TFPI is a multivalent Kunitz-type proteinase inhibitor that directly inhibits FXa and produces FXa-dependent feedback inhibition of the FVIIa–TF complex. It was recently demonstrated that Protein S (PS) plays the role of TFPI cofactor by enhancing the TFPI inhibition of factor Xa in vivo. Approximately 80% of plasma TFPI circulates as a complex with plasma lipoproteins, about 5–20% circulating as free TFPI. Under quiescent conditions, approximately 50–80% of intravascular TFPI is stored in association with the endothelium. Full-length TFPI α carried in platelets constitutes 8-10% of the total amount of TFPI in the blood, corresponding to a quantity comparable to that of soluble full-length TFPI α in the plasma. We searched for resistance to TFPI activity in patients who presented idiopathic venous thrombosis at a young age. Plasma sensitivity to TFPI was evaluated on the basis of diluted prothrombin time (dPT) measured in patients and in control plasma in the presence (W) and absence (Wo) of exogenous TFPI. At the same time, dPT was measured on a reference plasma to establish a normalized ratio termed TFPI NR and defined as (dPT wTFPI/ dPT Wo TFPI) patient or control / (dPT wTFPI/ dPT Wo TFPI) reference plasma. In an initial study, we found that TFPI resistance could be considered as a new coagulation abnormality that could be related to unexplained thrombosis. In a second study, we failed to demonstrate a role of TFPI resistance in patients with venous thrombosis, abnormal TFPI NR being more likely related to the non-respect of preanalytical conditions rather than to an inherited trait. However, in another study, we showed that inherited or acquired PS deficiency was responsible for a TFPI resistance, providing an ex vivo demonstration that PS is the cofactor of TFPI activity. We showed that this TFPI resistance existed throughout pregnancy and that it disappeared when PS returned to normal values after delivery. We evaluated this TFPI resistance as a possible marker of the risk of a gestational vascular complication (GVC) in 72 patients at risk of developing a GVC. TFPI NR did not differ between GVC+ patients (n =15) and GVC– patients (n = 57). High levels of Lipoprotein(a) (Lp(a) have been shown to be an independent risk factor for cardiovascular disease, lowering of these levels not being achievable by any treatment except possibly aspirin. An in vitro study showed that TFPI activity could be inhibited by Lp(a). We did not confirm this TFPI inhibition in vivo in 20 obese patients with coronary insufficiency who had either normal Lp(a) levels (≤ 0.3 g/L; n = 15) or high Lp(a) levels (≥ 0.3 g/L; n = 5) . Moreover, we found no effect of aspirin treatment on Lp(a) whatever the initial level of Lp(a). Haemophilia B patients bleed less than haemophilia A patients. We showed that this difference in bleeding profile could be explained by lower free TFPI levels in haemophilia B patients compared to haemophilia A patients. In an ongoing study, we showed that in haemophilia A patients there was a strong correlation between the different parameters of thrombin generation (TG) and free TFPI. We also showed, in a TG assay performed in platelet-rich plasma (PRP) with a low TF concentration, that LT was sensitive to free TFPI levels whatever the type of haemophilia and whatever theseverity of the disease. We demonstrated that blocking TFPI by an anti-TFPI Antibody (Ab) allows complete correction of the TG profile in PRP. We showed that it is of major importance to perform a TG assay in PRP in order to evaluate the efficacy of anti-TFPI Ab in correcting TG parameters in haemophilia patients