Respiratory disease and sero-epidemiology of respiratory pathogens in the working horses of Ethiopia

Abstract

Background: Pathogens are frequently implicated in equine respiratory disease. In Ethiopia, respiratory disease is a frequent cause for presentation at veterinary clinics and a priority concern for users of working horses. However, there is little existing literature on possible aetiologies.

Objectives: Determine prevalence of respiratory signs and exposure to major respiratory pathogens through a serological survey.

Study design: Cross-sectional.

Methods: Systematically selected horses from 19 sites in central Ethiopia were examined clinically and sampled once (August-December 2013). A face-to-face interview collected data on horses' management and history. Serological testing targeted equine influenza virus (EIV), equine herpesviruses-1 (EHV-1) and -4 (EHV-4), equine rhinitis viruses A (ERAV) and B (ERBV), equine arteritis virus (EAV) and Streptococcus equi subspecies equi (S. equi).

Results: Owners reported a recent history of coughing in 38% of horses and nasal discharge in 7%. No animals were observed coughing during examination but 6% had a nasal discharge. Antibodies towards S. equi, were most prevalent (8%, 33/350). Antibodies to EAV were confirmed in one animal (0.3%). Low antibody titres to EHV-1/4 and ERA/BV suggested prior exposure but antibodies to EIV were not detected. Multivariable, multilevel logistic regression analysis for risk factors associated with S. equi serostatus showed higher odds of seropositivity in younger animals and those working less frequently.

Main limitations: A single serological sample cannot describe dynamic changes in antibodies. Sampling horses at the place of work may result in healthy-worker bias.

Conclusions: S. equi may be endemic in this population and contributing, in part, to the occurrence of respiratory disease. Low prevalence of antibodies to viruses, with the exception of EIV, indicates these pathogens are present, but unlikely a predominant cause of respiratory signs and noninfectious causes of disease should also be investigated. Working horses in this region would be vulnerable to incursion of equine influenza.

Keywords: ELISA; Streptococcus equi; Africa; horse; prevalence; serology; strangles.

Figure 1

Map showing sampling sites and seroprevalence for S. equi. Red points indicate a site where no seropositive horses were detected. Blue points are proportional in size to the prevalence of S. equi positive animals detected at each location.

Figure 2

Smoothed distribution curves of clinical parameters for rested horses (n = 151), shown with comparison to published reference intervals for other populations. Upper and lower limits for normal for a UK horse population 23 (green). Comparison of haematological values to those from another working equid population in Pakistan (blue dashes) 24. 2.5 and 97.5% percentiles for this study population (grey dashes).