Evaluation of steam heat as a decontamination approach for SARS-CoV-2 when applied to common transit-related materials

Abstract

Aims:

The purpose of this study was to evaluate the efficacy of steam heat for inactivation of SARS-CoV-2 when applied to materials common in mass transit installations.

Methods and results:

SARS CoV-2 (USA-WA1/2020) was resuspended in either cell culture media or synthetic saliva, inoculated (~1 × 10⁶ TCID₅₀) onto porous and nonporous materials and subjected to steam inactivation efficacy tests as either wet or dried droplets. The inoculated test materials were exposed to steam heat ranging from 70°C to 90°C. The amount of infectious SARS-CoV-2 remaining after various exposure durations ranging from 1 to 60 s was assessed. Higher steam heat application resulted in higher inactivation rates at short contact times. Steam applied at 1-inch distance (~90°C at the surface) resulted in complete inactivation for dry inoculum within 2 s of exposure (excluding two outliers of 19 test samples at the 5-s duration) and within 2–30 s of exposure for wet droplets. Increasing the distance to 2 inches (~70°C) also increased the exposure time required to achieve complete inactivation to 15 or 30 s for materials inoculated with saliva or cell culture media, respectively.

Conclusions:

Steam heat can provide high levels of decontamination (>3 log reduction) for transit-related materials contaminated with SARS-CoV-2 using a commercially available steam generator with a manageable exposure time of 2–5 s.

Introduction

The World Health Organization declared severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) a global pandemic on 11 March 2020 (World Health Organization 2020). Due to the rapid onset of the coronavirus disease (COVID-19) pandemic, industries including the public transit sector had to make operational decisions with scant evidence regarding the mode of transmission, viral shedding, infectious dose, persistence, or decontamination of this novel pathogen in the transit environment. Since its onset, research including SARS-CoV-2 has determined that aerosol transmission or direct droplet contact is the primary route of disease transmission (Chia et al.2020, Guo et al.2020, Zhang et al.2020, Greenhalgh et al. 2021); additional research has also shown SARS-CoV-2 may remain infectious on surfaces for multiple days (Chin et al. 2020, Van Doremalen et al. 2020, Richter et al. 2022). Fomite transmission, although a potential secondary route of infection, presents challenges for industries such as the transit sector where many high-touch surfaces are present.

Many public sector industries (e.g. food, health, and schools) have been adversely impacted by the pandemic and have adopted mitigation strategies such as social distancing, masking, and improved ventilation, as well as routine surface disinfection methods in order to remain in service (Ebrahim et al. 2020, Ferioli et al. 2020,Morawska et al.2020, Rader et al. 2021).To maintain critical public services such as mass transit, the transportation sector implemented rigorous surface cleaning using EPA-approved (List-N) disinfectants (U.S. Environmental Protection Agency 2022). Several decontamination technologies (i.e. vapor phase hydrogen peroxide, formaldehyde) have been demonstrated effective for inactivating biological select agents (Rogers et al. 2005, Rogers et al. 2007, Rogers and Choi 2008, Calfee and Wendling 2015, Wood et al. 2016, Richter et al. 2018), including SARS-CoV-2 (Chin et al. 2020, Kratzel et al. 2020, Raeiszadeh and Adeli 2020, Ratnesar et al. 2020). However, the logistics and safety of scaling these technologies can be challenging for industries such as mass transit due to the scale of operations (e.g. equipment and facilities), as well as the daily number of people utilizing these services. Common challenges using chemical fumigants or liquids include dangers to human health, harm to the environment, and harm to the materials being decontaminated.

Alternative approaches to chemical disinfection are of interest to these public sectors to mitigate the associated dangers. The persistence of biological organisms in the environment has been previously studied and is largely influenced by the climate and the materials with which these biological organisms are in contact (Casanova et al. 2010, Calfee and Wendling 2012, Wood et al. 2016, Rogers et al. 2016, Wood et al. 2018, Richter et al. 2019, Biryukov et al. 2020, Chin et al. 2020, Kampf et al. 2020). More recently, the use of these environmental factors (e.g. temperature and humidity) was evaluated as an alternative decontamination approach for SARS-CoV-2 contaminated items found in libraries, archives, and museums and showed that elevating temperature resulted in increased SARS-CoV-2 inactivation rates (Richter et al. 2022). Steam heat has previously been shown to reduce viability of human coronavirus OC43 (Marchesi et al. 2021), MS2 phage (Fisher et al. 2011, Li et al. 2020, Zulauf et al. 2020), methicillin-resistant Staphylococcus aureus (MRSA) (Li et al. 2020), and influenza virus (Heimbuch et al. 2011, Lore et al. 2012). Steam has also been shown to be effective within minutes against SARS-CoV-2 when applied to N95 filtering facepiece respirators (Choi et al. 2020). These studies, however, did not evaluate the effects of steam against SARS-CoV-2 when applied to porous and non-porous transit relevant materials. Steam-generating devices are regulated in the United States by U.S. EPA under the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) if the manufacturing company intends to market the device with pesticidal claims (i.e. the device inactivates viruses). If a manufacturer is making these claims about a device, then they must have scientific data to support their claims. More information about pesticide devices can be found on the EPA pesticide registration website (https://www.epa.gov/pesticide-registration/pesticide-registration-manual-chapter-13-devices). The manufacturer of the garment steamer used in this research has not made pesticidal claims about the device.

The purpose of this study was to evaluate the decontamination efficacy of steam heat against SARS-CoV-2 when applied to porous and non-porous transit-related materials at distances feasible for real-world use ranging from 1 to 2 inches, over prescribed durations. The 2-inch distance exposes a larger area of the surface to the same steam condition, which would improve throughput of the steam cleaning. Due to the complicated logistics of the public transportation sector, we focused on high-touch items such as bus seat fabric, ABS plastic, and 301 stainless steel. The application of SARS-CoV-2 onto these materials was evaluated in multiple configurations, allowing direct comparison of both wet or dry droplets, as well as inoculation vehicle (e.g. cell culture media or synthetic saliva).These considerations allow for greater comparison between laboratory conditions and real-world applied uses, e.g. in the transit environment. This study presents data on the decontamination efficacy of SARS-CoV-2 both in terms of total virion recovery and log reduction (LR) and sought to identify steam exposure parameters where the virus was inactivated to levels that measured below the limit of quantitation (LOQ) of the tissue culture assay.

Materials and methods

Test organism

SARS-CoV-2 strain USA-WA1/2020 was obtained from BEI Resources and propagated in Vero (African green monkey kidney) clone E6 cells (BEI Resources, product No.NR-596). The cells were incubated at 37°C with 5% carbon dioxide (CO₂) in minimum essential medium (MEM; Corning, Cat. No. 10010-CV) supplemented with 10% fetal bovine serum (FBS; Gibco, Cat. No. 10082147) and 1% penicillin-streptomycin (PS; Gibco, Cat. No. 15140122) until ~90% cell confluency was achieved. Viral propagation was performed using a roller bottle method (Glasbrenner et al. 2021) using 2 ml of SARSCoV-2 stock at a multiplicity of infection (MOI) of 0.001, along with 5 ml of inoculation media and allowed to infect for 1 h at 37°C with 5% CO₂. After infection, 25 ml of complete MEM (2% FBS, PS) was added, and incubation continued for 36–48 h at 37°C with 5% CO₂ using a bottle roller (Thermo Scientific, Cat No. 11-676-343) at 5 revolutions per minute (rpm). Once cytopathic effect (CPE) was observed throughout the flask (1700 cm² expanded surface polystyrene roller bottle, Corning, Cat.No.43082), cells were removed by trypsinization (0.25% Trypsin-EDTA, Gibco, Cat. No. 25200-056). Harvested cells were vortexed for 2 minutes at maximum speed with a ratio of 1:7 sterile glass beads (Sigma-Aldrich, Cat. No. CLS72685) to cells and then centrifuged at 800 × g for 5 minutes at 4°C to remove any remaining cellular debris. The resulting supernatant was frozen at −80°C in single-use vials.

Test materials

Decontamination testing was conducted using common public transit relevant materials: 301 stainless steel (McMaster-Carr, 90415K664), ABS plastic (McMaster-Carr, 8586K551), and bus seat fabric (American Seating Company,00333UW1). Test coupons were cut to uniform length and width (1.9 cm × 7.5 cm) from a larger piece of material stock and prepared for testing by sterilization via electron beam (E-beam) irradiation at ~200 kilograys (kGy, E-beam Services Inc.). E-beam irradiated material coupons were sealed in 6 mil (0.006 inch) poly tubing (Uline, Cat. No. S-2940) to preserve sterility until the coupons were ready for use. Visual inspections of the physical integrity of the test coupons were performed prior to and after testing to assess any damage or change to the coupons.

Sample processing and data collection

All work with SARS-CoV-2 was conducted in a Biosafety Level 3 (BSL-3) laboratory. The test organism was evaluated in either an inoculation media matrix (MEM + 5% FBS + 1% PenStrep) or a simulated saliva matrix prepared according to ASTM E2721 using porcine as the mucin source (Heimbuch et al. 2011). The test organism in the inoculation media matrix was prepared by thawing the virus stock material at ambient temperature. Test organism in synthetic saliva was prepared by concentrating virus stock material in a 100 K molecular-weight cutoff (MWCO) protein concentrator (Thermo Scientific, Cat. No. PI88533) to ~0.2 ml, then adding ~5 ml of synthetic saliva.

Coupons were laid flat in a Class II biological safety cabinet (BSC) and inoculated with ~1 × 10⁵ median tissue culture infectious dose (TCID50) per coupon. A 100-microliter (μl) aliquot of test organism was dispensed as 10 droplets across the surface of the test coupons. For each type of material, three coupons were used to assess decontamination performance at each combination of testing conditions and timepoint tested. Three control coupons were also used to assess any loss of viability due to ambient desiccation. Material coupons were either allowed to dry for ~1 h in the BSC under ambient conditions [~22°C and 40% relative humidity (RH)] before testing or inoculated and tested immediately. Additionally, one coupon of each material and coupon type was used as a negative control (not inoculated) and included for each time point tested. The blank coupons controlled for potential cross-contamination during testing as well as for sterility and potential cytotoxic effects from the test coupons.

Test coupons were exposed to various combinations of steam application parameters as described in Table 1. Following inoculation and drying (if applicable), coupons were arranged flat on a custom metal rack in a Bio Safety Cabinet II (BSC II) which was able to provide an even distribution of steam heat (89.9°C ± 2.6°C and 70.0°C ± 3.9°C at 1- and 2-inch exposure distances, respectively) across the entire surface of the test material (N = 1 per exposure) as shown in Figs S1 and S2 in the supplemental information. A commercially available garment steamer (Jiffy Steamer, Model No. J-4000) was used to apply steam at the specified distance and contact time. At the specified timepoints, test, control, and blank coupons were collected, placed in 50-ml conical tubes, and extracted with 10 ml inoculation media by agitation on a platform shaker at 200 rotations per minute for 15 minutes. Extracts were then transferred to the 100 K MWCO protein concentrator and centrifuged at 3000 × g for ten minutes to ~0.5 ml. Bus seat fabric samples received 10 ml Hank’s Balanced Salt Solution (Thermo Fisher, Cat.No.88284) and were centrifuged again to ~0.5 ml to wash the concentrated extract of any residual chemicals from the material that may be cytotoxic to cell monolayers. Fresh inoculation media was added to all concentrated extracts to bring the final sample volumes to 2 ml, then all samples were filtered through a syringe filter (Corning, Cat. No. 43221) to remove any test material particles introduced during the extraction process. Approximately 0.2 ml from each washed and filtered extract sample was taken for RT-qPCR analysis and remaining volume was frozen at −80°C until use.

Table 1.

Overview of experimental design.

Test matrices	Distance (in)	Exposure duration (s)
	1	60
	1	30
	1	5
Inoculation media (wet)	1	2
Inoculation media (dry)	1	1
Synthetic saliva (wet)	2	60
Synthetic saliva (dry)	2	30
	2	15
	2	5
	1.5	15
	1.5	5

Test sample extracts were assayed for SARS-CoV-2 viability using a median TCID₅₀ assay in 96-well plates containing Vero E6 cells. Serial dilutions (5-fold) were completed in inoculation media and plated onto 80%–90% confluent Vero E6 monolayers, followed by evaluations for CPE 72–120 h postinfection. Quantification of infectious SARS-CoV-2 was determined via the Spearman–Karber method (Hamilton et al. 1977). The LOQ for the SARS-CoV-2 TCID₅₀ assay was 13.1 TCID₅₀ ml⁻¹ (1.12 log₁₀ TCID₅₀). Once below this threshold, the assay could no longer assign a quantitative value output; however, a qualitative assessment of the presence of infection could be observed through manual microscopic examination. Therefore, any values below LOQ, but positive for presence of virus, are assigned an arbitrary value of 10 TCID₅₀ ml⁻¹ (indicating positive) to allow it to be resolved from 0 (indicating negative) presence of viral infection in the host cells. An average is calculated for the values assigned to the five test coupons for each material per timepoint.

Polymerase chain reaction analysis

Samples were removed from frozen storage, thawed, and RNA was extracted using the Indispin Pathogen 96 QIAcube HT kit (Indical, Cat. No. SP54161) on the QIAcube HT system (QIAGEN Inc.). Approximately 0.2 ml of reserved liquid volume was used per sample. In each batch of samples extracted, an Isolation Positive (IP) containing the N1 gene target and an Isolation Negative (IN) not containing the N1 gene target were included as procedural controls and treated as all other test samples. A custom “cador Pathogen 96” QIAcube HT protocol was used to set up the QIAcube Prep Manager software, provided by QIAGEN to include off-board lysis and DNase digestion. Purified RNA extracts were stored at −80°C until RT-qPCR analysis.

The RT-qPCR analysis was completed using a Custom TaqMan Gene Expression Assay (Thermo Fisher, Cat. No. 4332079), a concentrated primer-probe mixture containing forward and reverse primers and a fluorescein amidite (FAM)-labeled probe using sequences specific to the SARS-CoV-2 nucleocapsid protein (N1) gene target. A master mix was prepared using the primer-probe reagent, TaqMan Fast Virus 1-Step Master Mix (Thermo Fisher, Cat. No. 4444436), and nuclease-free water (Thermo Fisher, Cat. No. AM9938). The master mix was loaded to a PCR 96-well plate (15 μl) and a no template control comprised of nuclease-free water was added in duplicate, then immediately capped. The RNA test samples (including the IN and IP) were loaded in duplicate and capped, then lastly, the reference standard (RS) was loaded in triplicate and capped. All samples were added using a 5-μl volume. The qualified RS was plated as an 8-point logarithmic dilution series ranging from 5.0 × 10⁰ to 5.0 × 10⁷ copies/reaction, generated with a custom synthetic RNA material (Biosynthesis) containing the N1 amplicon targeted by the primers and probe.

The RT-qPCR reaction was completed on a QuantStudio 6 Flex Real-Time PCR System (Thermo Fisher) using a fast run set to hold for 5 minutes at 50°C and 20 s at 95°C, followed by 40 cycles of 3 s at 95°C and 30 s at 60°C. Amplification of the N1 gene target was captured by the system at the end of each cycle by reading the level of fluorescence generated by the FAM-labeled probe, and the response was plotted over time to generate an amplification plot. The cycle threshold cutoff was set to 0.1 with an automatic baseline, and unknown test sample cycle threshold (C_T) values were generated in the QuantStudio software based on the cycle at which their amplification curve crossed the predetermined threshold setting. The C_T values were plotted against the RS curve to generate a total copy/reaction value for each well.

Results

Multiple tests were conducted to assess decontamination efficacy of steam on three test material surfaces. The mean amount of virus applied to each material across all tests was 5.46 ± 0.51 log₁₀ TCID₅₀ in media and 5.11 ± 0.70 log₁₀ TCID₅₀ in saliva (inoculum controls). The mean recovery of SARS-CoV-2 on immediately tested samples with wet droplets was 5.05 ± 0.66 log₁₀ TCID₅₀ in the media matrix and 4.50 ± 0.55 log₁₀ TCID₅₀ in the simulated saliva matrix. The mean recovery of SARS-CoV-2 samples after a 1-h dry time was 4.60 ± 0.67 log₁₀ TCID₅₀ for the media matrix and 3.78 ± 0.73 log₁₀ TCID₅₀ for the saliva matrix, resulting in an average LR resulting from natural attenuation of 0.86 and 1.33, respectively, when dried at ambient environmental conditions.

Complete decontamination (no recovery of infectious virus) was achieved when steam was applied to test coupons for 30 s at both distances tested (1 and 2 inch) and 60 s at a 1-inch distance for all materials, inoculum applications, and matrix types except for wet media inoculum on ABS plastic and stainless steel after 60 s but not after 30 s (Figs 1 and 2). Complete decontamination was also achieved in all but one sample type at a distance of 1.5 inches with a 15-s application as shown in Fig. 3.

Figure 1.

Efficacy data for 1-inch exposure distance. Light gray = pre-decon recovery, dark gray = post-decon recovery, and error bars = LR 95% CI; only positive values shown.

Figure 2.

Efficacy data for 2-inch exposure distance. Light gray = pre-decon recovery, dark gray = post-decon recovery, and error bars = LR 95% CI; only positive values shown.

Figure 3.

Efficacy data for 1.5-inch exposure distance. Light gray = pre-decon recovery, dark gray = post-decon recovery, and error bars = LR 95% CI; only positive values shown.

As steam application distance increased or application time decreased, decontamination efficacy was reduced. When application distance and time were decreased to 1 inch and 1 s, complete decontamination was only achieved on ABS plastic for both media matrices when dried, Fig. 1. For both matrices with a wet inoculum application, LR was smallest for ABS plastic, suggesting that the decontamination parameters are more effective on plastic when virus is dry. At these conditions, media and saliva matrices demonstrated similar LR patterns for each material and inoculum application. For the wet inoculum application, stainless steel had the highest LR (3.49 log₁₀ for media and 4.02 log₁₀ for saliva) for both matrix types.

At an application distance of 2 inches and a time of 5 s, infectious virus was recovered on all samples tested, Fig. 2. In both wet and dry media matrices, the largest LR was observed in seat fabric material (0.78 log₁₀ and 1.12 log₁₀, respectively), with minimal reduction for ABS plastic and stainless-steel materials. Higher decontamination efficacy was observed for the saliva matrix (2.21 log₁₀ average reduction for all three materials when inoculum was dried and 2.25 log₁₀ average reduction for all materials with a wet inoculum), indicating SARS-CoV-2 is more readily inactivated in synthetic saliva at these steam application parameters. Similar LRs were observed in ABS plastic and stainless steel for both wet and dry saliva. When application time was increased to 15 s (maintaining a 2-inch distance), complete decontamination was achieved on ABS plastic with a dry saliva inoculum and both seat fabric and stainless steel with a wet saliva inoculum. Decontamination efficacy increased 1.50–3.85 logs in samples where infectious virus was recovered in comparison to the 5-s application.

Complete decontamination was not achieved in any sample or matrix type at an application time of 5 s and distance of 1.5 inches, Fig. 3. Similar LR patterns were observed in wet applications of both matrices, with highest reduction in stainless steel and lowest reduction in seat fabric. Decontamination efficacy was higher for wet matrices than dry, with the exception of seat fabric with a dry media inoculum which achieved an LR of 2.48 log₁₀. At this application distance, decontamination efficacy increased when application time was extended to 15 s and complete reduction of infectious virus was achieved in all but one sample type (ABS plastic with a dry media inoculum).

A reverse transcriptase polymerase chain reaction or RT-PCR analysis was performed on a subset of samples to determine if decontamination also resulted in degradation of genetic material. Samples selected for analysis were bus seat fabric material with a dry saliva inoculum at a 1-inch decontamination application distance, Fig. 4. The average recovery for control samples was 5.32 ± 0.12 log₁₀ gene copies per test sample. For exposure durations of 1, 2, and 5 s, average gene copies for samples exposed to steam decontamination decreased as exposure time increased (reductions in gene copies of 0.89, 1.32, and 1.54 log₁₀ copies, respectively). Exposure durations of 30 and 60 s resulted in the largest reductions in gene copies, 3.05 and 1.84 log₁₀ copies, respectively.

Figure 4.

SARS-CoV-2 genomic recovery by RT-PCR with and without exposure to steam heat on bus seat fabric. Dark gray = pre-decon recovery, light gray = post-decon recovery, and error bars = 95% CI; only positive values shown.

Discussion

When evaluating decontamination strategies for large public service sectors such as transit authorities, the number of daily users as well as items used within these organizations may benefit from the use of non-chemical decontamination strategies to address potential surface contamination with SARS-CoV-2 virus. While fomite contamination is not considered the primary route of exposure for SARS-CoV-2, the nature of mass transit results in routine contact with high-touch surfaces, increasing the potential for community transmission (Pitol and Julian 2021). The use of environmental conditions (temperature and humidity) to decontaminate other microorganisms has been previously studied and offers advantages of safe deployment but requires lengthy contact times which limits application in a high turnover mass transit environment. Steam heat, on the other hand, has been shown to be effective against a range of organisms, leaves no residual hazardous chemicals, and requires much shorter exposure durations.

This study was designed to approximate both real-world conditions and at the same time provide information using systematic, highly controlled laboratory methods.SARS-CoV-2 was resuspended in a synthetic saliva formulation and used to replicate direct droplet inoculation that could result from a cough or sneeze (Wilson et al. 2021); while at the same time, another set of coupons was resuspended in cell culture media to provide for comparisons of other research using this method. The amount of virus applied to each test material was intended to represent a worst-case contamination scenario of direct droplet exposure (~5 log₁₀ TCID₅₀). The use of this high-titer virus for research purposes was recently corroborated when clinicians found high viral loads (>7 log₁₀ ml⁻¹) of the SARS-CoV-2 Delta variant in the nasopharyngeal swabs of vaccinated health care workers (Chau et al. 2021).

The results of this study have shown that the SARS-CoV-2 virus can be quickly and thoroughly inactivated using steam heat generated by a commercially available steam wand in as little as 2 s of exposure at a distance of 1 inch on dry droplets (average temp 90.7°C ± 2.03°C) to levels below the LOQ of the assay. Figures 5 and 6 visually summarize the measured SARS-CoV-2 recoveries as function of exposure time and distance for dry and wet inoculum, respectively. Complete inactivation of SARS-CoV-2 was somewhat material dependent with a complete inactivation more readily observed for the ABS plastic than the stainless steel and seat fabric, especially for the dry inoculum. This material effect may be attributed to differences in thermal response of the material to the steam with an expected lower thermal mass for ABS plastic than stainless steel and seat fabric.

Figure 5.

Dry inoculum log₁₀ TCID₅₀ recovery heatmap. Thick border boxes = shortest duration to achieve complete inactivation.

Figure 6.

Wet inoculum log₁₀ TCID₅₀ recovery heatmap. Thick border boxes = shortest duration to achieve complete inactivation.

Direct comparisons against other steam inactivation studies can be made to those by Marchesi et al. (2021) who used a dry steam unit that included a pressurized heating system with a 100°C–110°C steam temperature at the exit of the nozzle. In that study, full inactivation and an LR value <4 was reported for human coronavirus HCoV-OC43 after 4 s which is similar to the findings of this study for the SARS-CoV-2 virus on similar nonporous surfaces. Longer inactivation times of 20–30 minutes for SARS-CoV-2 were reported by Choi et al. using multicookers to apply moist heat at 65°C to N95 facepieces. Similarly, MS2 inactivation by steam as reported by Fisher et al. (2011) and Li et al. (2020) was complete after 45 s and 5 minutes as their shortest exposure times, respectively. These two MS2 inactivation studies did not provide data on shorter exposure times. The use of a handheld steamer allowed for use on objects that cannot be placed in a controlled steam environment (microwave and pressure cooker). Handheld steamers are also relatively low-cost and readily available. Steam also interacted favorably with a virus that resides on porous materials like the seat fabric, which are otherwise difficult to clean materials by disinfection liquids or UVC (Hessling et al. 2020).

There are two notable exceptions to the findings of this study as observed for the 1 inch, 5-s exposure for media (bus seat fabric) and saliva (ABS plastic), which resulted in low level ~1-log recovery. These are considered outliers since these same materials tested at 2 s of exposure were completely inactivated achieving average LR value of 4.4 ± 0.46 for all materials and inoculation vehicles. At this same distance, the wet droplets required longer contact times to achieve complete inactivation, ranging from 2 to 30 s. Increasing the exposure distance to 1.5″ (avg temp 71.3°C ± 1.40°C) resulted in increased exposure time required of 15 s with one exception on ABS plastic which was below the lower LOQ for the assay and resulted in 4.76 ± 0.85 LR. Increasing the distance further, to 2″ (avg temp 70.6°C ± 1.49°C) resulted in additional exposure time required to achieve complete inactivation of 30 s with two outliers at 60 s for wet media on ABS plastic and stainless steel (in both cases just one out of three replicate coupons held viable virus). These materials resulted in complete inactivation when tested at the shorter 30 s exposure duration and resulted in mean LR values of 3.9 ± 1.4 and 3.8 ± 1.1, respectively. When comparing data where complete inactivation was not achieved, it appears that dry inoculum was more readily decontaminated as compared to the wet inoculum. Additionally, there was no obvious connection between LR values and inoculation vehicle (media or saliva) or material type (porous vs. non-porous). This study showed that the SARS-CoV-2 virus RNA was detected using RT-PCR analysis for both control as well as test samples that had been completely inactivated. While longer exposure to steam heat resulted in a reduction of recoverable genetic material, for decontamination test samples, these results highlight that PCR analysis is not an ideal measure for potential risk of infection from fomite or site contamination, as the PCR assay does not distinguish between viable or inactivated virus. A limited statistical analysis of the RT-PCR results indicated that the number of gene copies after 30 s was not significantly different (standard t-test; P-value 0.12) from those after 60 s despite the apparent lower recovery after 30 s (Fig. 4).

These results provide much-needed scientific information that can be used to inform how transit authorities manage equipment and operations during the pandemic. Of note, this study suggests that low-cost procedures, including routine surface steam treatment, can be implemented to help reduce the risk of spreading SARS-CoV-2 via transit-related fomites. This study was conducted using clean materials. Presence of dirt and/or grime on surfaces may alter the outcomes of a steam application. Also, the porosity of a material may lead to the transport of a virus deeper into a porous material, which would make it more difficult to inactivate.

Significance and impact of the study.

These results provide information on the performance of steam heat against SARS-CoV-2 on surfaces. These data may aid response decision-makers and industry who manage surface decontamination as it relates to the COVID-19 pandemic.

Data availability

All data are publicly available at data.gov, DOI: 10.23719/1527898.