Search

Both type 2 diabetes (T2DM) and Bβ448Lys variant of fibrinogen are associated with dense fibrin clots, impaired fibrinolysis and increased cardiovascular risk. It was our objective to investigate whether BβArg448Lys adds to vascular risk by modulating fibrin network structure and/or fibrinolysis in diabetes. The primary aim was to study effects of BβArg448Lys on fibrin network characteristics in T2DM. Secondary aims investigated interactions between gender and BβArg448Lys substitution in relation to fibrin clot properties and vascular disease. Genotyping for BβArg448Lys and dynamic clot studies were carried out on 822 T2DM patients enrolled in the Edinburgh Type 2 Diabetes Study. Turbidimetric assays of individual plasma samples analysed fibrin clot characteristics with additional experiments conducted on clots made from purified fibrinogen, further examined by confocal and electron microscopy. Plasma clot lysis time in Bβ448Lys was longer than Bβ448Arg variant (mean ± SD; 763 ± 322 and 719 ± 351 seconds [s], respectively; p<0.05). Clots made from plasma-purified fibrinogen of individuals with Arg/Arg, Arg/Lys and Lys/Lys genotypes showed differences in fibre thickness (46.75 ± 8.07, 38.40 ± 6.04 and 25 ± 4.99 nm, respectively; p<0.001) and clot lysis time (419 ± 64, 442 ± 87 and 517 ± 65 s, respectively; p=0.02), directly implicating the polymorphism in the observed changes. Women with Bβ448Lys genotype had increased risk of cerebrovascular events and were younger compared with Bβ448Arg variant (67.2 ± 4.0 and 68.2 ± 4.4 years, respectively; p=0.035). In conclusion, fibrinogen Bβ448Lys variant is associated with thrombotic fibrin clots in diabetes independently of traditional risk factors. Prospective studies are warranted to fully understand the role of BβArg448Lys in predisposition to vascular ischaemia in T2DM with the potential to develop individualised antithrombotic management strategies.

Type 2 diabetes mellitus (T2DM) increases the risk of coronary thrombosis and both conditions are associated with altered fibrin clot properties. However, the influence of T2DM on fibrin clot properties in patients with coronary artery disease (CAD) remains unclear. We aimed to investigate the influence of T2DM on fibrin clot properties in patients with CAD. Fibrin clot structure and fibrinolysis were investigated in 581 CAD patients (148 with T2DM) using turbidimetric assays, confocal and scanning electron microscopy. Clots made from plasma and plasma-purified fibrinogen were studied, and plasma levels of inflammatory markers were analysed. T2DM patients had increased clot maximum absorbance compared with non-diabetic patients (0.36 ± 0.1 vs 0.33 ± 0.1 au; p=0.01), displayed longer lysis time (804 [618;1002] vs 750 [624;906] seconds; p=0.03) and showed more compact fibrin structure assessed by confocal and electron microscopy. Fibrinogen levels were elevated in T2DM (p< 0.001), but clots made from purified fibrinogen showed no differences in fibrin properties in the two populations. Adjusting for fibrinogen levels, T2DM was associated with C-reactive protein and complement C3 plasma levels, with the former correlating with clot maximum absorbance (r=0.24, p< 0.0001) and the latter with lysis time (r=0.30, p< 0.0001). Independent of fibrinogen levels, females had more compact clots with prolonged lysis time compared with males (all p-values< 0.001). In conclusion, T2DM is associated with prothrombotic changes in fibrin clot properties in patients with CAD. This is related to quantitative rather than qualitative changes in fibrinogen with a possible role for inflammatory proteins.

Patients with diabetes mellitus are at increased risk of cardiovascular morbidity and mortality. Atherothrombosis, defined as atherosclerotic lesion disruption with superimposed thrombus formation, is the most common cause of death among these patients. Following plaque rupture, adherence of platelets is followed by local activation of coagulation, the formation of a cross-linked fibrin clot and the development of an occlusive platelet rich fibrin mesh. Patients with diabetes exhibit a thrombotic risk clustering which is composed of hyper-reactive platelets, up regulation of pro-thrombotic markers and suppression of fibrinolysis. These changes are mainly mediated by the presence of insulin resistance and dysglycaemia and an increased inflammatory state which directly affects platelet function, coagulation factors and clot structure. This prothrombotic state is related to increased cardiovascular risk and may account for the reduced response to antithrombotic therapeutic approaches, underpinning the need for adequate antithrombotic therapy in patients with diabetes to reduce their cardiovascular mortality.

The aim of this study was to determine whether complement C3 is an indicator of coronary artery disease (CAD). We measured plasma C3 and CRP levels in 278 patients undergoing coronary angiography for typical symptoms of CAD and 269 healthy age and sex matched controls. C3 levels were significantly higher in patients compared with controls (1.15 g/l and 0.92 g/l respectively; p<0.001). In the patient group, C3 levels correlated with BMI, fasting glucose, HbA1c, fibrinogen, CRP and HDL in both men and women. CRP levels were also higher in patients compared with controls (1.14 mg/l and 0.86 mg/l respectively; p=0.005) and correlated with markers of the metabolic syndrome. In a logistic regression model including C3, smoking, hypertension, cholesterol and diabetes, C3 was independently associated with CAD with an odds ratio of 3.20 for a 1 SD increase in C3 levels. In contrast, CRP was not independently associated with CAD in a similar regression analysis. In conclusion, both C3 and CRP plasma levels are elevated in patients with symptoms of CAD. However, C3 seems to be a better indicator of CAD than CRP in this study, suggesting that C3 could be an additional marker for risk stratification in atherosclerosis.

The insulin resistance syndrome (IRS) is a clustering of atherothrombotic traits associated with increased vascular risk. We investigated the degree to which the phenotypic correlations between these traits are due to shared genetic and environmental factors. A multivariate genetic analysis was performed in 537 adults from 89 healthy white north European families. All traits showed significant heritability. BMI had significant genetic correlations with fasting insulin, systolic blood pressure (sBP), plasminogen activator activator inhibitor-1 (PAI-1) and fibrinogen and triglyceride. Fasting insulin had a significant genetic correlation with fibrinogen and triglyceride and Factor VII (FVII). Significant genetic correlations were shown between triglyceride and PAI-1, fibrinogen and FVII. PAI-1 and tissue plasminogen activator (t-PA) showed significant genetic correlation with sBP and with each other. Pleiotropy was demonstrated between fibrinogen and PAI-1, t-PA and FVII. Significant environmental correlations were also demonstrated. This study demonstrates pleiotropy between coagulation and fibrinolytic factors. Shared genetic and environmental factors influencing haemostatic, metabolic and anthropometric traits underlie the atherothrombotic nature of the IRS.

The aim of the study was to determine associations of factor VII:C levels in 140 South Asian stroke subjects and 143 first-degree relatives versus age-sex matched 146 control subjects without a personal or a family history of stroke subjects living in UK. There were no significant differences in Factor VII:C levels or FVII Msp I gene polymorphism (Arg-Gln 353) R353Q genotype frequency between the groups. R353Q genotype determined Factor VII:C levels in all the three groups. Factor VII:C levels correlated with triglycerides (patients, r = 0.23; relatives r = 27; control subjects, r = 0.24) and plasminogen activator inhibitor activity (patients, r = 0.30; relatives r = 0.22; control subjects r = 0.20) in all the three groups, but with insulin only in patients (p = 0.19). Circulating levels of Factor VII:C are determined by R353Q genotype and cluster with other risk factors associated with insulin resistance in South Asian ischaemic stroke patients, first-degree relatives and control subjects but are not related to stroke or a family history of stroke.

We aimed to investigate significant correlations of insulin resistance with thrombotic factors in South Asians with stroke. Correlations of Homeostasis Model Assessment (HOMA)(as a surrogate of insulin resistance) were analysed with 6 thrombotic factors in 140 South Asian patients with a history of confirmed (by computerised tomography) ischaemic stroke. Age and sex adjusted HOMA was correlated to waist-hip ratio (r = 0.38, p = 0.0001), triglycerides (r = 0.22, p = 0.03), systolic blood pressure (r = 0.21, p = 0.04), tissue plasminogen activator (t-PA) (r = 0.22, p = 0.04); plasminogen activator inhibitor 1(PAI-1) (r = 0.26, p = 0.02); fibrinogen (r = 0.25, p = 0.02); and factor VII antigen (r = 0.21, p = 0.06). On regression analysis, with HOMA as dependent variable and significant correlates as independent variables in the model, HOMA was independently associated with PAI-1 antigen. There is extensive clustering of metabolic and thrombotic factors with insulin resistance in South Asian patients with ischaemic stroke, which may contribute to high prevalence of vascular disease in this population.

We aimed to investigate whether increased levels of FXIII and/or a low prevalence of the protective Leu allele (of the Val34Leu FXIII polymorphism) occur in relatives of patients with severe CAD. 185 healthy male relatives aged 65 or less were recruited from 125 patients with multi-vessel CAD and compared to 185 healthy, age-matched controls. The relatives and controls were similar in terms of clinical parameters. FXIII B-subunit levels were elevated in relatives, 1.11 microg/mL (1.08-1.14), compared with controls, 1.00 microg/mL (0.97-1.04), P <0.0001 but FXIII A2B2 levels did not differ between the groups. There was a strong correlation between FXIII B-subunit and the insulin resistance syndrome, however, adjusted B-subunit levels remained significantly higher in relatives. There was no difference in genotype frequency at the FXIII Val34Leu polymorphism between relatives and controls. FXIII B-subunit levels are elevated in the relatives of CAD patients and this is independent of other cardiovascular risk factors.

Four common base-change polymorphisms have been found in the von Willebrand factor gene promoter: (-1793 C/G, -1234 T/C, -1185 G/A and -1051 A/G). All four polymorphisms are in strong linkage disequilibrium and recent reports have indicated these polymorphisms are associated with plasma vWF:Ag levels suggesting that one or more of these elements influence regulation of the vWF gene. We report that human endothelial cell-derived trans-acting factors display allelic preferences in binding activity to each polymorphic site. The common A allele variant of the -1051 polymorphism and the rarer A allele variant of the -1185 polymorphism provided specific binding of nuclear proteins. The G allele counterpart of these two variants did not produce any complex formation indicating that the nucleotide substitution at these positions alters the DNA binding ability of nuclear factors. The two alleles of the -1234 polymorphism produced two complexes with a similar migration pattern however stronger binding was found to the common T variant of this allele. Two specific complexes associated with the rarer G allele of the -1793 polymorphism, but only one associated with the C allele. Supershift experiments revealed that the trans-acting factor YYI recognised the slower migrating complex formed on the -1234 T/C and the -1051 A polymorphic sites with a strong binding preference for the -1234 T allele variant. The identification of YY1 as a component of the factors that recognise these elements suggests that this ubiquitous nuclear protein may play a role in the regulation of the vWF promoter.

Possession of the coagulation factor XIII Val34Leu (FXIIIVal34Leu) polymorphism is associated with protection against myocardial infarction (MI) in Caucasians, in the absence of features of insulin resistance. The role of this polymorphism in the UK Asian population, with its high prevalence of insulin resistance and ischaemic heart disease, is unknown. We investigated the frequency of genotypes at this polymorphism, and measures of circulating FXIII in a group of UK Asians attending for coronary angiography. Genotype at the FXIIIVal34Leu polymorphism was not associated with MI. FXIII B-subunit levels correlated with waist: hip ratio (r = 0.19, p <0.005), HbA1c (r = 0.18, p <0.05), fasting triglycerides (r = 0.21, p <0.005), total cholesterol (r = 0.29, p <0.0005) and PAI-1 antigen (r = 0.24, p <0.005). An association between FXIIIVal34Leu and FXIII cross-linking activity was confirmed in these subjects (one-way ANOVA p <0.0005). This evidence does not support the hypothesis that FXIIIVal34Leu is protective against MI in the UK Asian population. FXIII B-subunit levels are strongly linked to risk factors for cardiovascular disease, suggesting an underlying association with insulin resistance.

Increased Factor XIIa concentrations have been found in association with coronary artery disease. Recently, a common 46 C to T point mutation in exon I of the factor XII gene has been described which is associated with lower FXII clotting activity and lower zymogen levels in relation to possession of the T allele. It is not known whether this polymorphism relates to the phenotypes of FXIIa in vivo or to coronary artery disease. The aim of the study was to investigate the interaction of this polymorphism with FXIIa plasma levels and to study the prevalence of the polymorphism in 266 patients with suspected coronary artery disease characterised by angiography and in 185 healthy controls. FXIIa levels were strongly associated with FXII genotype with lower levels with increasing numbers of T alleles (p <0.0001). There was no difference between the prevalence of this polymorphism in patients with M1 compared to those without MI and controls and between all patients and controls (p > or =0.2, chi-square test). There was no association between extent of coronary artery disease (0, 1, 2, and 3 vessel disease) and FXII genotype. In conclusion, the common 46 C to T point mutation is strongly associated with FXIIa but the present study did not show an association with coronary artery disease. The role of this polymorphism in other thrombotic disorders such as ischemic stroke and venous thrombosis and its clinical significance in FXII deficient states remains to be investigated.

Activated blood coagulation factor XIII has an important role in the final stage of the clotting cascade by the covalent crosslinking of alpha- and gamma-fibrin chains. We have recently shown that a functional polymorphism in exon 2. codon 34 of the FXIII A-subunit gene is protective against myocardial infarction. To investigate the prevalence of three other common point mutations in the A-subunit gene (codon 564, C to T, 650 G to A and 651 G to C) and their association with FXIII activity and antigen levels, 275 patients with coronary artery disease and 196 controls were studied. There was no difference in the prevalence of the polymorphisms between patients and controls or between patients with or without MI. Only genotype at codon 564 was associated with FXIII activity with lower activities in subjects possessing the T allele. There was evidence of linkage disequilibrium between codon 34 and codon 564. These results suggest that FXIIIVal34Leu is the only common polymorphism in the coding region of the A-subunit gene of FXIII associated with coronary artery disease.

Raised factor VIII coagulant activity has been associated with Ischaemic Heart Disease and Myocardial Infarction (MI). We measured FVIII:C and FVIII:Ag levels in 288 white European patients characterised for extent of atheroma by coronary angiography and for a history of MI by WHO criteria, and 313 white European healthy controls. FVIII:Ag levels were higher in patients than controls after adjustment for confounding variables (1.17 IU/ml patients, 1.03 IU/ml controls [p <0.0005]), whereas FVIII:C levels were lower in patients (1.26 IU/ml patients, 1.45 IU/ml controls [p<0.0005]). FVIII:Ag levels were higher in patients with MI and with angiographic evidence of atheroma than controls. FVIII:Ag levels but not FVIII:C levels were a significant independent risk factor for MI in multivariate analysis (Odds Ratio 2.40[1.35-4.27] [p = 0.003]). No association of FVIII:Ag or FVIII:C levels and presence or extent of atheroma was found in multivariate analysis. In conclusion, this study found an independent association of FVIII:Ag but not FVIII:C with MI.

The aim of this study was to determine the association of tPA antigen levels with CAD and ischaemic stroke and whether associations are independent of levels of PAI-1 antigen. In subjects with CAD (n = 247) tPA was associated with the number of coronary arteries with > or = 50% stenosis, but this association was lost after adjustment for PAI-1, which was found to be the largest determinant of tPA levels in linear regression models and accounted for as much as 38% of the variation in levels. Levels of tPA were significantly higher in patients with a history of MI compared with those without, even after adjustment for covariates and PAI-1 (MI: 10.0 [9.4-10.6] ng/ml; no MI: 8.9 [8.5-9.4] ng/ml, p = 0.004). In a logistic regression model comparing patients with MI to patients without MI, the odds ratio for tPA levels in the upper quartile compared with the lowest quartile was 2.03 (1.33-3.10). Levels of tPA in subjects with ischaemic stroke (n = 338) were significantly higher than age matched healthy control subjects (n = 366) and again this difference remained after adjustment (patients: 10.4 [9.9-10.9] ng/ml; controls: 9.0 [8.7-9.3] ng/ml, p <0.0001). In a logistic regression model comparing patients with ischaemic stroke to healthy control subjects the odds ratio for tPA in the upper quartile compared with the lowest quartile was 4.23 (3.02-5.92). These data suggest that the associations of tPA with acute thrombosis are independent of levels of PAI-1 but the mechanisms whereby enhanced fibrinolysis may predispose to thrombosis remain unclear.

Women with coronary artery disease (CAD) have a prognosis at least as bad and possibly worse than men. Differences in classical risk factors do not fully account for these findings and there is evidence that circulating levels of haemostatic factors may predict CAD risk. In this study sex differences in haemostatic risk factors were examined in relation to coronary stenosis. 609 (420 men, 69%) subjects admitted for coronary angiography for suspected CAD were recruited. Levels of Factor VII:C (FVII:C), fibrinogen, plasminogen activator inhibitor-1 (PAI-1) and von Willebrand factor (vWF) were estimated in 296 subjects from one centre. Of these, women (n = 107) had higher levels of FVII:C (134% vs 117%, p < 0.0005), and fibrinogen (3.4 g/l vs 3.2 g/l p = 0.01) than men (n = 189) and these differences remained after adjusting for other covariates. In subjects with angiographically significant atheroma these differences in haemostatic factors (n = 50 for women vs n = 147 for men) were exaggerated, (FVII:C 139% vs 117, p < 0.0001, fibrinogen 3.7 g/l vs 3.3 g/l p = 0.003), PAI-1 (26.2 ng/ml vs 19.7 ng/ml, p = 0.02) with a trend towards higher levels of vWF in the women. Women with significant atheroma at angiography (n = 50) had higher levels of PAI-1 (25.0 ng/ml vs 13.4 ng/ml p < 0.0001) and vWF (1.25 IU/ml vs 1.06 IU/ml, p = 0.02) and a trend towards higher levels of both fibrinogen and FVII:C than women with normal or in significant coronary vessel disease (n = 57). Elevated circulating levels of PAI-1, vWF, fibrinogen and FVII:C in women with angiographically proven CAD may contribute to an adverse cardiovascular risk factor profile and the poorer prognosis in females than male patients with proven coronary artery disease.

Factor XIII when activated by thrombin, crosslinks fibrin, however its role in thrombotic disorders is unknown. A common point mutation (G-->T) in exon 2 of the A-subunit gene which codes for an amino acid change three amino acids from the thrombin activation site (Factor XIIIVal34Leu) is a candidate for a role in the pathogenesis of acute myocardial infarction. Factor XIII genotype frequencies were determined in a case-control study of 398 caucasian patients and 196 healthy controls. Patients had undergone angiography for investigation of coronary artery disease and were evaluated for a history of myocardial infarction. The prevalence of the mutation was lower in patients with myocardial infarction than without (32% vs. 50%), p = 0.0009 and than in controls (32% vs. 48%), p = 0.005. Patients possessing the mutation with a history of myocardial infarction had higher PAI-1 concentrations (mean, 27.9 vs. 16.7 ng/ml, p = 0.004) and the PAI-1 4G/4G genotype was commoner (43% vs. 26%, p = 0.03). There was no difference in PAI-1 4G/4G genotype (33% vs. 32%) and PAI-1 levels (mean, 21.0 vs. 20.9 ng/ml) in patients possessing wild type with MI compared to those without MI. These results indicate that the G-->T mutation coding for factor XIIIVal34Leu is protective against myocardial infarction and suggest a mechanism whereby elevated levels of PAI-1 may contribute to vascular risk.

Little data is available regarding the activated form of factor XIIa (FXIIa) in survivors of myocardial infarction. 292 Caucasian patients characterised for extent of coronary atheroma by angiography and for a past history of myocardial infarction and 77 healthy controls were included in the study. To investigate the relationship between coronary artery disease, activated factor XII and other circulating factors, we studied levels of FXIIa, cholesterol, triglycerides, fasting insulin, fibrinogen, FVII:C, t-PA antigen and PAI-1 antigen. Factor XIIa levels were higher in all patients [2.5 (2.3-2.6) ng/ml] and in patients with a history of MI [2.6 (2.4-2.9) ng/ml] than in controls [1.9 (1.7-2.1) ng/ml], p < 0.0001. In patients, FXIIa levels positively correlated with FVII:C, BMI, cholesterol, insulin, PAI-1 antigen, t-PA antigen and triglycerides. In controls FXIIa levels only correlated with PAI-1 antigen and triglycerides. FXIIa levels were strongly associated with extent of coronary stenosis: 2.8 (2.6-3.1) ng/ml and 2.6 (2.3-2.9 ng/ml) in those with 2 and 3 vessels stenosed compared to 2.1 (1.9-2.3) ng/ml in those with 0 vessel stenosed (p = 0.0004). Activated FXII relates to both extent of coronary atheroma and to a past history of myocardial infarction and clusters with features of insulin resistance.

Background: Elevated von Willebrand factor (vWF) is a risk factor in the development of acute myocardial infarction. The importance of vWF and factor VII:C in the pathogenesis of cerebrovascular disease (CVD) is poorly defined., Methods and Results: We studied 208 cases of stroke whose pathological type was defined by cranial computed tomography. Cerebral infarcts were grouped according to the Oxfordshire Community Stroke Project (OCSP) clinical classification. The results in patients were compared with 184 healthy reference subjects. In patients, vWF and FVIII:C levels were determined initially and after three months. Patients were followed prospectively for six months or until death. Levels of vWF and FVIII:C were elevated initially (1.86 IU/ml and 2.20 U/ml respectively) and after 3 months (1.51 IU/ml and 1.90 U/ml) compared with a healthy reference population (1.26 IU/ml and 1.49 U/ml p = 0.0001). In the initial sample, vWF was associated with age (p = 0.01). FVIII:C was related to age (p = 0.04), gender (p = 0.007 higher for females) and a history of diabetes mellitus (2.56 U/ml vs. 2.16 U/ml in non-diabetics, p = 0.008). Initial vWF levels were higher in subjects with large vessel disease (TACI/PACI) group compared with the small vessel disease (LACI) group [2.12 IU/ml, (n = 112) vs. 1.48 IU/ml (n = 59) respectively, p = 0.0001] and similarly in initial FVIII:C levels (2.43 U/ml vs. 1.87 U/ml, p = 0.0001). Analysis of six-month case fatality, vWF levels were associated with risk of death [p = 0.01, OR 1.73 (1.12, 2.66) for an increase of I U/ml], even after allowing for stroke type., Conclusion: The relationship of vWF with stroke mortality has not previously been described. Although we have not demonstrated a causal role for vWF in the pathogenesis of CVD, elevated circulating levels of vWF may be associated with increased risk of death following stroke. A prospective study would be required to establish whether vWF is predictive for the development of CVD.

PAI-2 is a fibrinolytic inhibitor produced predominantly by monocytes. Most PAI-2 is intracellular making study in clinical conditions difficult. Abnormalities in production may be associated with inflammation and fibrinolysis at sites of tissue damage such as the atherosclerotic plaque. PAI-2 gene variants have been described: variant A consists of Asn120, Asn404 and Ser413 and variant B consists of Asp120, Lys404 and Cys413. We designed a PCR-RFLP assay using primers spanning the region containing Asn/Lys404 and Ser/Cys413. Variant B contains an MwoI restriction site. We analysed 302 Pima Indians and 286 healthy Caucasian volunteers. To investigate relationships between genotype and vascular disease we analysed 333 Caucasian patients undergoing coronary angiography. Gene variant B was more common in the Pimas than in Caucasians (p < 0.0001). There was no significant difference in genotype distribution between the volunteers and patients. In the patients there was no association between genotype and either a history of MI or extent of coronary atheroma.

The exact role of the fibrinolytic system in the pathogenesis of stroke remains to be established. Elevated circulating levels of plasminogen activator inhibitor-1, the principle inactivator of the fibrinolytic system, have been related to the development of myocardial infarction. There is evidence that a polymorphism in the promoter region of the PAI-1 gene is associated with circulating PAI-1 levels. We studied a common single nucleotide insertion/deletion (4G/5G) polymorphism by PCR in 558 patients with stroke, the pathological type of which was established by cranial computed tomography, and in 172 controls. 4G/5G genotype and PAI-1 activity were investigated in relation to 1) stroke type and 2) mortality occurring within four weeks, three months and six months of the stroke. No difference in genotype frequency was observed when all cases of stroke were compared with controls nor between the clinically determined subtypes of cerebral infarction. PAI-1 activity was significantly higher in patients with stroke (n = 245) both at presentation (11.6 U/ml) and after three months (11.8 U/ml), in paired samples, than in control subjects (8.8 U/ml, p < 0.0001). Thirty-seven (6.2%), 86 (14.5%) and 122 (20.5%) patients had died within one, three and six months of admission, respectively. PAI-1 activity was independently associated with all-cause mortality at one and three months after stroke (p = 0.02 and p = 0.03 respectively), but not after six months. In this population the 4G/5G promoter polymorphism is not associated with an increased risk of stroke. PAI-1 levels were elevated at the time of acute stroke which persisted after three months. PAI-1 level but not genotype was associated with early mortality following stroke.

Aim: To identify whether levels of PAI-1 are elevated in relatives of NIDDM patients and the extent to which they relate to features of insulin resistance and to genotype at a common PAI-1 promoter polymorphism., Results: In 132 first-degree relatives of NIDDM patients and 151 controls PAI-1 activity was higher in relatives 14.4 U/ml than controls 9.3 U/ml (p < 0.0005) with higher body mass index 27.3 v. 24.7 kg/m2 p < 0.0005, fasting insulin 9.3 v. 7.6 mU/l p < 0.005 and triglyceride 1.4 v. 1.2 mmol/l p < 0.05. PAI-1 activity levels were higher in men and in smokers and showed a trend to being higher with increasing number of 4G alleles at the PAI-1 promoter polymorphism. After adjustment for age, sex, smoking, BMI and triglyceride levels, PAI-1 levels remained 26% higher in the NIDDM relatives (p = 0.01)., Conclusion: In relatives of NIDDM patients PAI-1 levels are elevated, presenting an additional mechanism for their increased cardiovascular risk.

Elevated Factor VII:C (FVII:C) levels are an independent risk factor for acute myocardial infarction. To examine the association between FVII:C levels, genetic and environmental factors in patients with a history suggestive of ischaemic heart disease, 270 patients were studied. FVII:C levels were significantly associated with the Msp I and promoter insertion polymorphisms. FVII:C correlated with environmental factors, and these correlations remained when analysed by genotype. In a multiple regression model, genotype, body mass index, cholesterol, triglycerides and gender remained as independent and significant predictors of FVII:C levels. There was no significant difference in FVII:C concentrations with severity of atheroma, or in patients with a history of myocardial infarction. These findings provide further evidence linking FVII:C with Msp I and promoter insertion genotypes and with environmental factors in established IHD, but suggest there is not a genotype-environment interaction or a relationship between FVII:C levels and severity of ischaemic heart disease.

The macrovascular complications of non-insulin-dependent diabetes mellitus (NIDDM) are related to the features of insulin resistance (IR). High Factor VII:C (FVII:C) levels are associated with increased cardiovascular risk and relate to a base change in the FVII gene detected by Msp I endonuclease, and also to an insertion polymorphism in the promoter region. To examine the association between FVII:C levels, genotype and features of IR, 95 NIDDM patients were studied. Genotype was related to FVII:C levels (M1M1 137%, n = 75; M1M2 and M2M2 114%, n = 20, p < 0.005; AA 136%, n = 71; Aa 119%, n = 21, p < 0.05), which is consistent with previous studies in healthy populations. FVII:C correlated with cholesterol (r = 0.51, p < 0.0005), insulin (r = 0.36, p = 0.002), triglycerides (r = 0.34, p = 0.001), age (r = 0.23, p < 0.005) and body mass index (r = 0.23, p < 0.05). When analysed by Msp I genotype, the stronger predictor of FVII:C levels, these correlations remained, with no difference in regression slopes. In a multiple regression model, genotype, cholesterol, insulin, and gender remained as independent predictors of FVII:C levels. In conclusion, FVII:C concentrations are elevated in NIDDM in relation to both FVII genotypes and features of IR.

Elevated levels of PAI-1 are found in coronary artery disease (CAD) and non-insulin-dependent diabetes (NIDDM). PAI-1 may be involved in the pathogenesis of CAD through suppression of fibrinolysis, alternatively the high levels may result from vascular damage. There is evidence that PAI-1 levels are related to genotype at a PAI-1 promoter polymorphism. Genotype at this 4G/5G polymorphism was determined in 160 NIDDM (90 males and 70 females) patients with (n = 38) or without (n = 122) clinical evidence of CAD. Levels of cholesterol were higher (6.5 vs 5.9 mM, p < 0.01) and PAI-1 tended to be higher (PAI-1 activity 23.0 vs 20.4 U/ml) with CAD. The frequency of the 4G/4G genotype was increased and the 5G/5G genotype decreased, in the group CAD compared to those without (p < 0.05). These results suggest that possession of the 4G/4G PAI-1 promoter genotype is a risk factor for the development of CAD in subjects with NIDDM.

To investigate the interaction of metabolic and genetic factors in relation to PAI-1, genotype was determined at a 4G/5G polymorphism in the PAI-1 gene promoter and a Hind III RFLP of the PAI-1 gene in 189 Caucasian NIDDM patients. PAI-1 levels were equivalent in each genotype group and PAI-1 activity correlated with fasting insulin (r=0.45), triglyceride (r=0.39) body mass index (r=0.44), cholesterol (r=0.17) and glucose (r=0.15). The regression slope (B) of PAI-1 activity on triglycerides was steeper in the 4G/4G group than the other two groups: 4G/4G B = 0.91, r = 0.62; 4G/5G B = 0.36, r=0.27; 5G/5G B=0.31, r=0.29 (difference between slopes p=0.02) and the association between PAI-1 activity and glucose remained only in the 4G/4G group (r=0.35). These results confirm the association of PAI-1 levels with the features of insulin resistance and indicate that the association between PAI-1 levels and both triglyceride and glucose is influenced by genotype in the region of PAI-1 gene promoter.

The effect of insulin-induced hypoglycaemia on plasma coagulant activity was studied in 11 subjects with well-controlled, uncomplicated type 1 diabetes. Thrombin generation was determined in plasma by a computer ex-vivo assisted chromogenic method and by the activated partial thromboplastin time (APTT). In addition, factor VIII:C, thrombin-antithrombin III (TAT) complex and fibrinopeptide A (FPA) levels were measured. Hypoglycaemia induced a rise in mean (SD) factor VIII:C concentrations from a baseline level of 1.13 (0.32) IU/ml to a peak 15 min after onset of symptoms and they remained increased at 90 min [1.54 (0.57) and 1.5 (0.54) IU/ml, p < 0.001 respectively]. A corresponding reduction in time to generate 50% maximal thrombin activity occurred from a pre-insulin value of 56 (6) s to a minimum reading of 46 (7) s at 15 min (p < 0.001) and remained low at 90 min [48 (6) s, p < 0.001]. APTT shortened from 43.3 (4.8) s to 40.1 (4.6) s at 30 min (p < 0.001) but did not fall below the normal range (37.6-42.7 s) and no significant changes in TAT or FPA levels were noted. Factor VIII:C correlated inversely with time to generate 50% maximal thrombin activity and APTT (r = -0.580, p < 0.001; r = -0.673, p < 0.001, n = 66, respectively). The results show that the rise in plasma factor VIII:C levels induced by hypoglycaemia is accompanied by accelerated rates of generation of thrombin in contact-activated plasma, though concentrations of FPA and TAT remain unchanged, although TAT complexes are not a sensitive marker of in vivo thrombin generation.

To identify whether genotype contributes to the difference in PAI-1 levels in type 1 and type 2 diabetic subjects and whether genotype relates to the development of retinopathy, a Hind III restriction fragment length polymorphism and two dinucleotide repeat polymorphisms were studied. In 519 Caucasian diabetic subjects (192 type 1, 327 type 2) and 123 Caucasian control subjects there were no differences in the frequency of the Hind III restriction alleles (type 1 vs type 2 vs control: allele 1 0.397 vs 0.420 vs 0.448; allele 2 0.603 vs 0.580 vs 0.552) nor in the allelic frequency at either dinucleotide repeat sequence. In 86 subjects with no retinopathy at 15 years or more from diagnosis of diabetes and 190 subjects with diabetic retinopathy there was no difference in the frequency of Hind III restriction alleles (retinopathy present vs retinopathy absent: allele 1 0.400 vs 0.467; allele 2 0.600 vs 0.533) nor in the allelic frequencies at either dinucleotide repeat sequence. The results indicate that there is no or minimal influence of the PAI-1 gene on either PAI-1 levels or the development of diabetic retinopathy in patients with diabetes mellitus.

In vitro studies have shown that insulin and IGF-1 releases the fibrinolytic inhibitor plasminogen activator inhibitor-1 (PAI-1) from cells of hepatic origin. To investigate the effects of IGF-1 on fibrinolysis: 1) cultured hepatoma cells were grown in the presence of IGF-1 and media collected for secreted PAI-1 and cells probed for PAI-1 mRNA, 2) 8 hypopituitary patients were treated with recombinant human growth hormone (rhGH) and 3) 5 type 2 diabetic patients were treated with recombinant human IGF-1 (rhIGF-1). Treatment of Hep G2 cells with IGF-1 (1000 ng/ml) increased secretion of PAI-1 from a median value of 80 ng/10(6) cells (range 21-91) to 144 ng/10(6) cells (range 128-169) after 24 h (p < 0.01). Synthesis of PAI-1 mRNA increased in a similar fashion. Treatment of hypopituitary patients with rhGH led to an increase in circulating IGF-1 from a mean value of 166 (range 41-324) ng/ml at baseline to 322 (77-575) ng/ml at 4 weeks and 259 (104-533) ng/ml after 8 weeks (p < 0.02). Despite this, no changes in circulating PAI-1 or fibrinolysis occurred. Type II diabetic patients treated with rhIGF-1 showed an increase in circulating IGF-1 from a mean value of 120 ng/ml (range 109-196), at baseline to 823 ng/ml (585-894) after 5 days. This also was not associated with changes in circulating PAI-1 or in fibrinolysis. The results confirm that IGF-1 induces the synthesis of PAI-1 in Hep G2 cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Regional circulating plasma levels of fibrinolytic activity were assessed in 15 patients undergoing cardiac catheterisation. The euglobulin clot lysis time (ECLT) was longer in the abdominal aorta (AA) than the inferior vena cava (IVC), (median difference -17.5 min, p = 0.008). This was associated with higher inhibition of plasminogen activator activity (PAI) in the AA than IVC, -1.75 IU/ml, p = 0.002. In the venous circulation the ECLT was higher in the peripheral venous sample than in the IVC, -25.5 min, p = 0.003, with higher PAI peripherally than in the IVC, -1.9 IU/ml, p = 0.001. There were no differences in ECLT, PAI, PAI-1:Ag or t-PA:Ag throughout the arterial circulation. These results demonstrate higher fibrinolytic activity with lower inhibitor activity in the venous compared to the arterial circulation. Within the venous circulation fibrinolytic activity is lower peripherally with increased inhibitor activity.

To investigate the effect of increasing FVIII:C in-vivo on coagulation ex-vivo, DDAVP was infused over 15 min in 10 volunteers and in-vitro thrombin generation measured. FVIII:C rose from 0.42 and 0.43 IU/ml before DDAVP to 1.38, 1.73 and 1.78 IU/ml at 15, 30 and 60 min respectively (p less than 0.001). A computer-assisted thrombin generation test was performed in defibrinated plasma using chromogenic substrate, S2238. Time to reach 50% maximal thrombin activity (T50/s) and lag phase of thrombin generation (lag/s) were measured. Lag shortened from 75 and 60 s before to 45 s during and after infusion (p less than 0.001). T50 shortened from 78.5 and 76.0 to 62.5, 60.0 and 58.5 s at times 15 (p less than 0.01), 30 (p less than 0.001) and 60 (p less than 0.001) min. FVIII:C correlated inversely with lag and T50 (r = -0.847, p less than 0.001, r = -0.826, p less than 0.001, n = 10) respectively. These findings show that acute elevations of FVIII:C in-vivo accelerate in-vitro thrombin production. This work suggests that elevated FVIII:C levels in-vivo may be important in thrombo-occlusive disease.

Hirudin is a specific, potent inhibitor of thrombin that may be a valuable antithrombotic agent. The aim of this study was to investigate the hypothesis that the haemostatic effects of DDAVP counteract the coagulation defect induced by hirudin. The effect of DDAVP was studied in vivo on the anticoagulant action of recombinant hirudin (CGP39393) in vitro. Blood samples were taken at intervals from 10 normal volunteers infused with DDAVP. Factor VIII:C rose from (mean) 0.68 IU/ml before DDAVP to 2.19 and 2.16 IU/ml after 30 and 60 min infusion, respectively. Samples taken during DDAVP infusion showed a dose related decrease in the hirudin (0.5 and 1.0 microM) induced prolongation of the APTT, that occurred at FVIII:C concentrations of up to twice normal. At higher concentrations of hirudin no effect on the APTT occurred. These results demonstrate that DDAVP infusion elevates factor VIII:C levels with an associated significant reduction in the anticoagulant effect of hirudin in vitro.

Factor VIII (FVIII) and plasminogen activator activity (PAA) rise during hypoglycaemia, and this might contribute to the vascular complications of diabetes. Similar changes in haemostasis accompany raised plasma levels of vasopressin (aVP) and adrenaline. To investigate the effects of these hormones on haemostasis during hypoglycaemia and the role of plasma insulin concentrations, eight insulin-dependent diabetic patients underwent controlled hypoglycaemia for 20 min and 13 diabetic patients were investigated during hyperinsulinaemia with blood glucose maintained at 8.0 mmol/l. During hypoglycaemia, insulin levels increased to median values of 114 mU/l, a VP rose from 0.5 to 4.4 (p less than 0.005) pg/ml and adrenaline from 0.4 to 4.4 nmol/l (p less than 0.005). FVIII coagulant activity (FVIII:C) rose from 0.75 to 1.09 IU/ml (p less than 0.01) and the ristocetin co-factor (FVIIIR:Co) and von Willebrand factor antigen (vWF:Ag) showed similar responses. PAA increased from 156 to 745 units (p less than 0.005). During hyperinsulinaemia, insulin rose following infusion from 24 to 52 and 118 mU/l, maintained for an hour at each level. Despite this, plasma aVP, FVIII:C, FVIIIR:Co, vWF:Ag and PAA remained unchanged. This study indicates that the marked changes in FVIII, vWF and PAA concentrations which accompany hypoglycaemia depend on low blood glucose and not raised plasma insulin. The response in probably mediated by increases in adrenaline and aVP, which are part of the physiological response to hypoglycaemia.

Vasopressin infusions in normal volunteers that produce concentrations in plasma comparable to those seen during stress, cause an increase in plasma factor VIII and shortening of the euglobulin clot lysis time (ECLT). We have investigated the relationship between endogenous vasopressin (aVP) release and haemostatic function in 7 patients undergoing major abdominal surgery. Blood samples were taken at nine intervals during the operative procedure. Plasma aVP levels peaked at median values of 51 pg/ml during bowel manipulation and remained elevated on the first post-operative day. Following, and in close temporal relationship with the rise in aVP there were increases in factor VIII coagulant activity, the ristocetin co-factor, von Willebrand antigen, plasminogen activator activity (10(6)/ECLT2) and fibrinopeptide A concentrations with shortening of the activated partial thromboplastin time. The relationship was similar to that seen following infusion of aVP in human volunteers. The results are consistent with the hypothesis that aVP is an important mediator of changes in haemostatic function which accompany stress and might contribute to the thrombotic risk associated with surgical operations.

High physiological concentrations of plasma vasopressin (aVP) when achieved by infusion cause an increase in plasma factor VIII coagulant activity and shortening of the euglobulin clot lysis time (ECLT). To investigate the effects of aVP on components of the fibrinolytic pathway and on thrombin generation, 9 healthy volunteers were infused with saline for 30 min followed by aVP for 1 hour and blood samples taken every 30 min for measurement of aVP, ECLT, tissue-type plasminogen activator (t-PA), t-PA inhibition (tPA-I), plasminogen activator inhibitor 1 (PAI-1 Ag), activated partial thromboplastin time (APTT), fibrinopeptide A (FPA), fibrinopeptide B 15-42 (FPB beta 15-42) and cross-linked fibrin breakdown products (XL-FDP). Plasma aVP rose to a median of 75 pg/ml after 90 min and fell to 13.8 pg/ml 30 min later. The APTT fell from 43.5 to 35 sec (p less than 0.01) but there was no change in plasma FPA or in XL-FDP. Plasminogen activator activity (10(6)/ECLT2) increased from 25 to 736 units (p less than 0.01) and t-PA from 200 to 1012 mIU/ml (p less than 0.01). tPA-I fell from 8.0 to 2.7 IU/ml at 90 min (p less than 0.05) but PAI-1 Ag remained unchanged. Plasma FPB beta 15-42 was 2.4 and 1.2 pmol/ml before infusion with aVP and showed a small rise to 3.5 pmol/ml after 60 min (p less than 0.05). The results show the effects of aVP on fibrinolysis are mediated by an increase in t-PA. In the absence of thrombin generation the rise in t-PA was not accompanied by changes in XL-FDP.