Factors Affecting Malaria Transmission By Vector Mosquito Populations In Western Kenya, With Special Reference To Altitude
Five populations of Anopheles vectors of malaria from different altitudes along a transect in Nyanza Province of western Kenya were studied over a period of two years, January, 1989 through to December, 1990. Each population was sampled by two methods: the pyrethrum-spray-sheet collection (PSC) technique for the day-resting, and the human-bait technique for the biting population (WHO 1975; Service, 1976). The following parameters were then determined: (1) species composition, (2) the relative density of indoor day-resting populations in inhabited houses, (3) the relative density of the biting populations (4) the man-biting rates and biting cycles between 19.00 and 07.00 hours, and for a full 24 hour day cycle, , (5) Blood feeding preferences, (6) parity rates, (7) sporozoite rates, and (8) entomological inoculation rates, i.e. man-biting rate x sporozoite rate. In the PSC technique, a. total of 192 house searches, were carried out by sampling once monthly from eight human-inhabited houses at each altitude. The anophelines were identified morphologically with the aid of keys. Siblings of the An. gambiae complex were separated by the cytogenetic identification of polytene chromosomes as described by Coluzzi & Sabatini (1967). Bloodmeal types were identified by the bloodmeal enzyme-linked immunosorbent assay (ELISA) techniquetxervice, 1986). . The human-bait catches were performed at three different altitudinal sites located at 1219 m (Ahero), 1350 m (Rota) and 1524' m (Oriwo). The protocol involved hourly catches for 12 hours of the night, and on six occasions throughout the 24 hours of day and night, by a team comprising two collectors seated indoors and two others seated outdoors with their legs exposed, and collecting mosquitoes from themselves using test-tubes with the aid of torch lights (WHO, 1975). The data was used for the analysis of the man-biting rates, biting cycles, longevity and parity rates. Sporozoite rates were determined by the ELISA technique described by Burkot et al. 1984 and Wirtz et al. 1985 and also by dissection for comparison; and were used to calculate the entomological inoculation rates. All the parameters were correlated to altitude, seasonal and climatic changes. The anopheline composition along the transect varied in species diversity and reduced in population abundance with rising altitude. Member species of the Anopheles gambiae complex An. gambiae s.s. and An. arabiensis were caught at all altitudes. An. funestus was not found in collections above 2100 m above sea level. At the lower altitude of 1219 m (Ahero), An. arabiensis existed as a homogeneous population, whereas at higher altitudes this species and An. gambiae s.s. existed as sympatric (mixed) populations, with the latter tending to be predominant. Small numbers of An. zeimanni, An. pharaoensis and An. coustani were also captured. The man-biting rates reduced with rising altitude from apeak of 108 bites/man/night for An. arabiensis at 1219 m, through 28.3 bites/man/night in a sympatric population, to less than 0.1 bites/man/night in a homogeneous population of An. gambiae s.s. at 1524 m. The man-biting rate for An. junestus, also reduced from' 69.3 bites/man/night to 65.1 bites/man/night and less than 0.1 bites/man/night at the respective altitudes . Altitude and seasonality affected the parity rates and age composition. An. arabiensis at 1219 m had a lower parity rate than An. gambiae s.s. at higher altitudes. However, the parity rate for An. funestus at 1350 m was higher than that of the same species at 1219 m. Age-composition studies showed that the percentage of age-groups at each gonotrophic cycle differed with altitude, with the duration of the gonotrophic cycle of the three vectors being longer at higher altitude. Although sporozoite rates were bound to differ with species, it was evident that altitude affected sporozoite development in a given species. While the mean sporozoite rate in An. arabiensis from 1219 m was 0.3%, it was significantly higher - 5.3% at 1350 m and 5.5% at 1524 m in its closest relative, An. gambiae s.s. The sporozoite rates in An. funestus also showed an increasing trend, from 1.9% at 1219 m to 4.2% at 1350 m and 4.0% at 1524 m. These differences occurred also in day resting populations, notably An. gambiae s.s. whose sporozoite rates increased from 3.7%',at 1524 m to 5.3% at 1829 m and 12.5% at 2134 m. Irrespective of altitude, no sporozoite infections were detected in An. zeimanni, An. pharoensis and An. coustani. In contrast to the sporozoite rates, the entomological- inoculation rate (EIR) reduced with increasing altitude. The EIR was intermittent and unstable at 1524 m, medium at 1350 m and intense at 1219 m. This phenomenon was consistent with parasitemia rates observed in school children during the same period. A comparison of the three vectors revealed that An. funestus consistently had a higher entomological inoculation rate at each altitude than the other two vectors. Bloodmeal analysis for pooled samples of the three vectors showed that at 1524 m feeding was predominantly on humans in contrast' to both human and bovid feeding exibited at 1350 m and 1219 m. Human feeding by An. arabiensis was only 28.8% in contrast to 63.5-72% for bovid feeding. However, human feeding was above 90% in both An. gambiae s.s. and An. funestus without significant variation in altitude. It is concluded that altitude, besides seasonal, ecological and climatic factors, appeared to have significant effect on malaria epidemiology. In particular, malaria transmission albeit very low and intermittent, occurred also at the high altitudes, formerly known popularly as the "white highlands", contrary to the long held notion that malaria did not occur in these "white highlands" of Kenya. The practical implications of these observations are that malaria control strategies against the mosquito vectors must be carefully selected in relation to altitude and their ecological diversity.