Association between canine leishmaniosis and Ehrlichia canis co-infection: a prospective case-control study

<h3>Background</h3><p>In the Mediterranean basin, <i>Leishmania infantum</i> is a major cause of disease in dogs, which are frequently co-infected with other vector-borne pathogens (VBP). However, the associations between dogs with clinical leishmaniosis (ClinL) and VBP co-infections have not been studied. We assessed the risk of VBP infections in dogs with ClinL and healthy controls.</p><h3>Methods</h3><p>We conducted a prospective case-control study of dogs with ClinL (positive qPCR and ELISA antibody for <i>L</i>. <i>infantum</i> on peripheral blood) and clinically healthy, ideally breed-, sex- and age-matched, control dogs (negative qPCR and ELISA antibody for <i>L</i>. <i>infantum</i> on peripheral blood) from Paphos, Cyprus. We obtained demographic data and all dogs underwent PCR on EDTA-blood extracted DNA for haemoplasma species, <i>Ehrlichia</i>/<i>Anaplasma</i> spp., <i>Babesia</i> spp., and <i>Hepatozoon</i> spp., with DNA sequencing to identify infecting species. We used logistic regression analysis and structural equation modelling (SEM) to evaluate the risk of VBP infections between ClinL cases and controls.</p><h3>Results</h3><p>From the 50 enrolled dogs with ClinL, DNA was detected in 24 (48%) for <i>Hepatozoon</i> spp., 14 (28%) for <i>Mycoplasma haemocanis</i>, 6 (12%) for <i>Ehrlichia canis</i> and 2 (4%) for <i>Anaplasma platys</i>. In the 92 enrolled control dogs, DNA was detected in 41 (45%) for <i>Hepatozoon</i> spp., 18 (20%) for <i>M</i>. <i>haemocanis</i>, 1 (1%) for <i>E. canis</i> and 3 (3%) for <i>A</i>. <i>platys</i>. No <i>Babesia</i> spp. or “<i>Candidatus</i> Mycoplasma haematoparvum” DNA was detected in any dog. No statistical differences were found between the ClinL and controls regarding age, sex, breed, lifestyle and use of ectoparasitic prevention. A significant association between ClinL and <i>E. canis</i> infection (OR = 12.4, 95% CI: 1.5–106.0, <i>P</i> = 0.022) was found compared to controls by multivariate logistic regression. This association was confirmed using SEM, which further identified that younger dogs were more likely to be infected with each of <i>Hepatozoon</i> spp. and <i>M</i>. <i>haemocanis</i>, and dogs with <i>Hepatozoon</i> spp. were more likely to be co-infected with <i>M</i>. <i>haemocanis</i>.</p><h3>Conclusions</h3><p>Dogs with ClinL are at a higher risk of co-infection with <i>E. canis</i> than clinically healthy dogs. We recommend that dogs diagnosed with ClinL should be tested for <i>E. canis</i> co-infection using PCR.</p>