Canine ehrlichiosis is caused by Ehrlichia canis, a rickettsial Gram-negative species in the family Anaplasmataceae (Dumler et al. 2007). Throughout the world, E. canis is transmitted by the brown dog tick, Rhipicephalus sanguineus (Harrus & Waner 2011; Nicholson et al. 2010), and is one of four species in the genus (E. canis, Ehrlichia chaffeensis, Ehrlichia ewingii and Ehrlichia ruminantium) that cause serious disease in animals and humans (Dumler et al. 2007; Louw, Allsopp & Meyer 2005; Murphy et al. 1998). Ehrlichia canis infection in dogs has been reported widely in Africa (Kamani et al. 2013; Kelly, Eoghain & Raoult 2004; Matthewman et al. 1993; Pretorius & Kelly 1998; Socolovschi et al. 2012), but not in Namibia specifically. Therefore, the aim of this study was to determine the seroprevalence of canine ehrlichiosis in central Namibia using a point-of-care assay.

The study was conducted at the Rhino Park Veterinary Clinic in Khomasdal, a north-western suburb of Windhoek, Namibia. Data were collected over a period of 8 weeks from dogs that presented to the clinic as well as from stray dogs under the care of the Windhoek branch of the Society for the Prevention of Cruelty to Animals (SPCA). In all cases, participation was voluntary and depended on ease of handling the patient and written informed consent from the patient's direct owner or guardian. The study protocol was approved by the Namibian Ministry of Agriculture, Water and Forestry.

At the clinic, veterinarians evaluated dogs for presenting with symptoms such as inappetence, lethargy and emaciation. Blood samples (2 mL – 5 mL) were obtained from central sites (e.g. the jugular vein) by venipuncture. After initial evaluation, these dogs were assigned to either of two groups, (1) those suspected of E. canis infection or (2) those with non-specific symptoms or that came to the clinic for elective procedures. At inclusion in the study, the sex and age of the dog were noted. Clinic veterinarians also evaluated 30 stray dogs from the local SPCA during the same period to provide another estimate of the potential prevalence of the disease in central Namibia.

The ImmunoComb kit (Biogal, Galed Laboratories, Israel), a solid-phase dot enzyme-linked immunosorbent assay (ELISA) based on a crude antigen of E. canis (Israeli strain), was used according to the manufacturer's instructions to detect IgG-specific antibodies. The manufacturer's cut-off value for positive samples was a titre ≥ 1:80. Reported sensitivity and specificity of this test are 96% and 87%, respectively, when compared with the ‘gold standard’ immunofluorescent antibody test for Nigerian strains of E. canis (Okewole & Adejinmi 2009). Similarly, sensitivity and specificity for Middle Eastern strains of E. canis are reported at 86% and 98%, respectively, with this ELISA (Harrus et al. 2002). Serum samples were assayed immediately after processing and the remainder of each sample was stored at -20 °C for further analysis.

Data were analysed using the PASW Statistics 18 (SPSS Inc, Chicago, IL) software package. Chi-square and Fisher exact tests were used to compare proportions between groups and t-tests and Analysis of variance (ANOVA) were used for comparisons of continuous variables. Age of animals was analysed in three groups: < 1 year, between 1 and 8 years, and > 8 years. A P-value of less than 0.05 was considered statistically significant for all tests.

A total of 106 dogs from various areas of Windhoek and the surrounding Khomas region were evaluated. Of these, 76 were dogs that visited the clinic and 30 were apparently healthy strays from the local SPCA. Ehrlichia canis infection was suspected for the majority (86.2%) of the clinic-evaluated dogs as they presented with inappetence, lethargy and emaciation. The majority of dogs sampled were males (53.5%) and younger than 8 years (92.7%) (Table 1). Ticks were found on 17 (22.4%) of the clinic-evaluated dogs, with 13 (76.5%) having clinical signs as opposed to 4 (23.5%) without clinical signs.

Of the 106 dogs tested, 57 (53.8%) tested seropositive for antibodies against E. canis at titres > 1:80. Dogs with symptoms of E. canis had a higher seroprevalence (86.6%) than dogs without symptoms (42.3%) or stray dogs (40.0%) (P = 0.00). Seroprevalence was similar for male and female dogs and did not differ significantly between the age groups (Table 1). Location of habitation was significant, with a high proportion of seropositive dogs coming from the northern or north-western parts of Windhoek (Khomasdal and Katatura areas) (Table 1). Ticks were found on more dogs from the northern or north-western areas (32.1%) than on dogs from other areas (15.4%), but the difference was not significant (P < 0.150).

This is the first serological study to establish E. canis as a major pathogen in dogs in central Namibia. The high seroprevalence (53.8%) of E. canis in the sample of Namibian dogs is comparable to results from studies in rural Zimbabwe (34%) (Kelly et al. 2004) and urban South Africa (42%) (Pretorius & Kelly 1998). The highest proportion of seropositive dogs came from the lower-income areas of Khomasdal and Katatura, which reflects a similar trend to that reported in Bloemfontein, South Africa (Pretorius & Kelly 1998). The only described vector of canine ehrlichiosis, Rhipicephalus sanguineus (the brown dog tick), is the principal tick species found on dogs throughout central and southern Namibia (Matthee et al. 2010). With a higher prevalence of ticks found on dogs from lower-income areas, this pattern may be due to a lack of access to tick control (Pretorius & Kelly 1998).

Although the involvement of only one veterinary clinic and the limited timescale and sample size may introduce several limitations, this study demonstrates that dogs in central Namibia are frequently exposed to Ehrlichia species, the most likely being E. canis. Based on the serological detection, further molecular investigation of Namibian dogs and their ticks is required to verify the presence of Ehrlichia species.