Safety of food additives.

Leading Arties CLJONL in the aetiology of thrombosis.Experimental evidence casts doubt on the possibility that stasis alone produces massive venous thrombosis, for blood usually remains fluid in vessels with a normal endothelial lining when they are isolated from the general circulation.Conversely, the formation of a throm- bus in static blood is observed in the presence of active intermediates of coagulation,3 whether they are produced locally in response to an alteration of the endothelium or at a remote site before they are cleared from the blood.Laboratory tests for the coagulation intermediates in the blood, and hence of a potentially "hypercoagulable" state,4 are imperfect.It has long been suspected that activation may occur in a small part of the circulation and yet produce little evidence of its existence in conventional tests applied to blood collected from an antecubital vein.The studies reported by Dr. John Bonnar and his colleagues in the B.M.j. this week at page 564 are important in offering a direct comparison of platelet, coagulation, and fibrinolytic measurements in blood collected simultaneously from the uterine vein and from a peripheral vein during caesarean section.Qualitatively similar but quantitatively much greater changes were found in the blood from the uterine vein.This confirms the interpretation of the remote changes as evidence for the stimulation of coagulation and fibrinolysis locally in the uterus and throws light on the clearance mechanism at work.The early shortening of the clotting time, kaolincephalin time, and partial thromboplastin time is related to the ap- pearance of increased factor VIII levels in the plasma, perhaps due to the activation of this factor by traces of thrombin.5This is followed by a fall in the concentration of platelets, coagulation, and fibrinolytic factors, as they are consumed more rapidly than they can be replaced.In each case pro- nounced changes in uterine blood are almost obscured in peripheral blood.Later, a rise is observed in the concentration of those factors which had fallen; the increased utilization could perhaps have stimulated an increased production of the factors.The study underlines the significance to be attached to raised levels of factor VIII in one-stage assays on peripheral blood as evidence for the existence of the local stimulation of coagulation somewhere in the body.6Evidence is presented to show there is simultaneous activation of coagulation and fibrinolysis within the uterus.Greater concentrations of the fibrin-degradation products of fibrinolysis in the uterine vein than in the arterial circulation have been observed previously at caesarean section.7The two enzyme systems responsible for the presence of fibrin in the body are closely related and appear to be activated by similar stimuli.8The steady accu- mulation of fibrin degradation products up to 10 to 14 days suggests that the stimulus to coagulation continues for relatively long periods.If active intermediates continue to circulate, they could contribute to the incidence of thrombosis in the puerperium.These observations do not diminish the importance of