ABSTRACT: BACKGROUND: With an Annual Parasite Incidence (API) of 132.1, in the high and moderate risks zones, the Maroni area of French Guiana has the second highest malaria incidence of South-America after Guyana (API = 183.54) and far above Brazil (API = 28.25). Malaria transmission is occurring despite strong medical assistance and active vector control, based on general WHO recommendations. This situation is generated by two main factors that are the social and cultural characteristics of this border area, where several ethnic groups are living, and the lack of understanding of transmission dynamics of the main mosquito vector, Anopheles darlingi. In this context, entomological data collected in two villages belonging to two different ethnic groups of the French border of the Maroni River, were retrospectively analysed to find out how the mosquito bionomics are related to the malaria transmission patterns. METHODS: Data were provided by human landing catches of mosquitoes carried out each month for two years in two villages belonging to two ethnic groups, the Amerindians Wayanas and the Aloukous of African origin. The mosquitoes were sorted by species, sex, date, hour and place of collection and processed for Plasmodium sp. parasite detection. The data were compiled to provide the following variables: human biting rates (HBR), parity rates (PR), numbers of infective bites (IB), entomological inoculation rates (EIR) and numbers of infected mosquitoes surviving enough to transmit (IMT). Spatial and temporal differences of variables between locations and during the night were tested by the Kruskall-Wallis analysis of variance to find out significant variations. RESULTS: The populations of the main mosquito vector An. darlingi showed significant variations in the spatial and temporal HBR/person/night and HBR/person/hour, IB/person/month and IB/person/hour, and IMT/village/night and IMT/village/hour. In the village of Loca (Aloukous), the IMT peaked from June to August with a very low transmission during the other months. The risks were higher during the first part of the night and an EIR of 10 infective bites per person and per year was estimated. In the village of Twenke (Wayanas), high level of transmission was reported all year with small peaks in March and October. The risk was higher during the second part of the night and an EIR of 5 infective bites per person and per year was estimated. CONCLUSION: For the first time in the past 40 years, the mosquito bionomics was related to the malaria transmission patterns in French Guiana. The peak of malaria cases reported from August to October in the Maroni region is concomitant with the significant peak of An. darlingi IMT, reported from the village of Loca where transmission is higher. However, the persistent number of cases reported all year long may also be related to the transmission in the Amerindian villages. The An. darlingi bionomics for these two close populations were found significantly different and may explain why a uniform vector control method is inadequate. Following these findings, malaria prevention measures adapted to the local conditions are needed. Finally, the question of the presence of An. darlingi sub-species is raised.
