Estimation of Seroprevalence and Predictors for Coxiella Burnetii Infection in Garbatulla, Isiolo County, Kenya.

Abstract

Coxiella burnetii is the etiological agent for Q fever, which affects people and animals globally. This pathogen is presumed to be endemic in both human beings and domestic animals in many countries Kenya included. It causes serious economic impacts and poses a zoonotic risk to people closely exposed to livestock. In Kenya, scientific information on the epidemiology of C. burnetii at the human-livestock interface in rural pastoral settings is limited. Therefore, I conducted a linked human-livestock survey in Garbatulla to estimate the seroprevalence and potential predictors for C. burnetii exposure in humans and livestock. The study employed a cross-sectional design with multistage random sampling to collect samples from humans and livestock. The main sampling units were the individual people and livestock from the randomly chosen households. In total, 683 humans were sampled from 242 households, and 2,157 livestock (cattle, camels, sheep, and goats) were sampled from 231 herds. These numbers were slightly higher than the calculated adjusted sample sizes of 538 humans and 1843 animals. Serum samples for human subjects were tested for antibodies against C. burnetii using SERION ELISA classic C. burnetii Phase II IgG ELISA kits (Virion/Serion, Germany), following the manufacturer's guidelines. In livestock, sera were screened for antibodies against C. burnetii using ID Screen Q Fever Indirect Multi-species Kit (IDvet™, Montpellier, France) according to the manufacturer's instructions. Optical densities and serum percentage positivity for sera antibodies were generated for both humans and livestock respectively. Data on potential predictors was collected using a structured questionnaire. The association between the independent variables and C. burnetii seropositivity in the study area was calculated using the chi-square test and odds ratios generated. Further, both univariable (p<0.20) and multivariable (p<0.05) analyses for human and livestock data were analyzed using generalized linear mixed-effects models (GLMMs) accordingly. This was to determine the association between the outcome variable and the independent variables both in humans and livestock. The overall estimated seroprevalence of C. burnetii was 47.9% (95% CI: 45.7%- 50.1%) in livestock and 44.7% (95% CI: 40.9% - 48.5%) in humans. In livestock, significant variation in seroprevalence per species was xii observed (p < 0.001). Further, species (goat) (OR=4.7, p=0.001), advanced age (OR=8.8, p=0.001), and spatial distribution (OR=1.5, p=0.024) were linked to C. burnetii exposure. In humans, the risk factors for human seropositivity included handling animals (OR=5.3, p=0.004), living in a household with a seropositive herd (OR=2.3, p=0.005), and being male (OR=2.2, p=0.001). This study documented evidence of high seroprevalence of C. burnetii amongst livestock and humans, which could result in serious public health implications. Therefore, there is a need to carry out further research using molecular analysis to determine antibodies due to active and past infections.