Spatiotemporal Variation in Plasmodium Falciparum Transmission in Selected Sites in Western Kenya

Abstract

Asymptomatic malaria infections are a threat to elimination of this vector-borne infectious disease. In many endemic regions, malaria transmission is seasonal. However, the impact of seasonality on Plasmodium falciparum (P. falciparum) gametocyte levels in peripheral blood and their transmission to local mosquito vectors are not well understood. Data describing these parasitological indices across regions of varying transmission intensity is scanty. In addition, malaria transmission can vary significantly over small geographic scales, but the drivers of this heterogeneity are not well understood. This study evaluated the impact of seasonality on P. falciparum transmission potential, trends in parasitological indices across areas of differential malaria transmission, and factors that might correlate small scale variation in transmission. Blood samples were collected from individuals living in Homa Bay County (low transmission) and Kisumu County (moderate transmission) in the dry season (n=1116) and rainy season (n=1743). In addition, blood samples were collected from approximately 150 individuals in each of 20 clusters in Busia County (high transmission) in rainy season. Blood samples were screened for P. falciparum parasites using quantitative polymerase reaction (qPCR) and microscopy. In Homa Bay and Kisumu the presence and density of blood gametocytes was measured by reverse transcription PCR (RT-qPCR). Differences in parasite and gametocyte densities across seasons were determined by unpaired t-test. Differences in the prevalence, proportion of submicroscopic and gametocyte positive infections across study sites were determined by χ2 test. A generalized linear mixed effect model was used to determine predictors of infections. Potential mosquito larval habitats and their number within 250 m of a household were determined by ArcMap. In Homa Bay and Kisumu, mean parasite densities did not differ in dry versus rainy season (P=0.562). Gametocyte densities were 3-fold higher in the rainy than dry season (rainy: 3.46 transcripts/uL blood, dry: 1.05 transcripts/uL, P<0.001). Parasite prevalence and densities, and gametocyte prevalence and densities were highest in the high transmission region. In contrast, the proportion of asymptomatic submicroscopic infections was highest in the low transmission region. Proportion of gametocyte positive infections did not differ across transmission intensities. In Busia County, across the 20 clusters, 3-folds and 4-folds variation in parasites prevalence by qPCR and microscopy respectively was observed. Three to 34 larval habitats per cluster, and 0-15 habitats within a 250m radius around households were observed. Low altitude, kitchen located indoors, open eaves, a lower level of education of the household head, younger age, and being male were significant predictors of higher prevalence. The number of habitats and their proximity to households was not a predictor for prevalence. In conclusion, parasites increase their investment in transmission in the rainy season, reflected by higher gametocyte densities. Seasonal changes of gametocytemia among infections need to be considered when designing malaria control measures. Pronounced variation in prevalence at small scales and the determinants need to be considered for malaria surveillance and control.