Yvonne Stirling

Haemostatic function and ischaemic heart disease: principal results of the Northwick Park heart study

The Northwick Park Heart Study (NPHS) has investigated the thrombotic component of ischaemic heart disease (IHD) by the inclusion of measures of haemostatic function. Among 1511 white men aged between 40 and 64 at the time of recruitment, 109 subsequently experienced first major events of IHD. High levels of factor VII coagulant activity and of plasma fibrinogen were associated with increased risk, especially for events occurring within 5 years of recruitment. These associations seemed to be stronger than for cholesterol, elevations of one standard deviation in factor VII activity, fibrinogen, and cholesterol being associated with increases in the risk of an episode of IHD within 5 years of 62%, 84%, and 43% respectively. Multiple regression analyses indicated independent associations between each of the clotting factor measures and IHD but not between the blood cholesterol level and IHD incidence. The risk of IHD in those with high fibrinogen levels was greater in younger than in older men. Much of the association between smoking and IHD may be mediated through the plasma fibrinogen level. The biochemical disturbance leading to IHD may lie at least as much in the coagulation system as in the metabolism of cholesterol.

HÆMOSTATIC FUNCTION AND CARDIOVASCULAR DEATH: EARLY RESULTS OF A PROSPECTIVE STUDY

Abstract Components of the haemostatic system which may be involved in the pathogenesis of ischaemic heart disease (IHD) were measured in the Northwick Park Heart Study. Of 1510 white men aged 40-64 at recruitment, 49 have since died. 27 died from cardiovascular disease (IHD in all but 3), 18 from cancer, and 4 from other causes. The mean recruitment levels of factor VIIc, factor VIIIc, and fibrinogen were significantly higher in those who died of cardiovascular disease than in those who survived. The independent associations of factor VIIc and fibrinogen with cardiovascular death were at least as strong as the association of blood cholesterol with cardiovascular death. A clustering of two or three high clotting-factor values (factor VIIc, factor VIIIc, and fibrinogen) was present at recruitment in 63% of those who died of cardiovascular disease, compared with 23% of those who survived. The clotting-factor results appeared to be specific for cardio- vascular disease: there was no evidence that high levels of factor VIIc, factor VIIIc, and fibrinogen were associated with death from cancer. The general epidemiology of fac- tor VIIc, factor VIIIc, and fibrinogen is consistent with their having a role in the pathogenesis of IHD.

EFFECTS OF CHANGES IN SMOKING AND OTHER CHARACTERISTICS ON CLOTTING FACTORS AND THE RISK OF ISCHAEMIC HEART DISEASE

The Northwick Park Heart Study (NPHS) has demonstrated associations of high levels of factor VII coagulant activity (VIIc) and of plasma fibrinogen concentration with the risk of subsequent ischaemic heart disease (IHD). In cross-sectional data from the 2023 white men in NPHS, lifetime duration of smoking was a determinant of initial plasma fibrinogen levels. Fibrinogen levels had apparently begun to fall soon after smoking was discontinued but it was over 5 years before they had returned to levels found in life-long non-smokers. In prospective data, smoking cessation and the adoption or resumption of smoking were associated with a decrease or an increase, respectively, of about 0.15 g/l in plasma fibrinogen. These changes would lower or raise the risk of IHD by about 20%. A switch from cigarettes to cigars was associated with a large increase in fibrinogen. A substantial part of the relation between smoking and IHD appears to be mediated through the fibrinogen concentration. Following changes in body mass, VIIc rose in those who had given up smoking and fell in those who resumed.

ROLE OF GENETIC VARIATION AT THE FIBRINOGEN LOCUS IN DETERMINATION OF PLASMA FIBRINOGEN CONCENTRATIONS

Three restriction fragment length polymorphisms (RFLPs) of the fibrinogen genes were used in 91 individuals to investigate the role of genetic variation at this locus in the determination of plasma fibrinogen. The strongest association was with a polymorphism detected with the beta-fibrinogen probe and the enzyme BclI. The probe detects two alleles, designated B1 and B2. The individuals with the genotype B1B1 had a mean fibrinogen of 2.74 g/l; those with B2B2 had a mean fibrinogen of 3.69 g/l (a level previously associated with a strongly increased risk of ischaemic heart disease); and those heterozygous for the two alleles, with the genotype B1B2, had a mean of 2.98 g/l. Genetic variation at the fibrinogen gene locus accounted for 15% of the total phenotypic variance in fibrinogen.

Antithrombin III and procoagulant activity: sex differences and effects of the menopause

Among participants in the Northwick Park Heart Study, antithrombin III activity was lower in pre‐menopausal women than in men of the same age. In the women, however, the menopause was associated with a significant increase in antithrombin III, mean levels in these older women then exceeding levels in men of the same age. The occurrence of the menopause was also accompanied by large increases in factor VII coagulant activity, VIIc, and in plasma fibrinogen, these increases being greater in those experiencing a natural menopause than in those whose menopause was artificial. Sex differences in antithrombin III may form part of the explanation for the observed differences between men and women in their experience of ischaemic heart disease (IHD) and also for the contrasting effects of oral contraceptives and of hormone replacement therapy on the risk of thromboembolic disease.

HÆMOSTATIC, LIPID, AND BLOOD-PRESSURE PROFILES OF WOMEN ON ORAL CONTRACEPTIVES CONTAINING 50 µg OR 30 µg ŒSTROGEN

Abstract In 15 women on oral contraceptives containing 30 μg œstrogen, mean values for factors II, VII, and x, fibrinogen, fibrinolytic activity, antithrombin III, cholesterol, and fasting triglycerides were intermediate between values for 63 women on preparations containing 50 μg œstrogen and those for 243 premenopausal women not on oral contraceptives. Mean blood-pressure levels, however, were higher in women on 30 μg than in those on 50 μg preparations. In 28 women on 50 μg preparations containing 3 mg or 4 mg norethisterone, mean values of factor VII, fibrinogen, fibrinolytic activity, cholesterol, fasting triglycerides, and systolic blood-pressure were higher than in 15 women whose preparations contained only 1 mg of norethisterone. A less consistent picture was found in women on 30 μg œstrogen preparations containing either 250 μg (10 women) or 150 μg (5 women) d -norgestrel. It is concluded that 30 μg œstrogen preparations probably result in smaller haemostatic and lipid changes than 50 μg preparations but that they may have a blood-pressure-raising effect attributable to the particular progestagen, d -norgestrel, used in 30 μg preparations. The safety of these 30 μg œstrogen preparations may thus depend partly on the balance between these two sets of effects. It is also concluded that norethisterone may have effects similar to those attributed to œstrogens.

Assay of factor VII activity by two techniques: evidence for increased conversion of VII to αVIIa in hyperlipidaemia, with possible implications for ischaemic heart disease

Factor VII was assayed in healthy adults and pregnant women by a coagulation method (VIIc) and a procedure (VIIt) based upon activation of factor X. Although VIIc and VIIt were highly correlated (r 0.8) they apparently measured different aspects of VII activity. This difference was related to plasma lipid concentrations. Plasma VIIc showed independent positive associations in vivo with VIIt, cholesterol and triglyceride concentrations, but was unaffected by in vitro adjustment of plasma lipoprotein concentrations. The difference between assays might be due to differing reactivities of VII. The VIIc assay measures VII in its in vivo proportions as the single‐chain protein and fully active double‐chain form (αVIIa). In VIIt, all VII is converted to αVIIa before measurement. Thus an increase in VIIc but not VIIt with increasing lipid concentrations reflects an increased proportion of VII as αVIIa, possibly secondary to activation of the contact system. This effect may explain at least part of the increased VIIc and normal VIIt in pregnancy, and the increased VIIc of hyperlipidaemias in general. The relative values of VIIc and VIIt are proposed as a measure of flux within the coagulation system, and as a measure of coagulability in hyperlipidaemia and other states.

Factor VII in an Industrial Population

Factor‐VII levels were measured in members of an industrial population participating in a long‐term prospective study of arterial disease. A modified semi‐automated method with an artificially prepared factor‐VII deficient plasma, one batch of freeze‐dried thromboplastin and one freeze‐dried pooled reference plasma was used throughout. All calculations were carried out on a programmable desk calculator.

CHANGES IN HÆMOSTATIC SYSTEM AFTER APPLICATION OF A TOURNIQUET

In 35 patients undergoing routine orthopaedic operations in which occlusive tourniquets were used there was a pronounced rise in fibrinolytic activity in the systemic circulation which lasted for at least 15 minutes after the release of the tourniquet; this response was seen after operations on both arms and legs. In contrast there was no increase in fibrinolytic activity in the systemic circulation associated with venous occlusion. Neither the application of a tourniquet nor venous occlusion resulted in changes in factors V or VIII, fibrinogen, or platelet-count. The application of a completely occlusive tourniquet might be a simple form of prophylaxis against deep-vein thrombosis and would avoid the disadvantages of using heparin.

An epidemiological study of the haemostatic and other effects of oral contraceptives.

Summary. Factors V, VII and VIII (each determined by biological assay), fibrinogen, platelet count and adhesiveness, and fibrinolytic activity were measured in 234 white pre‐menopausal women, of whom 57 (24%) were on oral contraceptives and 177 (76%) were not. Cholesterol, triglyceride and blood pressure levels were also recorded. In 20 of the women on oral contraceptives, and in an age‐matched group of 20 who were not, prothrombin, factor X, antithrombin III and α2‐macroglobulin levels were determined, and factors VII and VIII were also measured immunologically. For the majority of the variables studied, the differences between those using and not using oral contraceptives were greater in younger than older women; in the case of factor VII (biological assay) and fibrinogen, the differences between the regression slopes on age were statistically significant, and mean values were substantially higher in those on oral contraceptives. There was also a significant difference between regression slopes on age for cholesterol. Mean levels of prothrombin, factors VII (immunological assay) and X, triglycerides and blood pressure were significantly higher, and mean levels of antithrombin III significantly lower, in those on oral contraceptives compared with those not. Overall, fibrinolytic activity was significantly higher in the women on oral contraceptives; this difference was, however, almost entirely due to the greatly increased fibrinolytic activity of the non‐smokers on oral contraceptives, activity in the smokers on oral contraceptives being similar to that of the women not on these preparations. There were no significant differences in mean platelet count or adhesiveness, or in haemoglobin, packed cell volume, uric acid and blood sugar levels. Among the women on oral contraceptives, there was a significant negative correlation between factor VIII and fibrinolytic activity; this was largely due to five women all of blood groups A and B, in whom, besides high factor‐VIII levels and poor fibrinolytic activity, other variables (e.g. fibrinogen) were raised in a direction that might be expected to favour thrombogenesis. It is possible that it is those women whose fibrinolytic activity does not increase in order to compensate for the effects of oral contraceptives on clotting factors, lipids and blood pressure, who are at special risk of thromboembolic episodes. The differential effects of oral contraceptives by age must be borne in mind in evaluating the effects of these preparations on the haemostatic and lipid systems.