Main content area

A cross sectional study evaluating the prevalence of Coxiella burnetii, potential risk factors for infection, and agreement between diagnostic methods in goats in Indiana

Bauer, Amy E., Hubbard, Kirk R.A., Johnson, April J., Messick, Joanne B., Weng, Hsin-Yi, Pogranichniy, Roman M.
Preventive veterinary medicine 2016 v.126 pp. 131-137
Coxiella burnetii, DNA, Q fever, blood serum, cross-sectional studies, dairy breeds, diagnostic techniques, disease outbreaks, enzyme-linked immunosorbent assay, equations, etiology, farms, goat breeds, goat milk, goats, herds, meat, models, quantitative polymerase chain reaction, risk factors, screening, seroprevalence, zoonoses, Indiana
Coxiella burnetii is the etiologic agent of the zoonotic disease Q fever and is considered to be endemic in domestic ruminants. Small ruminants in particular are important reservoirs for human infection. Serologic and molecular methods are both available for diagnosis of infection with C. burnetii, but there has been little research evaluating the prevalence of this organism in small ruminants outside of the context of clinical disease outbreaks. The objectives of this study were to estimate seroprevalence of C. burnetii and the prevalence of shedding of C. burnetii DNA in milk by goats in Indiana, USA, to evaluate potential risk factors for association with C. burnetii exposure and shedding, and to assess the level of agreement between the enzyme-linked immunosorbent assay (ELISA) and real-time polymerase chain reaction (PCR) tests used to estimate prevalence. A total of 649 does over 1 year of age and not pregnant at the time of sampling were included in the study. Serum samples were collected from 608 does representing 89 farms. Milk samples were collected from 387 does representing 85 farms. Both milk and serum samples were collected from 356 does representing 80 farms. The estimated individual seroprevalence and shedding prevalence in milk adjusted for clustering were 3.1% (n=23/608, 95% CI: 1.2–7.0%) and 2.5% (n=9/387, 9.5% CI: 1.0–5.6%) respectively. Estimated adjusted herd level C. burnetii seroprevalence and herd level shedding prevalence were 11.5% (n=10/89, 95% CI: 6.4–20.1%) and 7.0% (n=6/85, 95% CI: 3.3–14.6%) respectively. Based on a generalized estimating equation model (GEE), meat breeds of goat had 7.0 times increased odds of shedding C. burnetii DNA in milk samples as compared to dairy breeds. Agreement between tests as determined by Cohen’s kappa was poor at both the individual (kappa=0.04, 95% CI: −0.1 to 0.2) and herd (kappa=0.2, 95% CI: −0.1 to 0.5) levels. This indicates that serologic screening alone is unlikely to prevent the introduction of does shedding C. burnetii into herds.