ABSTRACT
Background
Higher von Willebrand factor (VWF) levels and lower ADAMTS13 activity are linked to increased risk of atherosclerotic cardiovascular disease, but their association with atherosclerosis burden, a potential underlying mechanism, remains unclear. We investigated associations of these biomarkers with arterial calcification, a surrogate of atherosclerosis, in a general population.
Methods
In 2224 Rotterdam Study participants with measured VWF Antigen level (VWF:Ag) and ADAMTS13 activity (in percentage), computed tomography assessed calcification volume in the coronary arteries (CAC), aortic arch (AAC), and extracranial and intracranial carotid arteries (ECAC, ICAC) (mean age 69.5 ± 6.7 years; 51.5% female). Associations of VWF:Ag, ADAMTS13 activity, and their ratio (linear and quartiles) with calcification (linear and quartiles) were evaluated using multivariable linear and proportional odds logistic regression, adjusting for demographic and cardiovascular factors.
Results
Higher ADAMTS13 activity was significantly associated with lower volume of CAC (β [95% CI]: −0.04 [−0.08, −0.01]), AAC (−0.08 [−0.11, −0.03]), and ECAC (−0.05 [−0.09, −0.01]), but not in ICAC. A higher ADAMTS13:VWF ratio showed stronger inverse associations, extending to ICAC, suggesting a potential interplay between ADAMTS13 and VWF (β [95% CI]: CAC: −0.12 [−0.21, −0.02]; AAC: −0.20 [−0.30, −0.10]; ECAC: −0.08 [−0.18, −0.01]; ICAC: −0.11 [−0.21, −0.02]). The effect size of the ratio increased across quartiles at all artery‐specific locations, strengthening the observed relationship. Across levels of calcification, associations of the biomarkers remained generally unchanged.
Conclusions
Our findings support the potential relevance of ADAMTS13 activity in arterial calcification and, by extension, atherosclerosis. The observed associations may reflect both its balance with VWF and pathways independent of VWF. Future longitudinal and causal studies should determine whether ADAMTS13 influences the development and progression of arterial calcification through VWF‐dependent or VWF‐independent pathways, thereby clarifying its biological role and potentially identifying targets for preventive strategies against atherosclerotic vascular disease.
1. Introduction
Von Willebrand factor (VWF) is a plasma glycoprotein that plays a crucial role in hemostasis and thrombus formation, particularly by promoting platelet aggregation and stabilizing coagulation factor VIII [1, 2]. Higher circulating VWF exerts pro‐thrombotic effects, increasing the risk of arterial thrombus formation and acute cardiovascular events [3, 4, 5]. Epidemiological studies have also shown that higher VWF concentrations are associated with increased risk of atherosclerotic cardiovascular diseases, independent of traditional risk factors [5, 6, 7, 8, 9, 10]. This suggests a link between VWF and increased risk of atherosclerosis [11, 12].
VWF is primarily produced by endothelial cells [11] and higher plasma concentrations reflect endothelial activation or damage, a key early step in atherosclerosis. Human studies investigating the link between VWF and atherosclerosis are limited, but some real‐world data suggest associations with subclinical atherosclerosis in small groups, as assessed by carotid intima‐media thickness or carotid artery calcification, including in patients with asymptomatic cardiovascular disease [13], patients with rheumatoid arthritis [14] or patients with cerebrovascular events [15]. Large‐scale, population‐based studies are needed to evaluate the association of VWF with atherosclerosis across multiple vascular sites, particularly the coronary arteries, which reflect cumulative atherosclerotic burden [16].
VWF activity is regulated by the ADAMTS13, which cleaves ultra‐large VWF into smaller, less active forms, thereby preventing excessive platelet aggregation [17]. Attributed to this regulatory function, higher ADAMTS13 activity has been associated with a decreased risk of cardiovascular disease and acute events [18, 19]. Studying ADAMTS13 alongside VWF allows evaluation of their combined effect on outcomes [20, 21].
In the current study, embedded within the large population‐based Rotterdam Study, we examined the association between VWF and ADAMTS13 and arterial calcification, a surrogate marker of atherosclerosis, across multiple key arteries, including the coronary arteries, aortic arch, extracranial carotid arteries, and intracranial carotid arteries.
2. Materials and Methods
2.1. Settings and Population for Analysis
This study was conducted as part of the ongoing population‐based Rotterdam Study (RS), which aims to investigate the determinants and prevalence of chronic diseases among adults and the elderly [22]. The RS was approved by the Medical Ethics Committee of Erasmus MC (registration number MEC 02.1015) and the Dutch Ministry of Health, Welfare, and Sport (Population Screening Act WBO, license number 1071272–159521‐PG) and adheres to the principles of the Declaration of Helsinki. The RS is registered with the Netherlands National Trial Registry Platform and the WHO International Clinical Trials Registry Platform (ICTRP; www.who.int/ictrp/network/primary/en/) under shared catalogue number NTR6831. All participants have provided written informed consent to participate and allow access to their information from pharmacies.
The current study enrolled individuals from the RS research center who were invited to undergo a non‐contrast multi‐detector computed tomography (MDCT) scan to visualize arterial calcification in multiple arteries. Initially, 3229 individuals were invited to participate, and 2524 underwent the MDCT scan, resulting in a response rate of 78%. After excluding scans with image artifacts, 2413 individuals had complete examinations. Among these, a subset of 2399 individuals with available plasma measurements of VWF antigen (VWF:Ag) and ADAMTS13 activity comprised the current study population.
2.2. Assessment of VWF and ADAMTS13
Fasting venous blood samples were obtained in citrated tubes at the research center and stored at −80°C between 1996 and 2001. VWF:Ag levels were quantified using an in‐house ELISA, employing polyclonal rabbit anti‐human VWF antibodies (DakoCytomation, Glostrup, Denmark) for catching and tagging. The procedure showed a good intra‐assay coefficient of variation and an inter‐assay coefficient of variation (5.8% and 7.8%, respectively). VWF:Ag values, reported in IU/dL by multiplying the value in IU/mL by 10 [5]. ADAMTS13 activity was measured through a kinetic assay utilizing the fluorescence resonance energy transfer substrate VWF 73 (FRETS‐VWF73). Methods of measurement are described in more detail elsewhere [19]. Plasma samples were compared against a reference curve consisting of serial dilutions of pooled normal human plasma with an ADAMTS13 activity of 1 U/mL (George King Biomedical Inc., Overland Park, KS, USA). ADAMTS13 activity was expressed as a percentage of the reference plasma.
2.3. Assessment of Arterial Calcification
We used 16‐slice or 64‐slice MDCT scanners (Somatom Sensation 16 or 64; Siemens, Forchheim, Germany) between 2003 and 2006. We performed cardiac scan and scan of the aortic arch and carotid arteries which has been done using single breath‐hold and consecutive non‐overlapping 3‐mm‐thick slices. Scan protocol has been described previously [23]. We assessed coronary artery calcification (CAC), aortic arch calcification (AAC), extracranial carotid artery calcification (ECAC), and intracranial carotid artery calcification (ICAC). All calcification volumes are expressed in cubic millimeters. For assessment, two blinded reviewers quantified calcification using dedicated software (syngo Calcium Scoring, Siemens, Forchheim, Germany) with a threshold of 130 Hounsfield units. For ICAC, a semi‐automated scoring method was used that allowed for the manual segmentation of calcification in each consecutive CT slice. A detailed description of the quantification procedure has been described in more depth elsewhere [24, 25].
2.4. Other Measurements
Cardiovascular risk factors and other covariates were collected through questionnaire, physical examination, and blood sampling at the time of the CT scan. These factors encompassed confounders and modifiers, including age, sex, HDL cholesterol, non‐HDL cholesterol, waist‐to‐hip ratio (WHR), history of hypertension, diabetes mellitus, current smoking, prevalent CVD, use of vitamin K antagonists, and use of bisphosphonates. Methods for measuring cardiovascular risk factors are explained in detail in Method S1. The maximum proportion of missing covariate values was 15%. We applied multivariate imputation by chained equations (5 imputations and 10 iterations) to impute missing values using the “mice” package in R [26]. The distributions were similar in the imputed and non‐imputed datasets, and complete case analyses based on non‐imputed data showed similar results. The study population had complete data for the exposures (VWF:Ag and ADAMTS13 activity) and the outcome (calcification), so no imputation was required for these variables.
2.5. Statistical Analysis
To evaluate the associations of VWF and ADAMTS13 with atherosclerosis, we employed the following strategy. First, we assessed the separate associations of VWF:Ag and ADAMTS13 activity with calcification volume in each artery. Given the opposing biological roles of VWF and ADAMTS13, we additionally evaluated the ADAMTS13:VWF ratio, providing an integrated measure of their physiological balance and potential synergy [20, 21]. Multivariable linear regression was applied with calcification volume as the outcome. Because calcification volume was right‐skewed, values were log‐transformed after adding 1.0 mm3 to the non‐transformed values, followed by standardization to enable comparability across arteries. The values of VWF:Ag and ADAMTS13 activity were also standardized to allow direct comparison of their relative impact on the outcome. Both biomarkers were evaluated as continuous variables and also categorized into quartiles with an equal number of individuals per quartile. Associations of higher quartiles with calcification were assessed relative to the first quartile to examine potential dose–response relationships.
In a subsequent analysis, we assessed the biomarkers against quartile‐based ordinal levels of calcification volume to evaluate their relative effects across ordered categories of the outcome without assuming a constant relationship across all levels [27]. Multivariable ordinal regression was used to examine the effects of VWF:Ag, ADAMTS13 activity, and their ratio on calcification levels with an equal number of individuals across the calcification levels.
Three regression models were developed. Model 1 was adjusted for age, sex, and the time interval between VWF:Ag and ADAMTS13 activity measurement and the CT scan, to account for the previously reported linear increase of VWF and ADAMTS13 over time [28, 29]. Model 2 included additional adjustment for cardiovascular risk factors, including HDL cholesterol, non‐HDL cholesterol, smoking, WHR, diabetes, hypertension, as well as use of vitamin K antagonists, statins, bisphosphonates, and prevalent CVD. Model 3 further adjusted for fibrinogen, accounting for its regulatory role in platelet cohesion and its potential influence on the outcome. In a sensitivity analysis, we stratified participants according to the blood collection‐CT scan interval (shorter versus longer interval based on the median) and repeated the Model 2 regression analyses within each stratum. Separately, potential effect modification by sex and prevalent CVD history was evaluated by including interaction terms between the exposures and these factors in Model 2. All statistical analyses were performed using RStudio version 4.4.1 for Windows.
3. Results
3.1. Population Characteristics
Baseline characteristics of the study population are presented in Table 1. The mean age of participants was 69 years (SD±6.7), and 51.6% were female. The average time between the VWF and ADAMTS13 measurements and the CT scan was 4.59 years (SD±0.5). Calcification was prevalent, ranging from 73.0% in the ECAC to 92.5% in the AAC. The mean VWF:Ag level was 122.0 IU/dL (SD±48.4), and mean ADAMTS13 activity was 93.4% (SD±17.3). ADAMTS13 activity was significantly higher in females than in males (mean: 96.9% vs. 89.8%, p < 0.001; Table S1). Stratified by prevalent CVD, VWF:Ag concentrations were higher in participants with CVD (mean 130.7 vs. 121.5 IU/dL), whereas ADAMTS13 activity was slightly lower in this group (90.2% vs. 93.7%), with ranges indicating reduced variability among participants with CVD (Table S2).
TABLE 1.
Characteristics of the study population at CT scan.
| Total N = 2224 | ||
|---|---|---|
| Age |
Mean (SD) Median [Min, Max] |
69.00 (6.72) 67.00 [59.00, 98.00] |
| Current smoking | N (%) | 350 (15.7%) |
| Diabetes mellitus | N (%) | 293 (13.2%) |
| Hypertension | N (%) | 1121 (50.4%) |
| Waist to hip ratio |
Mean (SD) Median [Min, Max] |
0.911 (0.09) 0.908 [0.65, 1.21] |
| HDL‐cholesterol (mmol/L) |
Mean (SD) Median [Min, Max] |
1.44 (0.39) 1.39 [0.67, 3.59] |
| Non‐HDL‐cholesterol (mmol/L) |
Mean (SD) Median [Min, Max] |
4.24 (0.95) 4.19 [1.21, 7.77] |
| Statins use | N (%) | 547 (24.6%) |
| Bisphosphonates use | N (%) | 206 (9.3%) |
| Vitamin K antagonists use | N (%) | 235 (10.6%) |
| Prevalent CVD | N (%) | 146 (6.6%) |
| Fibrinogen (g/L) |
Mean (SD) Median [Min, Max] |
3.78 (0.86) 3.70 [1.40, 9.50] |
| VWF antigen (IU/dL) continous |
Mean (SD) Median [Min, Max] |
122.00 (48.40) 113.00 [29.00, 345.00] |
| ADAMTS13 activity (%) continous |
Mean (SD) Median [Min, Max] |
93.40 (17.30) 93.10 [5.00, 175.00] |
| VWF antigen levels categories | ≥ 50 (IU/dL), N (%) | 2196 (98.7%) |
| 30–50 (IU/dL), N (%) | 27 (1.20%) | |
| ≤ 30 (IU/dL), N (%) | 1 (0.0%) | |
| ADAMTS13 activity levels | ≥ 20%, N (%) | 2223 (100.0%) |
| 10%–20%, N (%) | 0 (0.0%) | |
| ≤ 10%, N (%) | 1 (0.0%) | |
| Blood collection to CT scan interval (years) |
Mean (SD) Median [Min, Max] |
4.59 (0.55) 4.58 [1.42, 7.9] |
| Presence of CAC | N (%) | 1821 (81.9%) |
| Presence of AOC | N (%) | 2058 (92.5%) |
| Presence of ECAC | N (%) | 1623 (73.0%) |
| Presence of ICAC | N (%) | 1829 (82.2%) |
| Volume of CAC (mm3) |
Mean (SD) Median [Min, Max] |
266.00 (540) 52.80 [0.00, 69.00] |
| Volume of AAC (mm3) |
Mean (SD) Median [Min, Max] |
728.00 (1240) 254.00 [0.00, 119.00] |
| Volume of ECAC (mm3) |
Mean (SD) Median [Min, Max] |
105.00 (210) 225.00 [0.00, 28.00] |
| Volume of ICAC (mm3) |
Mean (SD) Median [Min, Max] |
116.00 (187) 42.60 [0.00, 1450.00] |
Note: Diabetes was defined as a fasting glucose level ≥ 7 mmol/L and/or the use of anti‐diabetic medication. Hypertension was defined as a systolic blood pressure ≥ 160 mmHg and/or a diastolic blood pressure ≥ 100 mmHg, and/or the use of blood pressure‐lowering medication.
Abbreviations: AAC, Aortic arch calcification; Blood collection to CT scan interval, The time between blood collection (for measuring VWF and ADAMTS13 levels) and the CT scan; CAC, Coronary artery calcification; CT, Computed tomography; CVD, Cardiovascular diseases, defined as a history of heart failure, myocardial infarction, or stroke; ECAC, Extracranial internal carotid artery calcification; HDL, High‐Density Lipoprotein; ICAC, Intracranial internal carotid artery calcification; LDL, Low‐Density Lipoprotein; VWF, Von Willebrand factor.
3.2. Associations Between VWF:Ag, ADAMTS13 Activity, Their Ratio and Calcification Volume
There was no statistically significant association between VWF:Ag and calcification volume in any of the arterial sites (although all β coefficients > 0, Table 2, Section A). The effect estimate of VWF was generally unchanged when adjusted for covariates such as cardiovascular risk factors (Model 2) and fibrinogen (Model 3), compared with models without these adjustments (confounding < 2%). There was no statistically significant difference in the effects of VWF across the ADAMTS13 range (all p values for the interaction VWF*ADAMTS13 ≥ 0.39).
TABLE 2.
Multivariable linear regression analysis of the associations between von Willebrand factor antigen (Section A), ADAMTS13 activity (Section B), and their ratio (Section C) with the volume of calcification in different arteries.
| CAC Β (95% CI) | AAC Β (95% CI) | ECAC Β (95% CI) | ICAC Β (95% CI) | |
|---|---|---|---|---|
| A: VWF:Ag | ||||
| Model 1 (N = 2224) | 0.05 (0.01, 0.08)* | 0.05 (0.01, 0.09)* | 0.03 (−0.00, 0.07) | 0.05 (0.01, 0.09)* |
| Model 2 (N = 2224) | 0.03 (−0.01, 0.06) | 0.03 (−0.00, 0.08) | 0.02 (−0.02, 0.05) | 0.03 (−0.01, 0.06) |
| Model 3 (N = 2224) | 0.02 (−0.02, 0.06) | 0.02 (−0.01, 0.07) | 0.01 (−0.03, 0.05) | 0.02 (−0.02, 0.06) |
| B: ADAMTS13 activity | ||||
| Model 1 (N = 2224) | −0.03 (−0.01, 0.01) | −0.07 (−0.10, −0.03)* | −0.04 (−0.09, 0.00) | −0.03 (−0.07, 0.01) |
| Model 2 (N = 2224) | −0.04 (−0.08, −0.01)* | −0.08 (−0.11, −0.03)* | −0.05 (−0.09, −0.01)* | −0.04 (−0.08, 0.00) |
| Model 3 (N = 2224) | −0.04 (−0.08, −0.01)* | −0.07 (−0.10, −0.03)* | −0.05 (−0.09, −0.01)* | −0.03 (−0.07, 0.00) |
| C: ADAMTS13:VWF ratio | ||||
| Model 1 (N = 2224) | −0.02 (−0.26, −0.06)* | −0.24 (−0.33, −0.14)* | −0.14 (−0.24, −0.03)* | −0.16 (−0.26, −0.06)* |
| Model 2 (N = 2224) | −0.12 (−0.21, −0.02)* | −0.20 (−0.30, −0.10)* | −0.08 (−0.18, −0.01)* | −0.11 (−0.21, −0.02)* |
| Model 3 (N = 2224) | −0.10 (−0.20, −0.01)* | −0.18 (−0.27, −0.08)* | −0.07 (−0.17, −0.02) | −0.10 (−0.20, −0.00)* |
Note: Model 1: Adjusted for sex, age, and the time interval between blood collection and computed tomography. Model 2: Further adjusted for HDL cholesterol, non‐HDL cholesterol, current smoking, waist‐to‐hip ratio, diabetes mellitus, hypertension, use of vitamin K antagonists, statins, bisphosphonates, and prevalent cardiovascular diseases. Model 3: Further adjusted for fibrinogen. Statistically significant at p < 0.05. The asterisk * in the table denotes significance at this level.
Abbreviations: AAC, Aortic arch calcification; B, Beta coefficient; CAC, Coronary artery calcification; CI, Confidence interval; ECAC, Extracranial internal carotid artery calcification; ICAC, Intracranial internal carotid artery calcification; N, Number of individuals; VWF, Von Willebrand factor.
Overall, calcification volume decreased with increasing ADAMTS13 activity, reaching statistical significance for CAC, AAC, and ECAC, but not for ICAC (Table 2, Section B Model 2). This indicates that, for example, each one standard deviation increase in ADAMTS13 activity (in this study 17.3 points) was associated with an estimated 0.04 standard deviation decrease in the log‐transformed CAC volume, independent of other covariates. The associations of ADAMTS13 with calcification were generally unchanged after adjustment for cardiovascular risk factors (Model 2), and fibrinogen (Model 3) compared with models without adjustments for these factors (confounding < 1%).
The higher ADAMTS13:VWF ratio was associated with significantly lower calcification volumes across all artery‐specific locations (Table 2, Section C, Model 2). Compared with analyses of VWF:Ag or ADAMTS13 activity alone, the ratio consistently yielded larger effect sizes, indicating a stronger combined association (with the strongest associations observed in the AAC) and highlighting the potential synergistic effect between these two biomarkers in relation to the outcome.
3.3. Associations Between Quantiles of VWF:Ag, ADAMTS13 Activity, Their Ratio and Calcification Volume
Across increasing quartiles of VWF:Ag, increasing effect estimates were observed only for CAC (Table 3, Section A). For the ADAMTS13:VWF ratio, effect estimates increased across increasing quartiles for all artery‐specific locations, reaching statistical significance in the highest quartile (Q4) for both CAC and AAC (Table 3, Section C), with AAC showing the largest magnitude of association.
TABLE 3.
Multivariable linear regression analysis of the association between quartiles of von Willebrand factor antigen (VWF:Ag), ADAMTS13 activity, their ratio, and volume of calcification in different arteries.
| CAC Β (95% CI) | AAC Β (95% CI) | ECAC Β (95% CI) | ICAC Β (95% CI) | |
|---|---|---|---|---|
|
A: quartiles of VWF:Ag Model 2 | ||||
| Q1(29, 88) (N = 568) | Reference | Reference | Reference | Reference |
| Q2(88, 113) (N = 552) | 0.05 (−0.05, 0.15) | 0.11 (0.01, 0.21)* | 0.03 (−0.07, 0.14) | −0.06 (−0.16, 0.05) |
| Q3(113, 146) (N = 551) | 0.03 (−0.07, 0.13) | 0.05 (−0.05, 0.15) | 0.02 (−0.07, 0.11) | 0.02 (−0.08, 0.13) |
| Q4(146, 345) (N = 553) | 0.07 (−0.03, 0.17) | 0.11 (0.01, 0.22)* | 0.04 (−0.06, 0.15) | 0.05 (−0.05, 0.16) |
|
B: quartiles of ADAMTS13 activity Model 2 | ||||
| Q1(5, 82.8) (N = 557) | Reference | Reference | Reference | Reference |
| Q2(82.8, 93.1) (N = 555) | −0.09 (−0.20, 0.01) | −0.06 (−0.16, 0.03) | 0.01 (−0.09, 0.11) | −0.07 (−0.18, 0.03) |
| Q3(93.1, 104) (N = 556) | −0.11 (−0.21, −0.01)* | −0.16 (−0.26, −0.05)* | −0.09 (−0.20, 0.01) | −0.10 (−0.11, −0.09)* |
| Q4(104, 175) (N = 556) | −0.11 (−0.22, −0.01)* | −0.16 (−0.27, −0.05)* | −0.07 (−0.18, 0.04) | −0.08 (−0.20, 0.02) |
|
C: quartiles of VWF:ADAMTS13 ratio Model 2 | ||||
| Q1(0.282, 0.913) (N = 556) | Reference | Reference | Reference | Reference |
| Q2(0.913, 1.22) (N = 556) | −0.03 (−0.11, 0.05) | −0.03 (−0.13, 0.08) | −0.00 (−0.11, 0.10) | −0.06 (−0.17, 0.04) |
| Q3(1.22, 1.66) (N = 556) | −0.07 (−0.16, 0.03) | −0.08 (−0.18, 0.02) | −0.03 (−0.13, 0.07) | −0.08 (−0.19, 0.03) |
| Q4(1.66, 27.4) (N = 556) | −0.08 (−0.13, −0.01)* | −0.15 (−0.26, −0.04)* | −0.08 (−0.19, 0.03) | −0.10 (−0.21, 0.01) |
Note: Each quartile (Q1–Q4) contains approximately 25% of the total data points, with the first quartile (Q1) representing the lowest level and the fourth quartile (Q4) representing the highest level. Q1 was set as the reference group for the associations. Model 2 was adjusted for sex, age, current smoking, diabetes mellitus, HDL cholesterol, LDL cholesterol, hypertension, vitamin K antagonist use, statin use, bisphosphonate use, and prevalent cardiovascular diseases. Statistically significant at p < 0.05. The asterisk * in the table denotes significance at this level.
Abbreviations: AAC, Aortic arch calcification; B, Beta coefficient; CAC, Coronary artery calcification; CI, Confidence interval; ECAC, Extracranial internal carotid artery calcification; ICAC, Intracranial internal carotid artery calcification; N, Number of individuals; VWF, Von Willebrand factor.
3.4. Associations Between VWF:Ag, ADAMTS13 Activity, Their Ratio and Levels of Calcification
Analysis of the quartile‐based ordinal calcification levels showed no statistically significant associations between VWF and calcification levels across any artery‐specific locations (Table 4, Section A). In contrast, higher ADAMTS13 activity was associated with significantly lower calcification across all arteries (Table 4, Section B). For example, each one standard deviation increase in ADAMTS13 activity corresponded to an estimated 0.10 standard deviation decrease in log‐transformed calcification volume for CAC, independent of other covariates.
TABLE 4.
Multivariable partial proportional odds ordinal logistic regression analysis of the associations between von Willebrand factor antigen, ADAMTS13 activity, their ratio, and levels of calcification in different arteries.
| A: VWF:Ag | B: ADAMTS13 activity | C: ADAMTS13:VWF ratio | ||
|---|---|---|---|---|
| Model 2, quartiles of CAC | ||||
| Β (95% CI) | 0.06 (−0.02, 0.14) | −0.10 (−0.18, −0.02) | −0.28 (−0.48, −0.07) | |
| Intercepts Β (SE) | Q1(−1.45, −1.01)|Q2(−1.0, 0.17) | 7.22 (0.82) | 7.27 (0.81) | 6.85 (0.84) |
| Q2(−1.01, 0.17)|Q3(0.17, 0.83) | 8.54 (0.83) | 8.59 (0.82) | 8.17 (0.84) | |
| Q3(0.17, 0.83)|Q4(0.83, 2.14) | 9.91 (0.84) | 9.96 (0.83) | 9.55 (0.85) | |
| Model 2, quartiles of AAC | ||||
| Β (95% CI) | 0.08 (−0.00, 0.16) | −0.13 (−0.21, −0.04) | −0.45 (−0.66, −0.25) | |
| Intercepts Β (SE) | Q1(−2.17, −0.54)|Q2(−0.54, 0.22) | 10.05 (0.83) | 10.16 (0.82) | 9.45 (0.85) |
| Q2(−0.54, 0.22)|Q3(0.22, 0.73) | 11.38 (0.84) | 11.49 (0.83) | 10.78 (0.85) | |
| Q3(0.22, 0.73)|Q4(0.73, 1.88) | 12.76 (0.85) | 12.87 (0.84) | 12.16 (0.86) | |
| Model 2, quartiles of ECAC | ||||
| Β (95% CI) | 0.03 (−0.05, 0.11) | −0.11 (−0.20, −0.03) | −0.18 (−0.38, 0.01) | |
| Intercepts Β (SE) | Q1(−1.27)|Q2(−1.27, 0.14) | 7.48 (0.81) | 7.47 (0.80) | 7.20 (0.83) |
| Q2(−1.27, 0.14)|Q3(0.14, 0.86) | 8.64 (0.82) | 8.63 (0.81) | 8.36 (0.84) | |
| Q3(0.14, 0.86)|Q4(0.86, 2.28) | 9.95 (0.83) | 9.94 (0.82) | 9.67 (0.84) | |
| Model 2, quartiles of ICAC | ||||
| Β (95% CI) | 0.02 (−0.05, 0.10) | −0.08 (−0.16, −0.00) | −0.19 (−0.39, 0.01) | |
| Intercepts Β (SE) | Q1(−1.67, −0.63)|Q2(−0.63, 0.19) | 8.12 (0.82) | 8.12 (0.81) | 7.83 (0.83) |
| Q2(−0.63, 0.19)|Q3(0.19, 0.77) | 9.38 (0.82) | 9.38 (0.82) | 9.09 (0.84) | |
| Q3(0.19, 0.77)|Q4(0.769, 1.92) | 10.70 (0.83) | 10.70 (0.82) | 10.41 (0.85) | |
Note: Each quartile (Q1–Q4) contains approximately 25% of the total data points with the first quartile (Q1) representing the lowest level and the fourth quartile representing the highest level (Q4). The number of participants per quartile are as follows: For CAC, Q1 = 558, Q2 = 554, Q3 = 556, Q4 = 556; for AAC, Q1–Q4 = 556 each; for extracranial carotid calcification, Q1 = 601, Q2 = 512, Q3 = 557, Q4 = 554; and for intracranial carotid calcification, Q1–Q4 = 556 each. Single coefficients (for VWF, ADAMTS13, or their ratio) indicate that the proportional odds assumption holds for these variables, meaning their association with the outcome is consistent across all levels. Intercepts represent the log odds of being in a certain quartile versus the lower quartiles. Model 2 was adjusted for sex, age, current smoking, diabetes mellitus, HDL cholesterol, LDL cholesterol, hypertension, vitamin K antagonist use, statin use, bisphosphonate use, and prevalent cardiovascular diseases.
Abbreviations: AAC, Aortic arch calcification; B, Beta coefficient; CAC, Coronary artery calcification; CI, Confidence interval; ECAC, Extracranial internal carotid artery calcification; ICAC, Intracranial internal carotid artery calcification; N, Number of individuals; SE, Standard Error; VWF, Von Willebrand factor.
The higher ADAMTS13:VWF ratio was associated with lower calcification levels across all arteries, reaching statistical significance for CAC and AAC, but not for ECAC or ICAC (Table 4, Section B). Compared with analyses of VWF:Ag or ADAMTS13 activity alone, the ratio consistently yielded larger effect sizes. A progressive increase in the intercepts across higher quartiles indicated that progressively greater underlying risk is required to transition to higher calcification levels. When VWF:Ag, ADAMTS13 activity, and their ratio were evaluated separately across these quartiles (Table S2), effect sizes remained largely unchanged, suggesting that these biomarkers alone are unlikely to explain the reduced likelihood of progression to higher calcification levels, which is likely influenced by other confounders.
3.5. Sensitivity Analyses
In sensitivity analyses stratified by the blood collection‐CT scan interval, effect estimates were generally comparable between participants with shorter (1.42–4.58 years) and longer (4.58–7.90 years) intervals (Table S3). The 95% confidence intervals largely overlapped between strata, and no consistent pattern of differences in the direction or magnitude of associations between VWF:Ag, ADAMTS13 activity, their ratio, and calcification measures was observed.
In additional sensitivity analyses, the associations of VWF, ADAMTS13, and their ratio with calcification were not statistically different across sex or prevalent CVD (all p values for interaction ≥ 0.45). When stratified by sex or prevalent CVD, no consistent increase or decrease in effect sizes was observed across analyses or artery‐specific locations.
4. Discussion
In this population‐based study, higher ADAMTS13 activity was associated with lower calcification volumes in CAC, AAC, and ECAC. When ADAMTS13 activity was considered together with VWF:Ag (ADAMTS13:VWF ratio), we observed statistically significant associations with lower calcification across all arteries studied, with effect sizes for the ratio consistently larger than those of either marker alone. Across quartiles of the ratio, effect sizes increased with higher quartiles, observed at all artery‐specific locations, reaching statistical significance in the highest quartile (Q4) for CAC and, most notably, for AAC, where the effect sizes were overall largest. The absence of a statistically significant independent effect of VWF:Ag, alongside the significant effect of ADAMTS13 activity, suggests that ADAMTS13 may influence arterial calcification through mechanisms at least partly distinct from those of VWF. Nevertheless, the stronger associations observed for the ADAMTS13:VWF ratio suggest that considering both biomarkers together may better capture relevant biological variation related to arterial calcification and, by extension, atherosclerotic processes. This may reflect the biological balance between ADAMTS13 and VWF.
The observed significant association of ADAMTS13 activity with lower arterial calcification is consistent with previous studies linking higher ADAMTS13 activity to a reduced risk of atherosclerotic cardiovascular disease [6, 18]. Although the primary function of ADAMTS13 is known to be the cleavage of large VWF multimers, we did not observe an independent association of VWF with calcification in our study, as we did for ADAMTS13. These findings raise the hypothesis that ADAMTS13 may influence calcification, and by extension atherosclerosis, through mechanisms at least partially distinct from VWF. Supporting this hypothesis, basic science research in Adamts13‐deficient mice suggests that ADAMTS13 may have functions beyond thrombosis regulation and may be related to the contractile phenotype of vascular smooth muscle cells (VSMCs). Loss of the contractile VSMC phenotype and transition toward pathological phenotypes are important features of vascular remodeling and can contribute to vascular calcification and atherosclerotic lesion development [17, 30, 31]. In these mice, ADAMTS13 deficiency also enhances endothelial leukocyte adhesion and macrophage accumulation in plaques, leading to larger and more inflammatory atherosclerotic lesions [31]. These pathophysiological processes may provide a potential biological explanation for our findings and suggest mechanisms through which ADAMTS13 may be associated with vascular health beyond its established role in regulating VWF and thrombosis.
Building on our findings, future studies are needed to determine whether the association between ADAMTS13 activity and arterial calcification is explained by its effect on VWF or by pathways independent of VWF. Prospective studies with repeated measurements of ADAMTS13 and VWF and longitudinal assessment of arterial calcification are needed to clarify their temporal relationships with the development and progression of arterial calcification. Causal mediation analyses [32] could help assess the extent to which the association between ADAMTS13 and calcification is mediated through VWF, while population genetic approaches, such as Mendelian randomization [33] using genetic proxies for ADAMTS13 activity, may provide further insight into potential causal relationships paving the way for potential targeted therapeutic strategies. Identifying factors that enhance ADAMTS13 activity, examining the role of modifiable precursors affecting ADAMTS13, and exploring the potential to repurpose existing therapeutic strategies could be valuable areas for future research. Incorporating measures of VWF functional properties and multimer composition could provide information beyond VWF:Ag. In particular, high‐molecular‐weight VWF multimers (HMWM‐VWF), which represent the most functionally active forms of VWF [34], may be relevant for understanding whether specific highly functional forms of VWF are associated with arterial calcification. Studying populations with reduced HMWM‐VWF, such as selected patients with Type 2A or Type 3 von Willebrand disease and acquired von Willebrand syndrome [35] could provide an opportunity to investigate whether HMWM‐VWF is associated with the development or progression of arterial calcification despite the lack of an association with total VWF:Ag in our study. Similarly, factors that may influence the ability of ADAMTS13 to cleave VWF, such as interleukin‐6, thrombospondin‐1, and platelet factor 4 [36], could be assessed as potential modifiers of the association between ADAMTS13 activity and arterial calcification. This could help determine whether the effect of ADAMTS13 on VWF cleavage contributes to the observed association between ADAMTS13 activity and arterial calcification.
In our population‐based sample, we found no evidence that prevalent CVD modified the associations of VWF:Ag or ADAMTS13 activity with arterial calcification. VWF:Ag levels were higher among participants with prevalent CVD than among those without CVD (mean 130.7 vs. 121.5), consistent with the known increase in VWF in response to vascular injury and systemic inflammation. Higher VWF:Ag in participants with prevalent CVD may therefore partly reflect established vascular injury or systemic inflammation rather than the extent of calcification itself.
Although the association between ADAMTS13 activity and ICAC was not statistically significant, the direction of the association was consistent with that observed at the other artery‐specific locations. This may reflect the smaller effect estimate for ICAC, with a variance comparable to that observed for the other artery‐specific locations, resulting in a confidence interval that included the null. Differences in the pathophysiology of calcification may also contribute, as intracranial arterial calcification is more often located within or adjacent to the internal elastic lamina, whereas calcification in other artery‐specific locations may more commonly involve the intimal or medial layers [37, 38, 39].
Our study's strengths include a robust analytical strategy designed to address well‐supported biological hypotheses and the use of a large, population‐based cohort with high‐quality, image‐based measures of calcification in four vascular beds within the same individuals, enabling direct head‐to‐head comparisons. However, several limitations must be acknowledged. VWF plasma levels and ADAMTS13 activity may vary according to ABO blood group [40] and genetic variation [41, 42]. While accounting for these factors could explain some of the variability, they are not known to be associated with arterial calcification and are therefore unlikely to confound the observed associations. Additionally, bilirubin levels on the day of ADAMTS13 activity assessment were not available; therefore, we could not evaluate potential bilirubin‐related interference with the FRET‐based ADAMTS13 activity assay. However, the extent of such interference in a general population‐based cohort is unclear, as it is primarily a concern in individuals with markedly elevated bilirubin levels. Although VWF:Ag is widely used in population‐based studies and has shown relevance to clinical outcomes [4, 19, 43] it reflects the quantity of circulating VWF rather than its functional properties. Measures of VWF functional properties, such as platelet‐binding activity and multimer composition, were not available in our study and could provide complementary information on the functional state of VWF.
VWF and ADAMTS13 were measured on average 4.6 years (IQR 1.42–7.9) prior to the CT scans. These biomarkers may fluctuate due to transient conditions, such as inflammation or acute illness, and may also change over longer periods due to aging or underlying health status. Although we adjusted for age, the time interval between blood collection and imaging, and prevalent CVD, the biomarkers were not measured longitudinally or at the time of outcome assessment. Nevertheless, previous population‐based studies, including studies from the Rotterdam Study, have shown that single baseline measurements of VWF:Ag and ADAMTS13 activity can be associated with subsequent cardiovascular outcomes [4, 19, 43]. In line with this, our sensitivity analysis stratified by the interval between blood collection and CT showed generally comparable effect estimates for shorter and longer intervals. However, these findings do not address within‐person variability in VWF and ADAMTS13 over time. Repeated longitudinal measurements of VWF and ADAMTS13, combined with information on transient clinical or inflammatory conditions around the time of measurement, would be needed to characterize such variability and assess its potential impact on the observed associations. Finally, residual confounding remains possible despite adjustment for major known covariates. In addition, the predominantly Central European ancestry of the study population may limit the generalizability of our findings, and replication in populations with more diverse ancestral backgrounds is warranted.
5. Conclusion
Higher ADAMTS13 activity was associated with lower arterial calcification, whereas there was no evidence of an independent association between VWF and arterial calcification. Associations were stronger and more consistent when ADAMTS13 activity was considered relative to VWF, suggesting that both ADAMTS13 activity itself and its balance with VWF may be involved in atherosclerotic processes. Longitudinal studies incorporating repeated biomarker and calcification measurements, complemented by causal inference approaches, represent a valuable area for future research. Causal mediation analysis could assess the extent to which the potential effect of ADAMTS13 on the development and progression of arterial calcification is mediated through VWF. Mendelian randomization using genetic proxies for ADAMTS13 activity could further investigate its potential causal effect on arterial calcification. Together, these approaches could clarify the biological role of ADAMTS13 and identify potential targets for preventive strategies for atherosclerotic vascular disease, potentially including modifiable factors that influence ADAMTS13 activity or function.
Author Contributions
M.N.S.: study concept and design, developed the analysis plan, conducted the analyses, interpreted results, and drafted the manuscript, incorporating comments and revisions. F.W.G.L.: study conception and critical review of the manuscript. M.P.M.d.M.: manuscript review. M.K.: manuscript review. D.B.: study conception, methodological feedback, and supervision. All authors read and approved the final version.
Funding
This work was supported by the Dutch Heart Foundation (AtheroNeth consortium) and the Leducq Foundation (COMET Network), with MK as the recipient of funding support. All other authors did not receive any financial support for this work.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Table S1: Characteristics of the study population at CT by sex.
Method S1: Covariates assessment.
Table S2: Distribution of VWF and ADAMTS13 by prevalent cardiovascular disease status.
Table S3: Stratified multivariable linear regression analysis of the associations between von Willebrand factor antigen, ADAMTS13 activity, and their ratio with calcification volume in different arterial locations according to the blood collection‐CT scan interval.
Acknowledgments
The authors express their gratitude to the study participants and the dedicated staff of the Rotterdam Study (RS), including participating general practitioners, research assistants, pharmacists, and hospital personnel.
Nekouei Shahraki M., Leebeek F. W. G., de Maat M. P. M., Kavousi M., and Bos D., “Von Willebrand Factor and ADAMTS13 in Relation to Atherosclerosis in the General Population: The Rotterdam Study,” European Journal of Haematology 117, no. 6 (2026): 1637–1646, 10.1111/ejh.70309.
This work was performed at Erasmus Medical Center, Rotterdam, the Netherlands.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
References
- 1. Leebeek F. W. G. and Eikenboom J. C. J., “Von Willebrand's Disease,” New England Journal of Medicine 375, no. 21 (2016): 2067–2080. [DOI] [PubMed] [Google Scholar]
- 2. Reininger A. J., “Function of Von Willebrand Factor in Haemostasis and Thrombosis,” Haemophilia 14 (2008): 11–26. [DOI] [PubMed] [Google Scholar]
- 3. Spiel A. O., Gilbert J. C., and Jilma B., “Von Willebrand Factor in Cardiovascular Disease: Focus on Acute Coronary Syndromes,” Circulation 117, no. 11 (2008): 1449–1459. [DOI] [PubMed] [Google Scholar]
- 4. Sonneveld M. A. H., Franco O. H., Ikram M. A., et al., “Von Willebrand Factor, ADAMTS13, and the Risk of Mortality: The Rotterdam Study,” Arteriosclerosis, Thrombosis, and Vascular Biology 36, no. 12 (2016): 2446–2451. [DOI] [PubMed] [Google Scholar]
- 5. Wieberdink R. G., van Schie M. C., Koudstaal P. J., et al., “High Von Willebrand Factor Levels Increase the Risk of Stroke: The Rotterdam Study,” Stroke 41, no. 10 (2010): 2151–2156. [DOI] [PubMed] [Google Scholar]
- 6. Reardon B., Pasalic L., and Favaloro E. J., “The Intriguing Relationships of Von Willebrand Factor, ADAMTS13 and Cardiac Disease,” Journal of Cardiovascular Development and Disease 8, no. 9 (2021): 115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Rumley A., Lowe G. D. O., Sweetnam P. M., Yarnell J. W. G., and Ford R. P., “Factor VIII, Von Willebrand Factor and the Risk of Major Ischaemic Heart Disease in the Caerphilly Heart Study,” British Journal of Haematology 105, no. 1 (1999): 110–116. [PubMed] [Google Scholar]
- 8. Smith F. B., Lee A. J., Fowkes F. G. R., Price J. F., Rumley A., and Lowe G. D. O., “Hemostatic Factors as Predictors of Ischemic Heart Disease and Stroke in the Edinburgh Artery Study,” Arteriosclerosis, Thrombosis, and Vascular Biology 17, no. 11 (1997): 3321–3325. [DOI] [PubMed] [Google Scholar]
- 9. Whincup P. H., Danesh J., Walker M., et al., “Von Willebrand Factor and Coronary Heart Disease. Prospective Study and Meta‐Analysis,” European Heart Journal 23, no. 22 (2002): 1764–1770. [DOI] [PubMed] [Google Scholar]
- 10. Willeit P., Thompson A., Aspelund T., et al., “Hemostatic Factors and Risk of Coronary Heart Disease in General Populations: New Prospective Study and Updated Meta‐Analyses,” PLoS One 8, no. 2 (2013): e55175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Blann A., “Von Willebrand Factor and the Endothelium in Vascular Disease,” British Journal of Biomedical Science 50, no. 2 (1993): 125–134. [PubMed] [Google Scholar]
- 12. Van Galen K. P. M., Tuinenburg A., Smeets E. M., and Schutgens R. E. G., “Von Willebrand Factor Deficiency and Atherosclerosis,” Blood Reviews 26, no. 5 (2012): 189–196. [DOI] [PubMed] [Google Scholar]
- 13. Paramo J. A., Beloqui O., Colina I., Diez J., and Orbe J., “Independent Association of Von Willebrand Factor With Surrogate Markers of Atherosclerosis in Middle‐Aged Asymptomatic Subjects,” Journal of Thrombosis and Haemostasis 3, no. 4 (2005): 662–664. [DOI] [PubMed] [Google Scholar]
- 14. Ristić G. G., Subota V., Lepić T., et al., “Subclinical Atherosclerosis in Patients With Rheumatoid Arthritis and Low Cardiovascular Risk: The Role of Von Willebrand Factor Activity,” PLoS One 10, no. 8 (2015): e0130462. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Sonneveld M. A. H., van Dijk A. C., van den Herik E. G., et al., “Relationship of Von Willebrand Factor With Carotid Artery and Aortic Arch Calcification in Ischemic Stroke Patients,” Atherosclerosis 230, no. 2 (2013): 210–215. [DOI] [PubMed] [Google Scholar]
- 16. Alexopoulos N. and Raggi P., “Calcification in Atherosclerosis,” Nature Reviews Cardiology 6, no. 11 (2009): 681–688. [DOI] [PubMed] [Google Scholar]
- 17. Plautz W. E., Raval J. S., Dyer M. R., Rollins‐Raval M. A., Zuckerbraun B. S., and Neal M. D., “ADAMTS13: Origins, Applications, and Prospects,” Transfusion 58, no. 10 (2018): 2453–2462. [DOI] [PubMed] [Google Scholar]
- 18. Denorme F., Kraft P., Pareyn I., et al., “Reduced ADAMTS13 Levels in Patients With Acute and Chronic Cerebrovascular Disease,” PLoS One 12, no. 6 (2017): e0179258. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Sonneveld M. A. H., de Maat M. P. M., Portegies M. L. P., et al., “Low ADAMTS13 Activity Is Associated With an Increased Risk of Ischemic Stroke,” Blood 126, no. 25 (2015): 2739–2746. [DOI] [PubMed] [Google Scholar]
- 20. Roy P. and Esteve J., “Using Relative Risk Models for Estimating Synergy Between Two Risk Factors,” Statistics in Medicine 17, no. 12 (1998): 1357–1373. [DOI] [PubMed] [Google Scholar]
- 21. Pearce N. and Greenland S., “Confounding and Interaction,” in Handbook of Epidemiology (Springer, 2024), 1–31. [Google Scholar]
- 22. Ikram M. A., Brusselle G., Ghanbari M., et al., “Objectives, Design and Main Findings Until 2020 From the Rotterdam Study,” European Journal of Epidemiology 35 (2020): 483–517. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Bos D., van der Rijk M. J. M., Geeraedts T. E. A., et al., “Intracranial Carotid Artery Atherosclerosis: Prevalence and Risk Factors in the General Population,” Stroke 43, no. 7 (2012): 1878–1884. [DOI] [PubMed] [Google Scholar]
- 24. Bos D., Leening M. J. G., Kavousi M., et al., “Comparison of Atherosclerotic Calcification in Major Vessel Beds on the Risk of All‐Cause and Cause‐Specific Mortality: The Rotterdam Study,” Circulation: Cardiovascular Imaging 8, no. 12 (2015): e003843. [DOI] [PubMed] [Google Scholar]
- 25. van der Toorn J. E., Rueda‐Ochoa O. L., van der Schaft N., et al., “Arterial Calcification at Multiple Sites: Sex‐Specific Cardiovascular Risk Profiles and Mortality Risk—The Rotterdam Study,” BMC Medicine 18 (2020): 1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Van Buuren S. and Groothuis‐Oudshoorn K., “Mice: Multivariate Imputation by Chained Equations in R,” Journal of Statistical Software 45 (2011): 1–67. [Google Scholar]
- 27. Abreu M. N. S., Siqueira A. L., and Caiaffa W. T., “Ordinal Logistic Regression in Epidemiological Studies,” Revista de Saúde Pública 43 (2009): 183–194. [DOI] [PubMed] [Google Scholar]
- 28. Aillaud M. F., Pignol F., Alessi M. C., et al., “Increase in Plasma Concentrationof Plasminogen Activator Inhibitor, Fibrinogen, Von Willebrand Factor, Factor VIII: Cand in Erythrocyte Sedimentation Rate With Age,” Thrombosis and Haemostasis 55, no. 3 (1986): 330–332. [PubMed] [Google Scholar]
- 29. Liu W., Patel K., Wang Y., et al., “Dynamic and Functional Linkage Between Von Willebrand Factor and ADAMTS‐13 With Aging: An Atherosclerosis Risk in Community Study,” Journal of Thrombosis and Haemostasis 21, no. 12 (2023): 3371–3382. [DOI] [PubMed] [Google Scholar]
- 30. Feng Y., Li X., Xiao J., et al., “ADAMTS13: More Than a Regulator of Thrombosis,” International Journal of Hematology 104 (2016): 534–539. [DOI] [PubMed] [Google Scholar]
- 31. Gandhi C., Khan M. M., Lentz S. R., and Chauhan A. K., “ADAMTS13 Reduces Vascular Inflammation and the Development of Early Atherosclerosis in Mice,” Blood 119, no. 10 (2012): 2385–2391. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Richiardi L., Bellocco R., and Zugna D., “Mediation Analysis in Epidemiology: Methods, Interpretation and Bias,” International Journal of Epidemiology 42, no. 5 (2013): 1511–1519. [DOI] [PubMed] [Google Scholar]
- 33. Palmer T. M., Sterne J. A. C., Harbord R. M., et al., “Instrumental Variable Estimation of Causal Risk Ratios and Causal Odds Ratios in Mendelian Randomization Analyses,” American Journal of Epidemiology 173, no. 12 (2011): 1392–1403. [DOI] [PubMed] [Google Scholar]
- 34. Stockschlaeder M., Schneppenheim R., and Budde U., “Update on Von Willebrand Factor Multimers: Focus on High‐Molecular‐Weight Multimers and Their Role in Hemostasis,” Blood Coagulation & Fibrinolysis 25, no. 3 (2014): 206–216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Connell N. T., Flood V. H., Brignardello‐Petersen R., et al., “ASH ISTH NHF WFH 2021 Guidelines on the Management of Von Willebrand Disease,” Blood Advances 5, no. 1 (2021): 301–325. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. DeYoung V., Singh K., and Kretz C. A., “Mechanisms of ADAMTS13 Regulation,” Journal of Thrombosis and Haemostasis 20, no. 12 (2022): 2722–2732. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. van der Toorn J. E., Vernooij M. W., Ikram M. A., Kavousi M., and Bos D., “Progression of Arterial Calcifications: What, Where, and in Whom?,” European Radiology 34, no. 8 (2024): 5142–5152. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Fote G. M., Raefsky S., Mock K., Chaudhari A., Shafie M., and Yu W., “Intracranial Arterial Calcifications: Potential Biomarkers of Stroke Risk and Outcome,” Frontiers in Neurology 13 (2022): 900579. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Masuoka T., Hayashi N., Hori E., Kuwayama N., Ohtani O., and Endo S., “Distribution of Internal Elastic Lamina and External Elastic Lamina in the Internal Carotid Artery: Possible Relationship With Atherosclerosis,” Neurologia Medico‐Chirurgica 50, no. 3 (2010): 179–182. [DOI] [PubMed] [Google Scholar]
- 40. Whincup P. H., Cook D. G., Phillips A. N., and Shaper A. G., “ABO Blood Group and Ischaemic Heart Disease in British Men,” British Medical Journal 300, no. 6741 (1990): 1679–1682. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Jin S.‐Y., Tohyama J., Bauer R. C., Cao N. N., Rader D. J., and Zheng X. L., “Genetic Ablation of Adamts13 Gene Dramatically Accelerates the Formation of Early Atherosclerosis in a Murine Model,” Arteriosclerosis, Thrombosis, and Vascular Biology 32, no. 8 (2012): 1817–1823. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Kavousi M., Leebeek F. W. G., Felix J. F., Hofman A., and Witteman J. C. M., “Von Willebrand Factor Plasma Levels, Genetic Variations and Coronary Heart Disease in an Older Population,” Journal of Thrombosis and Haemostasis 10, no. 7 (2012): 1262–1269. [DOI] [PubMed] [Google Scholar]
- 43. Sonneveld M. A. H., Kavousi M., Ikram M. A., et al., “Low ADAMTS‐13 Activity and the Risk of Coronary Heart Disease–a Prospective Cohort Study: The Rotterdam Study,” Journal of Thrombosis and Haemostasis 14, no. 11 (2016): 2114–2120. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1: Characteristics of the study population at CT by sex.
Method S1: Covariates assessment.
Table S2: Distribution of VWF and ADAMTS13 by prevalent cardiovascular disease status.
Table S3: Stratified multivariable linear regression analysis of the associations between von Willebrand factor antigen, ADAMTS13 activity, and their ratio with calcification volume in different arterial locations according to the blood collection‐CT scan interval.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.