Despite the effectiveness of mass distribution of long-lasting insecticidal nets (LLINs) in reducing malaria transmission in Africa, in hyperendemic areas such as Burkina Faso the burden of malaria remains high. We here report the results of a 4-month survey on the feeding habits and Plasmodium infection in malaria vectors from a village in Burkina Faso one year following a national LLIN distribution programme. Low values of human blood index (HBI) observed in the major malaria vectors in the area (Anopheles coluzzii: N = 263, 20.1%; An. arabiensis: 5.8%, N = 103) are consistent with the hypothesis that LLINs reduced the availability of human hosts to mosquitoes. A regression meta-analysis of data from a systematic review of published studies reporting HBI and sporozoite rates (SR) for An. gambiae complex revealed that the observed SR values (An. coluzzii: 7.6%, N = 503; An. arabiensis: 5.3%, N = 225) are out of the ranges expected based on the low HBI observed. We hypothesize that a small fraction of inhabitants unprotected by bednets acts as a “core group” repeatedly exposed to mosquito bites, representing the major Plasmodium reservoir for the vectors, able to maintain a high risk of transmission even in a village protected by LLINs.

Background: Despite the overall major impact of long-lasting insecticide treated nets (LLINs) in eliciting individual and collective protection to malaria infections, some sub-Saharan countries, including Burkina Faso, still carry a disproportionately high share of the global malaria burden. This study aims to analyse the possible entomological bases of LLIN limited impact, focusing on a LLIN-protected village in the Plateau Central region of Burkina Faso. Methods: Human landing catches (HLCs) were carried out in 2015 for 12 nights both indoors and outdoors at different time windows during the highest biting activity phase for Anopheles gambiae (s.l.). Collected specimens were morphologically and molecularly identified and processed for Plasmodium detection and L1014F insecticide-resistance allele genotyping. Results: Almost 2000 unfed An. gambiae (s.l.) (54% Anopheles coluzzii and 44% Anopheles arabiensis) females landing on human volunteers were collected, corresponding to a median number of 23.5 females/person/hour. No significant differences were observed in median numbers of mosquitoes collected indoors and outdoors, nor between sporozoite rates in An. coluzzii (6.1%) and An. arabiensis (5.5%). The estimated median hourly entomological inoculation rate (EIR) on volunteers was 1.4 infective bites/person/hour. Results do not show evidence of the biting peak during night hours typical for An. gambiae (s.l.) in the absence of bednet protection. The frequency of the L1014F resistant allele (n = 285) was 66% in An. coluzzii and 38% in An. arabiensis. Conclusions: The observed biting rate and sporozoite rates are in line with the literature data available for An. gambiae (s.l.) in the same geographical area before LLIN implementation and highlight high levels of malaria transmission in the study village. Homogeneous biting rate throughout the night and lack of preference for indoor-biting activity, suggest the capacity of both An. coluzzii and An. arabiensis to adjust their host-seeking behaviour to bite humans despite bednet protection, accounting for the maintenance of high rates of mosquito infectivity and malaria transmission. These results, despite being limited to a local situation in Burkina Faso, represent a paradigmatic example of how

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.