Symptomatic onset of severe hemophilia A in childhood is dependent on the presence of prothrombotic risk factors

It has been recently suggested that the clinical phenotype of severe hemophilia A (HA) is influenced by co-inheritance with the factor V G1691A mutation. We therefore investigated 124 pediatric PUP patients with hemophilia (A: n = 111) consecutively admitted to German pediatric hemophilia treatment centers. In addition to factor VIII activity, the factor V (FV) G1691A mutation, the prothrombin (PT) G20210A variant, antithrombin, protein C, protein S and antithrombin were investigated. 92 out of 111 HA patients (F VIII activity < 1%) were suffering from severe HA. The prevalence of prothrombotic risk factors in children with severe HA was no different from previously reported data: FV G1691A 6.5%, PT G20201A 3.2%, and protein C type I deficiency 1.1%. No deficiency states of antithrombin or protein S were found in this cohort of hemophilic patients. The first symptomatic bleeding leading to diagnosis of severe hemophilia (< 1%) occurred with a median (range) age of 1.6 years (0.5-7.1) in children carrying defects within the protein C pathway or the PT gene mutation compared with non-carriers of prothrombotic risk factors (0.9 years (0.1-4.0; p = 0.01). The cumulative event-free bleeding survival was significantly prolonged in children carrying additionally prothrombotic defects (log-rank/Mantel-Cox: p = 0.0098). In conclusion, data of this multicenter cohort study clearly demonstrate that the first symptomatic bleeding onset in children with severe HA carrying prothrombotic risk factors is significantly later in life than in non-carriers.

Effects of diets containing olive oil, sunflower oil, or rapeseed oil on the hemostatic system

Various studies have already shown that the fatty acid composition of dietary fat has different effects on hemostasis and platelet function. However, knowledge on this topic is incomplete. In the present study, fifty-eight healthy students received either a 4-week rapeseed oil [high content of monounsaturated fatty acids (MUFA) and high n-3/n-6 PUFA ratio], an olive oil (high content of MUFA, low n-3/n-6 PUFA ratio) or a sunflower oil (low content of MUFA, low n-3/n-6 PUFA ratio) diet. In each group, effects on hemostatic parameters were compared with a wash-in diet rich in saturated fatty acids with respect to intermediate-time effects on the hemostatic system and platelet function. With the olive oil diet, a reduction of coagulation factors VIIc, XIIc, XIIa, and Xc was found, whereas sunflower oil led to lower values of coagulation factors XIIc, XIIa, and IXc. In all study groups levels of plasmin-alpha2-antiplasmin were lower in week 4 than at baseline. Lower fibrinogen binding on platelets was found after the sunflower oil diet, whereas expression of CD62 and spontaneous platelet aggregation were slightly higher after the olive oil diet. However, given the major differences in the fatty acid compositions of the diets, the differences between the groups with respect to hemostasis tended to be small. Therefore, the clinical significance of the present findings remains to be evaluated.

The plasminogen activator inhibitor (PAI)-1 promoter 4G/4G genotype is not associated with ischemic stroke in a population of German children. Childhood Stroke Study Group

OBJECTIVES

To investigate the relationship between an insertion/deletion (4G/5G) polymorphism of the plasminogen activator inhibitor (PAI)-1 gene and childhood patients with a past history of ischemic stroke.

METHODS

The PAI-1 4G/4G genotype and the coinheritance with lipoprotein (Lp) (a) levels, the factor V (FV) G1691A mutation, the prothrombin (PT) G20210A variant, and the methylene-tetrahydrofolate reductase (MTHFR) T677T genotype were studied in 198 Caucasian children with stroke and 951 controls (same age, sex and ethnical distribution). In a randomly selected subgroup of patients/controls (n=60) PAI-I activities have been investigated.

RESULTS

The distribution of the 4G/5G genotypes was no different in childhood stroke patients and controls, with a 4G allele frequency of 55.8% in patients compared with 53.8% in control subjects (P=0.49). The 4G/4G genotype compared with the remaining genotypes was present in 43 cases and 167 (17.6% vs. 21.7%; OR/CI: 1.30/0.89-1.98; P=0.3). PAI-1 activity was significantly elevated (P < 0.001) in the patient group.

CONCLUSIONS

Data presented here suggest that the 4G/4G genotype is not a major risk factor in the aetiology of childhood ischemic stroke.

Activation of phosphatidylinositol-specific phospholipase C by HDL-associated lysosphingolipid. Involvement in mitogenesis but not in cholesterol efflux

Our earlier studies demonstrated that high-density lipoproteins (HDLs) stimulate multiple signaling pathways, including activation of phosphatidylcholine-specific phospholipases C and D (PC-PLs) and phosphatidylinositol-specific phospholipase C (PI-PLC). However, only activation of PC-PLs was linked to the HDL-induced cholesterol efflux. In the study presented here, the role of HDL-induced PI-PLC activation was studied. In human skin fibroblasts, HDL potently induced PI-PLC as inferred from enhanced phosphatidylinositol bisphosphate (PtdInsP(2)) turnover and Ca(2+) mobilization. The major protein component of HDL, apo A-I, did not induce PtdInsP(2) turnover or Ca(2+) mobilization in these cells. Both HDL and apo A-I promoted cellular cholesterol efflux, whereas only HDL induced fibroblast proliferation. Inhibition of PI-PLC with U73122 or blocking intracellular Ca(2+) elevation with Ni(2+) or EGTA markedly reduced the extent of HDL-induced cell proliferation but had no effect on cholesterol efflux. In fibroblasts from patients with Tangier disease which are characterized by defective cholesterol efflux, neither HDL-induced PtdInsP(2) breakdown and Ca(2+) mobilization nor cell proliferation was impaired. HDL-induced fibroblast proliferation, PtdInsP(2) turnover, and Ca(2+) mobilization were fully mimicked by the lipid fraction isolated from HDL. Analysis of this fraction with high-performance liquid chromatography (HPLC) and time-of-flight secondary ion mass spectroscopy (TOF-SIMS) revealed that the PI-PLC-inducing activity is identical with two bioactive lysosphingolipids, namely, lysosulfatide (LSF) and sphingosylphosphorylcholine (SPC). Like native HDL, LSF and SPC induced PtdInsP(2) turnover, Ca(2+) mobilization, and fibroblast proliferation. However, both compounds did not promote cholesterol efflux. In conclusion, two agonist activities are carried by HDL. Apo A-I stimulates phosphatidylcholine breakdown and thereby facilitates cholesterol efflux, whereas LSF and SPC trigger PI-PLC activation and thereby stimulate cell proliferation.

Abdominal venous thrombosis in neonates and infants: role of prothrombotic risk factors - a multicentre case-control study. For the Childhood Thrombophilia Study Group

The factor V (FV) G1691A mutation, the prothrombin (PT) G20210A variant, the methylenetetrahydrofolate reductase (MTHFR) T677T genotype, together with fasting homocysteine (HCY) concentration, lipoprotein (Lp)(a), anti-thrombin (AT), protein C (PC), protein S (PS) and anti-cardiolipin antibodies were investigated in 65 consecutively recruited infants (neonate to < 12 months) with renal venous thrombosis (RVT; n = 31), portal vein thrombosis (PVT; n = 24) or hepatic vein thrombosis (HVT n = 10), and 100 age- and sex-matched healthy controls. FV G1691A was found in 14 babies (heterozygous: RVT n = 9, PVT n = 4; homozygous HVT n = 1) and five controls, the MTHFR TT677 genotype together with increased HCY in four infants with thrombosis (RVT n = 2; PVT n = 1; HVT n = 1) compared with one control, and the PT G20210A variant was present in one control only. PC type I deficiency was diagnosed in three patients (RVT n = 2; PVT n = 1) and AT deficiency in two patients (RVT n = 1; PVT n = 1). Three neonates with spontaneous thrombosis showed FV G1691A combined with Lp(a) and the FV G1691A was combined with the PT G20210A genotype in two infants. Additional triggering factors were reported in 27 patients (41.5%). The overall odds ratios (ORs) and 95% confidence intervals (CIs) with respect to the different thrombosis locations were: RVT (OR/CI: 10.9/3.85-31.1; P < 0.0001), PVT (5.47/1.7-17.6; P < 0.0007) and HVT (3.3/0.58-18.7; P = 0.18). The data presented here suggest that genetic prothrombotic risk factors also play an important role in abdominal venous thrombosis during infancy.

Symptomatic ischemic stroke in full-term neonates : role of acquired and genetic prothrombotic risk factors

BACKGROUND AND PURPOSE

The present multicenter case-control study was prospectively designed to assess the extent to which single and combined clotting factor abnormalities influence the onset of symptomatic ischemic stroke in full-term neonates.

METHODS

Lipoprotein (Lp)(a); the factor V (FV) G1691A mutation; the prothrombin (PT) G20210A variant; the methylenetetrahydrofolate reductase (MTHFR) T677T genotype; antithrombin; protein C; protein S; and anticardiolipin antibodies (ACAs) were investigated in 91 consecutively recruited neonatal stroke patients and 182 age- and sex-matched healthy controls.

RESULTS

Sixty-two of 91 stroke patients (68.1%) had at least 1 prothrombotic risk factor compared with 44 control subjects (24.2%) (odds ratio [OR]/95% confidence interval [CI], 6.70/3.84 to 11.67). An increased Lp(a) level (>30 mg/dL) was found in 20 patients and 10 controls (OR/95% CI, 4.84/2. 16 to 10.86); FV G1691A was present in 17 patients and 10 controls (OR/95% CI, 3.95/1.72 to 9.0); the PT G20210A variant was detected in 4 patients and 4 controls (OR/95% CI, 2.04/0.49 to 8.3); the MTHFR TT677 genotype was found in 15 patients and 20 controls (OR/95% CI, 1.59/0.77 to 3.29); and protein C type I deficiency was found in 6 neonates. Neither antithrombin deficiency nor protein S deficiency was found in the neonatal patients studied. Acquired IgG ACAs were found in 3 cases. Additional triggering factors, ie, asphyxia, septicemia, maternal diabetes, and perinatally acquired renal venous thrombosis, were reported in 54.0% of patients.

CONCLUSIONS

Besides acquired triggering factors, the data presented here suggest that genetic prothrombotic risk factors play a role in symptomatic neonatal stroke.

Prothrombotic risk factors in children with spontaneous venous thrombosis and their asymptomatic parents: a family study

The present study was designed to assess to what extent single and combined clotting abnormalities influence spontaneous vascular accidents in pediatric patients, and how the children affected differ in their prothrombotic risk profiles from their biological first-degree family members. In addition, this study was performed to investigate if relatively mild thrombophilic polymorphisms not leading to thrombosis in the parents cause severe clinical expression when coinherited with an established prothrombotic risk factor. The factor V (FV) G1691A mutation, the prothrombin (PT) G20210A variant, the methylenetetrahydrofolate reductase (MTHFR) T677T genotype, the plasminogen activator inhibitor (PAI)-1 promoter polymorphism, lipoprotein (Lp)(a), antithrombin, protein C, and protein S were investigated in 48 childhood patients aged neonate to <18 years (median 0.5 years) with spontaneous venous thromboembolism (SVT) compared with the carrier status of their first-degree family members. In 19 of the 48 patients (39.6%), one prothrombotic risk factor was diagnosed, and in 27 of the 48 subjects (56.3%) at least two prothrombotic defects/alleles. In the majority of cases with SVT, the FV G1691A mutation was involved either with a second mutated allele or combined with elevated Lp(a), the 4G/4G genotype of the PAI -1 promoter polymorphism, and the T677T MTHFR genotype. The rate of combined prothrombotic risk factors was significantly higher in childhood patients compared with their parents. In conclusion, based on the data presented here we suggest that early-onset SVT in childhood patients is mainly caused by combinations of at least two prothrombotic risk factors.

High-density lipoproteins inhibit fibrinogen binding on adenosine diphosphate-activated monocytes

High levels of fibrinogen and low levels of high-density lipoprotein (HDL) cholesterol were reported to be risk factors for coronary heart disease. CD11b/CD18, a fibrinogen-binding protein, is expressed on the surface of monocytes, which play a crucial role in the formation of atherosclerotic lesions. In the present study, we investigate the effects of antibodies against CD11b and CD18, as well as HDL3 and low-density lipoprotein (LDL) cholesterol on fibrinogen binding on monocytes. We find that binding of fibrinogen on monocytes activated with adenosine diphosphate (ADP) was reduced to 66.0+/-8.3% (mean +/- SD) in the presence of anti-CD11b antibodies (12.5 microg/ml; P < or = 0.02) and to 54.5+/-4.9% in the presence of anti-CD18 antibodies (20 microg/ml; P < or = 0.01), respectively. Fibrinogen binding on Cytochalasin-B-activated monocytes was reduced to 79.8+/-6.0% in the presence of anti-CD18 (20 microg/ml; P < or = 0.05). Incubation of ADP-activated monocytes with HDL3 (0.5 g/l) led to a lowering of fibrinogen binding to 65.0+/-6.6% (P < or = 0.05). No effect of HDL3 on fibrinogen binding was seen on Cytochalasin-B-activated monocytes. A slight, non-significant stimulatory effect of LDL on fibrinogen binding on ADP-activated but not on Cytochalasin-B-activated monocytes was additionally observed. Neither incubation with HDL3 or with LDL had a significant influence on ADP-activated cellular binding of anti-CD11b or anti-CD18 antibodies. The inhibition of fibrinogen binding on monocytes in the presence of HDL3 is a major new finding of this study. Since inhibition of fibrinogen binding in the presence of HDL might impair both monocyte recruitment to the arterial wall and foam cells formation, our findings suggests a novel mechanism by which HDL may prevent development of arteriosclerosis.

Prothrombotic risk factors in children with acute lymphoblastic leukemia treated with delayed E. coli asparaginase (COALL-92 and 97 protocols)

Hereditary prothrombotic risk factors have been shown to increase the risk of venous thrombosis in children treated with the combination of E. coli asparaginase and steroids. In the present study the role of prothrombotic risk factors in children with ALL treated according to the COALL study protocol was investigated in 108 consecutively recruited childhood patients. The prevalence rates of prothrombotic risk factors [factor V G1691A mutation, the prothrombin G20210A variant, the TT677 methylenetetrahydrofolate reductase genotype, deficiencies of protein C, protein S, antithrombin, elevated lipoprotein (a)] in this cohort were within the range reported for healthy Caucasians, and comparable to previously reported data for other leukemic patients. Venous thromboembolism occurred in 3 of the 108 children (induction n = 1; reinduction n = 2: 2.8%), and none of these children carried a prothrombotic risk factor. The results of the present study, suggest that the role of hereditary and acquired disturbances of coagulation in the development of thromboses might depend on the treatment regimen.

C-reactive protein and the severity of atherosclerosis in myocardial infarction patients with stable angina pectoris

BACKGROUND

Recent findings provide evidence for the importance of inflammatory processes in the pathogenesis of atherosclerosis. C-reactive protein was elevated in patients with peripheral artery disease, coronary heart disease and myocardial infarction compared to normal subjects.

METHODS

In 1112 male and 299 female survivors of myocardial infarction (mean age +/- SD, men, 50.4 +/- 9.5, women, 56.1 +/- 9.3), we investigated whether plasma C-reactive protein concentration is associated with the severity of coronary heart disease and generalized pre-clinical or clinically manifest arteriopathy. The control group consisted of 326 male and 138 female individuals matched for age without clinical symptoms of coronary disease. The severity of arteriosclerotic changes was determined for the extra-cranial brain-supplying arteries, abdominal aorta, pelvis and leg arteries. In myocardial infarction patients coronary angiography was performed. Laboratory analyses included determination of C-reactive protein, fibrinogen, D-dimer, HDL-cholesterol, LDL-cholesterol and triglycerides.

RESULTS

The following ranking of C-reactive protein concentrations was found: controls < or = patients after myocardial infarction without atherosclerosis < or = patients with myocardial infarction and pre-clinical atherosclerosis < or = patients with myocardial infarction and clinically manifest atherosclerosis. Additionally, our data showed a significant association between C-reactive protein concentrations and the angiographically detected degree of coronary heart disease.

CONCLUSIONS

As C-reactive protein is a marker of inflammatory processes, our results in patients with clinically manifest and early pre-clinical atherosclerosis support the hypothesis that inflammatory processes in the vessel wall participate in atherogenesis. Moreover, they support the hypothesis of a causal relationship between an acute phase reaction and the pathogenesis of atherosclerosis in coronary arteries and other parts of the arterial vessel system.

D609-phosphatidylcholine-specific phospholipase C inhibitor attenuates thapsigargin-induced sodium influx in human lymphocytes

Previously, we reported that the phosphatidylcholine-specific phospholipase C (PC-PLC) inhibitor tricyclodecan-9-yl xanthogenate (D609) potentiates thapsigargin-induced Ca(2+) influx in human lymphocytes. In the present study we examined the effect of D609 on the thapsigargin-induced Na(+) entry. We found that the early phase of the thapsigargin-induced increase in the intracellular Na(+) concentration (approx. 1-2 min after stimulation) was attenuated after preincubation of lymphocytes with D609. By contrast, thapsigargin-induced Na(+) influx was not affected in the presence butan-1-ol, which inhibits phosphatidylcholine-specific phospholipase D (PC-PLD). The thapsigargin-induced Na(+) influx could be mimicked by PC-PLC exogenously added to the lymphocyte suspension, whereas addition of PC-PLD had no effect. In addition, thapsigargin stimulated formation of the physiological PC-PLC products, diacylglycerol. Cell-permeable diacylglycerol analogue, dioctanoyl-glycerol (DOG), produced time- and concentration-dependent increase in the intracellular Na(+) concentration. Both thapsigargin- and DOG-induced Na(+) increases were not affected in the presence of Na(+)/H(+) antiport inhibitor, HOE609, or Na(+)/Ca(2+) antiport inhibitor, dimethylthiourea, as well as in the presence of Co(2+) and Ni(2+), which block store-operated Ca(2+) entry. By contrast, markedly reduced thapsigargin- and DOG-induced Na(+) influx were noted in the presence of flufenamic acid, which blocks the non-selective cation current (I(CRANC)). In conclusion, our results suggest that diacylglycerol released due to the PC-PLC activation contributes to the thapsigargin-induced Na(+) entry.

Phospholipase A(2) is involved in thapsigargin-induced sodium influx in human lymphocytes

Previously, we reported that emptying of intracellular Ca(2+) pools with endoplasmatic Ca(2+)-ATP-ase inhibitor thapsigargin leads to the Na(+) influx in human lymphocytes (M. Tepel et al., 1994, J. Biol. Chem. 269, 26239-26242). In the present study we examined the mechanism underlying the thapsigargin-induced Na(+) entry. We found that the thapsigargin-induced increase in Na(+) concentration was effectively inhibited by three structurally unrelated phospholipase A(2) (PLA(2)) inhibitors, p-bromophenacyl bromide, 3-(4-octadecyl)-benzoylacrylic acid (OBAA), and bromoenol lactone (BEL). The thapsigargin-induced Na(+) influx could be mimicked by PLA(2) exogenously added to the lymphocyte suspension. In addition, thapsigargin stimulated formation of arachidonic acid (AA), the physiological PLA(2) product. AA induced Na(+) entry in a time- and concentration-dependent fashion. Both, thapsigargin-induced Na(+) influx and AA liberation were completely inhibited in the presence of tyrosine kinase inhibitor genistein but not in the absence of extracellular Ca(2+). Collectively, these data show that thapsigargin-induced Na(+) entry is associated with tyrosine kinase-dependent stimulation of PLA(2).

Lipoprotein(a) serum concentrations and apolipoprotein(a) phenotypes in mild and moderate renal failure

High lipoprotein(a) (Lp(a)) serum concentrations and the underlying apolipoprotein(a) (apo(a)) phenotypes are risk factors for cardiovascular disease in the general population as well as in patients with renal disease. Lp(a) concentrations are markedly elevated in patients with end-stage renal disease. However, nothing is known about the changes of Lp(a) depending on apo(a) size polymorphism in the earliest stages of renal impairment. In this study, GFR was measured by iohexol technique in 227 non-nephrotic patients with different degrees of renal impairment and was then correlated with Lp(a) serum concentrations stratified according to low (LMW) and high (HMW) molecular weight apo(a) phenotypes. Lp(a) increased significantly with decreasing GFR. Such an increase was dependent on apo(a) phenotype. Only renal patients with HMW apo(a) phenotypes expressed higher median Lp(a) concentrations, i.e., 6.2 mg/dl at GFR >90 ml/min per 1.73 m2, 14.2 at GFR 45 to 90 ml/min per 1.73 m2, and 18.0 mg/dl at GFR <45 ml/min per 1.73 m2. These values were markedly different when compared with apo(a) phenotype-matched control subjects who had a median level of 4.4 mg/dl (ANOVA, linear relationship, P < 0.001). In contrast, no significant differences were observed at different stages of renal function in patients with LMW apo(a) phenotypes when compared with phenotype-matched control subjects. The elevation of Lp(a) was independent of the type of primary renal disease and was not related to the concentration of C-reactive protein. Multiple linear regression analysis found that the apo(a) phenotype and GFR were significantly associated with Lp(a) levels. Non-nephrotic-range proteinuria modified the association between GFR and Lp(a) levels. In summary, an increase of Lp(a) concentrations, compared with apo(a) phenotype-matched control subjects, is seen in non-nephrotic patients with primary renal disease even in the earliest stage when GFR is not yet subnormal. This change is found only in subjects with HMW apo(a) phenotypes, however.

Multicentre evaluation of combined prothrombotic defects associated with thrombophilia in childhood. Childhood Thrombophilia Study Group

To evaluate the role of multiple established and potential causes of childhood thrombophilia, 285 children with a history of thrombosis aged neonate to 18 years (first thrombotic onset) were investigated and compared with 185 healthy peers. APC-resistance (FV:Q506), protein C, protein S, antithrombin, heparin cofactor II (HCII), histidine-rich glycoprotein (HRGP), and prothrombin (F.II), factor XII (F.XII), plasminogen, homocysteine and lipoprotein (a) (Lp(a)) were investigated. In 59% of patients investigated one thrombotic defect was diagnosed, 19.6% showed two thrombotic risk factors, while in 21.4% of children investigated no risk factor could be identified. Single defects comprised established causes of inherited thrombophilia: FV:Q506 (homozygous n = 10, heterozygous n = 69), protein C (homozygous n = 1; heterozygous n = 31), heterozygous type I deficiency states of protein S (n = 7), antithrombin (n = 7) and homocystinuria (n = 6); potentially inherited clotting abnormalities which may be associated with thrombophilia: F.XII (n = 3), plasminogen (n = 2), HCII (n = 1), increased HRGP (n = 4); new candidate risk factors for thrombophilia: elevated plasma levels of Lp(a) (n = 26), F.II (n = 1). Heterozygous FV:Q506 was found in combination with heterozygous type I deficiency states of protein C (n = 2), protein S (n = 13), antithrombin (n = 8) and HCII (n = 1), increased Lp(a) (n = 13), and once each with elevated levels of F.II, moderate hyperhomocysteinemia, fibrinogen concentrations > 700 mg/dl and increased HRGP. In addition to the association with FV:Q506, heterozygous protein C type I deficiency was combined with deficiencies of protein S (n = 2), antithrombin (n = 1), and increased Lp(a) (n = 3). One patient showed protein C deficiency along with familially increased von Willebrand factor > 250%. Besides coexistence with FV:Q506 and protein C deficiency, protein S deficiency was combined with decreased F.XII and increased Lp(a) in one subject each. Furthermore, we found combinations of antithrombin deficiency/elevated Lp(a), hyperhomocysteinemia/Lp(a), deficiency of HCII/plasminogen, and plasminogen deficiency along with increased Lp(a) each in one. Increased prothrombin levels were associated with fibrinogen concentrations > 700 mg/dl and with HCII deficiency in one child each. Carrier frequencies of single and combined defects were significantly higher in patients compared with the controls.

CONCLUSION

In conclusion, data of this multicentre evaluation indicate that paediatric thromboembolism should be viewed as a multifactorial disorder.

Factor V G1691A and prothrombin G20210A in childhood spontaneous venous thrombosis--evidence of an age-dependent thrombotic onset in carriers of factor V G1691A and prothrombin G20210A mutation

Risk factors for venous thrombosis in adults are the prothrombin (PT) G20210A, the factor (F) V G1691A mutations and hereditary deficiencies of protein C, protein S and antithrombin. However, data are limited on the relevance of these risk factors for thrombosis in children and adolescents. We therefore investigated 119 patients aged 0-18 with spontaneous venous thrombosis and controls (n = 100) for the presence of the factor V G1691A mutation and the prothrombin G20210A variant with respect to thrombotic onset and thrombosis location. The following frequencies (patients vs. controls), odds ratios (OR), 95%-confidence intervals (CI) and p-values were found: FV G1691A, 19.3% vs. 5%, OR/CI 4.55/1.66-12.5, p = 0.0038 and prothrombin G20210A, 8.4% vs. 3%, OR/CI 2.96/0.8-11, p = 0.17. A combination of the FV G1691A mutation with the PT G20210A variant was found in 3 children (2.5% of cases) but only once in the controls. With a median (range) age of 2 years (0-17), carriers of the FV mutation were significantly younger compared with patients carrying the PT variant (16 years: 0-18, p < 0.001). Vascular accidents in carriers of the FV mutation occurred in deep veins of the leg (n = 11), cerebral veins (n = 4), renal veins (n = 3) and portal veins (n = 2). Patients with the PT mutation showed spontaneous thrombosis in the majority of cases in the deep veins of the leg (n = 5) and in the central nervous system (n = 2). Combined defects were found in a neonate with renal venous thrombosis and in two adolescents with deep vein thrombosis.

CONCLUSION

Data presented here suggest that the heterozygous FV mutation is the most commonly found prothrombotic risk factor responsible for spontaneous thrombosis during infancy and early childhood. In contrast, the PT G20210A variant is likely to be more important during puberty and adolescence.

The 677T genotype of the common MTHFR thermolabile variant and fasting homocysteine in childhood venous thrombosis

Controlled data on the association of MTHFR genotypes, hyperhomocysteinaemia and their interaction with factor V G1691A with childhood thrombosis are not yet available. Therefore we conducted a case-control study comparing 141 childhood patients with venous thrombosis with 345 healthy controls. The MTHFR C677T genotypes, FV G1691A and prothrombin G20210A were evaluated; in addition, fasting homocysteine concentrations were measured in a subgroup of 60 children and 80 healthy controls. 10.4% of the healthy control population showed the MTHFR TT genotype, 34.2% the CT genotype and 55.4% the CC variant. MTHFR genotypes account for fasting homocysteine concentrations in healthy controls (CC: 5.5 micromol/l (4-7.2); CT: 7 micromol/l (3.9-9.8); TT: 12.1 micromol/l (7.7-13.3)) with an upper age-specific 95th percentile of 8.3 micromol/l. The following frequencies (patients versus controls), odds ratios (OR) and 95% confidence intervals (CI) were found for single defects: MTHFR 677TT genotype (10.6% vs. 10.4%; OR/CI: 1.02/0.54-1.93; P = 0.99) and CT genotype (43.8% vs. 34.2%; OR/CI: 2.12/1.42-3.16; P = 0.0000). A combination of FV G1691A mutation and MTHFR 677CT genotype was found in 9.9% of patients and in 2.9% of the controls (OR/CI: 3.8/1.64-8.75; P = 0.027). Fasting homocysteine median (range) concentrations in the patient group were significantly higher than in the controls (7 micromol/l (3-23) vs. 5.5 micromol/l (3-8.4): P = 0.0004), and homocysteine concentrations >8.3 micromol/l were found in 40% of patients vs. 2.5% of the controls (OR/CI: 22/2.64-183; P = 0.0003). Conclusion Data of this childhood case-control study suggest that mildly elevated fasting homocysteine concentrations >8.3 micromol/l and the CT genotype of the MTHFR C677T variant are significant risk factors for venous vascular occlusion in children.

Lipoprotein (a) and genetic polymorphisms of clotting factor V, prothrombin, and methylenetetrahydrofolate reductase are risk factors of spontaneous ischemic stroke in childhood

Ischemic stroke is a rare event in childhood. In approximately one third of cases no obvious underlying cause or disorder can be detected. We investigated the importance of genetic risk factors of venous thromboembolism in childhood or stroke in adulthood as risk factors for spontaneous ischemic stroke in children. One hundred forty-eight Caucasian infants and children (aged 0.5 to 16 years) with stroke and 296 age-matched controls from the same geographic areas as the patients were analyzed for increased lipoprotein (a) [Lp(a)] levels >30 mg/dL; for the presence of the factor V (FV) G1691A mutation, the prothrombin (PT) G20210A variant, and the TT677 genotype of methylenetetrahydrofolate reductase (MTHFR); and deficiencies of protein C, protein S, and antithrombin. The following frequencies (patients v controls), odds ratios (ORs), and confidence intervals (CIs) of single risk factors were found: Lp(a) >30 mg/dL (26.4% v 4.7%; OR/CI, 7.2/3.8 to 13.8; P <.0001), FV G1691A (20.2% v 4%; OR/CI, 6/2.97 to 12.1; P <.0001), protein C deficiency (6% v 0.67%; OR/CI, 9.5/2 to 44.6; P =.001), PT G20210A (6% v 1.3%; OR/CI, 4.7/1.4 to 15.6; P =.01), and the MTHFR TT677 genotype (23.6% v 10.4%; OR/CI, 2.4/1.53 to 4.5; P <.0001). A combination of the heterozygous FV G1691A mutation with increased Lp(a) (n = 11) or the MTHFR TT677 genotype (n = 5) was found in 10. 8% of cases, but only 0.3% of controls (OR/CI, 35.75/4.7 to 272; P <. 0001). Increased Lp (a) levels, the FV G1691A mutation, protein C deficiency, the prothrombin G20210A variant, and the MTHFR TT677 are important risk factors for spontaneous ischemic stroke in childhood.

Role of genetic prothrombotic risk factors in childhood caval vein thrombosis

Childhood caval vein thrombosis has a high incidence especially in the first year of life. Besides deficiencies of protein C, protein S, antithrombin and plasminogen, the factor (F) V G1691A mutation, the prothrombin (PT) G20210A variant, the methylenetetrahydro-folate reductase (MTHFR) TT677 genotype, or increased lipoprotein (Lp) (a) > 30 mg/dl have emerged as important prothrombotic risk factors in childhood vascular accidents. 27 consecutive childhood patients with inferior caval vein thrombosis and 100 healthy age-matched controls were investigated for the presence of these prothrombotic risk factors with respect to the first thrombotic onset. In 19 out of 27, patients thrombosis occurred during infancy; the remaining vascular accidents were diagnosed during puberty. In 13 out of the 19 infants, vascular occlusion occurred spontaneously, five times associated with renal venous thrombosis. 68.4% of patients in the first year of life (n = 13) showed at least one prothrombotic risk factor. The FV mutation (heterozygous n = 4, homozygous n = 1). Lp (a) > 30 mg/dl and kringle 4 repeats < 28 (n = 4), MTH FR TT677 with mild hyperhomocysteinaemia (> 95th age-dependent percentile, i.e. 8.5 micromol/l: n = 3) and antithrombin deficiency type II (n = 1) were diagnosed with an overall odds ratio/95% confidence interval of 9.2/3.1-27.4. In the adolescent group, genetic risk factors were found in 50% of patients investigated (FV mutation (n = 1), PT variant (n = 3); odds ratio/95% confidence interval: 4.2/0.97-18.6).

CONCLUSION

Data presented here suggest that genetic prothrombotic risk factors play an important role in childhood caval vein thrombosis. Remarkably, during puberty and adolescence the predominant defect diagnosed was the PT G20210A variant, whereas the FV G1691A mutation had a higher incidence during infancy.

c-erbB-2/EGFR as dominant heterodimerization partners determine a motogenic phenotype in human breast cancer cells

Separate mechanisms for oncogenesis and metastasis have been postulated. We show here that prolonged and invasive cell migration, a key mechanism in cancer metastasis, is linked to c-erbB-2 signaling. Cell lines with c-erbB-2 and EGFR expression and transphosphorylation activity display a high transendothelial invasiveness in an endothelial-extracellular matrix model mimicking a capillary vessel wall in vitro. Tyrosine-phosphorylated c-erbB-2 receptors and EGFR are localized predominantly in areas of the cell with high membrane extension activity. On the molecular level, there is a subtle cross talk between the transmembrane signaling molecule c-erbB-2 and the actin cytoskeleton at multiple levels, including the generation of the second messenger PIP2 and the mobilization of the actin-regulatory protein gelsolin. Our data strongly suggest that c-erbB-2, especially in a heterodimer with EGFR, is closely involved in signaling pathways, inducing alterations in cell morphology that are required for a human breast cancer cell to become motile and conceivably metastatic.

The effects of six months of treatment with a low-dose of conjugated oestrogens in menopausal women

OBJECTIVE

Hormone replacement therapy (HRT) is usually prescribed as medium- to high-dose formulations. Little is known, however, about dose-dependency of oestrogen effects on plasma hormone levels, markers of cardiovascular risk in lipid metabolism and the haemostatic system, or markers of bone turnover.

SUBJECTS AND DESIGN

In an open trial, three groups of 12 or 13 healthy, non-obese postmenopausal women received conjugated equine oestrogens (CEE) for 6 months at doses of 0.3 mg/day (group 1), 0.6 mg/day (group 2) or 1.25 mg/day (group 3). From day 1 to day 10, CEE was administered alone, and from day 11 to day 21, in combination with 5 mg of medrogestone. Each treatment cycle was followed by a pause of 7 days. Fasting blood samples were obtained before treatment as well as on days 10, 21 and 28 of the first, third and sixth months on treatment. All results obtained on day 10 were grouped together as phase A, on day 21 as phase B, and on day 28 as phase C.

MEASUREMENTS

Plasma concentrations of oestradiol (E), dehydroepiandrosterone sulphate (DHEA-S), total testosterone (T), FSH, PRL, sex hormone binding globulin (SHBG), type I procollagen propeptide (PICP) and the cross-linked carboxyterminal telopeptide of type I collagen (ICTP), total cholesterol, HDL-cholesterol, triglycerides (TG), apolipoprotein (apo) A-1, apo B, lipoprotein(a)[Lp (a)], fibrinogen, factor VIIc and plasminogen activator inhibitor 1 (PAI-1) were evaluated with commercially available kits.

RESULTS

Dose-dependently, the three regimens increased E, SHBG and factor VIIc activity and decreased FSH, DHEAS, cholesterol, LDL-cholesterol and apoB. HDL-cholesterol and apoA-1 were slightly decreased in group 1 but increased in groups 2 and 3. The high CEE dosage in group 3 resulted in a significant increase of TG and decrease of Lp(a) and PAI-1. Markers of bone turnover were not significantly changed by any CEE dosage.

CONCLUSIONS

Six months of treatment with 0.3 mg/day of conjugated equine oestrogen significantly lowers serum levels of total cholesterol and LDL-cholesterol without causing the adverse increases of triglycerides or factor VIIc, which were observed at higher doses. However, this low-dose treatment did not yield the maximal LDL-cholesterol lowering effect. Moreover, the positive effects of HRT on HDL-cholesterol, apolipoprotein A-I, lipoprotein (a) and plasminogen activator inhibitor-1 required at least the medium dose of 0.6 mg conjugated equine oestrogens per day. Therefore, further studies are needed to determine which dose of conjugated equine oestrogens has the optimal effect on cardiovascular risk and bone turnover.

Systemic inflammatory parameters in patients with atherosclerosis of the coronary and peripheral arteries

Plasma concentration of markers of inflammation are increased in patients with atherosclerosis. However, it is unclear whether the pattern and magnitude of this increase vary with the site and extent of disease. In 147 patients undergoing semiquantitative coronary angiography, we measured the acute-phase reactants C-reactive protein (CRP) or serum amyloid A (SAA); the proinflammatory cytokine interleukin 6 (IL-6); the active and total fractions of the anti-inflammatory cytokine transforming growth factor-beta (TGF-beta); the macrophage activation marker neopterin; and the infection marker procalcitonin. Compared with 62 patients without either coronary artery disease (CAD) or peripheral artery disease (PAD), 57 patients with CAD but no PAD showed greater median CRP (0. 4 versus 0.2 mg/dL, P=0.004) and IL-6 (3.8 versus 1.6 pg/mL, P=0. 007) levels and a lower level of active-TGF-beta (57 versus 100 ng/mL, P=0.038). Moreover, CRP, IL-6, and neopterin levels showed a positive and the active TGF-beta level a negative correlation with the extent of coronary atherosclerosis. Compared with these 57 patients with CAD alone, 15 patients with PAD and CAD had higher median levels of SAA (17 versus 7 mg/mL, P=0.008), IL-6 (12 versus 4 pg/mL, P=0.002), neopterin (14 versus 11 mg/dL, P=0.006), and total TGF-beta (11834 versus 6417 ng/L, P=0.001). However, these strong univariate associations of markers of inflammation and atherosclerosis were lost in multivariate analysis once age, sex, and high density lipoprotein cholesterol or fibrinogen were taken into account. Increased plasma levels of CRP, SAA, IL-6, TGF-beta, neopterin, and procalcitonin constitute an inflammatory signature of advanced atherosclerosis and are correlated with the extent of disease but do not provide discriminatory diagnostic power over and above established risk factors.