Dynamics of the Human Infectious Reservoir for Malaria Determined by Mosquito Feeding Assays and Ultra-Sensitive Malaria Diagnosis in Burkina Faso
Background: Plasmodium falciparum gametocytes are essential for malaria transmission. Malaria control measures that aim at reducing transmission require an accurate characterization of the human infectious reservoir.
Methods: We longitudinally determined human infectiousness to mosquitoes and P. falciparum carriage by an ultrasensitive RNA-based diagnostics in 130 randomly selected inhabitants of an endemic area.
Results: At least 1 mosquito was infected by 32.6% (100 of 307) of the blood samples; in total, 7.6% of mosquitoes (916 of 12 079) were infected. The proportion of infectious individuals and infected mosquitoes were negatively associated with age and positively with asexual parasites (P < .001). Human infectiousness was higher at the start of the wet season and subsequently declined at the peak of the wet season (adjusted odds ratio, 0.52; P = .06) and in the dry season (0.23; P < .001). Overall, microscopy-negative individuals were responsible for 28.7% of infectious individuals (25 of 87) and 17.0% of mosquito infections (145 of 855).
Association of P. falciparum Infectiousness With Age and Season In Figure 3, the proportion of infected mosquitoes is presented in relation to Pfs25 mRNA QT-NASBA gametocyte density, age and season. The proportions of infectious individuals were 60.0% (24 of 40), 46.1% (47 of 102), 21.0% (20 of 95), and 12.9% (9 of 70) in children aged <5 or 5–14 years and in adults aged 15–30 or >30 years, respectively. After adjustment for covariates, age, season, and Pfs25 mRNA QT-NASBA gametocyte density remained significant predictors of P. falciparum infectiousness. The proportion of infectious individuals (Table 1) declined over seasons, with significant changes at the peak of the wet season (P = .06; borderline significance) and in the dry season (P < .001) compared with the start of the wet season (Table 3). Similarly, a seasonally dependent decrease in the percentage of infected mosquitoes was observed from the start of the wet season toward the dry season (P < .001), after correction for confounding factors (Table 3).
Conclusions: Our study reveals that the infectious reservoir peaks at the start of the wet season, with prominent roles for infections in children and submicroscopic infections. These findings have important consequences for strategies and the timing of interventions, which need to include submicroscopic infections and be implemented in the dry season.