Animal tuberculosis maintenance at low abundance of suitable wildlife reservoir hosts: A case study in northern Spain

Abstract

Animal tuberculosis (TB), which is caused by infection with members of the Mycobacterium tuberculosis complex (MTC), is a typical multi-host infection that flourishes at the livestock-wildlife interface. TB epidemiology is well characterized in the Mediterranean woodland habitats and Atlantic regions of southwestern Europe. However, much less is known about huge regions that do not form part of the two abovementioned settings, which have a low abundance of wild reservoirs. We hypothesized that MTC would be maintained in multi- rather than single-host communities in which wildlife would make a relatively low contribution to the maintenance of TB. Between 2011 and 2015, 7729 Eurasian wild boar (Sus scrofa) and 1729 wild ruminants were sampled for culture during hunting events on unfenced sites. In addition, 1058 wild ungulates were sampled on 23 fenced hunting estates. Infection prevalence data were modeled along with official data on cattle and goat TB, on livestock distribution and management, and on wild boar abundance. The mean individual MTC infection prevalence was 4.28% in wild boar, while the cattle skin test reactor percent was 0.17%. The prevalence of MTC infection in wild ungulates (mostly wild boar) from the fenced hunting estates was 11.6%. Modeling revealed that the main driver of TB in cattle was their management (beef; communal pastures). However, wild boar abundance, the prevalence of MTC infection in wild boar and the presence of fenced hunting estates also contributed to explaining cattle TB. The model used for goat TB identified communal pastures as a risk factor. The model for the prevalence of MTC infection in wild boar included wild boar abundance and communal pastures. We conclude that the MTC maintenance host community is most likely of a multi-host nature. While cattle and communal pastures pose the main risk regarding TB, it is also necessary to consider increasing wild boar densities and specific risks owing to fenced wildlife. We infer several management implications regarding wildlife management, the wildlife sampling strategy and laboratory testing, the peculiarities of fenced hunting estates, and the wildlife-livestock interface.