Relationships of malaria morbidity with exposure to Plasmodium falciparum in young children in a highly endemic area.

To study incidence of clinical Plasmodium falciparum malaria in relation to exposure to parasites, attendance of children less than eighteen months old at a village dispensary in a highly endemic area of Tanzania was recorded. Entomologic inoculation rates (EIRs), estimated as a function of time period and place of residence, exceeded one sporozoite positive bite per adult per night in some village neighborhoods during the wet season. Incidence of clinical P. falciparum malaria, defined either as fever with parasitemia or as fever with hyperparasitemia, increased with the EIR over the whole range of exposures. Each 10-fold increase in the EIR corresponded to a 1.6- fold increase in incidence of fever plus parasitemia (95% confidence interval 5 1.4-2.0). Therefore reduction of human-vector contacts will probably reduce morbidity incidence even at very high exposures. Incidence showed little relationship to estimated cumulative numbers of inoculations since birth, but decreased steeply with estimated cu- mulative time infected with trophozoites. This suggests that clinical immunity depends mainly on the extent of exposure to blood-stage antigens, not on the diversity of inocula seen, and thus temporary reductions in human-vector contacts are unlikely to result in subsequent increases in morbidity. Reducing human mosquito contact by using insecticide- impregnated mosquito bed nets is becoming an important component of malaria control strategies in sub-Saharan Af- rica. Such interventions reduce exposure to infectious mos- quito bites and their effectiveness therefore depends on the existence of a relationship between exposure to mosquito bites and morbidity and mortality. Impregnated bed nets have been shown to reduce both malaria morbidity 1-3 and mortality 4-6 in sites covering a range of different malaria endemicities. However, at the very highest exposures the re- lationships of morbidity and mortality to the frequency of bites from sporozoite-infected mosquitoes (the entomologic inoculation rate (EIR) 7 ) remain uncertain. 8 Children in a highly endemic area acquire clinical im- munity to malaria during the first few years of life. This immunity has been hypothesized to be the result of continual exposure to poorly immunogenic, conserved epitopes on asexual stages of the parasite. Evidence for this was the re- sults of artificial challenges during malaria therapy, which conferred partial clinical protection both to homologous and heterologous inocula. 9 Passive transfer studies have also demonstrated also that sera from west Africans can confer protection against east African P. falciparum parasites. 10 In contrast to this view, the slow pace at which anti-parasitic mechanisms develop has been used to suggest that plasmo- dial populations comprise many antigenically distinct geno- types, that clinical immunity is largely specific to genotypes, and that the child becomes immune as he or she is exposed sequentially to different genotypes. 11-14 This would imply that reduction in exposure by the use of impregnated bed nets could interfere with the acquisition of clinical immunity. Even when a short-term (e.g., 12-year follow-up) study in- dicated that the nets were protective, longer-term effects might be less favorable. 15