Thromboprophylaxis following caesarean section--a comparison of the antithrombotic properties of three low molecular weight heparins--dalteparin, enoxaparin and tinzaparin.

Pharmacological thromboprophylaxis is increasingly being used after caesarean section to prevent venous thromboembolism. Although a variety of low molecular weight heparins (LMWH) have been used no comparative study exists on their effects on the haemostatic system in this situation. Furthermore, their antithrombotic effect may be mediated through effects other than their inhibitory effect on activated factor X. We compared the plasma anti-factor Xa activity, plasma concentration of tissue factor pathway inhibitor (TFPI) and the reduction in plasma thrombin-antithrombin (TAT) complex concentration in 30 women randomised to receive either dalteparin 5,000 IU anti-Xa once daily (n = 10), enoxaparin 4,000 IU anti-Xa once daily (n = 10) or tinzaparin 50 IU/kg anti-Xa (average dose 3,650 anti-Xa units) once daily (n = 10) following caesarean section. Sampling occurred at 0, 1, 3, 6, 12 and 24 h relative to time of dosing. All preparations produced an increase in mean anti-Xa assay (p < 0.0001), a reduction in mean TAT (p < 0.05) and an increase in mean TFPI concentration (p <0.05). Analysis of variance (ANOVA) revealed a significant difference between the LMWHs in terms of mean anti-factor Xa activity (p < 0.005) and reduction in plasma TAT concentration (p < 0.005). Post hoc analysis indicated that the anti-Xa values of the groups receiving enoxaparin and dalteparin were significantly higher than those of the group receiving tinzaparin (p < 0.05), but not significantly different from each other. Post hoc analysis of the reduction in plasma TAT concentration showed the reduction to be significantly less in the group receiving enoxaparin compared to the dalteparin and tinzaparin groups (p < 0.05), which did not differ significantly from each other. There was no significant difference between treatment groups with regard to plasma concentration of TFPI. These findings demonstrate that LMWHs differ in their effects on haemostatic parameters including thrombin generation as assessed by TAT. The increase in TFPI may be an additional mediator of LMWH's antithrombotic effects. Although these findings demonstrate that LMWHs differ in their haemostatic effects, this does not necessarily infer a clinical difference between these agents.

Activated protein C sensitivity, protein C, protein S and coagulation in normal pregnancy.

A prospective study of activated protein C sensitivity, protein C, protein S, and other coagulation factors in 239 women during normal pregnancy was carried out. Protein C activity appeared unaffected by gestation, although an elevation of protein C activity was observed in the early puerperium. A fall in total and free protein S with increasing gestation was observed. Activated protein C sensitivity ratio (APC:SR) showed a progressive fall through pregnancy. This fall correlated with changes in factor VIIIc, factor Vc and protein S. 38% of subjects, with no evidence of Factor V Leiden or anticardiolipin antibodies, showed a low APC:SR (APC:SR <2.6) in the third trimester of pregnancy. Aside from a significant reduction in birth weight, no difference in pregnancy outcome was observed between these subjects and those with a normal APC:SR. Activated protein C sensitivity ratio, modified by pre-dilution of patient samples with factor V depleted plasma, showed no consistent trend with gestation.

Superficial vein thrombosis: incidence in association with pregnancy and prevalence of thrombophilic defects.

Superficial venous thrombotic (SVT) events are a feature of thrombophilic abnormalities, particularly those involving the protein C pathway. We have determined the incidence of SVT associated with pregnancy and the early postpartum period in a retrospective study involving 72000 deliveries. Fourty-nine cases occurring in 47 individuals were recorded, with an overall incidence of 0.68/1000 deliveries (95% CI 0.48-0.88). None had a previous history of deep vein thrombosis or pulmonary embolism. Most events occurred in the early postpartum period (0.54/1000 deliveries). Twenty-four/fourty-seven were screened for established thrombophilic abnormalities, with only 1 abnormality detected (FV(Leiden) heterozygote). Thrombophilia may play a minor role in the aetiology of SVT associated with pregnancy, although a larger study is required to confirm this.

Risk factors for pregnancy associated venous thromboembolism.

In an attempt to reduce the incidence of pregnancy associated venous thromboembolism (PA-VTE), some researchers have advocated screening of all women for the factor V(Leiden) mutation during early pregnancy. We have conducted a large retrospective study (over 72,000 deliveries) to determine if this would be useful. Sixty-two objectively confirmed venous thrombotic events (51 DVT, 11 PE) were recorded at two maternity units in the UK. The incidence of DVT was 0.71 per 1000 deliveries (95% CI 0.5-0.9) with 0.50 occurring in the antenatal period (95% CI 0.34-0.66) and 0.21 in the puerperium (95% CI 0.11-0.31). The incidence of PE was 0.15 per 1000 deliveries (95% CI 0.06-0.24), 0.07 antenatal (95% CI 0.01-0.13) and 0.08 in the puerperium (95% CI 0.02-0.14). Of these 62, 50 attended for follow-up and thrombophilia screening. 28% of all episodes of PA-VTE had no clinical risk factor for thrombosis or an identifiable thrombophilic abnormality. Deficiency of antithrombin was identified in 12% of individuals (95% CI 3-21) and the factor V(Leiden) mutation in 8% (95% CI 0.5-15.5). Based on estimates of the prevalence of the factor V(Leiden) mutation in the population, we estimate that the thrombotic risk for a woman during pregnancy or the puerperium with the defect is approximately 1 in 400-500. This figure would not lend support to the idea of random screening for the mutation in early pregnancy.

Asymptomatic coinheritance of heterozygous plasminogen deficiency and the factor VLeiden mutation.

Deficiency of plasminogen has been postulated by some authors as a possible thrombophilic abnormality, though this remains controversial. We have previously identified a cohort of individuals with plasminogen deficiency from a study to determine plasminogen levels within the general population. All were asymptomatic for thrombosis at initial identification. We followed this cohort over a 5-year period, with no venous thrombotic events recorded, although one patient did suffer a myocardial infarction. One family was identified with asymptomatic coinheritance of both plasminogen deficiency and the factor VLeiden mutation. There was no apparent venous thrombotic risk conferred upon any of our cohort of individuals by inheritance of plasminogen deficiency, and, in addition, the combination of the factor VLeiden mutation with plasminogen deficiency which was observed in three individuals did not result in thrombotic events. However, the cohort is small and we cannot entirely exclude plasminogen deficiency as a possible mild thrombophilic defect.

Isolated familial plasminogen deficiency may not be a risk factor for thrombosis.

Despite many reports of individuals with congenital plasminogen deficiency and thrombosis, there is still uncertainty whether heterozygous deficiency represents a real thrombophilic risk factor or simply a coincidental finding. We have addressed this issue by testing for plasminogen deficiency in a cohort of 9611 blood donors. Out of 66 donors with reduced plasminogen activity on two occasions 28 were shown to have a familial deficiency state (including 3 with dysplasminogenaemia). Our observed prevalence rate for familial plasminogen deficiency, calculated at 2.9/1000 (95% CI = 1.9-4.2 per 1000), was not significantly different from that calculated from published reports of congenital plasminogen deficiency in thrombotic cohorts (5.4/1000). Furthermore, with only two exceptions, all 80 donors and relatives with familial deficiency were asymptomatic with regard to thrombosis-including a 29 year old donor with suspected compound heterozygous hypoplasminogenaemia. These findings add further weight to the argument that familial heterozygous plasminogen deficiency, at least in isolation, does not constitute a significant thrombotic risk factor. However, it remains uncertain whether plasminogen deficiency, when combined with other thrombophilic conditions, may become more clinically important.

Low thrombosis rate seen in blood donors and their relatives with inherited deficiencies of antithrombin and protein C: correlation with type of defect, family history, and absence of the factor V Leiden mutation.

We have previously identified a group of blood donors with inherited deficiencies of either antithrombin (AT) or protein C who appear to have a relatively low thrombosis rate. In the 5 years that have elapsed since initial identification of these individuals, resistance to activated protein C (APC resistance), which is associated with the factor V Leiden gene mutation, has emerged as an important and highly prevalent inherited thrombophilic risk factor. We have followed 28 donors/relatives with deficiency of AT (median age 48 years, range 16-77) and 23 with deficiency of protein C (median age 44 years, range 15-79) over a period of 5 years. During the study period only one individual, who was previously symptomatic, has suffered a thrombotic event which occurred spontaneously whilst on warfarin. We have now excluded coinheritance of APC resistance due to the factor V Leiden mutation in our cohort. Our findings demonstrate that individuals with single inherited thrombophilic defects are not uncommon and are frequently asymptomatic. The absence of the factor V Leiden mutation may in part explain the low thrombosis rate observed, and lends support to the hypothesis that multiple thrombophilic defects may be necessary for the development of thrombosis.

Prevalence of protein C deficiency in the healthy population.

Reported prevalence rates for protein C (PC) deficiency in the population at large have varied widely. The differences presumably reflect the existence of an apparently high number of clinically recessive forms of the deficiency. In an attempt to document more precisely the prevalence of PC deficiency in the healthy population we have measured PC activity in just under 10,000 blood donors in the West of Scotland. After repeat testing of donors with low results and then further observation and selection, 32 donors were identified who had individual mean PC activities below the age- and gender-specific study reference range. Assessment of available first degree relatives, and also PC gene analysis in 23 of these donors, allowed identification of at least 14 with an inheritable deficiency (8 by both family study and gene analysis, 3 by family study alone and 3 by gene analysis alone). Two recurring and seven unique point mutations, only one of which has been previously described, were identified. The observed prevalence of inherited PC deficiency was 1.45 per 1000 (95% CI, 0.79/1000 to 2.43/1000). However after correcting for the possibility of missing some genuine inherited deficiencies we estimated the prevalence to be as high as 1 in 500. All cases of hereditary deficiency were asymptomatic with regard to thrombosis and none had a strong family history of thrombosis.(ABSTRACT TRUNCATED AT 250 WORDS)

A clinical and in vitro study on the possible interaction of intravenous nitrates with heparin anticoagulation.

It has been reported that intravenous nitrates inhibit the anticoagulant effect of heparin. This possible interaction has potentially serious implications for the management of patients with acute coronary ischemic syndromes. This possible interaction was assessed prospectively in a clinical and in an in vitro study involving 24 patients receiving both drugs for the management of unstable angina pectoris. There was a small inhibitory effect of intravenous glyceryl trinitrate or isosorbide dinitrate on the anticoagulant effect of heparin in 3 of 24 cases in vivo, as assessed by activated partial thromboplastin time measurements. Nitrates or propylene glycol had no effect on heparin activity in vitro. It was concluded that there may be an inhibitory effect of nitrates on anticoagulation in a small minority of patients, but close attention to detail in monitoring heparin anticoagulation is far more important.

Familial thrombophilia and activated protein C resistance: thrombotic risk in pregnancy?

An abnormal anticoagulant response in vitro to activated protein C (aPC) has been proposed as an aetiological factor in familial thrombophilia. It is postulated that this phenomenon is due to an inherited molecular defect of factor V resulting in poor inactivation by aPC. We conducted a family study when the proband presented in her second pregnancy with superficial phlebitis, a history of deep venous thrombosis and a family history of venous thromboembolic disease. No abnormality of antithrombin activity, protein C activity or deficiency of protein S were demonstrated in the family members tested. The proband had aPC ratios below the laboratory range on three consecutive occasions. In addition, her mother, who had a history of recurrent DVTs and a pulmonary embolus, and also an asymptomatic nulliparous sister, both had aPC resistance ratios below the laboratory range on consecutive samples. Further information about the combined risk of aPC resistance and pregnancy is needed before guidance on the management of affected women can be formulated.

Prevalence of antithrombin deficiency in the healthy population.

In a cohort of 9669 blood donors we have identified 16 cases of congenital AT deficiency (1 in 600) by way of family studies and AT gene analysis. Two donors had type I AT deficiency (prevalence 0.21 per 1000; 95% CI = 0.03/1000 to 0.75/1000), their families displaying a symptomatic phenotype. 14 donors had a type II deficiency (prevalence 1.45 per 1000; 95% CI = 0.79/1000 to 2.43/1000): one recurring and three unique mutations. None of these type II deficiencies appeared to confer a high thrombotic risk despite many of the affected individuals having experienced potentially prothrombotic challenges. The high frequency of these relatively asymptomatic variants may reflect a selection bias in the study population. However, their existence should not only add to our understanding of structure-function relationships of AT but may also influence our management of asymptomatic deficient individuals identified in epidemiological or presurgical screening programmes.

Protein C activity in healthy volunteers--influence of age, sex, smoking and oral contraceptives.

Interpretation of protein C (PC) levels in a given individual has several limitations. A normal PC activity does not necessarily exclude a genetic deficiency nor can a reduced level confirm it. Measuring PC amidolytic activity in 9,648 healthy blood donors has allowed identification of demographic factors which cause variation in PC activity and further hinder interpretation. PC activity displays a log normal distribution and significant variation with age. This is most marked in young adult males when mean PC activity rises from 0.86 iu/ml (15-19 years) to 1.04 iu/ml (45-49 years; P < 0.0001). Pre-menopausal females, who for most age ranges, have mean PC activity below their male contemporaries, show a less marked rise with age until the menopause when PC activity rises further. The use of hormonal contraceptive preparations is associated with an increase in mean PC activity of 0.05-0.08 iu/ml while smoking habit has no influence on PC activity. In view of these findings we strongly recommend the use of age and sex restricted reference ranges when interpreting PC activity.

Influence of demographic factors on antithrombin III activity in a healthy population.

Antithrombin III (AT III) activity has been measured in 9669 healthy blood donors (5525 male and 4144 female). The distribution of AT III is approximately 'normal' with mean 105.6 IU/dl and standard deviation 11.2; however, definite age and sex related variations are evident. Pre-menopausal females have lower mean AT III compared to their male contemporaries who have remarkably stable mean AT III until 45 years, after which there is a gradual decline. In contrast, post-menopausal females have higher mean AT III than both males of the same age and younger pre-menopausal females. Concurrent hormone replacement therapy inhibits this rise. The use of hormonal preparations is associated with a 4 IU/dl reduction of mean AT III in younger females but not in those over 30 years. Smoking may result in a mild increase in AT III of doubtful clinical significance. On-going genetic and family studies are expected to predict a prevalence rate of congenital AT III deficiency in excess of the previously reported figure of 0.02%. The authors consider these observed variations as minor and recommend the use of a single reference range for AT III activity, but that particular care be taken when interpreting results in pill-taking females and the elderly.

Plasminogen levels in healthy volunteers--influence of age, sex, smoking and oral contraceptives.

There is considerable doubt as to the importance of reduced plasminogen (PLG) activity as a risk factor for venous thrombosis. In the present study we have identified a wide range of PLG activities (25-200%) in a cohort of 9,611 blood donors. Males and females not taking hormonal contraceptives show a similar distribution of PLG, however, variation related to age appears to follow a different pattern in males and females. These differences are of doubtful clinical importance as are differences related to smoking. In contrast, females taking hormonal contraceptives or hormone replacement therapy (HRT) have up to 25% higher mean PLG levels in younger females but a less marked elevation (10%) is seen in 40-50 year olds. A PLG activity < 65% was recorded in 61 donors, none of whom appeared to have a history of thrombosis. These findings do not support the notion that reduced PLG is an important thrombophilic risk factor, however, further investigation of the donors with low PLG is required.

Evaluation of an automated prothrombin assay using a chromogenic substrate.

The performance of a commercially available assay for the determination of prothrombin by a chromogenic substrate method has been evaluated for use on an automated clinical analyser. The method was rapid and simple; accuracy, precision and sensitivity were satisfactory, while carry-over was negligible. Determinations were performed on 170 plasma samples from patients on long-term oral anticoagulant therapy, using the automated amidolytic method; comparison of the prothrombin levels with results obtained by the prothrombin time and Thrombotest methods provided highly significant correlations.

Preparation and evaluation of an antiserum to the carboxymethylated Aalpha chain of human fibrinogen.

Antiserum to the carboxymethylated Aalpha chain of human fibrinogen has been produced in rabbits and its reactions investigated in a double diffusion system. The antiserum reacts with the carboxymethylated Aalpha chains, but not with carboxymethylated gamma chains, nor with carboxymethylated Bbeta chains, provided that the antiserum had been absorbed with carboxymethylated Bbeta chains. The antiserum also reacts with Aalpha chains in the whole fibrinogen molecule, either in plasma or in the purified form, and spur formation was observed when intact fibrinogen was compared with high solubility fibrinogen, which contained little or no intact Aalpha chain.

Effect of anabolic steroids on plasma antithrombin III. alpha2 macroglobulin and alpha1 antitrypsin levels.

The effect of seven different anabolic steroids (Ethyloestrenol, Methenolone acetate, Norethandrolone, Methylandrostenediol, Oxymetholone, Methandienone, and Stanozolol) on three alpha-globulin antiprotease inhibitors of thrombin and plasmin was studied in men with ischaemic heart disease. In distinct contrast to the oral contraceptives, five of the six 17-alpha-alkylated anabolic steroids studied produced increased plasma Antithrombin III levels and five produced decreased levels of plasma alpha2-macroglobulin. The effect on plasma alpha1-antitrypsin levels was less clear-cut but three of the steroids examined produced significantly elevated levels. The increased plasma fibrinolytic activity which the 17-alpha-alkylated anabolic steroids induce is therefore unlikely to be secondary to disseminated intravascular coagulation.

Plasma fibrinolytic activity following oral anabolic steroid therapy.

Six anabolic steroids were assessed for their ability to enhance plasma fibrinolytic activity in males with ischaemic heart disease. Five 17alpha-alkylated steroids (Ethyloestrenol, Norethandrolone, Methandienone, Methylandrostenediol and Oxymetholone) were examined and all produced a significant increase in plasma plasminogen activator as measured by the euglobulin lysis time. The only non-17alpha-alkylated steroid studied (Methenolone acetate) failed to enhance fibrinolysis. The 17alpha-alkylated steroids studied all deserve more detailed evaluation of their long term effects on plasma fibrinolytic activity.