Prevalence, risk factors, and geographical distribution of zoonotic pathogens carried by peri-urban wild dogs

Abstract

Established wildlife populations within peri-urban and urban environments provide an opportunity for the spread of zoonotic pathogens to and from human-associated environments. Wild dogs, including dingoes (Canis lupus dingo), feral domestic dogs (Canis lupus familiaris), and their cross breeds (hybrids) are common across the Australian mainland, and hybrid populations are particularly frequent within peri-urban environments. Previous studies have shown that wild dogs carry zoonotic pathogens of public health significance, however, data relating to peri-urban wild dogs has been limited to a few studies with a small geographical range and / or small sample size. To address the potential public health implications of peri-urban wild dogs, information regarding pathogen diversity, prevalence, risk factors and geographical distribution are required. To investigate the diversity and prevalence of zoonotic pathogens in peri-urban wild dogs, a cross sectional survey was undertaken. Utilising council management programs and private trappers, 201 wild dogs were collected for necropsy. Analysis of faecal and blood samples indicated that helminth parasites were common, detected within 79.6 ± 5.4 % of wild dogs tested, however bacterial pathogens were much less prevalent. Echinococcus granulosus was the most prevalent parasite detected by necropsy, (prevalence 50.7 ± 6.9 %). Hookworms were found to comprise the second most common pathogen of zoonotic significance with an overall prevalence of 28.8% (± 7.1%). Molecular characterisation identified the majority as Ancylostoma caninum. Several pathogens of zoonotic interest were not detected during the study, namely Brucella suis and Neospora caninum. This study provided essential baseline data on the prevalence of pathogens carried by peri-urban wild dog populations in north-eastern Australia, which is required to inform the management of wild dogs in peri-urban areas. Characteristics of the most prevalent parasitic pathogens and their lifecycles suggested possible associations between the diet composition of wild dogs and their infection status. Whole stomachs were collected during necropsy and analysed for dietary composition and biomass. The majority of food items detected were mammalian prey species, most commonly: swamp wallaby (20.6 ± 6.08%); canine species (prey) (10.6 ± 4.62%); eastern grey kangaroo (10.0 ± 4.51 %); and deer species (10.0 ± 4.51%). Wild dogs that consumed swamp wallaby (OR 1.79, p<0.05) or unidentified species of macropods (OR 4.18, p<0.01) were found to be significantly more likely to be infected with E. granulosus than dogs who had consumed non-macropod species. Wild dogs that had consumed various bird (OR 7.80, p<0.01) or bandicoot species (OR 3.09, p<0.01) were found to be significantly more likely to be infected with hookworm compared to dogs that had consumed macropods. These findings suggest that diet composition is an important contributing factor to the infection status of wild dogs. To explore the geographical distribution of pathogens and to determine the risk factors associated with their presence, spatial modelling techniques were applied to the two most significant (based on prevalence in wild dogs and their potential impact on human health) parasites detected within the study. Using model-based geostatistics our results indicated significant geographical variation in the probability of E. granulosus carriage, with a strong propensity for clustering. None of the modelled physical environment or climate variables were significantly associated with the presence of E. granulosus, however the fitted model accounted for the majority of the spatial dependence. In the case of hookworm, our results demonstrated that there was no significant spatial variation in the residual probability of hookworm carriage. A prediction map was generated using a logit back transformation of the final non-spatial multivariable model. Overlaying the predictive maps of E. granulosus and hookworm indicated that regions of high predicted hookworm prevalence overlapped with low to moderate predicted E. granulosus prevalence. These results suggest that due to the significant public health risks of E. granulosus, an integrated program that controls both parasites in the regions of high predicted hookworm prevalence would be beneficial. Regions of high predicted prevalence for E. granulosus correspond with regions of low predicted prevalence for hookworm. In this case, management programs could exclusively target E. granulosus to efficiently control the risk on human health. This study describes the important role of wild dogs in the maintenance of zoonotic pathogens within peri-urban environments, identifies risk factors associated with pathogen carriage, and provides two predictive maps that have the potential to be used as spatial-decision supports tools to mitigate the public health risks of peri-urban wild dogs. The prediction maps will greatly assist land managers to best allocate their limited resources, and provide the opportunity for future research to assess the effectiveness of the applied control / management techniques on zoonotic pathogen prevalence in peri-urban wild dogs and their immediate habitat.