International Journal of Cardiology 187 (2015) 166–174
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Distribution, genetic and cardiovascular determinants of FVIII:c — Data from the population-based Gutenberg Health Study☆ M. Iris Hermanns a,⁎, Vera Grossmann a, Henri M.H. Spronk b, Andreas Schulz c, Claus Jünger c, Dagmar Laubert-Reh c, Johanna Mazur d, Tommaso Gori e, Tanja Zeller f,g, Norbert Pfeiffer h, Manfred Beutel i, Stefan Blankenberg f,g, Thomas Münzel e,j, Karl J. Lackner k, Arina J. ten Cate-Hoek b, Hugo ten Cate b, Philipp S. Wild a,c,j a
Center for Thrombosis and Hemostasis (CTH), University Medical Center of the Johannes Gutenberg-University Mainz, Germany Laboratory for Clinical Thrombosis and Hemostasis, Dept. of Internal Medicine, Cardiovascular Research Institute Maastricht (CARIM), Maastricht University Medical Center, Maastricht, The Netherlands c Preventive Cardiology and Preventive Medicine, Dept. of Medicine 2, University Medical Center of the Johannes Gutenberg-University Mainz, Germany d Institute for Medical Biostatistics, Epidemiology and Informatics, University Medical Center of the Johannes Gutenberg-University Mainz, Germany e Dept. of Medicine 2, University Medical Center of the Johannes Gutenberg-University Mainz, Germany f Clinic for General and Interventional Cardiology, University Heart Center Hamburg, Germany g DZHK (German Center for Cardiovascular Research), Partner Site Hamburg/Kiel/Lübeck, Hamburg, Germany h Department of Ophthalmology, University Medical Center of the Johannes Gutenberg-University Mainz, Germany i Department of Psychosomatic Medicine and Psychotherapy, University Medical Center of the Johannes Gutenberg-University Mainz, Germany j DZHK (German Center for Cardiovascular Research), Partner Site Rhine-Main, Mainz, Germany k Institute for Clinical Chemistry and Laboratory Medicine, University Medical Center of the Johannes Gutenberg-University Mainz, Germany b
a r t i c l e
i n f o
Article history: Received 24 November 2014 Received in revised form 18 February 2015 Accepted 20 March 2015 Available online 21 March 2015 Keywords: Venous thrombosis Arterial thrombosis Epidemiological studies FVIII:c reference values
a b s t r a c t Background: Elevated levels of FVIII:c are associated with risk for both venous and arterial thromboembolism. However, no population-based study on the sex-speciﬁc distribution and reference ranges of plasma FVIII:c and its cardiovascular determinants is available. Methods: FVIII:c was analyzed in a randomly selected sample of 2533 males and 2440 females from the Gutenberg Health Study in Germany. Multivariable regression analyses for FVIII:c were performed under adjustment for genetic determinants, cardiovascular risk factors and cardiovascular disease. Results and conclusions: Females (126.6% (95% CI: 125.2/128)) showed higher FVIII:c levels than males (121.2% (119.8/122.7)). FVIII:c levels increased with age in both sexes (ß per decade: 5.67% (4.22/7.13) male, 6.15% (4.72/7.57) female; p b 0.001). Sex-speciﬁc reference limits and categories indicating the grade of deviation from the reference were calculated, and nomograms for FVIII:c were created. FVIII:c was approximately 25% higher in individuals with non-O blood type. Adjusted for sex and age, ABO-blood group accounted for 18.3% of FVIII:c variation. In multivariable analysis, FVIII:c was notably positively associated with diabetes mellitus, obesity, hypertension and dyslipidemia and negatively with current smoking. In a fully adjusted multivariable model, the strongest associations observed were of elevated FVIII:c with diabetes and peripheral artery disease in both sexes and with obesity in males. Effects of SNPs in the vWF, STAB2 and SCARA5 gene were stronger in females than in males. The use of nomograms for valuation of FVIII:c might be useful to identify high-risk cohorts for thromboembolism. Additionally, the prospective evaluation of FVIII:c as a risk predictor becomes feasible. © 2015 Published by Elsevier Ireland Ltd.
Abbreviations: CAD, history of coronary artery disease; CHF, history of chronic heart failure; COPD, chronic obstructive pulmonary disease; CRP, C-reactive protein; CVD, history of cardiovascular disease; CVRF, cardiovascular risk factor; FVIII, coagulation factor VIII; FVIII:c, Factor VIII coagulant activity; GHS, Gutenberg Health Study; LDLR, low density lipoprotein receptor; LD, linkage disequilibrium; LRP1, low density lipoprotein receptor-related protein 1; MI, history of myocardial infarction; PAD, history of peripheral artery; SCARA5, scavenger receptor class A, member 5; SNP, single nucleotide polymorphism; STAB2, stabilin 2; STXBP5, syntaxin binding protein 5; vWF, von Willebrand factor; VTE, history of venous thromboembolism ☆ All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ⁎ Corresponding author at: CTH Center for Thrombosis and Hemostasis, Clinical Epidemiology, Center for Thrombosis and Hemostasis, University Medical Center of the Johannes Gutenberg-University Mainz, Langenbeckstr. 1, 55131 Mainz, Germany. E-mail address: [email protected]
http://dx.doi.org/10.1016/j.ijcard.2015.03.330 0167-5273/© 2015 Published by Elsevier Ireland Ltd.
M.I. Hermanns et al. / International Journal of Cardiology 187 (2015) 166–174
1. Introduction Circulating levels of the coagulation factor VIII (FVIII) have been associated with risk for both venous [1–5] and arterial thromboembolism [6–11]. The thrombogenic effect of FVIII emerges with increased protein concentrations, mediated via its catalytic effect on thrombin generation, also attenuating the anticoagulant effect of activated protein C . In the circulation, the majority of FVIII molecules (95%–98%) are in complex with von Willebrand factor (vWF) , and can be detected as FVIII coagulant activity (FVIII:c). The laboratory reference range of FVIII:c shows a large distribution. This variation is caused by multiple factors. Genetic factors seem to play an important role in the variability of FVIII:c. Although the heritability of FVIII plasma levels is high (near 40%–60%) , variations of FVIII gene (Xq28) explain only a minor proportion of this broad variance . Genetic studies show that, besides the vWF (12p12), the ABO blood group locus on chromosome 9q34 is the most important genetic determinant of plasma levels of FVIII:c . The ABO blood group determinants have been proposed to affect through different glycosylation patterns the processing and/or the clearance of vWF, which in turn binds and stabilizes FVIII . In most conditions, there is a concordant increase of FVIII and vWF:Ag levels . However, in patients with VTE only 50% of persistently high FVIII:c levels were associated with high vWF:Ag levels , indicating that vWF is not the only determinant for higher FVIII:c levels. High plasma levels of FVIII:c are found to be a result of increased synthesis or decreased clearance of this complex . Two multifunctional endocytic receptors mediate cellular uptake and subsequent degradation of FVIII, the low density lipoprotein receptor (LDLR)  and the LDLR-related protein 1 (LRP1) . Single nucleotide polymorphisms (SNPs) at the LDLR (19p13.2) and the LRP1 locus (12q13.3) have been associated with FVIII:c and the risk of coronary artery disease  or the risk of venous thromboembolism , respectively. The aim of the present study was to quantify determinants for plasma levels of FVIII:c including genetics, lifestyle and cardiovascular risk factors and disease in a large population-based sample. Finally, we aim to provide sex-speciﬁc reference ranges for FVIII:c in a population of European descent.
without biomaterial available (n = 27), 4973 individuals were successfully included in the analysis of FVIII:c. For the analysis of genetic determinants genome-wide association data were available in 4175 individuals. 2.3. Reference sample The reference group was deﬁned as a sample of apparently cardiovascular healthy subjects with a low risk for cardiovascular disease within the larger population sample. Hence, the reference sample included subjects without recognized (i.e., self-reported) history of myocardial infarction (MI), stroke, chronic heart failure (CHF), coronary artery disease (CAD), or peripheral artery disease (PAD) and without observed cardiovascular risk factors during the examination (i.e., normotensive, nonobese, nondiabetic, nonsmoking individuals without a family history of MI or stroke). To avoid the interference of an acute infection on FVIII:c levels, only individuals with a concentration of C-reactive protein (CRP) b 5 mg/L were included in the reference group. 2.4. Clinical assessment and laboratory analyses Clinical analyses were performed as published elsewhere [25,26]. The Framingham score (FH10y CVD score) was used to predict the probability of developing cardiovascular disease over a 10-year period . All laboratory measurements were carried out in a centralized set-up. Venous blood sampling was performed in supine position while the subject was in fasting state (i.e. overnight fast, if subject was examined before 12 a.m. and 5 hour fast, if subject was examined after 12 a.m.). FVIII:c was assessed in a 1-stage clot-based assay using an BCS XP® System (Siemens Healthcare Diagnostics, Marburg, Germany) and FVIIIdeﬁcient plasma. A commercial reference plasma (Siemens Healthcare) was used in each assay run for calibration. Additional assay control samples included a commercial high and low level control sample (control plasma N, control plasma P, Siemens Healthcare). C-reactive protein (CRP) was measured by immune-turbidimetry (Abbott, Wiesbaden, Germany).
2. Methods 2.5. Genotyping and imputation of single nucleotide polymorphisms 2.1. Research design The Gutenberg Health Study (GHS) is designed as a population-based, prospective, observational, single-center cohort study in the Rhein-Main region in western mid-Germany and includes a total of 15,010 individuals . The primary aim of the GHS is to evaluate and improve cardiovascular risk stratiﬁcation. The sample was drawn randomly from the governmental local registry ofﬁces in the City of Mainz and the district of Mainz-Bingen. The sample was stratiﬁed 1:1 for sex and residence (urban and rural) and in equal strata for decades of age. Individuals between 35 and 74 years of age were enrolled, and written informed consent was obtained from all participants. Details of the study protocol and the further purposes of the study are discussed elsewhere . The study was designed according to the tenets of the revised Helsinki protocol and protocol and sampling design were approved by the local ethics committee and by the local and federal data safety commissioners. Every participant underwent a comprehensive, standardized 5-hour clinical investigation. In addition to the clinical assessment, a large biobank has been established for biochemical and genetic analyses. All participants gave informed written consent to laboratory analyses, clinical examinations, sampling of biomaterial and the use of data records for research purposes. 2.2. Study sample We investigated baseline data of the ﬁrst 5000 subjects enrolled into the GHS between April 2007 and October 2008. After excluding subjects
Genotyping was conducted on the Affymetrix Genome-Wide Human SNP 6.0 Array (Affymetrix, Santa Clara, CA) according to manufacturer's recommendations. Genetic analysis was performed in 4175 individuals. Before genotype imputation, SNPs showing signiﬁcant (p b 10−4) deviation from the Hardy–Weinberg Equilibrium (HWE), with minor allele frequency (MAF) less than 1% or having a genotyping call rate b 98% were excluded. A two-step process was performed for the imputation. First, pre-phasing was done using the software MACH (v1.0.18c). Second, imputation of genotypes using the reference panel 1000G Phase I Integrated Release Version 2 Haplotypes (2010–11 data freeze, 2012-02-14 haplotypes) was performed with the software minimac (release 2012-03-14). The imputation results were ﬁltered using a minimum imputation quality score r2 of N 0.3 and a MAF threshold of 2%. Imputation results were summarized as “allele dosage”, deﬁned as the expected number of copies of the minor allele for each SNP (value 0, 1 and 2) and genotype. A ratio of observed to expected variance of the dosage statistic for each SNP was calculated for imputation quality. Association of SNPs described by previous GWAS [28,29] with FVIII levels was tested by means of genotype linear association analysis. Among others SNPs in syntaxin-binding protein 5 (STXBP5), scavenger receptor class A, member 5 (SCARA5) and stabilin 2 (STAB2) structural genes were described associated with FVIII:c [28,29]. For one of these probes, the GWAS SNP was present on the Affymetrix array used in this study. For 5 other probes, the GWAS SNP had one or several proxies in linkage disequilibrium (LD) on the array (see Table S6). To infer the genotypes of the ABO blood group, we used the
M.I. Hermanns et al. / International Journal of Cardiology 187 (2015) 166–174
SNPs rs657152, rs612169, rs651007, and rs8176672 (see Tables S7, S8). Homozygous carriers of the C allele of rs657152  or the A allele of rs612169  were used as a marker for the O blood type. rs651007 is in high LD (r2 = 0.96) with rs507666 which is described as surrogate for type A1 allele . The T allele of rs8176672 tags the B blood group allele . 2.6. Data management and statistical analyses All data of the present investigation underwent quality control by a central data management unit. Data were reviewed for completeness by predeﬁned algorithms and plausibility criteria. Descriptive data are presented by relative and absolute frequency for categorical, mean and standard deviation for continuous traits. Data were stratiﬁed by sex for analysis. To take the stratiﬁed nature of the sample into account, baseline characteristics (Table S1) were weighted according to the age and sex distribution in the study population. Regression coefﬁcients (betas) per increase in one unit from linear regression models are presented with corresponding p-values. All statistical comparisons were two-tailed, p-values b 0.05 were considered as relevant associations. To create nomograms, FVIII:c values were categorized according to sex-speciﬁc percentiles using quantile regression. The categories shown in Table S2 indicate the respective reference limits and grades of deviation from the reference limit. Statistical analysis was performed using R, version 3.0.1 (http://www.R-project.org). 3. Results 3.1. Sample characteristics The GHS study sample comprises 2533 (50.8%) male and 2440 (49.2%) female individuals with a median age of 56 years (interquartile range (IQR) 46–65 years) (see Table S1). From the sample of the ﬁrst 5000 subjects enrolled in the GHS, 13.48% (n = 674) were eligible for inclusion in the reference sample of apparently cardiovascular healthy persons. While there was a balanced sex ratio in the overall sample, the reference subsample showed a slight preponderance of female individuals (57.7%). As expected the reference sample had a right-skewed
distribution with a decreasing portion of older subjects (see Fig. S1). Distribution patterns of SNPs associated with O-blood group were comparable in study and reference sample and reﬂect the high prevalence of about 39% blood group O in the central European population. The risk factors included in the Framingham risk prediction algorithm to estimate the incidence of CVD were used to determine the prevalence of classic CVRFs . FH10y CVD score within GHS study sample was 11.49% (95% CI: 5.64/21.80) as expected with a higher risk in males than in females (17.76% (9.40/29.70) vs 7.31% (3.55/13.34) Table S1). 3.2. Distribution and determinants of plasma FVIII:c FVIII:c was measured in 4973 subjects with a mean value (SD) of 121% (36.7) and 127% (35.2) for males and females, respectively (Fig. 1). In general, means, medians, SD intervals, and 5th and 95th percentiles for FVIII:c were higher in females than in males. This sexdifference was also present in the reference sample (males: 111% (34.8), females: 120% (32.7)). FVIII:c plasma activity is increased in both, males and females, for all CVRFs, except for individuals smoking, that had comparable levels of FVIII:c to the reference sample. Males and females with a history of venous or arterial thrombotic events, e.g. venous thromboembolism (VTE), coronary artery disease (CAD), peripheral artery disease (PAD), showed an increased FVIII:c level compared to reference individuals. In general, all disease conditions analyzed accounted for increases of FVIII:c plasma activity that are analyzed further in more detail. There was a signiﬁcant linear correlation between age and the plasma level of FVIII:c (p b 0.0001). For this reason, statistical analyses were adjusted for age. In an age adjusted linear regression each CVRF, except family history of MI or stroke, was associated with FVIII:c (Table 1). Diabetes, obesity, hypertension and dyslipidemia accounted for absolute increases in activity of 14.5% (95% Cl: 9.82/19.3), 8.85% (5.48/11.8), 4.56% (1.57/7.56) and 4.80% (1.94/7.66) in males and 11.5% (5.32/ 17.7), 9.91% (6.64/13.2), 4.89% (1.89/7.88) and 3.77% (0.405/7.14) in females, respectively. In general, increases of FVIII:c levels in females were associated with a history of thrombotic events (stroke, ß = 16.8, p = 0.0038; PAD, ß = 10.5, p = 0.0042; CAD, ß = 12.8, p = 0.0080; VTE, ß = 10.6, p = 0.00083). The same applies to male individuals, that
Fig. 1. Sex-speciﬁc FVIII:c levels in the reference sample, the study sample and deﬁned subgroups. Presented are mean with the 95% conﬁdence interval for the respective sample. The reference sample is deﬁned as subjects without CVRF, CV diseases and acute inﬂammation (CRP N 5 mg/L), n = 674.
M.I. Hermanns et al. / International Journal of Cardiology 187 (2015) 166–174
Table 1 Sex-speciﬁc inﬂuence of CVRFs and diseases on FVIII:c levels presented as beta-estimates with 95% conﬁdence interval from linear regression models. Variable
Diabetes Obesity Smoking Hypertension Dyslipidemia FH of MI/stroke MI Stroke AF PAD CAD CHF VTE COPD Cancer
Beta coeff. (L 95%/U 95% CI)
Beta coeff. (L 95%/U 95% CI)
247 644 526 1423 938 876 119 58 98 109 173 40 78 116 201
14.5 (9.82/19.3) 8.65 (5.48/11.8) −5.57 (−9.03/−2.11) 4.56 (1.57/7.56) 4.80 (1.94/7.66) −0.608 (−3.52/2.31) 9.25 (2.61/15.9) 3.67 (−5.66/13.0) 4.40 (−2.84/11.6) 13.1 (6.22/19.9) 4.17 (−1.45/9.79) 9.30 (−1.82/20.4) 12.2 (3.94/20.4) 6.00 (−0.612/12.6) 7.03 (1.82/12.2)
b0.0001 b0.0001 0.0016 0.0028 0.0010 0.68 0.0063 0.44 0.23 0.00018 0.15 0.10 0.0037 0.075 0.0082
127 555 427 1129 538 946 36 36 38 93 53 37 131 122 245
11.5 (5.32/17.7) 9.91 (6.64/13.2) −7.26 (−10.9/−3.62) 4.89 (1.89/7.88) 3.77 (0.405/7.14) 1.67 (−1.16/4.49) 10.3 (−1.11/21.7) 16.8 (5.41/28.2) 3.21 (−7.88/14.3) 10.5 (3.31/17.7) 12.8 (3.35/22.3) 9.45 (−1.82/20.7) 10.6 (4.37/16.8) 0.752 (−5.52/7.03) 0.658 (−3.94/5.26)
0.00027 b0.0001 b0.0001 0.0014 0.028 0.25 0.077 0.0038 0.57 0.0042 0.0080 0.10 0.00083 0.81 0.78
Linear regression models with FVIII:c as dependent and the respective risk factor or disease as independent variable; all adjusted for age. Smoking was dichotomized into non-smokers (never smokers and former smokers) and current smokers. MI = myocardial infarction; AF = atrial ﬁbrillation; PAD = peripheral arterial disease; CAD = coronary artery disease; CHF = congestive heart failure; PE = pulmonary embolism; VTE = venous thromboembolism; COPD = chronic obstructive pulmonary disease; p-values are shown for the comparison of FVIII:c between event (CVRFs or disease) and non-event for male and female individuals adjusted for age. Data available in 99%. All data with a p-value below 0.05 are presented in bold.
exhibited increased FVIII:c levels with a history of MI (ß = 9.25, p = 0.0063), PAD (ß = 13.1, p = 0.00018), or VTE (ß = 12.2, p = 0.0037). Age adjusted linear regression revealed also the association of FVIII:c with other comorbidities like cancer (ß = 7.03, p = 0.0082) in men. 3.3. Nomograms Due to the signiﬁcant associations of CVRFs with FVIII:c levels a subsample of apparently cardiovascular healthy subjects was deﬁned as reference group. Nomograms relating FVIII:c to age were developed, partitioned by sex. Sex-speciﬁc reference limits and categories indicating the grade of deviation from the reference are shown in Fig. 2 (for O-and non-O blood type see Figs. S2–S3). The regression lines refer to the median of the reference category (dashed line), followed by the 95th percentile for the reference sample and the 98th, and 99th percentiles for the population sample (see Table S2). These percentiles can be used to discriminate the reference category (light green), and the categories mild (yellow), moderate (orange), and severe (red) deviation from the reference. Nomograms depicting the estimate for upper and lower percentiles of FVIII:c in fresh plasma samples and equations to calculate the sex-speciﬁc percentiles of FVIII:c according to age are presented in the supplement (Fig. S4, Tables S3–S4).
signiﬁcant associations were found in the ABO related SNPs and in only one SNP on the vWF locus, rs1063857. For the ABO-group related SNPs a higher increase per one % FVIII:c value was found in males than in females for rs651007 (tagging the A1-blood type) and rs8176672 (tagging the B-blood type) (ßmale = 23.6 (95% CI: 17.0/30.3) vs ßfemale = 12.6 (5.86/19.3) for major allele homozygotes of rs651007). In this respect, the decrease per one % FVIII:c activity was not as strong in males as compared to females for the O-blood group (ß male = − 9.29, p b 0.0001;
3.4. Genetic background of FVIII:c levels The distribution of all SNPs reported to be associated with FVIII:c in literature was comparable between males and females in the sample (see Table S9). SNPs were analyzed sex-speciﬁcally in regression models adjusted for age (see Table S10). Overall, 39.7% of males and 38.4% of females were detected as having blood group O. O-blood type accounted for 11% of FVIII:c variation when it was analyzed in a univariate linear model. When adding the covariates sex and age to the model, 18.3% of the variation was explained. The FVIII:c levels observed for O-blood type individuals were 24.4% (95% Cl: 26.5/22.3) lower than in individuals with non-O blood type. Multivariable analysis of all SNPs adjusted for age and all covariates explained 22.5% and 20.3% of the FVIII:c variation in male and female individuals, respectively. This analysis further showed that except for the SNPs in the STXBP5 gene (rs1496033, rs1992389), the LDLR gene (rs688), and the LRP1 gene (rs1799986) all other SNPs were shown to be signiﬁcantly associated with FVIII:c in females (Fig. 3). For males
Fig. 2. Nomograms: Age- and sex-speciﬁc levels of FVIII:c. The colored sections indicate the following: Upper limit of the reference range (light green, i.e. the 95th percentile of the reference sample) and categories indicating a mild (yellow, below 98th percentile in population sample), moderate (orange, below 99th percentile in population sample), and severe (red, above 99th percentile in population sample) deviation from the reference. The dashed line indicates the median of the reference sample. (For interpretation of the references to color in this ﬁgure legend, the reader is referred to the web version of this article.)
M.I. Hermanns et al. / International Journal of Cardiology 187 (2015) 166–174
ßfemale = − 15.8, p b 0.0001). In contrast, for SNPs in the vWF gene (rs1063857), the STAB2 gene (rs2271637) and in the SCARA5 gene (rs9644133) more signiﬁcant associations with FVIII:c were found in females (Table S10, Fig. 3). 3.5. Multivariable analysis of determinants of plasma FVIII:c Based on the ﬁndings above, determinants of FVIII:c levels were ﬁnally assessed by multivariable linear regression. Table 2 depicts the explained variability of FVIII:c plasma levels for different models. Age explained approximately 6% of the variation of FVIII:c (model 1). Adding CVRFs to the model, increased the R2 to 9.0% and 7.7% in men and women, respectively (model 2). An increase of FVIII:c level due to acute inﬂammation was taken into account by adding C-reactive protein (CRP), which accounted for another 3.9% in men and 3.0% in women (model 3). Including diseases to the model did not have a substantial effect on FVIII:c variation (model 4). The strong associations of SNPs in the ABO and vWF gene as genetic determinants of FVIII:c activity were conﬁrmed in the fully adjusted multivariable model 5. In total, the large model including all determinants explained 26% of the variation of FVIII:c in males and 24% in females (model 5) which was primarily driven by the genetic variants. In the large multivariable model (Table 3), strong independent determinants for FVIII:c were the age (ßfemale = 6.1; ßmale = 5.7, p b 0.0001), and among the genetic factors the ABO-blood type and one SNP (rs1063857) in the vWF gene (ßfemale = 8.4, p = 0.00011). Increase in 1 mg/L of the inﬂammation-marker CRP caused moderate increases of 1.7% FVIII:c in women and 1% in men (p b 0.0001). For CV
risk factors associations with FVIII:c were positive with diabetes (ßfemale = 10.7, p = 0.0011; ßmale = 5.99 p = 0.023) and negative with smoking (ßmale = − 8.7, p b 0.0001; ßfemale = − 7.2; p = 0.00011) in both sexes (Table 3). Obesity (ß = 4.15, p = 0.015), hypertension (ß = 3.16, p = 0.041), and dyslipidemia (ß = 3.01, p = 0.039) were signiﬁcant determinants of elevated FVIII:c in males only. PAD (ßmale = 8.41, p = 0.031; ßfemale = 7.53, p = 0.037) stayed independently associated with elevated FVIII:c in both sexes. Cancer (ß = 6.89, p = 0.0093) was, independently of the other factors, associated with increased FVIII:c levels in males. 4. Discussion 4.1. Inﬂuence of sex, age, CVRFs and disease on FVIII:c level Using a population-based sample from the Gutenberg Heart Study (GHS), we analyzed 2533 male and 2440 female individuals of a central European population for associations of sex, age, hereditary and cardiovascular risk factors and diseases with plasma levels of FVIII:c. Consistent with previous reports [6,33,34], FVIII:c varied signiﬁcantly among subjects. As already shown by others, FVIII:c levels increase with increasing age (linear correlation; p b 0.001) [33,35]. We could also conﬁrm that FVIII:c is positively associated with female sex [33,36–38]. However, to the best of our knowledge no study aimed to identify maximum and average ranges of plasma FVIII:c levels by age and sex in the population at large. Additionally, this study provides an accurate comparative analysis that quantiﬁes the cardiovascular and genetic determinants of plasma FVIII:c variation in a population-based observational sample.
Fig. 3. Inﬂuence of genetic variants on plasma FVIII:c. Forest plot for SNPs associated with plasma FVIII:c from a multivariable model. SNPs reported as being associated with FVIII:c were selected from literature. Linear regression was adjusted for age and for all SNPs listed. Regression coefﬁcients (beta) for SNPs are depicted for males and females with the corresponding p-value. *Participants with “−T” at rs8176672 and “TT” at rs8176672 were combined as the frequency of homozygous individuals for the minor allele was too low.
M.I. Hermanns et al. / International Journal of Cardiology 187 (2015) 166–174 Table 2 Determinants of FVIII:c variability in males and females. Model Variables in model
Adj. R2 Adj. R2 (male) (female)
0.129 0.130 0.264
0.107 0.104 0.244
3 4 5
Age [10 y] Model 1 + diabetes, obesity, smoking, hypertension, dyslipidemia, family history of MI or stroke Model 2 + CRP Model 3 + diseases Model 4 + rs1063857 (vWF), O-blood type, rs651007, rs8176672 (ABO)
Multivariable linear regression for FVIII:c as dependent variable with the following models: model 1, adjusted for age; model 2 as model 1, plus all cardiovascular risk factors; model 3 as model 2, plus CRP (C-reactive protein); model 4 as model 3, plus cardiovascular diseases and comorbidities (i.e. MI, stroke, AF, PAD, CAD, CHF, VTE, COPD, and cancer); model 5 as model 4 plus genetic variants signiﬁcantly associated with FVIII:c. Adjusted R2 is adjusted for the number of covariates in the model.
Previous studies cited the association of FVIII:c activity with CVRFs [33,34,36]. We corroborate the previous ﬁndings and showed that FVIII:c levels were notably associated positively with diabetes mellitus, obesity, hypertension and dyslipidemia and negatively with current smoking (at least with FVIII:c levels not notably different from the reference population in this study). Although chronic smoking is an important risk factor in relation to atherothrombosis , we and several other studies [34,36,40] found a negative association of FVIII:c with smoking. This ﬁnding suggests other biological mechanisms to be more important in smokers: Active smokers have an increased thrombogenic potential associated with signiﬁcant changes in the coagulation system, like increased plasma tissue factor  and plasma ﬁbrinogen . Additionally, in smoking-mediated athero-thrombotic diseases alterations in platelet function [43,44] and abnormally high resistance to ﬁbrinolysis [45,46] are described. Furthermore, smoking is known to cause dysfunction of endothelial cells and thus reduction of the FVIII-stabilizing vWF . A recent study supporting this theory displayed signiﬁcantly higher values for ﬁbrinogen, FXIII, and tissue
factor pathway inhibitor in active smokers; whereas FVII, FVIII, FXII, vWF, and thrombomodulin were decreased . Nomograms for FVIII:c plasma activity were developed to assess the individual probability of increased FVIII:c based on individually weighted variables, sex and age, with a subsample of cardiovascular healthy individuals serving as reference. The results obtained emphasize the importance of sex-speciﬁc nomograms relating FVIII:c to age. For example, for a 50 year old woman a FVIII:c value of 175% is still in the reference range, whereas 175% FVIII:c is a moderate deviation from the reference for a man of 35 years. On the other hand, a plasma FVIII:c of 195% demonstrates a severe deviation from the reference for a 35 year old woman, for a man of 50 this value accounts only for a mild deviation. Thus, for the interpretation of individual FVIII:c levels age- and sexspeciﬁc reference values should be taken into clinical practice. In addition to clear roles of FVIII:c in venous thrombosis [3–5,18], several studies suggest the relationship of FVIII:c with an increased risk for arterial thrombosis, particularly myocardial infarction and stroke [4,9,49,50]. Several (but not all) large prospective studies in healthy individuals report an association between elevated FVIII:c and the incidence of ischemic heart disease (Odds Ratio (OR) per 1 SD increase in FVIII:c: OR 1.0 (Atherosclerosis Risk in Communities (ARIC) study) ; OR 1.2 (Northwick Park Heart Study) ; OR 1.3 (Caerphilly Heart Study) ). In the Caerphilly Heart Study, 8.9% of patients with ischemic heart disease had FVIII:c levels exceeding 123 IU/dL (=123% of normal), with an associated relative risk of 1.78 after adjustment for age and smoking . ARIC investigators indicated an independent association of FVIII with mortality, but not CVD, in both men and women . In contrast, in elderly individuals FVIII was related to incident cardiovascular disease and death [52,53]. Here, FVIII was also signiﬁcantly associated with coronary heart disease events and mortality in men, and with stroke or transient ischemic attack in women . Recently published papers strengthen the importance of FVIII in risk assessment for recurrent thrombotic events following acute ischemic stroke [11,54]. In combination, FVIII and vWF levels may serve as clinically useful stroke biomarkers, as increases in both markers (FVIII+/vWF+) had increased odds of inpatient complications (OR 8.6; p = 0.013),
Table 3 Determinants of FVIII:c levels. Variable
Age [10 y] Diabetes Obesity Smoking Hypertension Dyslipidemia FH of MI/stroke MI Stroke AF PAD CAD CHF VTE COPD Cancer CRP [mg/L] rs1063857 rs657152 rs612169 rs651007
A/G vs A/A G/G vs A/A C/C A/A C/T vs C/C T/T vs C/C T N 1 vs C/C
Beta coeff. (L 95%/U 95% CI)
Beta coeff. (L 95%/U 95% CI)
5.67 (4.22/7.13) 5.99 (0.831/11.2) 4.15 (0.804/7.49) −8.36 (−11.8/−4.90) 3.16 (0.132/6.19) 3.01 (0.151/5.88) −0.752 (−3.64/2.14) 6.26 (−2.27/14.8) 7.01 (−2.73/16.7) 2.73 (−4.56/10.0) 8.41 (0.746/15.5) −1.81 (−8.70/5.07) −5.40 (−17.1/6.26) 2.60 (−5.84/11.0) 4.73 (−1.89/11.4) 6.89 (1.70/12.1) 0.99 (0.736/1.24) 3.13 (0.397/6.21) 4.09 (−0.406/8.59) −9.54 (−13.6/−5.51)
b0.0001 0.023 0.015 b0.0001 0.041 0.039 0.61 0.15 0.16 0.46 0.031 0.61 0.36 0.55 0.16 0.0093 b0.0001 0.026 0.075 b0.0001
6.15 (4.72/7.57) 10.7 (4.28/17.1) 2.42 (−1.12/5.95) −7.17 (−10.8/−3.56) 1.41 (−1.63/4.45) 0.,385(−3.00/3.77) 0.788 (−2.01/3.58) 4.34 (−9.86/18.5) 4.88 (−7.92/17.7) 1.48 (−12.0/14.9) 7.53 (0.454/14.6) 0.226 (−9.92/10.4) 2.72 (−11.0/16.4) 4.31 (−2.16/10.8) −0.725 (−7.23/5.78) −0.493 (−5.11/4.13) 1.70 (1.34/2.06) 5.54 (2.65/8.44) 8.41 (4.15/12.7) −14.1 (−18.1/−9.99)
b0.0001 0.0011 0.18 0.0001 0.36 0.82 0.58 0.55 0.46 0.83 0.037 0.97 0.70 0.19 0.83 0.83 b0.0001 0.00018 0.00011 b0.0001
17.0 (13.2/20.8) 22.7 (16.1/29.3) 16.9 (12.5/21.4)
b0.0001 b0.0001 b0.0001
11.6 (7.82/15.4) 13.1 (6.52/19.7) 10.6 (6.15/15.1)
b0.0001 0.00010 b0.0001
Multivariable linear regression for FVIII:c as dependent variable, model adjusted for all covariates. All data with a p-value below 0.05 are presented in bold. a O-blood type (individuals with SNP rs657152, haplotype C/C or rs612169 haplotype A/A). b Participants with “−T” at rs8176672 and “TT” at rs8176672 were combined because minor homozygous individuals were too sparse to be analyzed individually.
M.I. Hermanns et al. / International Journal of Cardiology 187 (2015) 166–174
neuroworsening (OR 3.2; p = 0.022) and a poor functional outcome (OR 2.87; p = 0.021) compared to patients with FVIII−/vWF− . Although the association of plasma levels of FVIII:c with the risk of venous and arterial thromboembolic disease has been widely recognized, FVIII:c has not been used as a diagnostic tool to identify patients at risk. This is in part due to the fact that baseline level of FVIII:c was found to vary considerably among individuals. In this regard, an evaluation of the individual probability of increased FVIII:c by the use of nomograms might be a helpful tool to assess the risk of thromboembolic disease. Not only in the nomograms but also for the determinants of FVIII:c notable sex-differences appeared. In a fully adjusted multivariable analysis including sex, age, CVRFs, main hereditary factors and diseases, most of the CVRF and genetic associations with FVIII:c were conﬁrmed. In females the strongest association observed was of elevated FVIII:c and diabetes. In males signiﬁcantly increased plasma FVIII:c was found for the CVRFs diabetes, obesity, and dyslipidemia. An association of elevated FVIII:c with CAD, MI and stroke was not afﬁrmed in our study, when a fully adjusted multivariable model was applied. The loss of association in the fully adjusted model suggests that primarily genetic and to a lower extent CV risk factors drive FVIII:c, but not additional (chronic) CV disease. Importantly, this population based study design does not include individuals with acute state of disease. The same applies to the association of FVIII:c with VTE in the fully adjusted model, although high levels of FVIII in patients with thrombosis are described to persist over time [1,19]. Nevertheless, in our fully adjusted multivariable model, PAD remained independently associated with elevated FVIII:c in males and females. PAD as such might be an exception as it reﬂects a distinct etiology of chronic athero-thrombotic disease over a large vascular bed compared to e.g. CAD, MI or stroke, which have a more heterogeneous background. Elevated FVIII:c was already shown in individuals with PAD in one population-based  and one case–control study . Therefore, additional studies based on the relation of elevated FVIII:c levels with PAD might be helpful to further explore the presence of unknown FVIII:c determinants for this particular disease type. 4.2. Genetic variation of baseline level of FVIII:c The interindividual variation of FVIII:c is caused by multiple factors, but genetic inﬂuence is signiﬁcant. One major genetic factor is ABO blood type, which is discussed in literature to affect FVIII activity primarily because of ABO impact on the FVIII carrier molecule vWF . There is a debate as to whether vascular disease risk relates to vWF or FVIII levels, or if they act independently [6,53,58]. In our study ABO (deﬁned as O type vs non-O type) accounted for 18.3% of FVIII:c variation when it was analyzed in addition to the covariates sex and age in a single linear model. In particular, FVIII:c levels are approximately 25% higher in individuals with non-O-blood group  that was shown sex-independently in this study. However, a clear sex-difference of SNP effects in the ABO gene should be considered. In SNPs tagging the A1 (rs651007) and B (rs817667)-blood types greater effects (ß-estimates, Fig. 3) on FVIII:c increase were found in males compared to females. Conversely, O blood type had greater allele dose-dependent effect on reduction of FVIII:c in females. The associations between FVIII:c levels and the ABO gene variants, and between ABO and vascular disease, suggest a possible mechanism behind the well-documented association between the ABO blood group and risk of vascular disorders . The ABO gene encodes a glycosyltransferase enzyme that catalyzes the transfer of different carbohydrate groups onto the H antigen, thus forming A and B antigens of the ABO system. In a systematic review and meta-analysis Wu and colleagues  conﬁrmed the linkage between vascular disorders (MI, angina, PAD, cerebral ischemia of arterial origin, and VTE) and non-Oblood group status. Non-O-blood groups have also been shown to be
risk factors for venous thrombosis  and, in a large prospective study, pulmonary embolism , suggesting coagulation as critical determinant. The association of ABO blood group with FVIII:c may account for this. The ABO blood groups show consistent associations with a large number of traits and may have broader impact on CVD than simply through modulation of thrombosis . It has been suggested that the increased glycosyltransferase activity in individuals carrying the A1 allele (rs651007 in high LD (r2 = 0.96) with rs507666) might have an effect on the shedding, clearance or secretion of adhesion molecules, thereby inﬂuencing their levels in the circulation [32,63]. Thus, the associations of rs651007 with FVIII:c that we found in both sexes in this study may represent a similar effect on secretion or clearance of FVIII:c. Additionally, the association with SNPs in SCARA5 and STAB2 genes presented in females generates the hypothesis that several loci may be involved in FVIII clearance or uptake. SCARA5 belongs to the group of scavenger molecules that participate in the clearance of various polyanionic macromolecules, pollution particles, and pathogens . STAB2 also acts as a scavenger receptor for heparin, acetylated lowdensity lipoprotein, pro-collagen propeptides and advanced glycation end products . The previously reported association of the rs688, a functional LDLR SNP, with FVIII:c , was not replicated in our study as also reported by a meta-analysis of GWAS in three population-based cohort studies . The same applies to the SNP rs1799986 (LRP1 gene), that was recently shown to be associated with higher FVIII plasma levels in heterozygote individuals . Although previous studies have cited an association of LRP1 polymorphisms with thrombosis [23,66], other studies have presented contradictory results for such an association [67,68]. Apart from the ABO locus our SNP-analysis was able to replicate three of the previously reported associations between SNPs and FVIII:c levels. Replicated associations include FVIII-associated SNPs in SCARA5, STAB2 and vWF genes [28,29]. Notably, these SNPs showed a more signiﬁcant association with FVIII activity in females. According to our ﬁndings the investigated SNPs collectively explained ~21% of the variation of FVIII:c mainly driven by the inﬂuence of the ABO locus. These observations suggest that there are additional genetic factors remaining to be identiﬁed and contributing to the hidden heritability of circulating FVIII:c. Limitations of our study are as follows: (a) the nomograms of FVIII:c have been determined for 5 year stored frozen plasma biosamples. Based on stability data investigated within the GHS (REFERENCE) we present estimated nomograms for FVIII:c in fresh plasma samples (Fig. S4). There might be potential effects of analyte instability that is reported to be inﬂuenced by the sampling procedure and storage conditions . However, a highly standardized setting was present in the current study. (b) There might be a misclassiﬁcation of diseases as these are assessed from medical reports and by self-reported data. (c) An association of FVIII:c with cardiovascular outcome was not reported. In conclusion, based on a large population-based data set, the present investigation provides a comprehensive characterization of the distribution and reference limits for FVIII:c in a central European sample. This study provides compelling evidence for a sex-speciﬁc, allele dosedependent strong association of variation at the ABO locus with circulating FVIII:c under consideration of a substantial number of determinants. The data support the use of age- and sex-speciﬁc reference limits for FVIII:c in clinical practice. The knowledge on the sex, age and blood group dependency of FVIII:c levels should be considered when interpreting FVIII:c levels.
Conﬂict of interest The authors report no relationships that could be construed as a conﬂict of interest.
M.I. Hermanns et al. / International Journal of Cardiology 187 (2015) 166–174
Acknowledgments First and foremost, we are extremely grateful to the survey participants who took the time to participate in the study. Without their participation and feedback, this study would not have been possible. We would like to extend a special thanks to the study and laboratory personnel for their help in the performance of the study, data collection, and for biomarker measurements. The team gratefully acknowledges the generous support of the research funds for providing us with the funding and opportunity to conduct this project. The Gutenberg Health Study is funded through the government of Rhineland-Palatinate (“Stiftung Rheinland-Pfalz für Innovation”, contract AZ 961-386261/ 733), the research programs “Wissen schafft Zukunft” and “Center for Translational Vascular Biology (CTVB)” of the Johannes GutenbergUniversity of Mainz, and its contract with Boehringer Ingelheim and PHILIPS Medical Systems, including an unrestricted grant for the Gutenberg Health Study. MIH, VG and PSW are funded by the Federal Ministry of Education and Research (BMBF 01EO1003). HTC is a Fellow of the Gutenberg Forschungskolleg of the Johannes Gutenberg-University Mainz, Germany. PSW has received research funding from Boehringer Ingelheim; PHILIPS Medical Systems; Sanoﬁ-Aventis; Bayer Vital; Daiichi Sankyo Europe; IMO Institute; Portavita; Federal Institute for Occupational Safety and Health (BAuA); Initiative “Health Economy”, Ministry of Health and Ministry of Economics, Rhineland-Palatinate; Federal Ministry of Education and Research; Federal Ministry of Health, Rhineland-Palatinate (MSAGD); Mainz Heart Foundation and received honoraria for lectures or consulting from Boehringer Ingelheim and Public Health, Heinrich-Heine-University Düsseldorf. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ijcard.2015.03.330. References  R.A. Kraaijenhagen, P.S. in't Anker, M.M. Koopman, P.H. Reitsma, M.H. Prins, A. van den Ende, et al., High plasma concentration of factor VIIIc is a major risk factor for venous thromboembolism, Thromb. Haemost. 83 (1) (2000) 5–9.  A.W. Tsai, M. Cushman, W.D. Rosamond, S.R. Heckbert, R.P. Tracy, N. Aleksic, et al., Coagulation factors, inﬂammation markers, and venous thromboembolism: the longitudinal investigation of thromboembolism etiology (LITE), Am. J. Med. 113 (8) (2002) 636–642.  M. Golder, J. Mewburn, D. Lillicrap, In vitro and in vivo evaluation of the effect of elevated factor VIII on the thrombogenic process, Thromb. Haemost. 109 (1) (2013) 53–60.  I. Bank, E.J. Libourel, S. Middeldorp, K. Hamulyak, E.C. van Pampus, M.M. Koopman, et al., Elevated levels of FVIII:C within families are associated with an increased risk for venous and arterial thrombosis, J. Thromb. Haemost. 3 (1) (2005) 79–84.  S. Ota, N. Yamada, Y. Ogihara, A. Tsuji, K. Ishikura, M. Nakamura, et al., High plasma level of factor VIII: an important risk factor for venous thromboembolism, Circ. J. 75 (6) (2011) 1472–1475.  A.R. Folsom, W.D. Rosamond, E. Shahar, L.S. Cooper, N. Aleksic, F.J. Nieto, et al., Prospective study of markers of hemostatic function with risk of ischemic stroke. The Atherosclerosis Risk in Communities (ARIC) Study Investigators, Circulation 100 (7) (1999) 736–742.  G.I. Rice, P.J. Grant, FVIII coagulant activity and antigen in subjects with ischaemic heart disease, Thromb. Haemost. 80 (5) (1998) 757–762.  A. Rumley, G.D.O. Lowe, P.M. Sweetnam, J.W.G. Yarnell, R.P. Ford, Factor VIII, von Willebrand factor and the risk of major ischaemic heart disease in the Caerphilly Heart Study, Br. J. Haematol. 105 (1) (1999) 110–116.  W.M. Lijfering, L.E. Flinterman, J.P. Vandenbroucke, F.R. Rosendaal, S.C. Cannegieter, Relationship between venous and arterial thrombosis: a review of the literature from a causal perspective, Semin. Thromb. Hemost. 37 (8) (2011) 885–896.  A. Samai, D. Monlezun, A. Shaban, A. George, L. Dowell, R. Kruse-Jarres, et al., Von Willebrand factor drives the association between elevated factor VIII and poor outcomes in patients with ischemic stroke, Stroke 45 (9) (2014) 2789–2791.  B.M. Gouse, A.K. Boehme, D.J. Monlezun, J.E. Siegler, A.J. George, K. Brag, et al., New thrombotic events in ischemic stroke patients with elevated factor VIII, Thrombosis 2014 (2014) 302861.  A.J. ten Cate-Hoek, A.W. Dielis, H.M. Spronk, R. van Oerle, K. Hamulyak, M.H. Prins, et al., Thrombin generation in patients after acute deep-vein thrombosis, Thromb. Haemost. 100 (2) (2008) 240–245.  P.J. Lenting, C.J. Vans, C.V. Denis, Clearance mechanisms of von Willebrand factor and factor VIII, J. Thromb. Haemost. 5 (7) (2007) 1353–1360.
 M. de Lange, H. Snieder, R.A. Ariens, T.D. Spector, P.J. Grant, The genetics of haemostasis: a twin study, Lancet 357 (9250) (2001) 101–105.  K.R. Viel, D.K. Machiah, D.M. Warren, M. Khachidze, A. Buil, K. Fernstrom, et al., A sequence variation scan of the coagulation factor VIII (FVIII) structural gene and associations with plasma FVIII activity levels, Blood 109 (9) (2007) 3713–3724.  J. O'Donnell, M.A. Laffan, The relationship between ABO histo-blood group, factor VIII and von Willebrand factor, Transfus. Med. 11 (4) (2001) 343–351.  P.V. Jenkins, J.S. O'Donnell, ABO blood group determines plasma von Willebrand factor levels: a biologic function after all? Transfusion 46 (10) (2006) 1836–1844.  P.W. Kamphuisen, J.C. Eikenboom, R.M. Bertina, Elevated factor VIII levels and the risk of thrombosis, Arterioscler. Thromb. Vasc. Biol. 21 (5) (2001) 731–738.  J. O'Donnell, A.D. Mumford, R.A. Manning, M. Laffan, Elevation of FVIII: C in venous thromboembolism is persistent and independent of the acute phase response, Thromb. Haemost. 83 (1) (2000) 10–13.  N. Bovenschen, K. Mertens, L. Hu, L.M. Havekes, B.J. van Vlijmen, LDL receptor cooperates with LDL receptor-related protein in regulating plasma levels of coagulation factor VIII in vivo, Blood 106 (3) (2005) 906–912.  E.L. Saenko, A.V. Yakhyaev, I. Mikhailenko, D.K. Strickland, A.G. Sarafanov, Role of the low density lipoprotein-related protein receptor in mediation of factor VIII catabolism, J. Biol. Chem. 274 (53) (1999) 37685–37692.  N. Martinelli, D. Girelli, B. Lunghi, M. Pinotti, G. Marchetti, G. Malerba, et al., Polymorphisms at LDLR locus may be associated with coronary artery disease through modulation of coagulation factor VIII activity and independently from lipid proﬁle, Blood 116 (25) (2010) 5688–5697.  R. Vormittag, P. Bencur, C. Ay, T. Tengler, T. Vukovich, P. Quehenberger, et al., Lowdensity lipoprotein receptor-related protein 1 polymorphism 663 C N T affects clotting factor VIII activity and increases the risk of venous thromboembolism, J. Thromb. Haemost. 5 (3) (2007) 497–502.  P.S. Wild, T. Zeller, M. Beutel, M. Blettner, K.A. Dugi, K.J. Lackner, et al., The Gutenberg Health Study, Bundesgesundheitsbl. Gesundheitsforsch. Gesundheitsschutz 55 (6-7) (2012) 824–829.  R.B. Schnabel, S. Wilde, P.S. Wild, T. Munzel, S. Blankenberg, Atrial ﬁbrillation: its prevalence and risk factor proﬁle in the German general population, Dtsch. Arztebl. Int. 109 (16) (2012) 293–299.  P.S. Wild, C.R. Sinning, A. Roth, S. Wilde, R.B. Schnabel, E. Lubos, et al., Distribution and categorization of left ventricular measurements in the general population: results from the population-based Gutenberg Heart Study, Circ. Cardiovasc. Imaging 3 (5) (2010) 604–613.  R.B. D'Agostino Sr., S. Grundy, L.M. Sullivan, P. Wilson, Group CHDRP, Validation of the Framingham coronary heart disease prediction scores: results of a multiple ethnic groups investigation, JAMA 286 (2) (2001) 180–187.  G. Antoni, T. Oudot-Mellakh, A. Dimitromanolakis, M. Germain, W. Cohen, P. Wells, et al., Combined analysis of three genome-wide association studies on vWF and FVIII plasma levels, BMC Med. Genet. 12 (2011) 102.  N.L. Smith, M.H. Chen, A. Dehghan, D.P. Strachan, S. Basu, N. Soranzo, et al., Novel associations of multiple genetic loci with plasma levels of factor VII, factor VIII, and von Willebrand factor: the CHARGE (Cohorts for Heart and Aging Research in Genome Epidemiology) Consortium, Circulation 121 (12) (2010) 1382–1392.  D. Teupser, R. Baber, U. Ceglarek, M. Scholz, T. Illig, C. Gieger, et al., Genetic regulation of serum phytosterol levels and risk of coronary artery disease, Circ. Cardiovasc. Genet. 3 (4) (2010) 331–339.  M.P. Reilly, M. Li, J. He, J.F. Ferguson, I.M. Stylianou, N.N. Mehta, et al., Identiﬁcation of ADAMTS7 as a novel locus for coronary atherosclerosis and association of ABO with myocardial infarction in the presence of coronary atherosclerosis: two genome-wide association studies, Lancet 377 (9763) (2011) 383–392.  G. Pare, D.I. Chasman, M. Kellogg, R.Y. Zee, N. Rifai, S. Badola, et al., Novel association of ABO histo-blood group antigen with soluble ICAM-1: results of a genome-wide association study of 6,578 women, PLoS Genet. 4 (7) (2008) e1000118.  M. Campos, A. Buchanan, F. Yu, M. Barbalic, Y. Xiao, L.E. Chambless, et al., Inﬂuence of single nucleotide polymorphisms in factor VIII and von Willebrand factor genes on plasma factor VIII activity: the ARIC Study, Blood 119 (8) (2012) 1929–1934.  A.R. Folsom, K.K. Wu, W.D. Rosamond, A.R. Sharrett, L.E. Chambless, Prospective study of hemostatic factors and incidence of coronary heart disease: the Atherosclerosis Risk in Communities (ARIC) Study, Circulation 96 (4) (1997) 1102–1108.  P.W. Kamphuisen, J.J. Houwing-Duistermaat, H.C. van Houwelingen, J.C. Eikenboom, R.M. Bertina, F.R. Rosendaal, Familial clustering of factor VIII and von Willebrand factor levels, Thromb. Haemost. 79 (2) (1998) 323–327.  M.G. Conlan, A.R. Folsom, A. Finch, C.E. Davis, P. Sorlie, G. Marcucci, et al., Associations of factor VIII and von Willebrand factor with age, race, sex, and risk factors for atherosclerosis. The Atherosclerosis Risk in Communities (ARIC) Study, Thromb. Haemost. 70 (3) (1993) 380–385.  E.J. Favaloro, S. Soltani, J. McDonald, E. Grezchnik, L. Easton, J.W. Favaloro, Reassessment of ABO blood group, sex, and age on laboratory parameters used to diagnose von Willebrand disorder: potential inﬂuence on the diagnosis vs the potential association with risk of thrombosis, Am. J. Clin. Pathol. 124 (6) (2005) 910–917.  P.W. Kamphuisen, J.C. Eikenboom, H.L. Vos, R. Pablo, A. Sturk, R.M. Bertina, et al., Increased levels of factor VIII and ﬁbrinogen in patients with venous thrombosis are not caused by acute phase reactions, Thromb. Haemost. 81 (5) (1999) 680–683.  B.S. Armour, T. Woollery, A. Malarcher, T.F. Pechacek, C. Husten, Annual smokingattributable mortality, years of potential life lost, and productivity losses—United States, 1997–2001, MMWR Morb. Mortal. Wkly Rep. 54 (25) (2005) 625–628.  A.R. Folsom, J.A. Delaney, P.L. Lutsey, N.A. Zakai, N.S. Jenny, J.F. Polak, et al., Associations of factor VIIIc, D-dimer, and plasmin–antiplasmin with incident cardiovascular disease and all-cause mortality, Am. J. Hematol. 84 (6) (2009) 349–353.
M.I. Hermanns et al. / International Journal of Cardiology 187 (2015) 166–174
 A. Sambola, J. Osende, J. Hathcock, M. Degen, Y. Nemerson, V. Fuster, et al., Role of risk factors in the modulation of tissue factor activity and blood thrombogenicity, Circulation 107 (7) (2003) 973–977.  M.D. Feher, M.W. Rampling, J. Brown, R. Robinson, W. Richmond, S. Cholerton, et al., Acute changes in atherogenic and thrombogenic factors with cessation of smoking, J. R. Soc. Med. 83 (3) (1990) 146–148.  Y. Fusegawa, S. Handa, Platelet aggregation induced by ADP or epinephrine is enhanced in habitual smokers, Thromb. Res. 97 (5) (2000) 287–295.  Y. Fusegawa, S. Goto, S. Handa, T. Kawada, Y. Ando, Platelet spontaneous aggregation in platelet-rich plasma is increased in habitual smokers, Thromb. Res. 93 (6) (1999) 271–278.  J.P. Collet, D. Park, C. Lesty, J. Soria, C. Soria, G. Montalescot, et al., Inﬂuence of ﬁbrin network conformation and ﬁbrin ﬁber diameter on ﬁbrinolysis speed: dynamic and structural approaches by confocal microscopy, Arterioscler. Thromb. Vasc. Biol. 20 (5) (2000) 1354–1361.  R.S. Barua, F. Sy, S. Srikanth, G. Huang, U. Javed, C. Buhari, et al., Effects of cigarette smoke exposure on clot dynamics and ﬁbrin structure: an ex vivo investigation, Arterioscler. Thromb. Vasc. Biol. 30 (1) (2010) 75–79.  P. Caponnetto, C. Russo, A. Di Maria, J.B. Morjaria, S. Barton, F. Guarino, et al., Circulating endothelial-coagulative activation markers after smoking cessation: a 12month observational study, Eur. J. Clin. Investig. 41 (6) (2011) 616–626.  G. Delgado, R. Siekmeier, T.B. Grammer, B.O. Boehm, W. Marz, M.E. Kleber, Alterations in the coagulation system of active smokers from the Ludwigshafen Risk and Cardiovascular Health (LURIC) Study, Adv. Exp. Med. Biol. 832 (2015) 9–14.  B. Tanis, A. Algra, Y. van der Graaf, F. Helmerhorst, F. Rosendaal, Procoagulant factors and the risk of myocardial infarction in young women, Eur. J. Haematol. 77 (1) (2006) 67–73.  V. Hernandez, N. Munoz, M.A. Montero, A. Camacho, F. Lozano, V. Fernandez, Acute myocardial infarction for thrombotic occlusion in patient with elevated coagulation factor VIII, Rev. Esp. Cardiol. 65 (7) (2012) 673–674.  T.W. Meade, J.A. Cooper, Y. Stirling, D.J. Howarth, V. Ruddock, G.J. Miller, Factor VIII, ABO blood group and the incidence of ischaemic heart disease, Br. J. Haematol. 88 (3) (1994) 601–607.  R.P. Tracy, A.M. Arnold, W. Ettinger, L. Fried, E. Meilahn, P. Savage, The relationship of ﬁbrinogen and factors VII and VIII to incident cardiovascular disease and death in the elderly: results from the Cardiovascular Health Study, Arterioscler. Thromb. Vasc. Biol. 19 (7) (1999) 1776–1783.  N.A. Zakai, R. Katz, N.S. Jenny, B.M. Psaty, A.P. Reiner, S.M. Schwartz, et al., Inﬂammation and hemostasis biomarkers and cardiovascular risk in the elderly: the Cardiovascular Health Study, J. Thromb. Haemost. 5 (6) (2007) 1128–1135.  T.R. Chang, K.C. Albright, A.K. Boehme, A. Dorsey, E.A. Sartor, R. Kruse-Jarres, et al., Factor VIII in the setting of acute ischemic stroke among patients with suspected hypercoagulable state, Clin. Appl. Thromb. Hemost. 20 (2) (2014) 124–128.  L.M. Reich, G. Heiss, L.L. Boland, A.T. Hirsch, K. Wu, A.R. Folsom, Ankle-brachial index and hemostatic markers in the Atherosclerosis Risk in Communities (ARIC) study cohort, Vasc. Med. 12 (4) (2007) 267–273.
 M. Sartori, E. Favaretto, C. Legnani, M. Cini, E. Conti, A. Amato, et al., Thrombophilic risk factors and peripheral arterial disease severity, Thromb. Haemost. 104 (1) (2010) 71–77.  A.J. Vlot, E.P. Mauser-Bunschoten, A.G. Zarkova, E. Haan, C.L. Kruitwagen, J.J. Sixma, et al., The half-life of infused factor VIII is shorter in hemophiliac patients with blood group O than in those with blood group A, Thromb. Haemost. 83 (1) (2000) 65–69.  N.A. Zakai, S.E. Judd, K. Alexander, L.A. McClure, B.M. Kissela, G. Howard, et al., ABO blood type and stroke risk: the reasons for geographic and racial differences in stroke study, J. Thromb. Haemost. 12 (4) (2014) 564–570.  O. Wu, N. Bayoumi, M.A. Vickers, P. Clark, ABO(H) blood groups and vascular disease: a systematic review and meta-analysis, J. Thromb. Haemost. 6 (1) (2008) 62–69.  K.L. Wiggins, N.L. Smith, N.L. Glazer, F.R. Rosendaal, S.R. Heckbert, B.M. Psaty, et al., ABO genotype and risk of thrombotic events and hemorrhagic stroke, J. Thromb. Haemost. 7 (2) (2009) 263–269.  B.M. Wolpin, C. Kabrhel, R. Varraso, P. Kraft, E.B. Rimm, S.Z. Goldhaber, et al., Prospective study of ABO blood type and the risk of pulmonary embolism in two large cohort studies, Thromb. Haemost. 104 (5) (2010) 962–971.  S. Zhou, I. Welsby, Is ABO blood group truly a risk factor for thrombosis and adverse outcomes? World J. Cardiol. 6 (9) (2014) 985–992.  M. Barbalic, J. Dupuis, A. Dehghan, J.C. Bis, R.C. Hoogeveen, R.B. Schnabel, et al., Large-scale genomic studies reveal central role of ABO in sP-selectin and sICAM-1 levels, Hum. Mol. Genet. 19 (9) (2010) 1863–1872.  J.R.M. Ojala, T. Pikkarainen, G. Elmberger, K. Tryggvason, Progressive reactive lymphoid connective tissue disease and development of autoantibodies in scavenger receptor A5-deﬁcient mice, Am. J. Pathol. 182 (5) (2013) 1681–1695.  E.N. Harris, P.H. Weigel, The ligand-binding proﬁle of HARE: hyaluronan and chondroitin sulfates A, C, and D bind to overlapping sites distinct from the sites for heparin, acetylated low-density lipoprotein, dermatan sulfate, and CS-E, Glycobiology 18 (8) (2008) 638–648.  G. Marchetti, B. Lunghi, C. Legnani, M. Cini, M. Pinotti, F. Mascoli, et al., Contribution of low density lipoprotein receptor-related protein genotypes to coagulation factor VIII levels in thrombotic women, Haematologica 91 (9) (2006) 1261–1263.  T.B. Mello, L.H. Siqueira, S.A. Montavao, M.C. Ozello, J.M. Annichino-Bizzacchi, Low density lipoprotein receptor-related protein polymorphisms are not risk factors for venous thromboembolism, Thromb. Res. 121 (5) (2008) 625–629.  N. Cunningham, M.A. Laffan, R.A. Manning, J.S. O'Donnell, Low-density lipoprotein receptor-related protein polymorphisms in patients with elevated factor VIII coagulant activity and venous thrombosis, Blood Coagul. Fibrinolysis 16 (7) (2005) 465–468.  F. Betsou, B. Roussel, N. Guillaume, J.J. Lefrere, Long-term stability of coagulation variables: protein S as a biomarker for preanalytical storage-related variations in human plasma, Thromb. Haemost. 101 (6) (2009) 1172–1175.