Patients presenting with non thrombocytopenic mucocutaneous haemorrhages (MHC) of hereditary nature are usually difficult to diagnose. It is widely accepted that most of these patients have a mild disorder of primary haemostasis, most frequently, type 1 von Willebrand disease (vWD) and platelet aggregation/ secretion defects (PASD). However, many patients presenting with clinically significant MCH have no demonstrable alterations in plasma von Willebrand factor and ex vivo platelet function. On the other hand, it is common to observe abnormalities in these tests in subjects with no abnormal haemorrhages. These observations suggest that the pathogenesis of the bleeding disorder in many of these patients is still unknown. We hypothesize that the concentration of other haemostatic factors influence or modulate the clinical expression of MCH, the bleeding symptoms resulting from the interactions and final balance among pro- and anticoagulant factors. Specifically, we propose that the overall activity of the fibrinolytic system is enhanced in patients as compared with controls, and that this hyperactivity contributes to bleeding. Among the determinants of fibrinolytic activity, in this project we propose to analyze the influence of plasma procarboxypeptidase U (proCPU, TAFI) levels and activity in patients consulting for MCH. Carboxypeptidase U, a recently discovered exopeptidase, circulates in plasma as an inactive proform, proCPU. Thrombin activates CPU, and the active form attenuates fibrinolysis by cleaving C-terminal lysine residues exposed on partially degraded fibrin by the action of plasmin. Theoretically, the lower the CPU activity, the higher the rate of fibrinolysis and thus the speed of fibrin clot degradation and removal. Thus, severely suppressed proCPU activation or concentration might be associated with a tendency to bleed, and accordingly, we propose that patients with MCH have a lower proCPU concentration and/or activity. In this context, low plasma proCPU would be considered a risk factor for haemorrhages and, as such, would potentiate the bleeding tendency in patients with vWD and PASD and high plasma proCPU would attenuate their bleeding tendency. Moreover, we hypothesize that patients with MCH, but with no specific diagnosis and those with prolonged bleeding time have lower plasma levels of proCPU than non-bleeder controls and patients with normal bleeding time.The prospective, controlled study, will be performed in patients consulting for MCH, which will be classified in different diagnostic categories (vWD, PASD, vWD+PASD, and patients with no specific diagnosis). Age and sex-matched healthy individuals, with no history of personal or familial bleeding will be used as controls. In all of them, plasma proCPU will be measured by antigenic and functional assays.