SummaryThe proteins of fibrinolysis have been quantified in human thrombi, to assess the balance between plasminogen activators and their major inhibitor PAI-1. The relative roles of PAI-1 and α2-AP were also examined since we have previously shown that both platelet PAI-1 and plasma α2-AP are important determinants of clot lysis in vitro. Extracts and sections were prepared from human thrombi for quantitative immunoassay and immunohistochemical staining respectively. PAI-1 and α2-AP were present at high concentrations. Levels of t-PA and t-PA-PAI-1 complex were relatively low. Staining confirmed the presence of abundant PAI-1, associated primarily with platelet material within the thrombus and also with fibrin. Staining for α2-AP was also intense and demonstrated strong association with fibrin. The α2-AP concentration was similar to its high plasma concentration, whereas PAI-1 levels were up to 30 times greater than that in circulating blood, suggesting that active recruitment of platelets contributes to the high PAI-1 concentration in thrombi.

No abstract

No abstract

SummaryWe have previously shown that whole blood clots prepared under static conditions are a poor model for human thrombi formed in vivo. The most striking contrast is in the PAI-1 content, some 100 times lower in static clots than in human thrombi. This study aimed to evaluate thrombi formed in an artificial circulation (Chandler loop) using whole blood augmented with platelets. Citrated blood was recalcified and placed in tubing, which was sealed to form a closed loop and circulated. Immunohistochemical staining revealed morphology very similar to human thrombi formed in vivo, comprising dense platelet-rich heads and fibrin-rich tails. Strong positive staining for PAI-1 was observed in fibrin-rich areas of both the head and the tail. The PAI-1 content of the thrombi increased significantly on augmentation of blood with isolated platelets, and reached levels comparable with those in human thrombi. These Chandler thrombi thus provide an appropriate model system for the study of thrombus lysis, in contrast to static blood clots.

Three consecutive patients with acute promyelocytic leukaemia who presented with severe haemorrhagic syndromes were studied and the findings contrasted with those of two patients with classical defibrination after electroshock or complicated labour. The leukaemic patients showed no depletion of fibrinogen. There was no evidence of disordered thrombin generation by either intrinsic or extrinsic pathway sufficient to account for their haemorrhage. All, however, showed strikingly enhanced fibrinolytic activity, which could have accounted for bleeding. This fibrinolytic disorder was characterized by free u-PA in the plasma and differed from that seen after classical defibrination, where free t-PA was observed. U-PA was found also in malignant promyelocytes, which may be the source of u-PA activity in the patients' plasma. Bleeding in promyelocytic leukaemia may be primarily a fibrinolytic disorder.