1. Introduction

Infectious animal viral diseases are a persistent problem affecting the livestockindustry throughout the world. This applies especially to those airborne diseases that are extremely difficult to manage through common biosecuritymeasures. Airborne diseases are transmitted by aerosols, which are liquid or solid particles suspended in the air and that act as a carrier for the infectious pathogens enabling them to move from infected animals to susceptible animals. Pathogens that are carried on aerosols are therefore called airborne pathogens. Once released from infected animals or facilities, airborne pathogens can remain airborne or settle on surfaces (fomites) due to gravity (Arruda et al., 2019). Aerosol transmission occurs when airborne pathogens are inhaled by a susceptible animal, resulting in an infection. Fomite transmission can occur when airborne pathogens cause infection after settling onto surfaces and gain entry into a susceptible animal through its interaction with the fomite. However, it can be difficult in animal settings to differentiate between infections caused by direct inhalation and indirect contact via fomites thus this paper refers to both transmission routes as airborne transmission, as suggested by Arruda et al. (2019).

Numerous devastating outbreaks of viral animal diseases have occurred in the past. For example, outbreaks of foot and mouth disease (FMD) in the UK in 2001 resulted in the loss of six million animals (Gloster et al., 2003). In Canada, outbreaks of highly pathogenic avian influenza (AI) affected 42 commercial poultry facilities (17 million animals; Bowes, 2007) in 2004 and 11 commercial poultry facilities (240,000 animals; Xu et al., 2016) in 2014. Porcine epidemic diarrhoea (PED) was introduced to the USA in 2014 and had caused infections in more than 3750 facilities by the end of 2017 (Niederwerder & Hesse, 2018). In 2018, the first case of African swine fever (ASF) in China was reported (Zhou et al., 2018), and it has been reported since in numerous Asian countries, such as Mongolia, Vietnam, Cambodia, North Korea, Laos, Philippines, Myanmar, Timor-Leste, South Korea (Mason-D'Croz et al., 2020), and Indonesia in 2019, as well as Papua New Guinea and India in 2020 (Woonwong, Tien, & Thanawongnuwech, 2020). Porcine reproductive and respiratory syndrome (PRRS) outbreaks in Canada and the USA occur frequently, with a 2011 study of its economic impacts on the USA pig industry estimating that the annual productivity losses in breeding and growing herds was 664 million US$ (Holtkamp et al., 2013). Because of their current relevance, and their ability to cause considerable detrimental damage to the livestock industry, these five animal diseases caused by viral pathogens were chosen to be the subject of this review.

The role and mechanisms of airborne transmission in spreading infectious viral diseases in the livestock industry are not well understood, and the level of scientific information available about their airborne transmission differs substantially for each specific virus. Arruda et al. (2019) recently performed a detailed review describing the research community's knowledge on airborne transmission of porcine reproductive and respiratory syndrome virus (PRRSV), as well as the knowledge gaps for airborne transmission of PRRSV. Arruda et al. (2019) differentiated studies as experimental, semi-experimental, and field studies and determined if the study results suggested airborne transmission of PRRSV. They reported that more research on airborne transmission of PRRSV is required, including genetic sequencing of PRRSV strains; field research of airborne transmission of European PRRSV; field research in more diverse environmental conditions; improvement in air sampling methodology and technology; air sampling of PRRSV under field conditions; and contribution of the deposition of aerosols towards environmental contamination. Thus, although it is one of the most researched infectious viral animal diseases in the context of airborne transmission, their review indicated that there was still more to learn about airborne transmission of PRRSV. It is inevitable that research gaps also exist for other infectious viral animal disease pathogens, such as African swine fever virus (ASFV), porcine epidemic diarrhoea virus (PEDV), avian influenza virus (AIV), and foot and mouth disease virus (FMDV). Additionally, Arruda et al. (2019) carried out their review from the perspective of an animal scientist. Understanding of aerosol transmission could be improved by analysing transmission from the perspective of an engineer. Specifically, the Arruda et al. review did not discuss the potential role of computer modelling in improving knowledge on aerosol transmission of PRRSV. A critical review of the knowledge of airborne transmission of important infectious animal viruses from an engineer's perspective will help researchers and producers to understand commonalities, and identify overall knowledge gaps in airborne transmission of viruses, as well as those gaps specific to particular animal viruses.

The objectives of this review were to: (1) review experimental studies of short range airborne transmission of infectious animal viruses; (2) review field research of long range airborne transmission of infectious animal viruses; (3) review modelling strategies that have been used to simulate or estimate airborne transmission in past outbreaks of infectious viral animal disease; (4) identify important factors that affect airborne transmission of infectious animal viruses; and (5) identify commonalities and knowledge gaps in understanding airborne transmission of the infectious animal viruses.

2. Focus of review

There are numerous routes of animal disease transmission: direct contact, oral, vector, indirect contact (fomite transmission), and airborne. Many diseases can be transmitted through multiple routes. In this review, we focused on four important porcine diseases with viral causative agents that have some evidence that they are airborne transmissible and also have had significant economic impact on livestock industries globally. These four diseases are: African swine fever (ASF); porcine epidemic diarrhoea (PED); porcine reproductive and respiratory syndrome (PRRS); and foot and mouth disease (FMD). Also included in this review is avian influenza (AI), which is also a significant animal disease and has been proven to be airborne transmissible. It is included in this review for reference. A brief description of the five causative agents (viruses) that cause these animal diseases is provided below.

ASFV is a large enveloped deoxyribonucleic acid (DNA) virus that belongs to the family Asfarviridae and genus Asfivirus (Beltrán-Alcrudo, Arias, Gallardo, Kramer, & Penrith, 2017; Galindo & Alonso, 2017; Pikalo, Zani, Hühr, Beer, & Blome, 2019). It has an average diameter of 200 nm (Galindo & Alonso, 2017). A general sign of ASFV is the “sudden death of pigs”, which occurs in pre-acute cases; however, death occurs 90–100% in acute cases, 30–70% of the time in subacute cases, and less than 30% in chronic cases (Beltrán-Alcrudo et al., 2017). The recent spread of ASFV into China and Vietnam significantly affected pork prices and local supply of pork (Woonwong et al., 2020).

PEDV is an enveloped, single-stranded ribonucleic acid (RNA) virus and its order, family, and genus are Nidovirales, Coronaviridae, and Alphacoronavirus, respectively (Kocherhans, Bridgen, Ackermann, & Tobler, 2001). PED is characterised by vomiting, watery diarrhoea, dehydration, and reduction in growth performance (Lin, Saif, Marthaler, & Wang, 2016). It causes high mortality in nursing pigs and low mortality in finisher and farrowing herds (Kochhar, 2017). As of February 2018, there were more than 3855 confirmed positive locations infected with PEDV in the United States (USDA, 2018).

AIV are influenza A viruses (IAV), which belong to the family Orthomyxoviridae(Webster, Bean, Gorman, Chambers, & Kawaoka, 1992). In general, IAV are single-stranded RNA viruses with a diameter of 80–120 nm and the RNA genome is divided into eight segments (Webster et al., 1992). IAV have two important envelope proteins; haemagglutinin (H), of which there are currently 18 known subtypes and neuraminidase (N), of which there are currently 11 known subtypes (Lycett, Duchatel, & Digard, 2019). IAV subtypes H1–H16 and N1–N9 are found in birds and these subtypes are considered AIV (Alexander, 2000; Lycett et al., 2019). AIV is often identified as lowly pathogenic avian influenza virus (LPAIV) or highly pathogenic avian influenza virus (HPAIV) in the literature, depending on the disease severity in the infected birds (Alexander, 2000).

PRRSV is an enveloped RNA virus with a diameter of 50–65 nm. Its order, family, and genus are Nidovirales, Arteriviridae, and Arterivirus, respectively (Cho & Dee, 2006). PRRSV causes respiratory disease in pigs of all ages, but it can cause more severe outcomes in younger pigs, such as respiratory failure and death (Egli, Thür, Liu, & Hofmann, 2001).

FMDV is a single-stranded, non-enveloped RNA virus that is 25–30 nm in diameter; it belongs to the family Picornaviridae and genus aphthovirus(Belsham, 1993; Malik et al., 2017). It has seven serotypes: A, O, C, SAT1, SAT2, SAT3, and Asia 1, but this system of organising the virus is considered obsolete (CFIA, 2013). FMDV infects cloven-hoofed animals (Donaldson & Alexandersen, 2002). Some examples of susceptible domestic animals are, but not limited to, pigs (Alexandersen & Donaldson, 2002), sheep (Gibson & Donaldson, 1986), and cattle (Donaldson, Gibson,& Oliver, 1987). FMD has a significant effect on the livestock industry globally, as it has direct and indirect costs on farmers (Rushton & Knight-Jones, 2012). As of 2011, FMD was found by Rushton & Knight-Jones (2012) to be “endemic in almost all developing countries” and farmers from these countries were likely the most affected by the increased costs associated with FMD.

3. Laboratory studies of short range transmission by aerosol

A fundamental question that often posed and does not have an easy answer is: what diseases are airborne transmissible? The contribution of airborne transmission towards spreading infectious disease is often elusive and debated in the literature. Many experimental studies of short range transmission of animal diseases have evaluated the airborne transmission of an infectious animal virus and are described in further detail in Sections 3.1 Controlled exposure of individual animals to aerosols, 3.2 Two-chamber tests for assessing aerosol transmission, 3.3 Multiple cage experiments. Short range airborne transmission here refers to transmission of disease pathogens within a single airspace (a chamber, room, or building). In studies reported in the literature, viral aerosols were generated naturally (shed) from animals that were either inoculated with the infectious virus or naturally infected, or generated artificially with aerosol generating devices filled with a stock solution of infectious virus; shedding is described in further detail in Section 6.2.

Overall, airborne transmission over short distances has been demonstrated for at least 1 strain of ASFV, PEDV, H9N2 AIV, H5N1 AIV, and PRRSV (Table 1) and several strains of FMDV (Table 2). Short range airborne transmission of AIV, PRRSV, and FMDV is well demonstrated, whilst more work is required to assess for airborne transmission of ASFV and PEDV. The different types of short-range airborne transmission are shown in Fig. 1.

1. Download : Download high-res image (340KB)
2. Download : Download full-size image

Fig. 1. Types of short range airborne transmission experiments seen in literature.

4. Field studies of long range transmission by aerosols

Long range airborne transmission occurs when virus aerosols are released from an infected source facility and cause infection in a downwind recipient facility. A diagram showing the basic process of long range airborne transmission from a source building to a downwind recipient building is shown in Fig. 2.

1. Download : Download high-res image (405KB)
2. Download : Download full-size image

Fig. 2. Basic explanation of long range airborne transmission between a source and recipient facility.

4.1. Airborne transmission from buildings with inoculated animals

These studies typically used a commercial livestock building to house inoculated animals for generating airborne infectious virus. The studies involved: (1) performing air sampling at the exhaust of a source building with inoculated animals to confirm shedding and release of viruses from the source building; (2) performing air sampling at downwind locations from the source building, demonstrating aerosol transport of the virus after exiting the infected facility; (3) monitoring and quantifying infection of susceptible recipient animals in a downwind building to demonstrate airborne transmission.

Research was conducted by a team at University of Minnesota in a mechanically-ventilated pig finisher facility that was ≈16 km away from other pig farms (Dee, Otake, & Deen, 2010; Dee, Otake, Oliveira, & Deen, 2009; Otake, Dee, Corzo, Oliveira, & Deen, 2010; Pitkin, Deen, & Dee, 2009). To summarise, the source building was populated with 300 grower-finisher pigs, of which 100 were inoculated with MN-184 at some point prior to air sample collection, and the infection in the building was sustained by repopulating with small groups of pigs (Dee et al., 2009, 2010; Pitkin et al., 2009). Dee et al. (2010) and Dee et al. (2009) also introduced mixed infection by inoculating 60 out of 300 pigs with Mycoplasma hyopneumoniae two weeks prior to PRRSV inoculation. In Otake et al. (2010), the population of 252 pigs infected with PRRSV and M. hyopneumoniae had existed for three years and two groups inoculated with either MN-1182 and MN-1262 were introduced to the building. Air sampling was performed using a liquid cyclonic collector. Air sampling occurred at the same time every day in the studies of Dee et al. (2010), Dee et al. (2009) and Pitkin (2009), but the time of day was not specified in the study of Otake et al. (2010).

Table 3 shows the range and average of infectious PRRSV concentrations measured in the exhaust air of the source building. The results indicated that PRRSV was not constantly emitted from the exhaust of a building with infected pigs, with the frequency of positive samples less than 34%. One exception was observed by Otake et al. (2010) that had a 100% positivity rate in the air samples collected. Otake et al. (2010) had a shorter sampling period (21-days) than the other studies, but also had three PRRSV variants present in the source building. These authors reported that the airborne concentration of infectious PRRSV at the exhaust varied with the sampling day, but was consistently ≥750 TCID50(50% Tissue Culture Infectious Dose) m−3 for MN-184 and, on the three days this variant was not present in the air, MN-1182 was present at levels >6.3 TCID50m−3 (Fig. 3).

Table 3. Ranges and average PRRSV concentrations of air samples collected from the exhaust of a mechanically ventilated finisher facility with 25–120 kg infected pigs.

Study	Sampling period	% Positive (# positive/# sampled)	Range of concentration (TCID50 m−3 air)a	Average concentration (TCID50 m−3 air)a
Dee et al. (2009)	50 days	34% (17/50)	22 to 1.2 × 106	1.6 × 105
Pitkin et al. (2009)	1 year	29% (55/190)	4.2 to 1.3 × 105	5.0 × 102
Otake et al. (2010)	21 days	100% (21/21)	6.3 to 2.6 × 105	3.6 × 104
Dee et al. (2010)	1st year	12% (38/324)	4.2 to 1.8 × 104	1.7 × 103
	2nd year	22% (69/312)	4.2 to 1.3 × 105	1.0 × 104

a

50% Tissue Culture Infectious Dose (TCID50) per m3 air, air samples collected with a liquid cyclonic collector (Midwest Micro-Tek), calculated from the PRRSV concentration in the solution, volume of the collection fluid (5 mL), sampling rate (400 L/min), and sampling time (30 min).

1. Download : Download high-res image (152KB)
2. Download : Download full-size image

Fig. 3. PRRSV in exhaust air from mechanically-ventilated finisher building containing infected pigs (Otake et al., 2010).

Dee et al. (2009) and Otake et al. (2010) collected air samples for PRRSV at various downwind distances from the source building (Table 4). The transport, dilution, and downwind concentration of airborne viruses, like other air pollutants, depend on environmental/meteorological conditions (ambient temperature, relative humidity, and solar irradiation), which not only affect the aerosol dispersal but also the stability/survival of PRRSV in the atmosphere (Hermann et al., 2009). Meteorological conditions that reduce the dilution of infectious PRRSV aerosols will increase the likelihood of detecting infectious PRRSV downwind of the virus source. In Dee et al. (2009) and Otake et al. (2010), the optimal conditions for airborne transmission, in terms of shedding of PRRSV and meteorological conditions, occurred, allowing for detection of infectious PRRSV at various long distances downwind of the source building (Table 4). Upwind concentrations of PRRSV were not available in these studies, but were likely negligible as the nearest (∼16 km away) pig site was negative for PRRSV at the time of the study (Dee et al., 2009).

Table 4. Downwind distance and infectious PRRSV concentration collected in air samples. Airborne concentrations were measured on different dates of the study periods.

Study	Downwind distance (direction)a	Airborne concentration of PRRSV (TCID50m−3 air)b
Dee et al. (2009)	4.7 km (NW)	1.25 × 101
	4.7 km (NW)	1.71 × 101
	4.7 km (NW)	2.17 × 102
	4.7 km (NW)	3.25 × 102
Otake et al. (2010)	2.3 km (NW)	4.17 × 102
	2.3 km (SE)	7.08 × 102
	4.6 km (SW)	1.29 × 102
	6.6 km (NW)	2.50 × 101
	9.1 km (SE)	1.75 × 101

a

NW = northwest, SE = southeast, SW = southwest.

b

50% Tissue Culture Infectious Dose (TCID50), air samples collected with a liquid cyclonic collector (Midwest Micro-Tek), calculated from the PRRSV concentration in the solution, volume of the collection fluid (5 mL), sampling rate (400 L/min), and