Independent association of von Willebrand factor with surrogate markers of atherosclerosis in middle-aged asymptomatic subjects


José A. Páramo, Atherosclerosis Research, University Clinic, Avda Pio XII 36, 31008 Pamplona, Spain. Tel.: 34-948296397; fax: 34-948296500; e-mail:


Summary.  Von Willebrand factor (VWF) is considered a reliable marker of endothelial damage. Plasma levels of VWF were measured in 857 middle-aged subjects (80.4% men) free of clinically overt atherosclerotic disease, in relation to cardiovascular risk factors, carotid intima-media thickness (IMT) and microalbuminuria, two established surrogate markers of atherosclerosis. There was linear trend for the increase of VWF, carotid IMT (p < 0.001) and microalbuminuria (p = 0.018). The association between VEF and both markers remained statistically significant after adjusting for cardiovascular risk factors and inflammatory markers (p < 0.01). In conclusion, VWF was independently associated with both structural and functional surrogates of atherosclerosis in asymptomatic subjects, thus representing a systemic biomarker of subclinical atherosclerosis.


Atherosclerosis is a systemic chronic inflammatory disease affecting the intima of large- and medium-size arteries of various vascular beds. Endothelial dysfunction/damage, a marker of atherosclerosis risk associated with a higher rate of atherothrombotic events [1], can be assessed by measurement of plasma levels of von Willebrand factor (VWF) [2,3], a complex glycoprotein that plays an important role in hemostasis by promoting platelet adhesion and aggregation and acting as a carrier of coagulation factor VIII [4]. Prospective epidemiological studies have shown VWF levels to be independently associated with atherothrombotic events – either coronary heart disease or ischemic stroke [5,6]. In addition, elevated VWF predicts cardiovascular mortality in survivors of acute myocardial infarction (AMI) and diabetic patients [7]. However, limited data are available regarding its possible relationship with non-invasive measurements commonly used as surrogates for the presence of atherosclerosis. We, therefore, hypothesized that VWF might be a valuable marker of subclinical atherosclerosis in asymptomatic subjects.

In 857 consecutive middle-aged subjects (80.4% men) with cardiovascular risk factors but free from clinically overt atherosclerosis, we measured the plasma levels of VWF and high-sensitivity C-reactive protein (hs-CRP) by specific immunoassays (Liatest VWF, Diagnostica Stago, Asnières, France; and Immulyte hs-CRP, DPC, Los Angeles, CA, USA, respectively), as well as serum cholesterol, glucose and creatinine clearance by standard biochemical techniques. In all subjects, we assessed the carotid intima-media thickness (IMT) by ultrasonography [8] and microalbuminuria (≥17 mg albumin g−1 creatinine in males and ≥25 mg g−1 creatinine in females), two established structural and functional surrogate markers of atherosclerosis, respectively [9,10]. Multivaried linear regression analyses were used to assess the independent relationship between VWF and both carotid IMT and microalbuminuria after adjusting for traditional risk factors and inflammatory markers.

The average age of patients was 54 ± 10 years, 51% were dyslipidemics, 32.2% hypertensives, 11.4% diabetics and 35% smokers. Treatments include antihypertensive drugs (16.2%), statins (15.8%) and oral antidiabetics (7.1%). In the overall population, there was a significant positive bivariate correlation between plasma VWF with age (P < 0.001), body mass index (BMI) (P = 0.037), systolic blood pressure (P = 0.008), glucose (P < 0.001), and total cholesterol (P < 0.001). VWF levels also correlated significantly with hs-CRP (P = 0.03). As shown in Fig. 1, when considering tertiles of VWF, there was a linear trend for the increase in VWF, carotid IMT (P < 0.001) and microalbuminuria (P = 0.018), without significant differences in the anti-atherosclerotic treatments across VWF tertiles.

Figure 1.

 Box plot showing the carotid IMT (top) and microalbuminuria (bottom) in asymptomatic subjects in relation to plasma VWF tertiles (median and 75th percentile are shown).

Both surrogates of subclinical atherosclerosis also correlated with age, BMI, systolic and diastolic blood pressure and glucose (P < 0.001 for all parameters). A multivaried linear regression analysis was, therefore, performed to assess the association of VWF with both carotid IMT and microalbuminuria after adjusting for these potential confounders. As there is some evidence that VWF levels may reflect a systemic inflammatory state, further adjustment was also made for hs-CRP, as a marker of microinflammation. As shown in Table 1, the association between VWF and IMT remained highly significant (P < 0.001) after adjusting for the effects of cardiovascular risk factors and CRP. Similarly, VWF was significantly associated with microalbuminuria after adjustment for the same potential confounders (P < 0.01).

Table 1.   Association of VWF with carotid IMT (mm) and microalbuminuria (mg g−1 creatinine) as dependent variables, in multiple linear regression analysis
 BSE(B)P-value*Partial R2 (%)
  1. SBP, systolic blood pressure; BMI, body mass index; VWF, von Willebrand factor; CRP, C-reactive protein; B, regression coefficient; SE(B), standard error; R2, partial correlation coefficient after adjustment.

  2. *Adjusted for age, sex, BMI, SBP, cholesterol, glucose levels, smoking and CRP.

Carotid IMT
 Age (years)5.80.001<0.00112.2
 Sex (F/M)67.10.014<0.0012.9
 SBP (mmHg)1.40.001<0.0012.8
 VWF (U dL−1)0.30.0010.0021.3
 Smoking (N/Y)34.00.0120.0041.1
 SBP (mmHg)7.90.001<0.0019.7
 Glucose (mg dL−1)3.70.001<0.0014.8
 VWF (U dL−1)0.60.0010.0031.1
 Smoking (N/Y)76.60.0330.0210.6
 Sex (F/M)82.60.0650.0420.5

von Willebrand factor has generally been regarded as a reliable marker of endothelial dysfunction/damage [2,3], an early event in the process of atherogenesis. An enhanced release of this molecule by the endothelium could contribute to atherosclerosis development and progression [6,11]. In a large series of asymptomatic subjects free of overt atherosclerotic disease, VWF levels were independently associated with carotid IMT and microalbuminuria, two well-established surrogates of underlying subclinical atherosclerosis. Our results strongly suggest that VWF may represent a systemic marker of subclinical atherosclerosis and are consistent with the hypothesis that VWF might provide an important pathophysiological link between endothelial cell dysfunction/damage and surrogates of diffuse atherosclerosis in different regions of the arterial tree. This is, to our knowledge, the first report showing a simultaneous and independent association of VWF with both structural and functional surrogates of atherosclerosis in asymptomatic subjects. Measurement of VWF might be an useful marker in the identification of asymptomatic subjects who may benefit from anti-atherosclerotic and/or antithrombotic strategies aimed to reduce circulating VWF plasma levels [12].


This work was funded through the agreement between FIMA and the ‘UTE project CIMA’, University of Navara (Pamplona, Spain). Partial support was also provided by the Pfizer Foundation, the Department of Health, Gobierno de Navarra (Pamplona, Spain) and RECAVA.

Declaration of conflicts of interest

The authors declare that they have no conflicts of interest regarding this study.