Seroprevalence of brucellosis in Mississippi shelter dogs

Canine brucellosis is an emerging disease and compatible with a One Health management approach. Previous research has found higher prevalence of Brucella canis in stray dog populations than in owned animals, and shelter dogs may represent a zoonotic risk to pet owners. Dogs may also contract other Brucella spp., including Brucella suis, which is carried by some feral swine in the United States and poses a public health risk. Diagnostic tests for Brucella spp. are imperfect. Misclassification of disease status can result in serious repercussions for canine and human health including the unnecessary euthanasia of false positive dogs or failure to identify and remove false negative dogs from susceptible populations. Correct interpretation of any diagnostic test requires knowledge of the pre-test probability of disease in the population, therefore the objective of this cross-sectional study was to estimate the seroprevalence of B. canis and B. suis in Mississippi shelter dogs to guide evidence-based diagnostic testing and inform policy recommendations. Banked serum samples from 571 dogs collected in 2016-2017 as a representative sample of the Mississippi shelter dog population were tested for B. canis using a rapid slide agglutination test (RSAT) and for B. suis using a buffered acidified plate agglutination test. No dogs were seropositive for B. suis antibodies. Twenty-eight dogs (4.9%) were seropositive for B. canis antibodies on the RSAT, with 13 dogs (2.3%) remaining positive when retested with the addition of 2-mercaptoethanol to increase specificity. Test prevalence by shelter ranged from 0 to 8.6%. True prevalence was estimated using stochastic modeling to account for test performance and clustering of dogs by shelter. Approximately 65% of modeled shelters did not have seropositive dogs. For shelters where B. canis was present, the mean modeled seroprevalence was 17.8%. This cross-sectional study reveals important information regarding the distribution of B. canis seroprevalence in Mississippi shelter dogs. Current diagnostic tests lack the sensitivity needed to correctly identify individual infected dogs, but population testing may provide a reasonable estimate of disease. Eradication or control measures may be most efficiently allocated to shelters where canine brucellosis has been identified to effectively minimize transmission among dogs and to humans.