditions. A) EBOV stability in water at 2 environmental temperatures. Virus concentration was reduced at a significantly faster rate in 27°C water than in 21°C water (p = 0.0001). B) Stability in drying or liquid EBOV-spiked human blood samples at 2 environmental conditions. Virus concentration was reduced at a significantly faster rate by drying than in liquid blood at both conditions (p<0.0001 for each condition). No significant difference between reduction rates in virus titer in drying human blood at both conditions was found (p = 0.92). Triplicate samples were taken at each time point. Error bars indicate mean ± SEM virus titer. Dashed line indicates the limit of detection for the assay. An analysis of covariance equivalent test was used to compare linear regression models and determine differences in virus reduction rates. TCID 50 , 50% tissue culture infectious dose.