Spread of Infection (Epidemiology)

The behaviour of MAP in populations of sheep is perhaps best understood through two case stud­ies where the organism was introduced into re­gions where it was previously unknown: Iceland and Australia.

Sheep were introduced to Iceland in the 9th century and remained free of MAP until 20 Karakul sheep were imported from Germany in 1933. In fact, three slowly developing infec­tious diseases evaded quarantine controls due to their long incubation periods: paratuberculosis, maedi/visna and jaagsiekte. Paratuberculosis was diagnosed 5 years later, and 440 farms were infected in the subsequent 20 years. In 1940, the disease appeared in cattle that were grazed with sheep, but fewer cows than sheep were af­fected. It later occurred in goats and a reindeer (Fridriksdottir et al., 2000). Mortality rates per annum in sheep averaged 8-9% but were as high as 40% on some farms. Sheep and dairy cattle grazed together on pastures during the summer and were housed together during the colder months. Thus, there was probably a very high level of cross-species MAP contamination. Control measures based on fencing and zon­ing based on prevalence levels, along with test and cull, failed to impede spread. Destocking of all sheep in some zones was then tried, but the infection was probably maintained in cat­tle, as healthy sheep introduced 1 year after the destocking programme succumbed to paratu­berculosis. Finally, vaccination of lambs was introduced, and this is still practised to prevent clinical disease. Paratuberculosis in Iceland is caused by MAP-S, which could not be routinely cultured; it was identified using molecular tech­niques from archives of histological paraffin blocks (Whittington et al., 2001).

Australia has experienced a similar ‘slow epizootic' of paratuberculosis in sheep. The most likely scenario for Australia involved importa­tion of infected sheep from New Zealand prior to 1958 or in the 19 70s, as live sheep were not imported in the intervening period (Sergeant, 2001). The first case of ovine paratuberculosis was diagnosed in New South Wales in 1980 (Seaman et al., 1981), years after its likely in­troduction. The disease was most unlikely to have been widespread in sheep flocks at that time as there was a strong system of passive surveillance in Australia (provided by regional veterinary laboratories) at no cost to farmers. For this reason, it is likely that the rate of spread of paratuberculosis increased exponentially after about 1980, associated with the pattern of trade of live sheep in Australia. By the mid- 1990s, the disease was of significant economic impact on farms where it had been present for some years. Economic losses of 6.4-8.5% in the gross income margin have been recorded on farms where ovine paratuberculosis mortal­ity ranged from 6.2-7.8% (Bush et al., 2006). By 2000, there were 823 known infected flocks in New South Wales, Victoria, Tasmania and South Australia (Sergeant, 2001). Thousands of other flocks were suspected to be infected, based on purchase of sheep from infected farms and shared farm boundaries. Western Australia, which was geographically isolated, was the last state to notify infection; at the time of detection in 2004, infection was deemed likely to have been present for 7 years or longer; seven flocks with over 82,000 sheep were infected and 144 more were under suspicion (Sunderman, 2004).

A national disease control programme commenced in Australia in 1999; it employed pooled faecal culture and abattoir surveillance to determine infection and set up an assurance programme to identify and protect flocks that did not have the disease (Sergeant, 2001).

Strains of MAP from sheep, cattle and many other animals can be clearly divided on genetic grounds into two lineages called MAP-C and MAP-S. A more complete description of these groups, their alternative designations and strain characterization is given elsewhere (see Chapter 6, this volume). The MAP-S, which was responsible for paratuberculosis in sheep in Iceland and Australia, is also prevalent in New Zealand. In other countries it may be displaced by MAP-C, where either sheep are uncommon or cattle are prevalent. Sheep are also suscepti­ble to the MAP-C strain and it has been a com­mon finding in sheep in Europe (de Juan et al., 2005; Sevilla et al., 2008; Florou et al., 2009). The difficulty of cultivation of MAP-S strains probably leads to underestimation of its distribu­tion and abundance. Cattle, goats and deer can become infected with MAP-S strains (O'Brien et al., 2006; Mackintosh et al., 2007; Moloney and Whittington, 2008; Sevilla et al., 2008), but cattle and deer appear to be more resistant to infection and clinical disease associated with MAP-S. In Australia, cattle have become infect­ed as calves if exposed to heavily infected sheep, but it is an uncommon infection (Whittington et al., 2001; Moloney and Whittington, 2008).

A detailed investigation on one infected farm in Australia revealed a slow rate of trans­mission and clustering of infection within age classes of sheep for 7 years. Susceptibility of young sheep to lower levels of MAP and the long incubation period before faecal shedding may explain this (Rast and Whittington, 2005). Once contamination rates build up, this pattern may be lost as sheep of any age become infected. This results in increased prevalence rates, with in­fected sheep possibly shedding MAP sooner after infection, but this is yet to be proven. One rea­son for different levels of prevalence between af­fected flocks is variation in stocking rate during lambing (Dhand et al., 2007). A higher stocking rate leads to higher levels of contamination, causing a greater risk of exposure of lambs, and this was associated with higher levels of infec­tion (Dhand et al., 2007).

12.7