Buffalo, bush meat, and the zoonotic threat of brucellosis in Botswana

Abstract

Background: Brucellosis is a zoonotic disease of global importance infecting humans, domestic animals, and wildlife. Little is known about the epidemiology and persistence of brucellosis in wildlife in Southern Africa, particularly in Botswana.

Methods: Archived wildlife samples from Botswana (1995-2000) were screened with the Rose Bengal Test (RBT) and fluorescence polarization assay (FPA) and included the African buffalo (247), bushbuck (1), eland (5), elephant (25), gemsbok (1), giraffe (9), hartebeest (12), impala (171), kudu (27), red lechwe (10), reedbuck (1), rhino (2), springbok (5), steenbok (2), warthog (24), waterbuck (1), wildebeest (33), honey badger (1), lion (43), and zebra (21). Human case data were extracted from government annual health reports (1974-2006).

Findings: Only buffalo (6%, 95% CI 3.04%-8.96%) and giraffe (11%, 95% CI 0-38.43%) were confirmed seropositive on both tests. Seropositive buffalo were widely distributed across the buffalo range where cattle density was low. Human infections were reported in low numbers with most infections (46%) occurring in children (<14 years old) and no cases were reported among people working in the agricultural sector.

Conclusions: Low seroprevalence of brucellosis in Botswana buffalo in a previous study in 1974 and again in this survey suggests an endemic status of the disease in this species. Buffalo, a preferred source of bush meat, is utilized both legally and illegally in Botswana. Household meat processing practices can provide widespread pathogen exposure risk to family members and the community, identifying an important source of zoonotic pathogen transmission potential. Although brucellosis may be controlled in livestock populations, public health officials need to be alert to the possibility of human infections arising from the use of bush meat. This study illustrates the need for a unified approach in infectious disease research that includes consideration of both domestic and wildlife sources of infection in determining public health risks from zoonotic disease invasions.

Figure 1. Mapped distribution of survey results for brucellosis antibodies among buffalo in Botswana in relation to average cattle counts (orange color ramp) and average buffalo counts (blue color ramp; Inset A); Inset B illustrates the distribution of FPA positive animals identified in this study (red squares) relative to historical samples screened by Cooper and Carmichael (1974; purple boxes).

Green shaded areas are gazetted as conservation land use such as national parks, reserves, and wildlife management areas.

Figure 2. Estimated buffalo population numbers for Northern Botswana (1988–2006) from dry season aerial surveys conducted by the Department of Wildlife and National Parks.

Figure 3. Cattle distribution mapped across Africa corrected to 2000 country estimates from the Food and Agricultural Organization (FAO) Gridded Livestock Production and Health Atlas (GLiPHA) livestock data set (Inset A) .

Changes in cattle density from 2000 to 2005 are illustrated in the Inset B (low to high, blue to red color ramp).

Figure 4. Seasonal changes in total biomass (livestock and wildlife combined) in northeastern Botswana from aerial animal surveys conducted by the Department of Wildlife and National Parks in the dry season (A-1990, B-1999) and wet season (C-1990, D-1999).

Red cells represent increases in total biomass above the annual mean. Blue cells represent decreases below the annual mean. The green line represents the Chobe National Park boundary for reference. Note: the park is not fenced and wildlife populations occur throughout the area at different densities and intensity of overlap with humans.

Figure 5. Map of Botswana showing annual range of greenness (NDVI) from temporal Fourier processed AVHRR satellite data illustrating the extreme variation in vegetation associated with rainfall, particularly in northern Botswana.

Red areas indicate zones with the most dramatic change in vegetation over the year. The green area indicates the location of Chobe National Park (referred to in Figure 4).