Canine infectious respiratory disease: New insights into the etiology and epidemiology of associated pathogens

Abstract

Canine infectious respiratory disease (CIRD) is a syndrome where multiple viral and bacterial pathogens are involved sequentially or synergistically to cause illness. There is limited information regarding the prevalence of pathogens related to CIRD in the United States as well as the role of co-infections in the pathogenesis of the syndrome. We aimed to conduct a comprehensive etiologic and epidemiologic study of multiple CIRD agents in a diverse dog population using molecular methods and statistical modeling analyses. In addition, a novel probe-based multiplex real-time PCR was developed to simultaneously detect and differentiate two species of Mycoplasma (M. canis and M. cynos). Canine adenovirus, canine distemper virus, canine parainfluenza virus, coronavirus, influenza A virus (H3N2 and H3N8), Bordetella bronchiseptica, M. canis, M. cynos and Streptococcus equi subsp. zooepidemicus were investigated in specimens from clinically ill and asymptomatic dogs received at the Athens Veterinary Diagnostic Laboratory. Results showed low occurrence of classical CIRD agents such as B. bronchiseptica, canine adenovirus and distemper virus, while highlighting the potential role of emerging bacteria such as M. canis and M. cynos. Statistical modeling analyses of CIRD pathogens emphasized the impact of co-infections on the severity of clinical presentation, and showed that host factors, such as animal age, are the most important predictors of disease severity. This study provides new insights into the current understanding of the prevalence and role of co-infections with selected viruses and bacteria in the etiology of CIRD, while underscoring the importance of molecular diagnosis and vaccination against this disease.

Fig 1. Rate of detection of canine infectious respiratory disease pathogens detected by PCR assays in a veterinary diagnostic laboratory.

(A) Total detection of pathogens in respiratory specimens from 2011 to 2017. (B) Rate of detection according to season. Two seasons, cold and warm, were determined based on the average temperatures of the state of Georgia (USA). Streptococcus equi subsp. zooepidemicus was not displayed in graphs B, C and D because all samples were negative for this pathogen. Error bars represent 95% confidence intervals. Data were analyzed using Fisher`s Exact test: * P < 0.05, **P < 0.01, ***P < 0.001.

Fig 2. Rate of detection of canine infectious respiratory pathogens by age (A) and clinical presentation (scores) (B).

Age class: puppyhood (0 to 12 months, n = 158), adolescence (1 to 3 years, n = 160), adulthood (4 to 8 years, n = 119) and seniors (≥ 8 years, n = 99). Sample set included in the PCR panel in 2017 (canine parainfluenza virus, Mycoplasma canis and M. cynos): puppyhood n = 42, adolescent n = 30, adulthood n = 20, senior n = 19 (A). Clinical score 0 (asymptomatic dogs, n = 52), score 1 (cough or sneeze or nasal discharge, n = 213), score 2 (signs from score 1 in addition to fever or lethargy/depression or inappetence or pneumonia, n = 223) (B). All samples were negative for Streptococcus equi subsp. zooepidemicus. Error bars represent 95% confidence intervals. Data were analyzed using Fisher`s Exact test and Bonferroni correction: * P < 0.05, **P < 0.01, ***P < 0.001.

Fig 3. Visualization of CIRD co-infections interactions in Network 3D, n = 95 (Williams, 2010; Yoon et al., 2004).

The network shows strong connectivity between CPIV+M. canis, CPIV+M. cynos, CPIV+M. cynos+M. canis, CPIV+B. bronchiseptica, B. bronchiseptica+M. cynos+CPIV, CPIV+M. cynos+M. canis.