If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The predictive value of VWF for MACE is unknown in patients with asymptomatic carotid stenosis.
•
This large cohort study investigated the predictive value of VWF for MACE in 811 patients with carotid stenosis.
•
VWF levels within the highest quartile were independently associated with a 2.1-fold increase in MACE.
•
This cohort study in 811 patients with carotid stenosis showed that high VWF-levels predicted MACE, which may be useful for risk stratification.
Abstract
Background and aims
Von Willebrand factor (VWF) plays an important role in thrombogenesis and mediates platelet adhesion particularly under high shear stress. Such conditions are generally found in stenotic arteries and can eventually cause myocardial infarction or stroke. We aimed to study whether levels of VWF antigen (VWF:Ag) predict future major adverse cardiovascular events (MACE) in patients suffering from carotid artery stenosis.
Methods
Patients with atherosclerotic carotid artery disease defined by the presence of nonstenotic plaques or any degree of carotid stenosis were prospectively enrolled. Concentrations of VWF were measured by enzyme immunoassay.
Results
VWF:Ag levels were more stable after 4 freeze-thaw cycles, when compared to VWF activity, and we showed similar concentrations of VWF in citrated plasma and serum (±4%). Levels of VWF:Ag predicted future cardiovascular events in 811 patients with carotid stenosis independent of known cardiovascular risk factors. Patients with VWF:Ag concentrations in the 4th quartile had a 44% event rate after an average 3-year follow up and a hazard ratio of 2.15 (95% confidence interval 1.46–3.16; p < 0.001).
Conclusions
High concentrations of VWF:Ag predict major cardiovascular events in patients with carotid stenosis, and given their high event rate may be useful for risk stratification of such patients.
]. On the other hand, the central role of von Willebrand factor in thrombosis is particularly apparent in blood flow conditions with high shear stress, such as found in capillaries under physiologic conditions or in stenosed arteries [
]. VWF mediates the first step in platelet thrombus formation, i.e. adhesion of platelets in the bloodstream to the luminal surface of blood vessels at sites of vascular injury under conditions of moderate to high shear force [
The PLAT Study: hemostatic function in relation to atherothrombotic ischemic events in vascular disease patients. Principal results. PLAT Study Group. Progetto Lombardo Atero-Trombosi (PLAT) Study Group.
Hemostatic factors and the risk of myocardial infarction or sudden death in patients with angina pectoris. European Concerted Action on Thrombosis and Disabilities Angina Pectoris Study Group.
]. Concordantly, platelet aggregation under high shear rates is enhanced by increased plasma VWF levels in patients with acute myocardial infarction when compared with controls [
]. Finally, genetic studies have demonstrated that VWF is not only a biomarker for atherosclerosis, but is actually involved in the pathogenesis of stroke [
It is unknown whether VWF can predict cardiovascular outcomes in patients with subclinical carotid atherosclerosis. Given the large body of evidence suggesting a role of VWF in arterial thrombosis, we hypothesized that VWF antigen concentrations are a predictor of cardiovascular outcome in patients with carotid atherosclerosis. Therefore we also hypothesized that VWF levels predict the occurrence of major adverse cardiovascular events in patients suffering from carotid artery stenosis.
2. Materials and methods
2.1 Study designs and patients
Inclusion and exclusion criteria of the ICARAS design have been published elsewhere [
]. In brief, subjects with atherosclerosis in their carotid arteries, who, at the time of screening were clinically asymptomatic, were enrolled between March 2002 and March 2003. Prevalent carotid atherosclerosis was defined as the presence of either non-stenotic plaques or any degree of carotid stenosis. Patients underwent duplex ultrasound investigations of the extracranial carotid arteries. Patients were excluded if any of the following factors were present: active infectious or inflammatory disease, symptomatic carotid artery stenosis necessitating revascularisation therapy, recent operations or endovascular interventions (recent is classified as within the last 14 days), bilateral carotid occlusions, with bilateral stent implantation or bilateral carotid endarterectomy, myocardial infarction (MI), coronary revascularisation, stroke, or peripheral vascular surgery during the preceding 6 months. The rationale behind this was the assumption that acute cardiovascular events may affect laboratory measures and thus it may reflect rather the acute situation than the chronic nature of atherosclerosis [
]. The study complied with the Declaration of Helsinki and was approved by the institutional review board of the Medical University of Vienna. All patients gave their written informed consent. Definitions have been published previously, for all the clinical data (diabetes mellitus, arterial hypertension, hyperlipidemia etc.) [
811 samples of 1268 prospectively collected samples, stored at −80 °C, were available from Caucasian patients with carotid atherosclerosis. 203 patients (16%) were lost to clinical follow-up and of 254 patients no serum samples were available. The 457 patients who had to be excluded did not significantly differ from the patients who were included in terms of baseline and demographic parameters (age, sex, frequency of atherothrombotic risk factors, cardiovascular comorbidities, patients’ medical history, family history, results of health assessments and physical examinations, and degree of carotid stenosis; data not shown). Patients were evaluated by duplex sonography and followed up clinically for a median of 3.1 years in order to quantify the occurrence of major adverse cardiovascular events (MACE), a composite of stroke (n = 43), myocardial infarction (n = 35), percutaneous coronary intervention (n = 59), coronary artery bypass graft (n = 35) and death (n = 96). MACE was defined as the primary study end point.
2.3 Loss of VWF antigen vs. activity after multiple freeze/thaw cycles
The analyst was blinded to all clinical and ultrasound data. The ELISA was performed following the manufacturer's protocol. We compared two different commercially available ELISA-kits, one measuring VWF antigen levels and one measuring VWF activity (both REAADS from Corgenix, Haemochrom Diagnostica GmbH, Essen, FRG) [
A dose ranging phase I/II trial of the von Willebrand factor inhibiting aptamer ARC1779 in patients with congenital thrombotic thrombocytopenic purpura.
], using plasma and serum samples from three healthy volunteers. We froze samples and subjected them to five freeze-thaw cycles. After freezing the samples in the freezer at −20°, samples were thawed at 37 °C in a water bath incubator for 20 min before analyzing VWF antigen and VWF activity with ELISA assay. The same samples were used in both ELISAs. The median coefficient of variation for inter- and intra-assay variability was less than 5%.
2.4 Plasma vs. serum VWF levels
As only serum samples were available, we compared VWF antigen levels between citrated plasma and serum in order to confirm that serum is suitable as demonstrated previously [
]. We used samples of patients with atherosclerosis (coronary, carotid, peripheral arteries, MI, stroke, critical limb ischaemia) and an ESR (erythrocyte sedimentation rate) < 20 mm/h (to exclude the cofounder of gross inflammation) and of healthy controls.
Additionally, plasma VWF antigen levels were measured in healthy volunteers for the purpose of another study at our department. Baseline values were 97% (95% CI 69-125) which was in agreement with our results and we used these values as normal VWF antigen range.
2.5 Statistical methods
Statistics were calculated as previously described [
]. To obtain clinically useful measures, levels of VWF were categorized in quartiles. Continuous data are presented as median and interquartile range (range from the 25th to the 75th percentile). Discrete data are given as counts and percentages. Analysis of variance and the χ2 test were used for comparisons between quartiles and Spearman correlation coefficients (Rs) for comparison of discrete data, as appropriate. The log-rank test was used for comparison between groups. Event-free survival probabilities were estimated using the Kaplan–Meier method. Univariable and multivariable Cox proportional hazards models were applied to assess the association between levels of VWF levels and the occurrence of MACE. The multivariable model included the following variables: age (years), sex (male/female), history of myocardial infarction (binary), history of stroke (binary), peripheral arterial disease (binary), body mass index (kg/m2), hypertension (binary), diabetes mellitus (binary), serum creatinine (mg/dL), glycohemoglobin A1 (%), levels of triglycerides (mg/dL), total cholesterol levels (mg/dL), low density lipoprotein cholesterol levels (mg/dL), and statin treatment (binary) (Supplementary Table 5). The selection of the variables was defined a priori and is based on current guidelines for cardiovascular risk prediction. All the variables listed above were included in every multivariable Cox proportional hazard model used for this study. Results of the Cox models are presented as hazard ratios (HR; 95% confidence interval [CI]). We assessed the overall model fit using Cox–Snell residuals.
We also tested the proportional hazard assumption for all covariates using Schoenfeld residuals (overall test) and the scaled Schoenfeld residuals (variable-by-variable testing). Interactions between VWF and the degree of carotid artery stenosis were tested by entering interaction terms in the Cox proportional hazard regression models. A 2-sided p value of <0.05 was considered significant. All calculations were performed with SPSS (version 20.0 SPSS Inc) and the STATA11 software package (StataCorp) for Windows.
3. Results
3.1 Pre-study experiments
Measuring VWF activity and VWF antigen level after multiple freeze-thaw cycles, we found an average decrease in VWF activity of 19% compared to less than 1% decrease of VWF antigen levels (Supplementary Table 1). As the patient samples may have been exposed to similar cycles, the VWF antigen ELISA was selected as the appropriate measurement.
Before performing the main analyses we investigated possible differences in concentrations of VWF antigen between citrated plasma and serum. While atherosclerotic patients had ~30% higher VWF levels than controls irrespective of the matrix (Supplementary Table 2), there were no differences in the levels of VWF antigen between serum and citrate plasma (p = 0.752), and there was an excellent correlation between VWF antigen levels measured in these two matrices in the patient group (r = 0.98) and also in the control group (r = 0.95).
3.2 VWF antigen levels and MACE
Baseline demographic data and clinical characteristics of 811 patients with carotid stenosis are given in Table 1. Von Willebrand Factor antigen concentrations according to the degree of carotid artery stenosis are given in the Supplementary Table 3. The number of patients who suffered a major adverse cardiovascular event (MACE) during a median follow up of 3.1 years (IQR 2.5–3.5) was 235, which comprises 22.1% of the entire study population. A total of 43 cases of stroke, 32 cases of myocardial infarction, 59 cases of percutaneous coronary intervention, 35 cases of the coronary artery bypass graft and 96 cases of death were recorded.
Table 1Demographic characteristics of patients.
Variable
Age, years
69.0 (61.2–76.2)
Female gender
322 (38.5%)
Hypertension
570 (68.1%)
Family history of cardiovascular disease
457 (54.6%)
Previous peripheral arterial disease
366 (43.7%)
Current smokers
217 (25.9%)
Previous myocardial infarction
204 (24.4%)
Diabetes
194 (23.2%)
Previous stroke
130 (15.5%)
Body Mass Index (kg/m2)
26.1 (24.0–28.7)
HbA1c% (glycated haemoglobin)
6.0 (5.6–6.6)
Triglycerides (mg/dl)
147 (107–216)
Total cholesterol (mg/dl)
205 (175–236)
Low Density Lipoprotein- cholesterol (mg/dl)
118 (94–146)
Highly sensitive C- reactive protein (mg/dl)
0.29 (0.14–0.64)
Serum creatinine (mg/dl)
1.06 (0.93–1.23)
VWF antigen level (%)
168 (122–227)
*Data presented as counts (%) or as median values and 25th and 75th percentiles for continuous values. VWF, von Willebrand Factor.
The Cox regression model (Table 2) showed that independent predictors of outcome were high levels of VWF, previous myocardial infarction, family history of cardiovascular disease and high levels of highly sensitive-C reactive protein (hs-CRP). Patients within the highest quartile of VWF antigen levels had a 2.15-fold 95% confidence interval 1.46–3.16; p < 0.001 higher risk for MACE as compared to those in the first quartile (Fig. 1). A sensitivity analysis was performed including discrete VWF antigen levels as a continuous variable rather than quartiles, which showed the same effect (Supplementary Table 4). Although there was obviously a substantial overlap, patients with a higher degree of carotid stenosis had on average ~10% higher VWF levels than those with less carotid narrowing (Table 3; plaques only, or <30% stenosis). This resulted in a relatively weak correlation between VWF levels and the degree of stenosis (Spearman ranks correlation test: Rs = 0.08; p = 0.024).
Table 2Independent predictors of major adverse coronary events in patients with carotid stenosis.
Whether VWF is a predictor for major adverse cardiovascular events (MACE) in patients with carotid stenosis has not previously been investigated. Our study shows that VWF levels within the highest quartile are associated with a 2.1-fold increase in events. The incidence of MACE was 22% after a median follow up of three years in our study. This rather high incidence rate can be explained by the inclusion and exclusion criteria. As patients with previous cardio- or cerebrovascular disease were not excluded, this could be the reason for this high incidence rate. Demographic characteristics of our patients showed a high rate of hypertension, previous myocardial infarction, peripheral arterial disease and diabetes, which are all risk factors for MACE and a likely explanation for coronary events happening not in close proximity of the carotid stenosis.
A recent study also looked at the incidence of MACE in subjects with asymptomatic mild and moderate carotid artery stenosis [
]. The incidence of MACE in that study was substantially lower when compared with our study. This was likely due to their exclusion of patients with previous cardiovascular disease.
In the general population VWF levels are a weak predictor of cardiovascular disease [
High plasminogen activator inhibitor and tissue plasminogen activator levels in plasma precede a first acute myocardial infarction in both men and women: evidence for the fibrinolytic system as an independent primary risk factor.
]. Nevertheless, results from studies in high-risk populations often show substantially stronger associations. The patients with the highest risk are likely those with acute coronary syndromes. Patients with VWF levels in the 4th quartile had a 6-fold higher rate of recurrent MACE, and fatal events (approximately ~10%) were only seen in the 3rd and 4th quartile of VWF-levels [
] and VWF antigen concentrations were an independent predictor of poor outcome after cardiopulmonary resuscitation and were associated with increased adverse outcome rates in patients with cardiac arrest risk scores of 2–3 [
Plasma concentrations of von Willebrand factor and intracellular adhesion molecule-1 for prediction of outcome after successful cardiopulmonary resuscitation.
]. Similar to patients with cardiovascular disease, the association between VWF levels and stroke appears to be stronger in acute settings. The relative risk of ischemic stroke was highest in individuals in the upper quartile of VWF:Ag (odds ratio, 3.2; 95% CI, 1.4 to 7.5) compared with individuals in the lowest quartile [
]. Patients with a high VWF antigen level had a higher risk of unfavorable outcomes when compared with patients with a low VWF antigen level (hazard ratio: 2.15; 95% CI, 1.46–3.16) [
]. The hazard ratio for stroke was 1.79 (95% CI, 1.19–2.45) in those individuals with VWF levels in the 4th quartile of the population-based Rotterdam study which included approximately 6000 elderly subjects (aged ≥ 55years) [
] which has been shown to correlate with VWF levels. Sonneveld et al. examined the association between the extent of atherosclerosis, determined by the calcification volume in both the aortic arch and the carotid arteries and VWF antigen levels in patients suffering from TIA or ischemic stroke [
]. Patients with calcifications in the aortic arch and/or carotid arteries had significantly higher VWF antigen levels compared with patients without calcifications in the aortic arch and/or carotid arteries [
]. Levels of VWF antigen linearly increased with increasing aortic calcification volume. We found a similar, but weaker correlation between levels of VWF antigen and the severity of carotid stenosis. The stronger correlation in Sonneveld's work may be attributed to their use of more sophisticated imaging procedures.
The role of VWF in atherosclerosis is controversial and was experimentally addressed in different animal models. First, atherogenesis was investigated in three different groups of pigs after diet-induced atherosclerosis. Atherosclerosis developed in both the control group and heterozygous VWF deficient pigs whereas the severe homozygous VWF deficient animals were resistant to atherosclerosis [
von Willebrand factor and occlusive arterial thrombosis. A study in normal and von Willebrand's disease pigs with diet-induced hypercholesterolemia and atherosclerosis.
]. Atherogenesis was examined after an atherogenic diet in two groups of mice, one lacking low-density lipoprotein receptor and VWF and the other one lacking low-density lipoprotein receptor but with normal VWF. Fatty streaks in the aortic sinus of mice lacking VWF were 40% smaller and contained fewer monocytes than in controls; further, the distribution of lesion in the aortas was different in different regions [
]. In addition, endothelial-cell derived VWF and platelet-derived VWF were investigated separately in genetically modified mice. Endothelial-cell derived VWF, but not platelet-derived VWF contributed to VWF-dependent atherosclerosis [
In contrast to this evidence that severe VWF deficiency protects against atherogenesis in animals, it seems harder to prove that in human studies. Evidently, data in humans are scarce and methodological weaknesses exist. Some studies indicate that VWF protects against atherosclerosis [
Common carotid intima-media thickness and von Willebrand factor serum levels in rheumatoid arthritis female patients without cardiovascular risk factors.
Association of von Willebrand factor deficiency with prevalent cardiovascular disease and asymptomatic carotid atherosclerosis: the Atherosclerosis Risk in Communities Study.
Low paraoxonase 1 arylesterase activity and high von Willebrand factor levels are associated with severe coronary atherosclerosis in patients with non-diabetic stable coronary artery disease.
Do hemophilia A and von Willebrand disease protect against carotid atherosclerosis? A comparative study between coagulopathics and normal subjects by means of carotid echo-color Doppler scan.
]. However, humans with severe VWD are commonly treated with VWF replacement therapy which could possibly influence the findings by neutralizing any protective effective of VWD [
]. In summary, VWF appears to have a role in atherosclerosis in different species although its contribution to arterial thrombosis may be better documented. In that regard recently developed VWF inhibitors may be interesting drug candidates [
] increase VWF secretion, so that one may hypothesise that these conditions and atherosclerosis could be possible reasons for increased VWF levels in our population. Increased production and secretion of VWF from endothelium in case of inflammation has been previously shown [
Endothelial cell-derived von Willebrand factor is the major determinant that mediates von Willebrand factor-dependent acute ischemic stroke by promoting postischemic thrombo-inflammation.
]. Nonetheless, the effects of inflammation on secretion and production by megakaryocytes has not been investigated, except for interleukin-11 (IL-11).
The explanation for that could be the difference in the amount of secreted VWF from endothelium and platelets. The VWF secretion is predominantly endothelium-dependent, whereas the contribution of platelets is limited for several reasons [
] which likely reflects enhanced VWF transcription in megakaryocytes too. Additionally, recombinant IL-11 increased VWF antigen and activity in patients with mild von Willebrand disease [
]. which could mean that VWF is secreted from platelets and megakaryocytes. So in principle at least some cytokines could stimulate VWF release from platelets and megakaryocytes and influence inflammation generating atherosclerosis in our population.
Before analyzing VWF antigen concentrations, we investigated the influence of up to four freeze-thaw cycles on laboratory results. The samples from the ICARAS study have been frozen and thawed at least once. Two previous studies have indicated that freezing and subsequent thawing does not influence the stability of VWF antigen in plasma samples [
], but we observed a 19% decrease in VWF activity after 4 freeze-thaw cycles. This difference in results could be assay dependent. Although 15 years elapsed between patient enrollment and measurements of VWF Ag, the measured levels are in accordance with the literature and expected for this patient group [
The samples available for this study were serum samples, and for this reason, we explored possible differences between VWF antigen in serum or plasma samples. One previous study showed no major differences in paired samples comparing plasma vs serum of 81 patients [
]. Our analyses showed that VWF levels in citrated plasma were comparable to those in serum. Therefore, serum appears to be an appropriate matrix, and serum was also used in a recent elegant study on the prognostic utility of VWF in amyloidosis [
As described, patients included in this study suffered from carotid stenosis. Annual stroke rates in patients with carotid artery occlusion range from 0 to 5% [
]. In our study comprising 811 patients, there were not enough stroke cases to examine the relationship between VWF and stroke events. In addition, substantial proportions of our patients had previously suffered from cardiovascular events. In our study, only those who had suffered cardiovascular events in the last six months were excluded. Therefore, we cannot extrapolate our data to patients with “isolated” carotid stenosis. Additionally, VWF data were not adjusted for ABO blood groups, as we lacked information about the blood groups of the participants. Nevertheless, the differences in the MACE events between the blood groups are probably driven by the VWF levels. Moreover, high-molecular-weight VWF multimers were not investigated or measured.
4.4 Conclusion
High concentrations of VWF:Ag predict major cardiovascular events in patients with carotid stenosis, and given their high event rate may be useful for risk stratification of such patients.
Financial support
This study was supported by a grant from the Austrian Science Funds (FWF; grant# SFB54–P04).
Author contributions
K. D. Kovacevic performed the analyses and wrote the draft paper. F. J. Mayer did the statistical analyses and wrote the statistical part of the paper. M.Hoke, E. Minar, M. Schillinger, C. Binder were responsible for the conduct the main ICARAS study. N. Buchtele contributed to the idea, analysis of the literature and writing of the background, and G. Obermayer was responsible for sample handling, storage and analysis. A. Blann was responsible for the study and performed the analysis in the atherosclerotic patients validating the matrix. B. Jilma conceived the study, organized funding and contributed to the writing of the manuscript. All authors critically reviewed the manuscript and contributed to its content.
Conflicts of interest
The authors declared they do not have anything to disclose regarding conflict of interest with respect to this manuscript.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
The PLAT Study: hemostatic function in relation to atherothrombotic ischemic events in vascular disease patients. Principal results. PLAT Study Group. Progetto Lombardo Atero-Trombosi (PLAT) Study Group.
Hemostatic factors and the risk of myocardial infarction or sudden death in patients with angina pectoris. European Concerted Action on Thrombosis and Disabilities Angina Pectoris Study Group.
A dose ranging phase I/II trial of the von Willebrand factor inhibiting aptamer ARC1779 in patients with congenital thrombotic thrombocytopenic purpura.
High plasminogen activator inhibitor and tissue plasminogen activator levels in plasma precede a first acute myocardial infarction in both men and women: evidence for the fibrinolytic system as an independent primary risk factor.
Plasma concentrations of von Willebrand factor and intracellular adhesion molecule-1 for prediction of outcome after successful cardiopulmonary resuscitation.
von Willebrand factor and occlusive arterial thrombosis. A study in normal and von Willebrand's disease pigs with diet-induced hypercholesterolemia and atherosclerosis.
Common carotid intima-media thickness and von Willebrand factor serum levels in rheumatoid arthritis female patients without cardiovascular risk factors.
Association of von Willebrand factor deficiency with prevalent cardiovascular disease and asymptomatic carotid atherosclerosis: the Atherosclerosis Risk in Communities Study.
Low paraoxonase 1 arylesterase activity and high von Willebrand factor levels are associated with severe coronary atherosclerosis in patients with non-diabetic stable coronary artery disease.
Do hemophilia A and von Willebrand disease protect against carotid atherosclerosis? A comparative study between coagulopathics and normal subjects by means of carotid echo-color Doppler scan.
Endothelial cell-derived von Willebrand factor is the major determinant that mediates von Willebrand factor-dependent acute ischemic stroke by promoting postischemic thrombo-inflammation.
31477-7&id=fx1.jpg){kind=link}
31477-7&id=gr1.jpg){kind=link}