Molecular epidemiology of Theileria parva in cattle, factors associated with the positivity in northern Malawi, and efficacy of traditional medicine.

INTRODUCTION: East Coast fever (ECF), caused by Theileria parva, remains endemic in Malawi. ECF vaccines and chemical treatments are regarded as expensive by small-holder farmers. The aims of this study were to clarify the epidemiology of ECF and to evaluate the efficacy of ethnoveterinary medicine, Synadenium cupulare on T. parva infection. METHODS: A cross-sectional study was conducted in three districts of northern Malawi (Chitipa, Karonga, and Mzimba). A structured questionnaire on feeding style, treatment of ECF, and tick control practices was administered to 79 farmers, and blood samples and animal information were collected from 550 cattle. PCR using p104 gene-specific assay was used to screen for T. parva, and positive samples were subjected to semi-nested PCR using Tp1 and Tp2 primers to characterize their genetic traits. Descriptive statistics were conducted. Global spatial clustering was tested using Moran's I. Univariable and multivariable statistical analyses were performed on farm- and animal-level factors associated with T. parva positivity. RESULTS: The overall farm- and animal-level prevalence was 78.5% (62/79 farms, 95% CI: 67.8-86.9%) and 28.4% (156/550 animals, 95% CI: 24.6-32.3%), respectively. Phylogenetic trees suggested 4 variants of Tp1 and 6 variants of Tp2 circulations among cattle populations in Malawi. Moran's I showed clustering of T. parva negative farms only in Karonga District (p < 0.001). Farm-level T. parva prevalence was significantly higher in the farms conducting extensive grazing (81.9%) than those not (42.9%, p = 0.035). At the animal level, the final multivariable generalized linear mixed-effects model included age and treatment types. For age, the odds of T. parva positivity were significantly lower in adults (odds ratio [OR] = 0.47, p = 0.003) compared with calves, but not for weaners (OR = 0.78, p = 0.316). Regarding treatment, the odds of T. parva positivity were significantly higher in untreated cattle with symptoms of ECF (OR = 2.38, p = 0.027) compared with asymptomatic cattle, but no significant increase was observed for those treated with oxytetracycline (OR = 0.99, p = 0.971) or S. cupulare (OR = 0.59, p = 0.133). DISCUSSION: Future ECF control should integrate traditional ethnoveterinary medicine for calves.