Cathepsin G (Cath G), a serine-protease found in neutrophils, has been reported to have effects that could either facilitate or impede coagulation. Thrombin generation (CAT method) was chosen to study its overall effect on the process, at a plasma concentration (240 nM) observed after neutrophil activation. Coagulation was triggered by tissue factor in the presence of platelets or phospholipid vesicles. To help identify potential targets of Cath G, plasma depleted of clotting factors or of inhibitors was used. Cath G induced a puzzling combination of two diverging effects of varying intensities depending on the phospholipid surface provided: accelerating the process under the three conditions (shortened clotting time by up to 30%), and impeding the process during the same thrombin generation time-course since thrombin peak and ETP (total thrombin potential) were decreased, up to 45% and 12%, respectively, suggestive of deficient prothrombinase. This is consistent with Cath G working on at least two targets in the coagulation cascade. Our data indicate that coagulation acceleration can be attributed neither to platelet activation and nor to activation of a clotting factor. When TFPI (tissue factor pathway inhibitor) was absent, no effect on lag time was observed and the anticoagulant activity of TFPI was decreased in the presence of Cath G. Consistent with the literature and the hypothesis of deficient prothrombinase, experiments using Russel's Viper Venom indicate that the anticoagulant effect can be attributed to a deleterious effect on factor V. The clinical relevance of these findings deserves to be studied.