Discussion
The present study aimed at providing a general picture of the exposure to BoDV in horses in Spain. The aim was to collect a representative sample of autonomous communities throughout the country. However, this was only possible in 5 of the 9 communities sampled, most of them in central Spain (a high risk area in terms of transmission of equine infectious diseases). This might be attributable to the fact that horses being exported to continental Europe or abroad often stop in central Spain (Cruz et al., 2016b) and because there the majority of national and international equestrian competitions are held (Deloitte, 2013).
Up to present BoDV causing clinical disease has never been reported in horses in Spain and the only information published is a case of an Appaloosa pony showing neurological signs and being seropositive to EHV-1 and BoDV. However in the latter case BoDV was not isolated and no molecular tests were carried out (Vitale et al., 2018). This is the first study to define BoDV seroprevalence in breeding and sport horses in Spain. Even though Spain was thought to be free from BoDV, our results showed that the virus may be endemic in this country. The determined seroprevalence (33.1%; 95% CI 29.0-37.3) was even higher than the seroprevalence in horses from areas with clinical disease in endemic countries like Germany (22.5%) (Richt and Rott, 2001) or other countries like USA (2.7%) (Kao et al., 1993), Finland (2.4%) (Kinnunen et al., 2007), France (8.9%) (Galabru et al., 2000) or Turkey (4.9%) (Yesilbag et al., 2012). The high BoDV seroprevalence found in our study contrasts with the fact that no clinical cases have been reported up to date. This can be explained by subclinical infection (Richt, 2007) or misdiagnosed disease due to the fact that BoDV is not considered as a possible cause for neurological signs in horses in Spain. This report would suggest the need to test for BoDV in horses showing neurological signs, along with Equine Herpesvirus type 1 (EHV-1) or West Nile virus (WNV).
The seropositive horses in this study showed moderate to high titres, with up to 14.6% of seropositives presenting a titre of ≥1:2560. High titres have been restricted to clinical disease, and some reports have demonstrated that titres decrease relatively fast and are hardly detectable in a subacute or chronic disease (Dauphin and Zientara, 2001). The high titres shown by the horses in our study would therefore either suggest a recent (or frequent) exposure to BoDV or the fact that antibody titres against BoDV could remain high for a long time, in agreement with a study reported in Japan, where horses were found to be continuously seropositive for a period of four years (Inoue et al., 2002a). Further studies investigating possible hosts like shrews or other rodents would be necessary to corroborate a high prevalence of BoDV in them (Hilbe et al., 2006) which would enable a repeated exposure to BoDV in Spanish horses.
Regarding BoDV seroprevalence by autonomous community, no significant differences were found between the different sampled areas (p= 0.87). The slight differences observed could be due to bias in sampling, and in order to establish an influence of the area on BoDV seroprevalence, a more extensive sampling covering all areas in Spain should be carried out. Furthermore, the population of shrews and other rodents in Spain should be investigated for the presence of the virus, in order to point out risk areas in Spain for this disease (Encarnacao et al., 2013).
The results of our study show that BoDV is endemic and stable through the years in Spain, since there were no significant differences between the seroprevalence in the two sampling periods (2011-2013 and 2015-2016) (p= 0.55). Breeding horses sampled in both periods showed a slight increase in the seroprevalence in 2015-2016, although this increase was not significant (p= 0.53) and could be a reflection of BoDV dynamics or fluctuation in wild rodents (Richt, 2007, Encarnacao et al., 2013, Kinnunen et al., 2013).
Similarly to other studies, our results determined that BoDV seropositivity was not associated to sex (p= 0.85) (Inoue et al., 2002b, Inoue et al., 2002a). The only study that found a higher seroprevalence in female horses was carried out in Italy using circulating immunocomplexes (CIC) ELISA tests, which are not considered the gold standard test for BoDV diagnosis (Pisoni et al., 2007).
Even though we did not find statistically significant differences (p= 0.17), seroprevalence in breeding horses (which are often kept in the field) was higher than in sport horses (kept in stables). Further studies would be necessary to prove or exclude this association, since several studies have found higher BoDV seroprevalence rates in stabled horses, probably associated to a higher presence of reservoir hosts (Richt, 2007, Inoue et al., 2002b). A higher presence of the virus in breeding horses, already seen by Pisoni et al. (2007) in Italy would entail the need to investigate this population and the reservoir hosts present in the fields of Spain, since the risk of vertical transmission of the virus and the occurrence of abortion has been documented in the literature (Hagiwara et al., 2000).
In the population of breeding horses of our study, factors associated to BoDV seropositivity were the climate, the age and the time of the year when the horses were sampled (Table 2). Horses residing in an area with a cold semi-arid climate (BSK climate from the Köppen-Geiger Climate Classification) (Rubel and Kottek, 2010) were 3 times more likely to be BoDV seropositive. This factor has already been reported elsewhere, and is probably related to the living conditions for reservoir hosts, which prefer a cold climate with low precipitations (Encarnacao et al., 2013).
Our results showed a lower seropositivity in adult horses (7 to 14 years old) (OR= 0.31, 95% CI 0.16 to 0.63) and in older horses (15 to 24 years old) compared to young horses (1 to 6 years old), However the decrease was not significant in the oldest horses (p = 0.69). In contrast other viruses like Equine Viral Arteritis virus or EHV-1 showed a higher seroprevalence in older individuals in the same population (Cruz et al., 2016a, Cruz et al., 2016b). In addition, other BoDV studies showed no association between age and seropositivity rate (Pisoni et al., 2007, Inoue et al., 2002b). The decrease in the seroprevalence in older individuals could be related to a decrease in the antibody titres after infection, suggesting that SP breeding horses would be infected in the early years, when they are in the field, and not in stables.
The literature has shown that BoDV clinical disease peaks in the period between March and June due to infection during the autumn-winter when the horses are stabled (Richt, 2007, Kinnunen et al., 2013, Durrwald and Ludwig, 1997), and that antibodies rise late in the course of infection, along with clinical disease (Richt et al., 2000). However, in our study, the breeding horses sampled from February to June (supposedly the period associated to clinical disease and high antibody titres) were 2.7 times less likely to be BoDV seropositive than horses sampled from July to January (OR= 0.37, 95% CI 0.19 to 0.73). This is interesting and agrees with the idea that, in Spain, breeding SP horses would be more likely to encounter BoDV when they are young, and while they are out in the fields during the spring-summer, showing a rise in their antibodies during the autumn-winter. Therefore, further studies in order to characterise BoDV reservoir hosts may help to elucidate BoDV transmission in horses in Spain.
Given the recent association of BoDV-1 with eight cases of fatal encephalitis in humans in areas of BoDV endemicity in Germany (Niller et al., 2020), our results would rise the concern regarding the zoonotic potential of this virus in Spain; therefore, BoDV should be included in the differential diagnosis of encephalitis in humans.
In conclusion, the present study provides the first report on the endemicity of BoDV in apparently asymptomatic breeding and sport horses in Spain. We suggest that Borna Disease should be considered as a differential diagnosis to Equine Herpesvirus Myeloencephalopathy or West Nile of a horse suffering from neurological symptoms, and that human cases of encephalitis should also be tested for BoDV. Epidemiological studies at the local levels are paramount to forecast disease risks and understand the transmission and the reservoir hosts involved, in order to implement preventive measures.