Tissue Factor, Its Inhibitor, and the Thrombogenicity of Two New Synthetic Membranes

Coronary atherosclerotic-plaque thrombosis is a key event in the pathogenesis of unstable angina and myocardial infarction. Although plaque rupture or fissuring frequently occurs in atherosclerosis, only a small proportion of ruptured plaques develop thromboses. Tissue-factor antigen and activity were measured in atherectomy samples from 50 consecutive patients with coronary artery disease (stable angina n = 19, unstable angina n = 24, and myocardial infarction n = 7). Median tissue-factor antigen and activity concentrations were significantly higher in plaques from patients with unstable angina and myocardial infarction than in those from patients with stable angina (antigen: 66.1 pg/mg [interquartile range 43.8-82.5] vs 32.4 pg/mg [9.8-43.4], p = 0.0001; activity: 0.22 mU/mg [0.17-0.41] vs 0.13 mU/mg [0.05-0.16], p = 0.0004). Tissue-factor, an initiator of the coagulation cascade, may account for the different thrombotic responses to the rupture of human coronary atherosclerotic plaques.

Tissue factor pathway inhibitor (TFPI) is a multivalent, Kunitz-type plasma proteinase inhibitor that regulates tissue factor-induced coagulation. TFPI directly inhibits activated factor X and, in a factor Xa-dependent fashion, produces feedback inhibition of the factor VIIa/tissue factor catalytic complex. The properties of this rediscovered inhibitor appear, at least in part, to explain the clinical requirement for both the extrinsic and intrinsic pathways of the cascade and waterfall theories of blood clotting and have led to a reformulation of the coagulation mechanism. In the revised hypothesis, factor VIIa/tissue factor is responsible for the initiation of coagulation, but owing to TFPI-mediated inhibition, sustained hemostasis requires the persistent and amplified procoagulant action of intrinsic factors VIII, IX, and XI.

The kallikrein-kinin system activation by contact with a negatively charged surface has been promulgated to be responsible for hypersensitivity reactions. However, to explain the low frequency and heterogeneity of hypersensitivity reactions, we hypothesized that not only the electronegativity of the membrane, but also other physicochemical parameters could influence the activation of the contact phase system of plasma assessed by the measurement of kallikrein activity and bradykinin concentration. Plasma kallikrein activity using chromogenic substrate (S2302) and plasma bradykinin concentration (enzyme immuno assay) were measured during the perfusion of human plasma (2.5 ml/min) through minidialyzers mounted with six different membranes [polyacrylonitrile (PAN) from Asahi (PANDX) and from Hospal (AN69), polymethylmethacrylate (PMMA) from Toray, cellulose triacetate (CT) from Baxter, cuprophane (CUP) from Akzo and polysulfone (PS) from Fresenius]. A direct relationship was shown between the electronegativity of the membrane assessed by its zeta potential and the activation of plasma during the first five minutes of plasma circulation. With the AN69 membrane, the detection of a kallikrein activity in diluted plasma but not in undiluted samples confirmed the importance of a protease-antiprotease imbalance leading to bradykinin release during the first five minutes of dialysis. With PAN membranes, the use of citrated versus heparinized plasma and the use of various rinsing solutions clearly show a dramatic effect of pH on the kallikrein activity and the bradykinin concentration measured in plasma. Finally, increasing the zeta potential of the membrane leads to a significant increase of plasma kallikrein activity and bradykinin concentration. Our in vitro experimental approach evidences the importance of the control of these physicochemical factors to decrease the activation of the contact system.