Despite the relatively high frequency of imported malaria in metropolitan France, the transmission of malaria by transfusion is exceptional. The screening of donations to determine those at risk is performed by an interview, and by the testing of serology for defined groups of donors. However, the exclusion of a candidate 'at risk' as a blood donor, by a pre-donation interview, is not completely mastered and the discrimination by biological examination lacks sensitivity, as much for methodological reasons as for reasons linked to the complex parasitic pathogenic agent (Plasmodium ssp.), as for the specific host defence system. The risk of introducing an unsafe-potentially dangerous (transfusion-transmitted malaria is often lethal)-element into the transfusional circuit is not completely covered. Is serology testing the most adequate test to avoid the risk of infected donations, in particular by Plasmodium falciparum; what are the alternatives and what will be the eventual added-costs of the biological qualification of such donations? The transfusional risk linked to Plasmodium seems, however, to be reduced to a minimum, concerning the circulation of plasma, which could represent an alternative for donors at real risk (rare) and those with a supposed risk (relatively numerous). (C) 2004 Elsevier SAS. Tons droits reserves.