Von Willebrand Factor and ADAMTS13 in Relation to Major Adverse Limb Events in Peripheral Artery Disease

Abstract

Purpose

This study aimed to investigate the relationship between the two biomarkers, von Willebrand factor (VWF) and ADAMTS13, and major adverse limb events (MALE) in patients with peripheral artery disease (PAD).

Materials and Methods

After obtaining informed consent, baseline blood samples were collected from 48 PAD patients aged 60-75 years who were undergoing assessment, surveillance, medical management, or consideration for surgical intervention. VWF and ADAMTS13 antigen levels were measured by enzyme-linked immunosorbent assay (ELISA). Patients were monitored prospectively for subsequent MALE, defined as lower-limb revascularization or major amputation. Biomarker levels were compared between patients who did and did not experience subsequent MALE using Mann–Whitney U-tests, and time to MALE was evaluated using Kaplan–Meier analyses. This study was approved by the Unity Health Toronto Research Ethics Board.

Results

Twenty-four patients (50%) experienced MALE after baseline blood sampling. These patients had significantly lower VWF antigen levels (median [interquartile range]: 17,817.18 [32,339.16] ng/mL vs. 46,175.95 [75,284.64] ng/mL, P=0.026) and lower VWF/ADAMTS13 ratios (P=0.008) than those without MALE. Kaplan–Meier analyses comparing biomarker values above versus below the cohort median showed non-significant trends toward a higher cumulative incidence of MALE during follow-up for lower VWF, higher ADAMTS13, and lower VWF/ADAMTS13 ratios. In an exploratory analysis excluding patients with baseline chronic limb-threatening ischemia, below-median VWF/ADAMTS13 ratios were associated with a higher cumulative incidence of MALE during follow-up (log-rank P=0.018).

Conclusion

Contrary to our original hypothesis, lower baseline VWF levels and VWF/ADAMTS13 ratios were associated with subsequent MALE in this high-risk PAD cohort, which may reflect increased VWF consumption in severe, diffuse atherosclerosis. This study contributes to the limited literature on VWF and ADAMTS13 in PAD in relation to MALE, as previous research has largely focused on major adverse cardiovascular events. Larger studies in patients with similar risk profiles are required to validate these findings.

INTRODUCTION

Peripheral artery disease (PAD) affects approximately 200 million people globally [1]. Patients with PAD are at risk of major adverse limb events (MALE), defined as surgical or endovascular lower extremity revascularization or major (above-ankle) amputation. In the COMPASS trial of antithrombotic therapy for patients with stable vascular disease, 2.0% of patients with PAD developed MALE over a median follow-up of 21 months [2]. Several clinical factors contribute to MALE risk, including coronary artery disease, smoking, diabetes mellitus, hypertension, and a history of revascularization [3-7]. In COMPASS, patients with a history of revascularization had a higher incidence of MALE (3.8%). Antithrombotic strategies to prevent MALE in PAD include low-dose anticoagulation combined with aspirin; however, this approach is associated with a higher bleeding risk than aspirin alone. Therefore, the potential benefits of intensifying antithrombotic therapy must be balanced against bleeding risk.

Antiplatelet agents are well established for preventing thrombotic events in PAD. Current antiplatelet agents primarily target pathways of platelet activation and aggregation. However, initial platelet adhesion to the vessel wall is mediated by von Willebrand factor (VWF). Following vascular injury, VWF binds to several components of the exposed subendothelium (including collagen), stabilizes factor VIII, and promotes platelet aggregation to facilitate coagulation [8]. ADAMTS13 is a major regulator of VWF activity [9]. Selectively interfering with the interaction between VWF and platelets may therefore reduce platelet-driven thrombus formation, either as an alternative or adjunct to conventional antiplatelet therapy.

Currently, the contribution of VWF and ADAMTS13 to thromboembolic events in PAD remains unclear. Our study aimed to evaluate this relationship by measuring VWF and ADAMTS13 antigen concentration in a cohort of PAD patients across various stages of disease, including those with and without chronic limb-threatening ischemia (CLTI), and those with or without recent revascularization. We compared baseline VWF and ADAMTS13 levels between individuals who experienced subsequent MALE after enrollment and those who did not. Given the real-world clinical setting, we evaluated associations between baseline biomarkers and subsequent MALE.

Based on the existing literature and the established role of VWF in thrombosis, we hypothesized that elevated VWF and reduced ADAMTS13 concentrations would be associated with MALE after baseline sample collection.

MATERIALS AND METHODS

1) Ethical considerations

Blood sampling, clinical data collection, and laboratory analyses were performed under a protocol approved by the Unity Health Toronto Research Ethics Board (No. REB#16-375). For the present analysis, we used baseline blood samples that had been prospectively collected at the index study visit (when written informed consent was obtained) and stored in an existing biobank at St. Michael’s Hospital, Toronto, Canada. The biobank included samples from PAD patients requiring assessment, surveillance, medical management, or consideration for surgical intervention, collected through the vascular surgery outpatient clinics or emergency departments at St. Michael’s Hospital and McMaster University.

Written informed consent was obtained from all patients. The biobank consent permits secondary use of samples for current and future PAD-related biomarker research, and this study was conducted within this approved scope. Biobank samples and associated clinical data were provided to the investigators in a coded, de-identified format using a non-identifiable study ID. All identifying information was retained securely at Unity Health Toronto and was not accessible to investigators at any stage of the analysis.

2) Patient selection

Patients within a restricted age range (60-75 years) were selected, since age is known to influence VWF levels. Female and male patients were included, including those with CLTI or stable claudication at baseline, and those with or without a history of lower limb revascularization.

3) Blood samples and laboratory analyses

Ten-milliliter blood samples were collected from the antecubital vein using a 21-gauge needle into 3.2% buffered sodium citrate vacutainer tubes. Samples were taken to the laboratory and processed within one hour. Plasma was separated by centrifugation at 1,000×g for 10 minutes at 4°C and stored at –80°C until analysis, with only one freeze-thaw cycle.

A single baseline blood sample was collected at the index study visit, at which written informed consent was obtained, during a vascular surgery outpatient clinic visit or hospitalization when the patient was clinically stable. The sample was stored in the biobank and later retrieved for biomarker analysis. Patients were prospectively monitored for MALE following the baseline blood draw.

ADAMTS13 and VWF antigen levels were measured using a Human ADAMTS13 Quantikine® enzyme-linked immunosorbent assay (ELISA) Kit (R&D Systems Inc.) and a Human VWF ELISA Kit (Thermo Fisher Scientific Inc.), respectively.

4) Clinical data collection

Data collected at the index study visit included demographics, medical history, and concomitant medications. Patients were followed prospectively for subsequent MALE, defined as lower limb arterial revascularization (surgical or endovascular) or major (above-ankle) amputation occurring after baseline blood collection. Because some participants were enrolled during vascular surgery assessment, revascularization events may have included elective procedures; therefore, the study was designed to assess associations rather than definitive prediction of spontaneous events.

5) Statistical analysis

Baseline characteristics (age, sex, hypertension, diabetes mellitus, coronary artery disease, smoking status, and prior revascularization) and biomarker measurements are described for the overall cohort and stratified by subsequent MALE status. Continuous variables are reported as mean (standard deviation) or median (interquartile range [IQR]), as appropriate, and categorical variables as number (%).

To assess associations between baseline characteristics and VWF or ADAMTS13 levels, linear regression was used for continuous variables and Mann–Whitney U-tests for binary variables. Chi-squared tests (or Fisher exact tests, as appropriate) were used to compare baseline characteristics between those who did and did not experience MALE. Group comparisons of baseline VWF levels, ADAMTS13 levels, and VWF/ADAMTS13 ratios by subsequent MALE status were performed using Mann–Whitney U-tests.

Kaplan–Meier curves were generated to describe time to MALE events for all PAD patients and non-CLTI patients separately, comparing groups with VWF and ADAMTS13 concentrations and VWF/ADAMTS13 ratios above versus below the cohort median. Log-rank (Mantel–Cox) tests were used to compare survival curves, with statistical significance assessed between groups.

RESULTS

Forty-eight PAD patients were included in the study. Baseline characteristics are shown in Table 1. Twenty-four patients experienced MALE after baseline blood sampling. Patients who subsequently developed MALE were more likely to have CLTI at baseline than those who did not (P=0.046). No other statistically significant differences in baseline characteristics were observed between patients who did and did not develop MALE. No clear association between VWF or ADAMTS13 levels and age, sex, diabetes, hypertension, coronary artery disease, smoking status, or history of revascularization was observed (Supplementary Figs. 1, 2).

Table 1.

Baseline characteristics

Baseline characteristic	Total population (n=48)	PAD with subsequent MALE (n=24)	PAD without subsequent MALE (n=24)	P-value
Age (y)	68.5 (63.0-72.3)	67.5 (63.8-74.0)	67.9±4.8	0.675
Male	30/48 (62.5)	16/24 (66.7)	14/24 (58.3)	0.551
Diabetes mellitus	26/48 (54.2)	15/24 (62.5)	11/24 (45.8)	0.247
Hypertension	42/48 (87.5)	20/24 (83.3)	22/24 (91.7)	0.383
Coronary artery disease	11/48 (22.9)	5/24 (20.8)	6/24 (25.0)	0.731
Smoking
Non-smoker	4/48 (8.3)	2/24 (8.3)	2/24 (8.3)	1.000
Former smoker	30/48 (62.5)	14/24 (58.3)	16/24 (66.7)	0.551
Current smoker	14/48 (29.2)	8/24 (33.3)	6/24 (25.0)	0.525
Revascularization <3 months	11/48 (22.9)	5/24 (20.8)	6/24 (25.0)	0.731
CLTI at baseline sampling	12/48 (25.0)	9/24 (37.5)	3/24 (12.5)	0.046

Values are presented as median (interquartile range), mean±standard deviation, or number (%).

PAD, peripheral artery disease; MALE, major adverse limb events; CLTI, chronic limb-threatening ischemia.

Participants who subsequently developed MALE had lower VWF antigen levels than those without MALE (median [IQR]: 17,817.18 [32,339.16] vs. 46,175.95 [75,284.64] ng/mL, respectively; P=0.026) (Fig. 1). ADAMTS13 concentration did not differ significantly between groups (P=0.106) (Fig. 1), but the VWF/ADAMTS13 ratio was lower in patients who developed MALE than in those who did not (median [IQR]: 20.22 [34.33] vs. 59.26 [113.48], respectively; P=0.008) (Fig. 1).

Fig. 1.

Baseline VWF and ADAMTS13 antigen levels in PAD patients with and without subsequent MALE. (A) VWF antigen. (B) ADAMTS13 antigen. (C) VWF/ADAMTS13 ratio. VWF, von Willebrand factor; PAD, peripheral artery disease; MALE, major adverse limb events.

When time to MALE was compared between patients with biomarker values above versus below the cohort median for VWF, ADAMTS13, and VWF/ADAMTS13 ratio, no statistically significant differences were observed (log-rank P=0.073, P=0.064, and P=0.069, respectively) (Fig. 2). Nevertheless, the Kaplan–Meier curves suggested that patients with lower VWF, higher ADAMTS13, and a lower VWF/ADAMTS13 ratio may have a higher cumulative incidence of MALE during follow-up (Fig. 2). In an exploratory analysis excluding patients with CLTI at baseline sampling, those with a below-median VWF/ADAMTS13 ratio had a higher cumulative incidence of MALE during follow-up (log-rank P=0.018) (Fig. 3).

Fig. 2.

Kaplan–Meier curves for time to subsequent MALE in PAD patients stratified by biomarker values above versus below the cohort median. (A) VWF antigen. (B) ADAMTS13 antigen. (C) VWF/ADAMTS13 ratio. P-values are from the log-rank (Mantel–Cox) test. VWF, von Willebrand factor; PAD, peripheral artery disease; MALE, major adverse limb event.

Fig. 3.

Kaplan–Meier curves for time to subsequent MALE in non-CLTI PAD patients stratified by biomarker values above versus below the cohort median. (A) VWF antigen. (B) ADAMTS13 antigen. (C) VWF/ADAMTS13 ratio. P-values are from the log-rank (Mantel–Cox) test. VWF, von Willebrand factor; CLTI, chronic limb-threatening ischemia; PAD, peripheral artery disease; MALE, major adverse limb events.

DISCUSSION

In our cohort of 48 patients with PAD, the rate of MALE was high (50%). Patients who developed MALE exhibited significantly lower baseline VWF antigen concentrations than those who did not (P=0.026). The VWF/ADAMTS13 ratio was also significantly lower in the patients with subsequent MALE (P=0.008).

Kaplan–Meier analysis showed a non-significant trend toward earlier MALE onset among PAD patients with VWF levels and VWF/ADAMTS13 ratios below the cohort median. The Kaplan–Meier curves for VWF and VWF/ADAMTS13 ratio began to separate early during follow-up, after an initial period with high event rates in both groups, with clearer divergence emerging within the first few months after baseline sampling. For ADAMTS13, separation appeared earlier during follow-up.

Our findings that PAD patients who developed MALE had lower VWF levels and lower VWF/ADAMTS13 ratios were unexpected, given that previous studies have reported associations between higher VWF levels and cardiovascular events. For example, Green et al. [10] investigated VWF and ADAMTS13 levels in PAD patients leading up to cardiovascular events, and reported a weak trend toward increased VWF/ADAMTS13 ratios in the 10 months before such events. Kalinin et al. [11] examined the level and activity of various hemostatic markers, including VWF, in PAD patients before and after endovascular treatment. They found that elevated levels of VWF antigen and activity were linked to the onset of myocardial infarction and fatalities one year following lower extremity endovascular procedures.

We considered whether confounding factors may have contributed to our unexpected result. Established clinical risk factors for MALE in PAD, including smoking, hypertension, diabetes, and coronary artery disease [3-7], have also been associated with elevated VWF levels [12-17]. In addition, smoking has been associated with decreased ADAMTS13 levels [18], which would be expected to increase the VWF/ADAMTS13 ratio. Thus, these factors would be expected to bias results toward higher VWF levels or higher VWF/ADAMTS13 ratios among higher-risk patients, which is opposite to our observations. In our cohort, we did not observe a clear association between baseline clinical factors (including smoking, age, and sex) and VWF or ADAMTS13 levels that would plausibly explain the observed inverse association; however, given the modest sample size, residual confounding cannot be excluded. While previous studies suggest that oral coagulants and antiplatelet agents have, at most, minor effects on VWF antigen levels [19,20], we were not able to adjust for the use of these agents in relation to VWF levels or outcomes.

Our cohort included 12 patients with CLTI who may have had planned procedures at the time of sampling. In an exploratory analysis excluding this subgroup, a below-median VWF/ADAMTS13 ratio was significantly associated with a higher cumulative incidence of MALE during follow-up. However, given the observational design and limited sample size, planned procedures and other unmeasured factors cannot be excluded as contributors to the observed associations.

While previous studies, such as those by Green et al. [10] and Kalinin et al. [11], have primarily focused on major adverse cardiovascular events and biomarkers in PAD, our study focused on MALE and yielded contrasting associations. We hypothesize that in patients with severe, diffuse atherosclerosis affecting larger peripheral arteries, there may be increased VWF consumption or clearance, leading to lower circulating antigen levels. Such consumption may explain the inverse association between VWF antigen levels and subsequent MALE risk observed in our PAD cohort. We further hypothesize that increased consumption of VWF in the aorta and proximal leg arteries may be associated with downstream thromboembolism leading to MALE. Because only antigen levels were measured and functional assays (e.g., activity or multimer distribution) were not available, mechanistic inferences remain speculative. Accordingly, our findings should be interpreted as exploratory and hypothesis-generating.

This cohort represents a particularly high-risk clinical population, as evidenced by the high MALE rate of 50%. Patients were enrolled from vascular surgery outpatient clinics and the emergency department where they were attending for assessment, surveillance, medical management, or consideration for intervention, rather than from a general PAD population. Within this cohort, 25.0% had CLTI at baseline, 22.9% had undergone lower limb revascularization within the preceding three months, and diabetes, hypertension, and current smoking were highly prevalent. These characteristics may explain the elevated incidence of MALE observed in this study. Larger studies involving patients with similar risk profiles are needed to validate these findings and to clarify the potential role of the VWF–ADAMTS13 axis in MALE.

Currently, a VWF-platelet interaction inhibitor, caplacizumab, is approved for use in acquired thrombotic thrombocytopenic purpura, a condition characterized by excess VWF activity. Recombinant ADAMTS13 (TAK-755) has been evaluated clinically and approved as enzyme replacement therapy for congenital thrombotic thrombocytopenic purpura, a condition of severe ADAMTS13 deficiency. Although these indications are distinct from PAD, they illustrate that the VWF–ADAMTS13 axis is therapeutically modifiable. If future studies clarify a pathophysiological role of VWF-platelet axis in MALE, the potential relevance of such concepts in PAD could be investigated.

CONCLUSION

This study identified an unexpected inverse association between baseline VWF antigen levels and VWF/ADAMTS13 ratios and subsequent MALE in high-risk PAD patients, which may reflect increased VWF consumption in severe atherosclerosis. Further research is warranted to confirm these findings and to elucidate the underlying mechanisms and potential clinical relevance in PAD.

SUPPLEMENTARY MATERIALS

Supplementary data can be found via https://doi.org/10.5758/vsi.250103.

Funding Statement

FUNDING This work was supported by the Canadian Institutes of Health Research, the Natural Sciences and Engineering Research Council, and the National Heart, Lung and Blood Institute of the National Institutes of Health (R01HL172780). This work was also funded through support from the Blair Foundation. The funder did not play a role in the design or conduct of this research.