Recent reports have documented that ventricular assist devices (VADs) diminish circuiting plasma von Willebrand factor (vWF). It has been suggested that in patients with a VAD, the loss of high-molecular-weight vWF multimers produces an acquired von Willebrand syndrome,1–8 a bleeding diathesis.
vWF, a 20,000 kDa multimeric glycoprotein, is assembled from 225 kDa monomers. At sites of vascular injury, turbulent blood flow and altered shear stress trigger conformational activation of vWF multimers. Subsequently, vWF binds to subendothelial collagen and platelets to produce the primary hemostatic plug.9
VADs cause turbulent flow and alter shear stress profiles.10,11 As a result, the blood of patients with a VAD is exposed to nonphysiologic shear stress. Consequently, it has been speculated that acquired von Willebrand syndrome in VAD patients may be the result of shear-induced destruction of vWF.1–8 However, the mechanistic pathway(s) of vWF degradation and the role of the vWF-cleaving metalloproteinase, ADAMTS-1312, are unknown.
The objective of this study was to develop an in vitro system to study the molecular mechanisms of VAD-induced vWF impairment. With this model, we tested the following hypotheses: 1) degradation of high-molecular-weight vWF multimers is independent of endothelial cell function, 2) ADAMTS-13 participates in this process, 3) mechanical demolition participates in this process, 4) platelet activation participates in this process. As a result, decreased availability of vWF leads to a predisposition for bleeding episodes during mechanical circulatory support.
Whole Bovine Blood Preparation
Fresh, single-donor, whole bovine blood was obtained less than 24 hours before experimentation. Individual units from the single-donor source were mixed to create a homogenous blood pool. The pooled blood was diluted with Plasma-Lyte solution (Abbott Laboratories, Deerfield, IL) per American Society for Testing and Materials (ASTM) guidelines13 to obtain a hematocrit of 24 ± 1% and was filtered through a 20 μm blood filter (Baxter Y-type blood administration set, Deerfield, IL). Heparin (3,000 unit bolus, plus titration to effect) was administered to maintain an activated clotting time > 300 seconds according to ASTM guidelines.13
Mock Circulatory Loop Development
Mock circulatory loops (n = 4; Figure 1) were designed with a 75 cm 3/8” connective Tygon tubing (Medtronic, Minneapolis, MN) and a 1 L reservoir with a blood sampling port as previously described.14 A clinically approved, sterilized paracorporeal continuous-flow VAD (Rotaflow, Maquet Cardiovascular, Wayne, NJ; Centrimag, Thoratec, Pleasanton, CA) was placed in series with each loop. A c-clamp resistor (Fischer Scientific, Pittsburgh, PA) was placed distal to the pump to control pump head pressure and total loop flow at ~4 L/min. An ultrasonic flow probe (Transonics, Ithaca, NY) and proximal and distal fluid-filled pressure transducers (BD, Franklin Lakes, NJ) were placed to measure volumetric flow through the loop and pump inlet and outlet pressures, respectively.
Each circulatory loop was primed with 750 ml of anticoagulated, whole bovine blood as previously described.14 VADs were operated continuously for 6 hours. Whole blood samples were drawn in 6 ml ethylenediaminetetraacetic acid (EDTA) blood tubes (BD, Franklin Lakes, NJ) at baseline (before VAD support) and every 60 minutes for 6 hours. Samples were centrifuged at 3,000 rpm for 10 minutes at 15°C. Plasma aliquots were snap-frozen in liquid nitrogen and stored at −80°C.
Analysis of High-Molecular-Weight vWF Multimers
High-molecular-weight plasma vWF multimers were analyzed by standard immunoblotting techniques as previously described.15 Plasma samples were combined 1:10 with loading buffer (10 mM Tris-HCl, 1 mM EDTA, 2% sodium dodecyl sulfate (SDS), 8 M urea, bromophenol blue, pH 8.0). Samples were heated in a 60°C water bath for 20 minutes. Samples were loaded equally into 1.75% agarose-SDS gels (8.3 × 7 cm × 1.5 mm thick), and electrophoresis was performed at 30 V for 9 hours at 4°C in a 0.05 M Tris buffer (0.384 M glycine, 0.1% SDS, pH 8.35). Proteins were transferred at 30 V for 16 hours at 4°C to an Immobilon-P polyvinylidene fluoride (PVDF) membrane by wet electroblotting in phosphate buffer (0.2 M phosphate, 0.04% SDS, pH 7.4). Blots were blocked with milk-blocking buffer (1 M Tris-buffered saline (TBS), 5% dried milk powder, pH 7.6) for 1 hour and were washed three times in TBS-Tween (1 M TBS, 0.1% Tween-20, pH 7.6). Blots were incubated with a rabbit antihuman vWF primary antibody (1/500; Dako, Carpinteria, CA) overnight at 4°C, washed three times in TBS-Tween, and incubated with goat antirabbit IgG horseradish-peroxidase-conjugated secondary antibody (1/3000; Cell Signaling, Danvers, MA) for 1 hour. Blots were developed with ECL Plus Western blotting detection reagents (GE Healthcare, Piscataway, NJ) and scanned with a Typhoon 9400 Variable-Mode Imager (GE Healthcare, Piscataway, NJ).
Analysis of ADAMTS-13
Plasma ADAMTS-13 levels were analyzed by standard immunoblotting techniques. Plasma samples were combined 1:10 with loading buffer (10 mM Tris-HCl, 1 mM EDTA, 2% SDS, 8 M Urea, bromophenol blue, pH 8.0). Samples were heated in a 95°C water bath for 5 minutes. Samples were loaded equally into 7% polyacrylamide-SDS gels (1.5 mm thick, 10% SDS, 0.5 M Tris, 10% APS, and 10 μl tetramethylethylenediamine), and electrophoresis was performed at 40 V for 7 hours in a 0.025 M Tris buffer (0.192 M glycine, 0.1% SDS, pH 8.4). Proteins were transferred at 30 V for 12 hours to an Immobilon-P PVDF membrane by wet electroblotting at 4°C in a 0.025 M Tris buffer (0.192 M glycine, 0.04% SDS, 20% methanol, pH 8.4). Blots were blocked with milk-blocking buffer (1 M TBS, 5% dried milk powder, pH 7.6) for 1 hour and were washed three times in TBS-Tween (1 M TBS, 0.1% Tween-20, pH 7.6). Blots were incubated with rabbit antihuman ADAMTS-13 primary antibody (1/1000; Abcam, Cambridge, MA) overnight at 4°C, washed three times in TBS-Tween, and incubated with a goat antirabbit IgG horseradish-peroxidase-conjugated secondary antibody (1/3000; Cell Signaling, Danvers, MA) for 1 hour. Blots were developed with ECL Plus Western blotting detection reagents and were scanned with a Typhoon 9400 Variable-Mode Imager. Total protein was quantified with ImageQuant software (GE Healthcare, Piscataway, NJ) as the average staining intensity divided by total staining area.
Analysis of Low-Molecular-Weight vWF
Low-molecular-weight plasma vWF multimers were analyzed with a 1.75% agarose-SDS/7% polyacrylamide-SDS 30/70% stacking gel. Plasma samples from baseline and after 360 minutes of VAD support were prepared, electrophoresed, transferred, and developed as previously described for the analysis of high-molecular-weight vWF multimers. Total protein was quantified with ImageQuant (GE Healthcare, Piscataway, NJ) as the average staining intensity divided by total staining area.
Plasma Platelet Factor 4 ELISA Assay
Plasma platelet factor 4 (PF4) levels were measured in duplicate with a sandwich ELISA PF4/CXCL4 kit (USCNK, Wuhan, P.R. China) according to manufacturer instructions. The microtiter plate was precoated with monoclonal antibody specific to bovine PF4. Plasma samples were incubated in the coated wells for 2 hours at 37°C. Wells were emptied and incubated for 1 hour at 37°C with a biotin-conjugated polyclonal antibody preparation specific for PF4. Microplate wells were aspirated and washed three times before incubation with a solution of avidin-conjugated horseradish peroxidase for 30 minutes at 37°C. Washing was repeated. Tetramethylbenzidine was added, and the reaction was incubated for 20 minutes. The reaction was stopped with 1 M H2S04, and the signal was measured with a colorimetric microplate reader (Synergy Mx Microplate Reader, Biotek, Winooski, VT) at 450 nm.
GraphPad, version 4.00 (Prism, La Jolla, CA) was used to perform statistical analyses and plot data. Paired Student’s t-tests were used to compare low-molecular-weight vWF, ADAMTS-13, and PF4 levels at baseline versus 60, 120, 180, 240, and 360 minutes of VAD support. A p < 0.05 (95% confidence) was considered statistically significant. All data were presented as mean ± standard error. Plasma ADAMTS-13 and vWF fragment quantities were reported as percentage change (%Δ).
Time-Course of High-Molecular-Weight vWF Degradation
High-molecular-weight vWF multimers were analyzed with agarose-SDS gel electrophoresis and immunoblotting. Within 120 minutes, the highest-molecular-weight multimers decreased, and lower-molecular-weight multimers increased (Figure 2A, dashed boxes). Over the course of the experiment, this trend continued.
ADAMTS-13 was analyzed with polyacrylamide-SDS gel electrophoresis and immunoblotting. The ADAMTS-13 major and minor bands, which likely represented endothelial-derived plasma and platelet isoforms of ADAMTS-13, respectively,16 increased significantly from baseline (14 ± 4%, 12 ± 3%; p < 0.05; Figure 2B). Total plasma ADAMTS-13 increased significantly from baseline (13 ± 3%, p < 0.05).
Low-Molecular-Weight vWF Degradation
Low-molecular-weight vWF multimers were analyzed with an agarose-SDS/polyacrylamide-SDS stacking gel (Figure 3A). High-molecular-weight vWF multimers deposited in the agarose portion of the stacking gel and exhibited the same pattern of decreased high-molecular-weight multimers during VAD support as previously observed. In the polyacrylamide portion of the gel, bands that represented multiple low-molecular-weight vWF fragments emerged. When compared with baseline, 6 hours of VAD support increased the 140 kDa (5 ± 1%, p < 0.05) and 176 kDa (8 ± 3%, p = 0.09) ADAMTS-13 cleavage products, 225 kDa (33 ± 15%, p = 0.06) vWF monomers, and 310 kDa (34 ± 15%, p = 0.07) nascent vWF polypeptide monomers with vWF propeptide.
Plasma Platelet Factor 4
Platelet activation was quantified with a plasma PF4 ELISA (Figure 4). Over the course of 6 hours of VAD support, plasma PF4 progressively increased from baseline (21 ± 7%, p = 0.05) and indicated that the activation of platelets and release of PF4 from platelet α-granules into the plasma had occurred.
To date, this study reports the first mechanistic data on vWF degradation during mechanical circulatory support. In an in vitro circulatory loop, we observed the time-course and potential mechanism(s) of vWF degradation with a clinically approved continuous-flow VAD. Within 6 hours of uninterrupted VAD support, findings included: 1) degradation of high-molecular-weight and accumulation of low-molecular-weight vWF in the plasma, 2) small increases in plasma ADAMTS-13, 3) increases in ADAMTS-13-dependant (140, 176 kDa) and ADAMTS-13-independent (225, 310 kDa) vWF fragments, and 4) increases in plasma PF4, a marker of platelet activation. These data suggest that degradation of vWF during mechanical circulatory support develops rapidly and as a result of multiple mechanisms that occur in parallel. Figure 5 summarizes the proposed pathways of vWF degradation.
Recent reports have documented the rapid loss of high-molecular-weight vWF multimers in patients within hours5,7 of the implantation of a VAD. Our data confirm that the pathophysiological mechanism(s) that trigger degradation of vWF occur quickly. However, up until now, pathways of vWF degradation have not been elucidated.
It is well established that a plasma metalloprotease, ADAMTS-13, proteolytically cleaves vWF continuously in the blood.12 During this process, peptide fragments of 140 and 176 kDa are produced from vWF.17,18 In the current study, after 6 hours of support with a VAD, vWF fragments that were approximately 140 and 176 kDa increased by 5% and 8%, respectively. This finding suggested that ADAMTS-13 was partially responsible for the degradation of high-molecular-weight vWF multimers during VAD support.
It is interesting that we also observed a modest increase in plasma ADAMTS-13 protein. Platelets were the likely source. In vivo, ADAMTS-13 is manufactured primarily by hepatic stellate cells19 and by vascular endothelium.20 Yet, the liver and blood vessels were not present in our in vitro mock circulatory loop. But α-granules in platelets, which were present in the loop, also contain ADAMTS-13.16,21 During activation, platelets release multiple factors from α-granules to initiate autocrine and paracrine cascades that lead to platelet aggregation. PF4, a hemostatic chemokine, is one of these factors.22 Therefore, to determine whether the source of additional ADAMTS-13 was from platelets, we measured plasma PF4 as a marker of α-granule release. Indeed, over the course of the experiments, plasma PF4 increased significantly. This finding suggested that during VAD support, activated platelets released the contents of α-granules, which included PF4 and ADAMTS-13, into the plasma. Consequently, we speculate that in patients with a VAD, chronic activation of platelets may trigger a continuous, low-level release of ADAMTS-13 from platelet α-granules, increase plasma ADAMTS-13, and result in greater enzymatic cleavage of endogenous plasma vWF. It is well documented that mechanical circulatory support in humans23–25 and in bovids26,27 activates platelets, which provides further credibility to this hypothesis.
Plasma ADAMTS-13 increased by approximately 10%. Although this change was statistically significant, plasma ADAMTS-13 protein levels may still have been within the range of normal. The clinical relevance of such a modest increase is unknown. Indeed, clinical data have suggested that ADAMTS-13 does not play a major role in acquired von Willebrand syndrome during mechanical circulatory support.2,4,6 However, the studies were conducted in small subgroups of patients without appropriate controls, and published data were difficult to interpret. For example, in one study, plasma ADAMTS-13 during VAD support was 1.20 ± 0.40 AU/L.2 It was reported that vWF impairment was not caused by elevated ADAMTS-13. However, this range of reported values exceeded the reported range of normal (reference interval of 0.75–1.40 AU/L). Of additional concern, paired pre-VAD baseline values were not available for comparison. In another study, six patients had normal values of ADAMTS-13 activity (median 89%; range 65–121%) and antigen (median, 0.98 μg/ml; range 0.61–1.67 μg/ml).6 However, a reference interval was not provided, and again, paired baseline samples were not available as a control for comparison. In a third study, ADAMTS-13 levels did not change in a cohort of 10 patients with a continuous-flow VAD but increased in 2 patients with a pulsatile device.4 Our in vitro data suggest that ADAMTS-13 plays at least a minor role in VAD-induced von Willebrand syndrome, but is unlikely the dominant mechanism of vWF degradation. As such, the effects of mechanical circulatory support on ADAMTS-13 still remain to be determined in a large cohort of patients with paired controls. In these studies, it will be necessary to measure not only plasma ADAMTS-13 levels but also ADAMTS-13 activity, which may be enhanced by shear stress.28
Nonetheless, ADAMTS-13 cleavage of vWF likely did not account for the large accumulation of 225 kDa fragments that we observed. The 225 kDa band increased by approximately 30% and suggested that a separate and more robust mechanism was largely responsible for the degradation of vWF. The 225 kDa fragments represent vWF monomer subunits that are assembled into high-molecular-weight multimers.9 During the assembly process, monomers are linked to each other by disulfide bonds near the amino-terminus “head to head”. The resulting dimers are then linked to each other by disulfide bonds “tail to tail” and yield multimers that may exceed 20,000 kDa in size.9 Consequently, high-molecular-weight vWF multimers contain numerous disulfide bridges. It should be noted that disulfide bonds are 40% weaker than carbon–carbon covalent bonds. As a result, a mechanical weak point exists between each vWF monomer. With this in mind, our data suggest that during VAD support, vWF multimers are dismantled into individual monomer subunits. The mechanism by which this occurs is unknown. However, we speculate that high shear stress generated by the VAD may tear monomers apart at the sites of disulfide linkages. This hypothesis has not been tested but is conceptually appealing. If this proves to be the case, it will be important to determine the shear stress threshold after which mechanical demolition of vWF occurs and to establish at which point this phenomenon becomes biologically apparent and clinically relevant.
Platelets are also an important source of vWF.29,30 To determine the status of vWF from platelets, we measured the 310 kDa vWF fragment. This immature peptide, constructed from a single 225 kDa monomer and the 97 kDa vWF propeptide, is generated in megakaryocytes during vWF assembly and is stored with high-molecular-weight vWF in platelet α-granules.31 As such, the 310 kDa band served as a marker of nascent vWF release from platelets. We observed a 30% increase in the plasma 310 kDa band. This suggested that activated platelets released vWF stores into the plasma. This finding is especially clinically important. Platelets contain ultra-high-molecular-weight vWF.31 This form of vWF produces better hemostasis than vWF secreted constitutively from endothelial cells. During episodes of bleeding, theses platelet stores, which may be as much as 10% to 25% of systemic vWF,30 are available to augment primary hemostasis. Our data suggest that patients with a VAD may not have this important vWF reserve. Chronic platelet activation that leads to the release and degradation of ultra-high-molecular weight vWF may deplete a backup source of vWF that is normally available and mobilized during primary hemostasis.
Over the past 2 decades, mechanical circulatory support has evolved into a standard therapy for patients with end-stage heart failure.32–35 With nearly 2,000 patients studied, the Interagency Registry for Mechanically Assisted Circulatory Support documented a recent 2 year survival rate of 85%.35 Although long-term support is routinely achieved, better clinical outcomes are necessary before VAD support is more widely accepted.
In VAD patients, bleeding remains an important source of morbidity and mortality.32 Although mild impairment of prothrombotic pathways may represent a potential clinical benefit to prevent thromboembolism with a VAD, a question arises: How much impairment of vWF may be tolerated before clinical sequelae are observed? In humans, > 50% vWF deficiency is suggestive (not diagnostic) of von Willebrand syndrome but rarely causes episodes of bleeding.36 Our data suggest that devices that are less traumatic to blood may reduce the mechanical demolition of vWF. It remains to be determined how much vWF impairment may be tolerated in patients with a VAD.
Our findings may also have important implications for the mechanism(s) of bleeding in patients with Heyde’s syndrome. In these patients, high shear stress through a stenotic aortic valve reduces vWF and causes gastrointestinal bleeding.37 Indeed, the mechanistic pathways that predispose to bleeding episodes in patients with VADs may be identical in patients with severe aortic stenosis.
In this study, we have identified four potential therapeutic targets: plasma ADAMTS-13, platelet activation, release of platelet ADAMTS-13 and ultra-high-molecular weight vWF from platelet α-granules, and mechanical demolition of vWF each play a role in VAD-induced acquired von Willebrand syndrome. Additional studies are needed to investigate therapies that modify these biological markers.
Our results should be interpreted in light of study limitations. These experiments were performed in an in vitro system with bovine blood from healthy calves. Our avascular mock circulatory loop did not reproduce specific flow conditions of the arboreal anatomy of the circulation or contain endothelial cells, which are a major source of vWF in the body, and likely play a role in the bleeding diathesis observed in patients undergoing mechanical circulatory support. As such, these findings should be extrapolated with caution to patients with heart failure.
The artificial interior surface of our mock circulatory loop may have influenced our results. Blood contact with an artificial surface activates numerous pathways that affect the coagulation cascade, platelet function, and blood–protein interactions. However, this may simultaneously be a limitation and a strength of our study. The specific effects of an artificial surface of vWF/ADAMTS-13 interactions are largely unknown. Consequently, our results may be partially interpreted with this in mind and with potentially important clinical correlates. For example, blood contact with the artificial surface of the mock circulatory loop simulated the clinical condition in which circulating blood contacts the interior artificial surface of a VAD. Clinically, blood interactions with the artificial surface of the device may contribute to vWF activation and subsequent cleavage by ADAMTS-13. Indeed, it is possible that the artificial surface of the device also directly contributes to vWF demolition. If this is the case, blood interactions with an artificial surface provide an additional potential pathway for future investigation.
A potential limitation was the small number of loops studied (n = 4). When a sample size is small, there is a risk that the measured effects were driven by chance findings in one sample (Type I statistical error). It is encouraging that all loops demonstrated consistent and robust biological changes. Nonetheless, because of the small sample size, we may have failed to detect important effects (Type II statistical error). Multiple borderline statistically significant p values (p < 0.10) suggested that this was the case.
Shear stress enhances the proteolysis of vWF in normal plasma.28 We did not measure ADAMTS-13 activity. In future studies, it will be important to determine the effects of VAD shear stress on ADAMTS-13 activity. Similarly, defining the relative contribution of shear stress versus platelet activation versus endogenous ADAMTS-13 cleavage versus artificial-surface activation may be important in understanding (and ultimately managing) the bleeding diathesis observed in patients with a VAD. Additional studies in which plasma alone or washed platelets alone are studied in loops with and without an endothelial lining may assist in elucidating the impact of each mechanism.
A pulsatile device was not included as a test group or as a control. Evidence suggests that pulsatile VADs may also produce acquired von Willebrand syndrome, but at a lower incidence than continuous-flow VADs.1,5,7 Additional studies are needed to quantify the severity of acquired von Willebrand syndrome with a pulsatile-flow versus a continuous-flow VAD.
Limitations notwithstanding, this study contained a number of strengths that included clinically approved VADs, multiple overlapping measurements, and the identification of potential mechanistic pathways of acquired von Willebrand syndrome in patients with VADs as well as Heyde’s syndrome in patients with critical aortic stenosis.
For the first time, we have demonstrated the potential mechanisms of vWF degradation during mechanical circulatory support. Mechanical demolition of high-molecular-weight vWF likely plays a significant role. In parallel, ADAMTS-13, the vWF metalloproteinase, constitutively cleaves endogenous plasma vWF and vWF newly released from activated platelets. Further experiments are needed to define the shear stress threshold after which platelet activation and vWF degradation occur with a VAD. The effect of blood contact with an artificial surface may also contribute to vWF degradation and may be an additional topic for future investigation.
The authors acknowledge and thank Dr. Stanley D’Souza, Dr. Steven Koenig, Dr. Srinivas Sithu, Dr. Guruprasad Giridharan, Mike Sobieski, and Mickey Ising for their assistance. The authors also thank Maquet Cardiovascular for providing RotoFlow devices and controllers used in this study.
1. Geisen U, Heilmann C, Beyersdorf F, et al. Non-surgical bleeding in patients with ventricular assist devices could be explained by acquired von Willebrand disease. Eur J Cardiothorac Surg. 2008;33:679–684
2. Klovaite J, Gustafsson F, Mortensen SA, Sander K, Nielsen LB. Severely impaired von Willebrand factor-dependent platelet aggregation in patients with a continuous-flow left ventricular assist device (HeartMate II). J Am Coll Cardiol. 2009;53:2162–2167
3. Crow S, Chen D, Milano C, et al. Acquired von Willebrand syndrome in continuous-flow ventricular assist device recipients. Ann Thorac Surg. 2010;90:1263–1269 discussion 1269
4. Crow S, Milano C, Joyce L, et al. Comparative analysis of von Willebrand factor profiles in pulsatile and continuous left ventricular assist device recipients. ASAIO J. 2010;56:441–445
5. Heilmann C, Geisen U, Beyersdorf F, et al. Acquired von Willebrand syndrome in patients with ventricular assist device or total artificial heart. Thromb Haemost. 2010;103:962–967
6. Meyer AL, Malehsa D, Bara C, et al. Acquired von Willebrand syndrome in patients with an axial flow left ventricular assist device. Circ Heart Fail. 2010;3:675–681
7. Heilmann C, Geisen U, Beyersdorf F, et al. Acquired Von Willebrand syndrome is an early-onset problem in ventricular assist device patients. Eur J Cardiothorac Surg. 2011;40:1328–1333 discussion 1233
8. Heilmann C, Geisen U, Beyersdorf F, et al. Acquired von Willebrand syndrome in patients with extracorporeal life support (ECLS). Intensive Care Med. 2012;38:62–68
9. Sadler JE. Biochemistry and genetics of von Willebrand factor. Annu Rev Biochem. 1998;67:395–424
10. Markl M, Benk C, Klausmann D, et al. Three-dimensional magnetic resonance flow analysis in a ventricular assist device. J Thorac Cardiovasc Surg. 2007;134:1471–1476
11. Mizunuma H, Nakajima R. Experimental study on shear stress distributions in a centrifugal blood pump. Artif Organs. 2007;31:550–559
12. Crawley JT, de Groot R, Xiang Y, Luken BM, Lane DA. Unraveling the scissile bond: how ADAMTS13 recognizes and cleaves von Willebrand factor. Blood. 2011;118:3212–3221
13. ASTM. . Standard practice for assessment of hemolysis in continuous flow blood pumps. Designation. 2005:F1841–1897
14. Giridharan GA, Sobieski MA, Ising M, Slaughter MS, Koenig SC. Blood trauma testing for mechanical circulatory support devices. Biomed Instrum Technol. 2011;45:334–339
15. Budde U, Schneppenheim R, Plendl H, Dent J, Ruggeri ZM, Zimmerman TS. Luminographic detection of von Willebrand factor multimers in agarose gels and on nitrocellulose membranes. Thromb Haemost. 1990;63:312–315
16. Liu L, Choi H, Bernardo A, et al. Platelet-derived VWF-cleaving metalloprotease ADAMTS-13. J Thromb Haemost. 2005;3:2536–2544
17. Dent JA, Galbusera M, Ruggeri ZM. Heterogeneity of plasma von Willebrand factor multimers resulting from proteolysis of the constituent subunit. J Clin Invest. 1991;88:774–782
18. Turner NA, Nolasco L, Ruggeri ZM, Moake JL. Endothelial cell ADAMTS-13 and VWF: Production, release, and VWF string cleavage. Blood. 2009;114:5102–5111
19. Uemura M, Tatsumi K, Matsumoto M, et al. Localization of ADAMTS13 to the stellate cells of human liver. Blood. 2005;106:922–924
20. Turner N, Nolasco L, Tao Z, Dong JF, Moake J. Human endothelial cells synthesize and release ADAMTS-13. J Thromb Haemost. 2006;4:1396–1404
21. Suzuki M, Murata M, Matsubara Y, et al. Detection of von Willebrand factor-cleaving protease (ADAMTS-13) in human platelets. Biochem Biophys Res Commun. 2004;313:212–216
22. Kaplan KL, Broekman MJ, Chernoff A, Lesznik GR, Drillings M. Platelet alpha-granule proteins: studies on release and subcellular localization. Blood. 1979;53:604–618
23. Radovancevic R, Matijevic N, Bracey AW, et al. Increased leukocyte-platelet interactions during circulatory support with left ventricular assist devices. ASAIO J. 2009;55:459–464
24. Houël R, Mazoyer E, Boval B, et al. Platelet activation and aggregation profile in prolonged external ventricular support. J Thorac Cardiovasc Surg. 2004;128:197–202
25. Dewald O, Schmitz C, Diem H, et al. Platelet activation markers in patients with heart assist device. Artif Organs. 2005;29:292–299
26. Baker LC, Davis WC, Autieri J, et al. Flow cytometric assays to detect platelet activation and aggregation in device-implanted calves. J Biomed Mater Res. 1998;41:312–321
27. Snyder TA, Watach MJ, Litwak KN, Wagner WR. Platelet activation, aggregation, and life span in calves implanted with axial flow ventricular assist devices. Ann Thorac Surg. 2002;73:1933–1938
28. Tsai HM, Sussman II, Nagel RL. Shear stress enhances the proteolysis of von Willebrand factor in normal plasma. Blood. 1994;83:2171–2179
29. Howard MA, Montgomery DC, Hardisty RM. Factor-VIII-related antigen in platelets. Thromb Res. 1974;4:617–624
30. McGrath RT, McRae E, Smith OP, O’Donnell JS. Platelet von Willebrand factor–structure, function and biological importance. Br J Haematol. 2010;148:834–843
31. Sporn LA, Marder VJ, Wagner DD. Inducible secretion of large, biologically potent von Willebrand factor multimers. Cell. 1986;46:185–190
32. Hunt SA. Mechanical circulatory support: new data, old problems. Circulation. 2007;116:461–462
33. Bartoli CR, Dowling RD. The future of adult cardiac assist devices: novel systems and mechanical circulatory support strategies. Cardiol Clin. 2011;29:559–582
34. Slaughter MS, Rogers JG, Milano CA, et al.HeartMate II Investigators. Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med. 2009;361:2241–2251
35. Kirklin JK, Naftel DC, Kormos RL, et al. Third INTERMACS Annual Report: the evolution of destination therapy in the United States. J Heart Lung Transplant. 2011;30:115–123
36. Torres R, Fedoriw Y. Laboratory testing for von Willebrand disease: Toward a mechanism-based classification. Clin Lab Med. 2009;29:193–228
37. Vincentelli A, Susen S, Le Tourneau T, et al. Acquired von Willebrand syndrome in aortic stenosis. N Engl J Med. 2003;349:343–349