In vitro analysis of mycobacterial pathogenicity or host susceptibility has traditionally relied on the infection of macrophages, the target cell of mycobacteria, despite difficulties reproducing their antimycobacterial activity. We have employed alternative models, namely whole blood and leukocytes in plasma, from QuantiFERON negative individuals, and performed infections with the pathogenic M. tuberculosis, the less pathogenic M. avium, M. kansasii and M. chelonae and the occasionally pathogenic M. gordonae and M. bovis. The anticoagulant used in blood extraction, heparin or EDTA, had a major influence in the outcome of the infection. Thus, while in the heparinized models a similar number of bacteria were enumerated in the inoculum and after seven days, in the presence of EDTA a killing effect was observed, despite the inhibitory effect of EDTA on cellular functions like the production of cytokines or reactive oxygen species (ROS). A special case was the rapidly growing mycobacteria M. chelonae, that multiplied in heparinized models but was eliminated in models with EDTA. We verified that EDTA is not responsible for the bactericidal effect, but acts as a bacteriostatic agent. Further work will determine whether blood derived models are a better alternative to the classical macrophage.