Epidemiology of Foot and Mouth Disease, Brucellosis and Leptospirosis at the Livestock Wildlife Interface Area in Maasai Mara, Kenya

Abstract

Intact ecosystems can regulate the transmission of infectious diseases by maintaining the diversity of species in equilibrium. However, human driven land use change is a major driver of environmental change and can affect the emergence and the transmission dynamics of infectious diseases. This study investigated how land use change, and hence the extent of wildlife-livestock interactions affect the transmission dynamics of infectious diseases in the Maasai Mara ecosystem in Kenya, using foot and mouth disease virus (FMDV), Brucella spp. and Leptospira spp. as case study pathogens. Three ecological zones were selected along transect from the Maasai Mara National Reserve (MMNR) to settled areas with different landuse types and varying resulting levels of wildlife-livestock interactions. Areas surrounding the

MMNR and wildlife conservancies represented zone 1 (areas with intensive wildlife-livestock

interactions) while areas between 20-40 km away from the MMNR constituted zone 2 with

moderate wildlife-livestock interactions. Zone 3 was represented by areas more than 40 km

away

from the MMNR where wildlife-livestock interactions are low. A participatory epidemiological study was first conducted to determine peoples’ perceptions on prevalence, seasonality, and impacts on livestock production of infectious diseases of concern to pastoralists in the defined zones. For this objective, four villages were purposively selected in zone 1 and another two in zone 2. Data were collected in focus group discussions (FGDs) using participatory epidemiological methods and with each group having 8-13 participants. A cross-sectional study was also conducted to determine the seroprevalences of Brucella spp., Leptospira spp. and foot and mouth disease (FMD) among cattle herds raised in the area. Five villages were purposefully selected; two in zone 1, another two in zone 2 and one in zone 3. A total of 1,170 cattle sera were collected from 390 herds distributed across the zones and tested for antibodies against Brucella abortus, Leptospira interrogans serovar hardjo and non-structural 3ABC proteins (NSPs) of FMDV using commercially available EnzymeLinked Immunosorbent Assay (ELISA) kits. All sera samples were further tested for serotypespecific antibodies of FMDV using Solid Phase Competitive ELISA (SPCE) kits (IZSLER,

Italy). The specific FMDV serotypes targeted included, A, O, South African Territory [SAT]

1 and SAT 2, known to be endemic in Kenya. Data on putative risk factors for transmission

of

the targeted pathogens were collected for each sampled herd using a household questionnaire. A compartmental baseline model with Susceptible-Exposed-InfectedRecovered

(SEIR) epidemiological classes was also developed to simulate the theoretical transmission of FMDV between cattle and sheep hosts. Thereafter, a series of deterministic SEIR models were fitted using the baseline model framework to evaluate the effects of reactive and pre-emptive vaccination strategies in reducing the cumulative incidence of FMD. The results of the participatory study showed that groups associated wildlife presence with malignant catarrhal fever (MCF), FMD, East coast fever (ECF), African animal trypanosomiasis (AAT), and anthrax, but they did not identify such linkages with goat pox, bovine ephemeral fever (BEF), salmonellosis and bovine cerebral theileriosis (BCT). When data from all sites were combined for impact matrix scoring, MCF, anthrax, FMD, contagious bovine pleuropneumonia (CBPP), ECF and AAT, in decreasing order, were considered to cause the highest economic losses in livestock production. A Kruskal–Wallis test revealed a significant difference in FMD annual prevalence between cattle age groups (p < 0.001) and was the highest in animals > 4 years (median score of 32.5, range; 10-50). FMD had the highest impact on milk production, but in relation to its treatment costs, it was ranked second to CBPP. The overall apparent animal-level seroprevalences of Leptospira spp., Brucella spp and FMD were 23.5% (95% CI; 21.1-26.0), 36.9% (95% CI; 34.1-39.8) and 83.8% (95% CI; 81.5–86.2), respectively. Zones 1 and 2 (closer to the MMNR) had significantly higher seroprevalence of Leptospira spp. than zone 3 (χ2 = 7.0, df = 2, p = 0.029), while for Brucella spp., the seroprevalence was higher in zone 1 than in zones 2 and 3 (χ2 = 25.1, df = 2, p < 0.001). The seroprevalence of FMD was also higher in zone 1 than zones 2 and 3 (χ2 = 116.1, df = 2, p < 0.001). In decreasing order, the overall seroprevalences of FMDV serotypes A, SAT 2, O and SAT 1 were 26.3% (95% CI; 23.5-29.2), 21.4% (95% CI; 18.8-24.0), 21.2% (95% CI; 18.7-23.9) and 13.1% (95% CI; 11.1-15.3), respectively. The distribution of these serotypes differed significantly between zones (p < 0.05) except for SAT 2 serotype (χ2 = 0.90, df = 2, p = 0.639). Both serotypes A and O were more prevalent in zones 1 and 2 than zone 3 (low interface area) while serotype SAT 1, was higher in zone 3 compared to other zones. The results of multivariable analyses identified animal sex (female) and zones (high interface area) as significant predictors (p < 0.05) of animal-level seropositivity of Brucella spp. while for Leptospira spp., important predictors of animal-level seropositivity were animal sex (female), zones (moderate interface area) and herds utilizing a communal grazing reserve. For FMD, animal sex (female), raising of cattle in zones with moderate and high wildlife-livestock interactions; mixing of cattle from multiple herds at watering points and pastoral husbandry practices were all identified as significant predictors of animal-level seropositivity. The results of the vaccination scenario analyses against FMD indicated that both cattle and sheep hosts should be vaccinated in the Maasai Mara ecosystem for vaccination campaigns to have the desired effect on this disease. Reactive vaccination of cattle alone, for example, with 100% coverage reduced the cumulative incidence of FMD by 4.23% and 0.04% in cattle and sheep populations, respectively, while when both host species were targeted, the cumulative incidence of FMD decreased by 4.43% and 2.15% in cattle and sheep hosts, respectively. The results also showed that reactive mass vaccination can substantially reduce the cumulative incidence of FMD if implemented immediately at the onset of the outbreak and with high coverage to compensate for the low vaccine efficacy. The cumulative incidence of FMD, for example, reduced by 3.18% and 1.61% in cattle and sheep populations, respectively, at day 1 of reactive mass vaccination with 75% coverage compared to 0.009% (cattle) and 0.005% (sheep) at day 15 with the same coverage. Overall, pre-emptive mass vaccination was found to be more effective in reducing FMD cumulative incidence than reactive vaccination even when the former was implemented at low coverage. At a low coverage of 25% in both cattle and sheep hosts, pre-emptive vaccination reduced the cumulative incidence of FMD by 75.5% and 10.3% in cattle and sheep populations respectively, when implemented 5 days before the outbreak. This study provides information on disease priorities that occur in the surveyed zones in the Mara ecosystem and which the locals must consider when accessing key ecosystem services such as water and pasture. The seroprevalences of Brucella spp., Leptospira spp. and FMD in cattle were higher in areas with moderate to high wildlife-livestock interactions than those with rare interactions. The observed differences in the seroprevalences of the targeted pathogens between zones can be considered while instituting routine disease control programs.