High seroconversion rate to Rift Valley fever virus in cattle and goats in far northern KwaZulu-Natal, South Africa, in the absence of reported outbreaks

Abstract

Background: Rift Valley fever (RVF) is a mosquito-borne zoonotic disease characterized in South Africa by large epidemics amongst ruminant livestock at very long, irregular intervals, mainly in the central interior. However, the presence and patterns of occurrence of the virus in the eastern parts of the country are poorly known. This study aimed to detect the presence of RVF virus (RVFV) in cattle and goats in far northern KwaZulu-Natal province and to estimate the prevalence of antibodies to the virus and the incidence rate of seroconversion.

Methodology: Cross-sectional studies were performed in communally farmed cattle (n = 423) and goats (n = 104), followed by longitudinal follow-up of seronegative livestock (n = 253) 14 times over 24 months, representing 160.3 animal-years at risk. Exposure to RVFV was assessed using an IgG sandwich ELISA and a serum neutralization test (SNT) and seroconversion was assessed using SNT. Incidence density was estimated and compared using multivariable Poisson models and hazard of seroconversion was estimated over time.

Principal findings: Initial overall seroprevalence was 34.0% (95%CI: 29.5-38.8%) in cattle and 31.7% (95%CI: 22.9-41.6%) in goats, varying by locality from 18-54%. Seroconversions to RVFV based on SNT were detected throughout the year, with the incidence rate peaking during the high rainfall months of January to March, and differed considerably between years. Overall seroconversion rate in cattle was 0.59 per animal-year (95% CI: 0.46-0.75) and in goats it was 0.41 per animal-year (95% CI: 0.25-0.64), varying significantly over short distances.

Conclusions/significance: The high seroprevalence in all age groups and evidence of year-round viral circulation provide evidence for a hyperendemic situation in the study area. This is the first study to directly estimate infection rate of RVFV in livestock in an endemic area in the absence of reported outbreaks and provides the basis for further investigation of factors affecting viral circulation and mechanisms for virus survival during interepidemic periods.

Fig 1. A map of the study area showing major rivers and temporary wetlands, with the locations of the nine diptanks where cattle were sampled and the area where goats were sampled for the cross-sectional study.

The map was constructed for this manuscript in Esri ArcGIS 10.2 using country boundaries from Esri ArcGIS Online, diptank coordinates collected during the study, and river [39], wetland [40] and protected area boundary [41] data available under a Creative Commons Attribution (CC BY 4.0) license.

Fig 2. Actual monthly rainfall recorded at Ndumo Game Reserve (NGR) in the study area, with 30-year average.

Totals above the bars indicate rainfall for each year (July—June). The blue arrow indicates the study period (June 2016—June 2018).

Fig 3. Rift Valley fever virus seroprevalence in cattle and goats in far northern KwaZulu-Natal determined using IgG ELISA and serum neutralization test: risk surfaces created by interpolation using ordinary kriging.

The map was constructed for this manuscript in Esri ArcGIS 10.2 using country boundaries from Esri ArcGIS Online, diptank coordinates collected during the study, and river [39], wetland [40] and protected area boundary [41] data available under a Creative Commons Attribution (CC BY 4.0) license.

Fig 4. Rift Valley fever virus seroconversion rate in cattle and goats in far northern KwaZulu-Natal between June 2016 and June 2018, determined by serum neutralization test: Risk surfaces created by interpolation using ordinary kriging.

The map was constructed for this manuscript in Esri ArcGIS 10.2 using country boundaries from Esri ArcGIS Online, diptank coordinates collected during the study, and river [39], wetland [40] and protected area boundary [41] data available under a Creative Commons Attribution (CC BY 4.0) license.

Fig 5. Seroconversion rate of cattle to Rift Valley fever virus in far northern KwaZulu-Natal between June 2016 and June 2018, determined by serum neutralization test and expressed as numbers of seroconversions per animal-year, plotted using the derivative of the kernel-smoothed Nelson-Aalen cumulative hazard estimator.

Fig 6. Seroconversion rate of goats to Rift Valley fever virus in far northern KwaZulu-Natal between February and December 2017, determined by serum neutralization test and expressed as numbers of seroconversions per animal-year, plotted using the derivative of the kernel-smoothed Nelson-Aalen cumulative hazard estimator.