Relative risks of the uncontrollable (airborne) spread of FMD by different species

A.I.DONALDSON, S.ALEXANDERSEN, J.H.S0RENSEN, T.MIKKELSEN

THE 2001 epidemic of foot-and-mouth disease (FMD) in the UK has highlighted the ability of the disease to spread rapidly and widely. The highly contagious nature of FMD is a reflection of the wide range of species which are susceptible, the enormous quantities of virus liberated by infected animals, the range of excretions and secretions which can be infectious, the stability of the virus in the environment, the multiplicity of routes of infection and the very small doses of virus which can initiate infection in susceptible hosts (Sellers 1971,Thomson 1994).

The most common mechanism by which FMD is spread is the movement of infected animals and subsequent direct transmission of virus to susceptible animals in exhaled droplets and droplet nuclei. This is an especially common means of spread among ruminant species and from pigs to ruminants. The next most common mechanism is by the movement of contaminated animal products such as meat, offal, milk, and so on. Pigs, consuming infected meat or offal as swill, and calves, drinking infected milk, are the species most likely to be infected by these routes. FMD virus can also be transmitted mechanically, for example, by contaminated milking machines, by vehicles, especially those used for transporting animals, and by people (Sellers 1971,Donaldson 1987).

In the event of an outbreak of FMD, the imposition of movement restrictions around infected holdings and the application of disinfectants should prevent the spread of disease by the mechanisms mentioned above. All of those mechanisms are, therefore, controllable, provided the source of infection has been identified.

There is, however, one mechanism of spread which is not controllable by humans -airborne spread. Under certain epidemiological and climatic conditions, FMD virus can be spread by the wind (Henderson 1969,Hugh-Jones and Wright 1970,Tinline 1970,Sellers and Forman 1973,Gloster and others 1981,1982). The airborne spread of FMD virus from infected animals can be prevented if they are housed in facilities equipped with absolute air filtration, as used in a biosecure laboratory, but this is not applicable in the field where the spread of FMD virus by the wind is uncontrollable. Computer models have been developed which can simulate the airborne spread of FMD and these have been used successfully under operational conditions in the UK and Italy to analyse the risk of airborne virus dissemination and the probable direction and distance of spread (Donaldson and others 1982,Maragon and others 1994). Since the early 1990s,a series of more sophisticated, faster models have been developed which can predict the spread of FMD virus. They can be linked to other systems so that the predicted plumes can be modified according to the topographical conditions, and the output data displayed in a format which is easily analysed and interpreted. Models are available which can simulate spread over short or long distances (Sorensen and others 2000,2001).

During the 2001 FMD epidemic in the UK, the Government's Chief Scientific Adviser, Professor D. King, declared, 'There are two key elements in the strategy for controlling the foot-and-mouth disease epidemic that involve culling livestock. The top priority is to cull all animals susceptible to foot-and-mouth disease (sheep, cattle and pigs)on infected farm holdings within 24 hours. At the same time, contacts between infected premises and any other place where animals are kept are followed up. After that it is essential to eliminate disease risk on neighbouring farms which share a boundary within 48 hours. This strategy is based on advice from epidemiologists -experts in the development of infectious diseases. Several groups of epidemiologists have used computer models to predict the course of the disease' (King 2001). In the field, however, the identification of contiguous premises for the implementation of the contiguous ring-culling policy proved difficult in many instances and was challenged by farmers.

This short communication describes experimental data and simulated modelling of the airborne spread of FMD from different species and different numbers of infected animals, to assess the relative risks of uncontrollable airborne spread to susceptible animals located at various distances from the source-infected animals. The results are examined in the context of the current disease control strategies.

The authors summarised published data for the minimum infectious doses of airborne FMD virus for different species during a 24-hour exposure period (Sorensen and others 2000). The minimum dose of airborne FMD virus required to infect pigs was re-investigated experimentally because the earlier published data (Terpstra 1972) were based on the use of artificially generated aerosols and a relatively insensitive virus assay system. It was found that a dose of more than 800 TCID50 of the O1, Lausanne strain of virus was required to establish infection and cause disease in a pig (Alexandersen and others 2001). The above data were used to estimate the concentration of virus in the air within a plume which would be needed to infect different species over a 24-hour exposure period (Table 1).

TABLE 1 :Minimum doses of airborne foot-and-mouth disease virus required to infect different species during a 24-hour exposure period

Species

Minimal dose*

TCID50 $

Inhalation rate

m3/24 hours #

Threshold plume concentration to infect TCiD50/m 3

Cattle

10

150

0.07

Pigs

>800

50

>16

Sheep

10

15

0.7

*The dose to infect cattle and sheep is given as the 'minimal' dose to cause clinical disease in those species (Gibson and Donaldson 1986,Donaldson and others 1987),while the dose for pigs is given as the 50 per cent minimal infectious dose to cause clinical disease. These are not absolute quantities but represent low probabilities of infection. It is possible that doses less than a minimal dose could be infectious if a large number of cattle or sheep were exposed (Sutmoller and Vose 1997).Pigs are relatively resistant to airborne infection (S.Alexandersen I.Brotherhood, A.I.Donaldson, unpublished observations),so fractions of a minimal dose are unlikely to infect that species

$ TCID50 Bovine thyroid tissue culture 50 per cent dose end point

# Average inhalation rate for a 90 to 100 kg pig,40 to 80 kg sheep and 500 to 700 kg steer

(Tenney 1970,Alexandersen and others 2001)

Cattle, sheep and pigs all excrete airborne FMD virus over a four-to five-day period, excreting maximally in the early acute stage of the disease (Sellers and Parker 1969),but pigs are by far the most potent source of airborne virus. The amount of virus excreted varies with the virus strain (Donaldson and others 1970). Airborne virus is excreted mainly in exhaled breath and originates from the upper respiratory tract initially and later from the lower respiratory tract (Donaldson and Ferris 1980). The authors measured the maximum amounts of airborne virus excreted by pigs and sheep infected with the UKG 34/2001 isolate from the current FMD epidemic in the UK. The amounts recovered over a 24-hour period were 10 6.1 TCIDSO per pig and 10 4.3 TCID50 per sheep (S.Alexandersen, A.I.Donaldson, unpublished observations). The amount of airborne virus excreted by cattle was not determined, but previous studies with a range of different strains of FMD virus have shown that the amounts excreted by cattle and sheep are similar (Sellers and Parker 1969,Donaldson and others 1970).

Using these data and the Rimpuff atmospheric dispersion model (Mikkelsen and others 1984,1997,Sorensen and others 2000,2001), simulations were performed to determine the variation between species and the effect that different numbers of animals at the source would have on the distance that a plume could travel and still be infectious. The parameters chosen were: source species of pigs, sheep or cattle; a virus output of 1,10,100 or 1000 animals; a constant wind direction; a wind speed of 5 m/second; a high atmospheric stability; no precipitation; and a relative humidity above 55 per cent. These are the meteorological conditions which are most favourable for the airborne dispersion of FMD virus (Gloster and others1981,1982, Sorensen and others 2001)and, thus, represent the worst case scenario. The simulations did not include the effect of topography. Obstacles such as hills and mountains would cause a plume to deviate, and structures such as urban areas and forests would cause turbulence and a dilution of the particle concentration. These effects would reduce the distance of transmission (Sellers and Forman 1973). The results are shown in Table 2.Rimpuff simulations were run using as input the amount of airborne virus excreted by 1,10,100 or 1000 animals of each species in 24 hours and optimal meteorological conditions for airborne spread. The distances shown are those at which the virus concentration in a plume would be sufficient to infect each of the species shown.

The data in Table 2 show that the highest risk of airborne spread is for cattle and sheep downwind of a farm with infected pigs. One hundred affected pigs could transmit sufficient virus to infect cattle, the species most susceptible to airborne infection (Donaldson and others 1987) up to 2 km way. Cattle located further away could also be at risk since he minimal doses to infect (Table 1) are not absolute quantities but represent low probabilities of infection. Therefore, a fraction of a minimal dose could be infectious, especially if a large number of cattle were exposed (Sutmoller and Vose1997). The distances over which plumes of virus originating from cattle or sheep present a risk would be much less. The simulations predict that it would require 100 cattle or sheep to be infected at source for an infectious dose to travel a distance of 0.2 km and be sufficient to infect cattle. In the FMD situation in the UK, the level of clinical surveillance is such that it is unlikely that the number of infected cases in a cattle herd would be as high as 100.Furthermore,the emerging serological evidence suggests that infection has progressed slowly through sheep flocks during the UK epidemic and so there is a low probability that 100 sheep would be in the early acute phase of infection or clinical disease, simultaneously (R.P.Kitching, unpublished observations).

TABLE 2:Effect of species and number of animals excreting virus on the risk for different species downwind

Species excreting virus

Distance (km) downwind at which species are at risk

Cattle

Sheep

Pigs

1000 infected animals

Pigs

6

2

<0.2

Cattle

0.7

0.2

<0.1

Sheep

0.7

0.2

<0.1

100 infected animals

Pigs

2

0.4

<0.1

Cattle

0.2

<0.1

<0.1

Sheep

0.2

<0.1

<0.1

10 infected animals

Pigs

0.5

0.1

<0.1

Cattle

<0.1

<0.1

<0.1

Sheep

<0.1

<0.1

<0.1

1 infected animal

Pig

<0.1

<0.1

<0.1

Steer

<0.1

<0.1

<0.1

Sheep

<0.1

<0.1

<0.1

The slaughter and disposal of the animals on infected premises within 24 hours of reporting has been given the highest priority in the current campaign. There is experimental evidence to support this policy, especially for cases involving pigs, since they are the most dangerous emitters of airborne virus (Sellers and Parker 1969, Donaldson and others 1970). Culling pigs experimentally infected with FMD virus was shown by Sellers and others (1971) to lead to a marked reduction in the amount of virus in the atmosphere, although some airborne virus persisted for at least 24 hours. They recommended that to minimise the risk of airborne spread, affected pigs should be slaughtered first, followed by affected cattle and then sheep.

The importance of slaughtering animals during the same day as diagnosis has been shown for epidemics in the UK and Denmark. By contrast, where this period has been delayed, for example, in 1997 in Taiwan, the epidemic was larger and lasted longer (Westergaard 1982, Haydon and others 1997, Howard and Donnelly 2000). However, Yang and others (1999) have suggested that there were several additional reasons for the scale of the epidemic, including the inability of the government to close livestock markets, the high density of pig farms and the inadequate quantities of vaccine available.

Culling within 48 hours of susceptible animals in all premises contiguous to the infected premises is the second highest priority measure in the 2001 FMD campaign. This policy follows advice from biomathematicians who predicted that the animals in contiguous premises would be at risk of being infected due to their proximity to infected premises and so they should also be slaughtered before they became a source of infection and contributed to further spread of the disease (Ferguson and others 2001). The modellers did not define the mechanism of 'local' spread, but assumed that it would happen as a statistical probability. Furthermore, the infectivity and transmission parameters used by the modellers were based on an average hypothetical species. Given the very wide variation between different species in terms of the quantities of virus excreted, their susceptibility to infection, and the routes by which they are likely to be infected, the modelling of the spread of FMD using an average species is an over-simplification, and in certain circumstances would generate inaccurate forecasts.

The results presented in this short communication indicate that, when disease is diagnosed and movement control is fully implemented around an infected premises, the animals on contiguous premises should not be at risk from uncontrollable spread, that is, from airborne spread, unless (a)there are pigs or very large numbers of cattle or sheep on the affected premises with early clinical signs, and (b)the concentration of virus in the plume was at the same or higher concentration than the threshold concentration required to infect them. Under those 'ideal' circumstances for airborne spread, the species at risk downwind would be sheep and cattle. Pigs under a plume would be unlikely to be at risk since very high doses of airborne virus are required to infect them (Donaldson and Alexandersen 2001). The action taken on contiguous premises should, therefore, be determined by the species at risk on those premises. In the case of sheep which may have been exposed to an infectious plume of virus, culling would be justified since FMD in that species is often mild or inapparent (Donaldson and Sellers 2000) and so clinical surveillance would be of limited value in determining whether a flock was infected or not. For cattle, intensified clinical surveillance would be an appropriate alternative to immediate culling, since FMD in that species is easily recognised and any cases should be quickly identified and eliminated before there was a risk of infectious plumes of virus being generated. For pigs, provided that the possibility of any dangerous contacts had been eliminated (see below), on-going clinical surveillance would be appropriate but no other special actions would be justified. These recommendations for contiguous premises should not over-ride the requirement for clinical surveillance, possibly supplemented by serological investigations on other holdings within the 3 km protection and 10 km surveillance areas and the controlled area, if declared.

These actions should not preclude a careful investigation of the infected premises to determine whether direct contact may have taken place between animals at farm boundaries and to review the movements from the premises during the previous three weeks. This should determine if there have been any contacts with contiguous farms or other distant holdings. If such links are found, then an assessment should be made of the degree of risk and the results obtained should determine the actions taken. A high risk would justify the culling of animals on a contiguous farm or a more distant farm as dangerous direct contacts a low risk may call for only increased clinical surveillance. The action taken should be guided by the species at risk, the local circumstances and whether clinical and/or serological surveillance is appropriate.

The implementation of the 48-hour contiguous herd culling policy has resulted in the slaughter of hundreds of thousands of animals and created severe disposal problems. The potential benefits of culling all animals on all contiguous premises within 48 hours should be weighed against the likelihood that many of the contiguous premises did not contain infected animals, the impact of having to dispose of the resultant animal carcases and the diversion of very limited veterinary resources and support staff from surveillance activities.

ACKNOWLEDGEMENTS

Geoffrey Hutchings, Nigel Ferris, Luke Fitzpatrick, Nigel Tallon and Darren Nunney of the Institute for Animal Health, Pirbright, are thanked for their excellent technical assistance. Paul Kitching, Tony Garland, Tony Forman, Phil Wilkinson and David Mackay are thanked for their helpful comments on the manuscript. The research was supported by the UK Ministry of Agriculture,Fisheries and Food and the Danish Ministry of Transport.

References

ALEXANDERSEN,S.,BROTHERHOOD,I.&DONALDSON,A.I.(2001) Natural aerosol transmission of foot-and-mouth disease virus to pigs:minimal infectious dose for strain Or Epidemiology and Infection (In press)

DONALDSON,A.I.(1987)Foot-and-mouth disease:the principal features. Irish Veterinary Journal 41,325-327

DONALDSON,A.I.&ALEXANDERSEN,S.(2001)Relative resistance of pigsto infection by natural aerosols of FMD virus.Veterinary Record 148,

DONALDSON,A.I.&FERRIS,N.P.(1980)Sites of release of airborne foot-and-mouth disease virus from infected pigs.Research in Veterinary Science 29,315-319

DONALDSON,A.I.,GIBSON,C.R,OLIVER,R.,HAMBLIN,C.& KITCHING,R.P.(1987)Infection of cattle by airborne foot-and-mouth disease virus:minimal doses with Ol and SAT 2 strains.Research in Veterinary Science 43,339-346

DONALDSON,A.I.,GLOSTER,J.,HARVEY,L.D.J.&DEANS,D.H.(1982) Use of prediction models to forecast and analyse airborne spread during the foot-and-mouth disease outbreaks in Brittany,Jersey and the Isle of Wight in 1981.Veterinary Record 110,53-57

DONALDSON,A.L,HERNIMAN,K.A.J.,PARKER,J.&SELLERS,R.F(1970)Further investigations on the airborne excretion of foot-and-mouth disease virus.Journal of Hygiene,Cambridge 68,557-564

DONALDSON,A.I.&SELLERS,R.F.(2000)Foot-and-mouth disease.In Diseases of Sheep.3rd edn.Eds W.B.Martin,I.D.Aitken.Oxford,Blackwell Science,pp 254-25

FERGUSON,N.,DONNELLY,C.&ANDERSON,R.(2001)The foot-and-mouthepidemic in Great Britain:pattern of spread and impact of interventions. Published online April 12;10.1126/science.1061020 (Science Express Reports)

GIBSON,C.F.&DONALDSON,A.I.(1986)Exposure of sheep to natural aerosols of foot-and-mouth disease virus.Research in Veterinary Science 41,45-49

GLOSTER,J.,BLACKALL,R.M.,SELLERS,R F &DONALDSON,A.I.(1981) Forecasting the airborne spread of foot-and-mouth disease.Veterinary Record 108,370-374

GLOSTER,J.,SELLERS,R.F.&DONALDSON,A.I.(1982)Long distance transport of foot-and-mouth disease virus over the sea.Veterinary Record 110,47-52

HAYDON,D.T.,WOOLHOUSE,M.E.J.&KITCHING,R.P.(1997)An analysis of foot-and-mouth disease epidemics in the UK.IMA Journal of Mathematics Applied in Medicine and Biology 14,1-9

HENDERSON,R.J.(1969)The outbreak of foot-and-mouth disease in Worcestershire.An epidemiological study:with special reference to spread of the disease by wind-carriage of the virus.Journal of Hygiene 67,21-33

HOWARD,S.C.&DONNELLY,C.A.(2000)The importance of immediate destruction in epidemics of foot and mouth disease.Research in Veterinary Science 69,189-196

HUGH-JONES,M.E.&WRIGHT,P.B.(1970)Studies on the 1967-68 foot-and-mouth disease epidemic.The relation of weather to the spread of disease.Journal of Hygiene 68,253-271

KING,D.(2001)Controlling the foot-and-mouth epidemic -briefing from the Government's Chief Scientific Adviser.MAFF website:http://www.maff.gov.uk/animalh/diseases/fmd/latest/king.ht.Date accessed April 4,2001

MARAGON,S.,FACCHIN,E.,MOUTOU,E,MASSIRIO,L,VINCENZI,G.&DAVIES,G.(1994)The 1993 Italian foot-and-mouth disease epidemic:epidemiological features of the four outbreaks identified in Verona province (Veneto region).Veterinary Record 135,53-57

MIKKELSEN,T,LARSEN,S.E.&THYKIER-NIELSEN,S.(1984)Description of the Ris0 puff diffusion model.Nuclear Technology 67,56-65

MIKKELSEN,T, THYKIER-NIELSEN,S., ASTRUP,P., SANTABARBARA,J.M, S0RENSEN, J.H., RASMUSSEN, A.,ROBERTSON, L., ULLERSTIG,A., DEME,S., MARTENS,R., BARTZIS,J.G .&PASLER-SAUER,J .(1997)METRODOS:a comprehensive atmospheric dispersion module.Radiation Protection Dosimetry 73,45-56

SELLERS,R.F.(1971)Quantitative aspects of the spread of foot-and-mouth disease.Veterinary Bulletin 41,431-439

SELLERS,R.F.&FORMAN,A.J.(1973)The Hampshire epidemic of foot-and-mouth disease.Journal of Hygiene 7,15-34

SELLERS,R.E,HERNIMAN,K.A.J.&DONALDSON,A.I.(1971)The effects of killing or removal of animals affected with foot-and-mouth disease on the amounts of airborne virus present in looseboxes.British Veterinary Journal 127,358-365

SELLERS,R.F.&PARKER,J.(1969)Airborne excretion of foot-and-mouth disease virus.Journal of Hygiene,Cambridge 67,671-677

S0RENSEN,J.H., JENSEN,C.O., MIKKELSEN,T., MACKAY,D.K.J. & DONALDSON,A.I. (2001) Modelling the atmospheric dispersion of foot-and-mouth disease for emergency preparedness.Physics and Chemistry of theEarth B 26,93-97

S0RENSEN,J.H.,MACKAY,D.K.J.,JENSEN,C.O.&DONALDSON,A.I (2000)An integrated model to predict the atmospheric spread of foot-and-mouth disease virus.Epidemiology and Infection 124,577-590

SUTMOLLER,P.&VOSE,D.J.(1997)Contamination of animal products:the minimal pathogen dose required to initiate infection.Revue Scientifique et Technique -Office International des Epizooties 16,30-32

TERPSTRA,C.(1972)Pathogenesis of foot-and-mouth disease in experimentally infected pigs.Bulletin I'Office International des Epizooties 77,859-874

THOMSON,G.R.(1994)Foot-and-mouth disease.In Infectious Diseases of Livestock with Special Reference to Southern Africa.Eds J.A.W.Coetzer,G.R.Thomson,R.C.Tustin.Cape Town,Oxford University Press,pp 825-852

TINLINE,R.(1970)Lee wave hypothesis for the initial pattern of spread during the 1967-68 foot-and-mouth epizootic.Nature 227,860-862

WESTERGAARD,J.M.(1982)Report on the eradication of foot-and-mouth disease on the islands of Funen and Zealand,Denmark 1982.Copenhagen, Danish Veterinary Service,pp 39-41

YANG,P.C.,CHU,R.M.,CHUNG,W.B.&SUNG,H.T.(1999) Epidemiological characteristics and financial costs of the 1997 foot-and-mouth disease epidemic in Taiwan.Veterinary Record 145,731-734604

Copyright 2001 British Veterinary Association

The Veterinary Record, May 12, 2001