Air and surface measurements of SARS-CoV-2 inside a bus during normal operation

Transmission pathways of SARS-CoV-2 are aerosol, droplet and touching infected material. The diffusion of the virus contagion among people is easier in indoor location, but direct detection of SARS-CoV-2 in air or on surfaces is quite sparse, especially regarding public transport, while it would be important to know how and if it is safe to use them. To answer these questions we analysed the air and the surfaces most usually touched by passengers inside a city bus during normal operation, in order to understand the possible spreading of the virus and the effectiveness of the protective measures. The measurements were carried out across the last week of the lockdown and the first week when, gradually, all the travel restrictions were removed. The air and surface samples were analysed with the RT-PCR for the detection of SARS-CoV-2 virus. After two weeks of measurements and more than 1100 passenger travelling on the bus the virus was never detected both on surfaces and on air, suggesting that the precautions adopted on public transportation are effective in reducing the COVID-19 spreading.


Introduction
The diffusion of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) affected over 100 countries in a matter of weeks, therefore on January 30th the World Health Organization (WHO) declared the COVID-19 epidemic a Public Health Emergency of International Concern [1]. Italy has been one of the most affected countries with more than 230000 infected people and more that 33000 deaths and was the first country in Europe to proceed with a total lockdown (so called phase 1, started on 9 March 2020). The government decided to impose strong restrictions in the whole Country closing schools, public places (such as restaurants or cafés) and shops, allowing only the basic necessity stores (such as supermarkets and pharmacies) and relative activities to remain open [2]. The huge increase of infected people resulted, on 13 March, more strict measures including transport rationalization, with a strong reduction of public transport, maintaining only a minimum level of services [3]. A protocol between the Italian Ministry of Infrastructure and Transport and the Italian Ministry of Health, together with trade organizations and trade unions representatives, established anticontagion rules and actions and promoted cleaning and disinfection procedures for public transport with the aim to contain the COVID-19 spreading and to ensure the safety of workers and travellers in the transport and logistics sectors [4]. One of the main measures recommended was the recurrent cleaning and disinfection of frequently touched surfaces such as handles and rails because of the potential environmental stability of SARS-CoV-2 that, according to some reports, could span from up to three hours, in the post-aerosolisation air, to about 24 hours on cardboard and about three days on plastic and stainless steel [5]. Recent studies have shown the possible airborne transmission of the virus in public places by asymptomatic people [6,7]. In addition, the research findings suggest reducing the number of people in the same ambient and carry out control actions to limit the pandemic expansion [6]. For buses and trains sanitation were recommended virucidal licensed products, containing sodium hypochlorite, or those containing ethanol (at least 70%), after cleaning with a neutral detergent [8].
In Italy, the end of the lockdown was planned to finish gradually, with different dates for the reduction of constrains as function of the infection risks of the different activities. Starting on 18 May 2020, phase 2 began with a Decree of the President of the Council of Ministers [9] that included guidelines for public transport, establishing general rules such as: 1) reduction of the passengers' number inside the buses, 2) interpersonal distance of one meter, 3) rear door boarding in order to protect drivers, 4) only distanced seats permitted, 5) passengers must frequently sanitize hands and 6) obligation to wear facial masks [9]. The local governments, following the national guidelines of the DPCM [9], established the exact operational rules for the local public transportation system. In details, the Abruzzo Region (Central Italy), where this study was carried out, defined that: 1) the maximum number of passengers on-board buses must not exceed 40% of the total seats and 15% of standing places, if provided; 2) standing places must be marked with a signal on the ground 3) by 18 May 2020 at least 50% of the services performed before the reduction due to COVID-19 is reactivated, reaching the 70% within and not beyond 31 May 2020 [10].
The analysis of the air and surfaces in indoor environments are crucial to better understand the SARS-CoV-2 spreading and airborne transmission, to improve the assessment of the risks for doctors and health-care operators [6]. However, these observations are very limited and mostly confined to hospital environments [11,12]. Results of risk models assessing the airborne transmission of the virus in different indoor environments such as restaurants, post offices, pharmacies, supermarkets and banks, suggest the key role of air ventilation, but simulations of more confined environments like city buses, trolleybuses, trams or trains, are missing [13]. For this reason, we analysed both the air and samples taken from the surfaces most frequently touched of a city bus.

Materials and methods
The study was conducted from 12 to 22 May 2020 in Chieti, a town in the Abruzzo region that is the fifth Italian region for mortality due to COVID-19, with an infection fatality rate (deaths / cases) of 12.1% [14]. In Abruzzo, as of 28 May 2020, are reported 3237 cases of infected people, 820 in the province of Chieti, which is 0.213% of the total population [15]. In the present study, the environment inside the line number 1 trolleybus (local transportation system) was monitored. This line is the most important of the town in terms of number of passengers and since it covers a route of 20 km with 50 stops from downtown to the University Campus and the Santa Annunziata Hospital and back.
Samples of air inside the bus were taken every day of the two observational weeks, excluding the weekends, during the line 1 shift (5 routes) that started at 12.00 and finished at 18:30. The bus was operated following the rules established in the DPCM [9] and the DPRC [10]. Two microbiological gelatine membrane sample filters of 80 mm diameter were installed on board: one close to the ticket machine, the other on the rear part of the bus (Fig 1). These filters are the proper support for the detection of SARS-CoV-2 virus, to be analysed with the RT-PCR [11]. These microbiological gelatine membrane filters were successfully tested, at the Clinic of Infectious Diseases of the S.S. Annunziata Hospital in Chieti, to check their performance in detecting the virus in air [16].
A constant flow sampler (AMS Analitica model HE BASIC PLUS), fed by the trolleybus electrical power supply, ensured an air flow of 24 l/min to each filter. All the air samples were gathered during the 6.5 hours daily operation of the bus, therefore the total volume sampled was 18.72 m 3 . One air sample, as a control reference, was taken without passengers overnight, for 21 hours, with the bus in the hangar.
Surface samples were carried out with swabs, wet with physiological solution, on five points of the bus, that are those more frequently touched by the passengers, according to the experience of the bus drivers. The sides of the ticket machine and four stop buttons are the sample points selected. The surface sampled covered all the surfaces of the stop button (about 30 cm 2 ) and about 40 cm 2 for each side of the ticked machine (Fig 1). The samples for each observation day and each surface were taken before the beginning of the bus shift, to have a reference, and immediately after the end of the shift. Cleaning and sanitation inside the bus are carried out daily with a generic detergent followed by a sanitation with sodium hypochlorite 0.1% and ethanol 70%. The sanitation was carried out after the collection of the surface samples. Once a week the bus is further sanitised, after the routine cleaning and sanitation describe above, with an electric aerosol applicator that delivers, for 1 minute, highly oxidizing, non-foaming acid disinfectant, containing 56 g/l of peracetic acid, 12 g/l of hydrogen peroxide and 56 g/l of acetic acid. Moreover, once a week, the entire trolleybus cabin is ozonized for 10 minutes.
Surface and air samples were collected and delivered, immediately after gathering them, to the Microbiology and Molecular Genetics laboratory of the Center for Advanced Studies and Technology (CAST), University "G. d'Annunzio" of Chieti-Pescara to be analyzed through RT-PCR technique. The collected samples (wet swabs and microbiological gelatine membrane) were inserted on 2 cc of physiologic solution and transported to the laboratory. On arrival at the research lab, specific real-time reverse transcriptase-polymerase chain reaction (RT-PCR) (TaqMan™ 2019-nCoV Assay Kit v2; Thermo Fisher Scientific, Italy) targeting RNA-dependent RNA polymerase was used to detect the presence of SARS-CoV-2 [17,18]. This technique uses 3 genes: ORF1ab, N gene and S gene to quantify the viral load with a number of cycles for the fluorescent signal to cross the threshold in RT-PCR. The threshold is 5000, baseline is 5 and cut-off is 37 cycles. Lower values of the number of cycles means higher viral load. According to the TaqMan 2019-nCoV Assay Kit v2 [19], samples are considered "Positive" when at least two genes have a cycle threshold value < 37; if cycle threshold value is 'Undetermined' or >37 for two or all the genes, then the result of the sample is "Negative".

Results and discussion
During the whole observation period about 1100 passengers travelled on the trolleybus set up for the observations reported here, with an average of 123 people for each measurement shift as shown in details in Table 1. All the surfaces samples were negative for two or all the genes to RT-PCR analysis with no technical replicates for SARS-CoV-2 virus ('undetermined' or >37). Similarly, the same results were obtained for all the air samples during the whole study period, including the overnight air sample taken without passengers with the bus in the hangar. Moreover, in all the samples (surface, air) was never detected a prevalence of a single gene that

Total passengers 1107
The air samples were carried out on two sample sites (F1 and F2, see could be considered as a variant, nor was observed a gene difference between samples collected before the bus shift and those collected after. These results mean that none of the samples, both on the bus surfaces and indoor air, resulted 'Positive' to the SARS-CoV-2 virus. Unfortunately, we could not test each passenger for SARS-CoV-2, therefore we do not know exactly how many infected people travelled on the trolleybus during our observations. A recent work, based on the analyses of data from different parts of the world (China, Italy, US, Greece) and diverse situations, suggests that the asymptomatic people infected by SARS-CoV-2 are between 40% and 45% of the population [20]. Considering a conservative estimation of 30% asymptomatic people infected, since 123 passengers travelled on average on each bus shift, we estimated that about 37 infected and asymptomatic people potentially touched the surfaces that we sampled at the end of the journeys and breathed inside the bus while our instrument was sampling the indoor air. Under this hypothesis we can argue that the requirements of cleaning up hands, using a dispenser of alcoholbased sanitizer at the bus entrance door, seem to keep the surfaces and the air inside the bus safe and free from the virus. The hands sanitizing was a strict requirement to get on the bus, requested by the DPCM guidelines [9], at the same time the bus company requested also to wear gloves, as a further precaution. The rule of wearing a facial mask during the travel, and the recommendation to keep the windows open to allow high air ventilation, probably prevent the virus diffusion in the air inside the bus. These results are in agreement with different model simulations that recommend facial masks to combat the virus spread in aerosols and droplets by asymptomatic people [21]. Moreover, the air ventilation, that model simulations showed to be important to reduce the risk of virus transmission in different indoor environments [13], is confirmed to be essential also in a more confined location like inside a bus.

Conclusion
The end of the lockdown, imposed to contain the COVID-19 infection outbreak, is entailing a growing number of people that restart the usual daily activities including travelling on public transport. Our observations inside a bus showed that the air and all the surfaces samples were not infected by SARS-CoV-2 virus. Even if it was not possible to test the passengers for the virus but considering that the asymptomatic people infected could be more than 30% [20], we can expect potential infections inside the bus. Whether or not the number of infected passengers was about 30%, our findings confirm that the measures established for public transport in terms of sanitation, air ventilation and interpersonal precautions (facial mask, distancing, hands hygienisation) are effective, at least during this study, to keep healthy and COVID-free the environment inside the buses.