Pathophysiology of von Willebrand disease

INTRODUCTION — von Willebrand disease (VWD) is characterized by quantitative or qualitative abnormalities of von Willebrand factor (VWF), a crucial protein in hemostasis. Heritable VWD can be caused by pathogenic variants in the VWF gene, which lead to an impairment in the synthesis, secretion, clearance, or function of VWF. Acquired von Willebrand syndrome (AVWS) can be caused by several conditions that affect VWF levels and activity.

The pathophysiology of VWD and AVWS will be reviewed here.

Separate topic reviews discuss:

●Diagnosis of VWD – (See "Clinical presentation and diagnosis of von Willebrand disease".)

●Treatment of VWD – (See "von Willebrand disease (VWD): Treatment of major bleeding and major surgery" and "von Willebrand disease (VWD): Treatment of minor bleeding, use of DDAVP, and routine preventive care".)

●Diagnosis and treatment of AVWS – (See "Acquired von Willebrand syndrome".)

INITIAL DESCRIPTION — In 1926, Erik von Willebrand described the first patient with the disease that now bears his name. The patient and her family members, many of whom were also affected, lived on the Åland Islands in the Gulf of Bothnia (in the Baltic Sea, between Finland and Sweden).

The proband was severely affected and had multiple episodes of mucosal bleeding that led to her death at the age of 13 years. Four of her 11 siblings were also severely affected; 66 family members were studied. Von Willebrand named the disorder "hereditary pseudohemophilia" because he recognized the autosomal inheritance pattern, which contrasted with the X-linked inheritance pattern seen in hemophilia A and B [1].

After an assay for factor VIII became available in the 1950s, patients with VWD were found to have decreased levels of factor VIII [2]. Their bleeding could be treated with transfusion of plasma or partially purified preparations of factor VIII; these preparations are now known to contain VWF, which binds factor VIII in the circulation [3]. (See 'Stabilization of factor VIII' below.)

Considerable debate remained about the nature of the deficient factor in individuals with VWD, and there was uncertainty as to whether one or two proteins were involved. These questions were answered when the genes for factor VIII and VWF were cloned in the 1980s [4-8]. (See 'VWF gene' below.)

VWF GENE — The VWF gene is located on the short arm of chromosome 12; it is 178 kilobases (kb) in length and contains 52 exons [6-9]; the VWF messenger ribonucleic acid (mRNA) is approximately 9 kb. A pseudogene on chromosome 22 includes exons 23-34 of the VWF gene that correspond to regions of the authentic gene encoding domains A1, A2, and A3 [10,11]. (See 'Domain structure' below.)

Genotype-phenotype correlations and inheritance patterns — Inherited VWD has been classified into three types (table 1) [12-14]:

●Type 1 is the most common, accounting for approximately 75 percent of patients. It represents a partial quantitative deficiency of VWF. Type 1 VWD is a complex genetic trait with contributions from many genes, including VWF, genes that regulate VWF synthesis and clearance, and genes that affect ABO blood group. (See "Principles of complex trait genetics" and 'Synthesis and multimerization' below and 'Clearance and control of plasma VWF levels' below.)

Type 1 VWD is generally caused by heterozygosity for a variant that decreases VWF production, although in some individuals a causative variant cannot be found. It shows an autosomal dominant transmission pattern with incomplete penetrance and variable expressivity. Type 1C (increased VWF clearance) was added to the 2021 International Society of Thrombosis and Hemostasis (ISTH) classification of VWD [15-17]. (See 'Type 1 (reduced VWF)' below.)

●Type 2 is characterized by qualitative abnormalities of VWF. Four subtypes have been identified: 2A, 2B, 2M, and 2N. Type 2 is generally caused by heterozygosity for a variant that alters VWF function. The transmission pattern is usually an autosomal dominant, although type 2N and some individuals with other subtypes show autosomal recessive transmission. (See 'Type 2 (dysfunctional protein; 2A, 2B, 2M, 2N)' below.)

●Type 3 is very rare, characterized by total deficiency of VWF with severe bleeding manifestations. This is generally caused by homozygosity or compound heterozygosity for variants that abolish VWF production. The transmission pattern is autosomal recessive. (See 'Type 3 (undetectable VWF)' below.)

Clinical features and laboratory findings that distinguish the types of VWD are presented separately. (See "Clinical presentation and diagnosis of von Willebrand disease", section on 'Summary of VWD types' and "Clinical presentation and diagnosis of von Willebrand disease", section on 'Additional testing to characterize (classify) the type of VWD'.)

Spectrum of pathogenic variants — The VWF gene is highly polymorphic, and numerous polymorphisms (benign variants) contribute to the large variation in normal VWF levels [18,19]. Polymorphisms in non-VWF genes also contribute to the variation in VWF levels; ABO blood group determinants, which affect the rate of VWF clearance, account for approximately 25 percent of the variance in VWF levels [20]. (See 'Clearance and control of plasma VWF levels' below.)

Pathogenic variants in the VWD gene are genetic changes (point mutations, deletions, insertions) that either decrease VWF levels below the normal range or result in a dysfunctional protein with activity below the normal range [21,22]. Compilations of VWF variants can be found on the Leiden Open Variation Database (LOVD) or ClinVar.

A compilation of VWF variants with explanatory notes on >750 unique variants was published in 2017 [23].

Type 1 (reduced VWF) — Type 1 VWD is characterized by partial deficiency of VWF (VWF antigen and/or activity <30 percent, or 30 to 50 percent with a positive bleeding history). (See "Clinical presentation and diagnosis of von Willebrand disease", section on 'Diagnostic confirmation'.)

Hundreds of pathogenic variants in the VWF gene, mostly missense mutations, have been described in individuals with type 1 VWD [22-31]. Variants affecting introns have also been described [31-33]. It is likely that additional variants remain unidentified.

These variants can reduce VWF levels in several ways [23].

●Decreased production – VWF production can be decreased by a null allele, which can result from a nonsense mutation, frameshift mutation, small deletion or insertion, or splice site mutation. A deletion in the VWF promoter that alters the binding of transcription factors has also been identified [28]. Variants causing mRNA degradation have been described in several patients [34]. A variant with deletion of exons 4 and 5 causes reduced expression from the unaffected allele [35]. (See 'Synthesis and multimerization' below.)

●Decreased secretion – Some variants interfere with intracellular transport of VWF and/or cause VWF to be retained in different cellular compartments including endoplasmic reticulum, Golgi, or Weibel-Palade bodies [27,36-39]. Some of these variants affect cysteine residues and may modify disulfide bond formation, causing inhibition of VWF multimer assembly and retention of abnormal VWF intracellularly. One study identified the Y1584C variant (substitution of cysteine for tyrosine at amino acid 1584), which leads to intracellular VWF retention, in 10 of 70 (14 percent) Canadian kindreds with type 1 VWD, suggesting a founder effect [39].

●Increased clearance (type 1C) – VWF levels will be lower if clearance from the circulation is increased. Variants that increase VWF clearance often affect D3 domain (see 'Domain structure' below), although variants are dispersed throughout the gene, with higher frequencies at the amino-termini and carboxyl-termini [21,40]. An example is the Vicenza mutation, the most commonly seen variant that leads to increased clearance [41]. Some investigators refer to type 1 VWD that occurs by this mechanism as type 1C VWD. The Vicenza variant, which has larger-than-normal multimers and an abnormal multimer banding pattern, is a type 1C variant.

An assay for variants that increase clearance uses detection the VWF propeptide (VWFpp) and that is cleaved and released in equimolar concentration with mature VWF, and determination of the ratio of VWFpp to the mature VWF protein (VWFpp:VWF:Ag ratio) [15,16,29,40-45]. Variants with increased clearance show an increased VWFpp:VWF:Ag ratio, as clearance of VWFpp is unaffected while the clearance of VWF:Ag (the mature protein) is increased. A ratio >3 has been used to designate "rapid clearance" of VWF [40]. However, information about VWD half-life can also be obtained from a DDAVP trial, which is preferred for clinical management because it also provides information about the therapeutic effect of DDAVP. (See "Clinical presentation and diagnosis of von Willebrand disease", section on 'Specialized tests for VWD' and "von Willebrand disease (VWD): Treatment of minor bleeding, use of DDAVP, and routine preventive care", section on 'DDAVP trial'.)

The clinical presentation of type 1 VWD varies from mild to severe as determined by bleeding symptoms. Some individuals with type 1 disease are asymptomatic and have been discovered in studies investigating a relative with type 3 disease. In addition to only one functioning allele, the presence or absence of bleeding within type 1 kindreds may be due to interactions with variants in other genes that reduce VWF expression or activity or that affect functions of other hemostatic proteins such as specific platelet collagen adhesion receptors [46,47]. (See "Platelet biology and mechanism of anti-platelet drugs", section on 'Collagen receptors (GPIa/IIa and GPVI)'.)

Most type 1 VWD is autosomal dominant. This is the most common type of VWD, accounting for approximately 75 percent of cases.

Type 2 (dysfunctional protein; 2A, 2B, 2M, 2N) — There are several subtypes of Type 2 VWD that differ in which aspect of VWF function is affected. Because the protein abundance can be normal, these dysfunctional variants are characterized by a discrepancy between protein level (VWF:Ag) and platelet-dependent VWF activity, such as ristocetin cofactor activity (VWF:RCo, often used in the United States) or binding to mutant GPIb (VWF:GPIbM, used in many countries). The ratio of VWF:RCo to VWF:Ag can be used to identify type 2; a ratio of <0.7 is often seen. (See "Clinical presentation and diagnosis of von Willebrand disease", section on 'Derived ratios to aid classification'.)

●Type 2A (reduced large multimers) – Type 2A is characterized by a reduction in high molecular weight VWF multimers, due to reduced dimerization or multimerization. This is caused by many types of variants including missense mutations, deletions, insertions, and frameshift mutations; approximately 90 percent are missense mutations [48]. Many affect cysteine residues, indicating the importance of disulfide bonds in VWF multimerization. Many are located in the CK, D2, or D3 regions; D2 is in the propeptide region that is cleaved off to produce the mature monomer (figure 1) [49-58]. Some type 2A variants cause intracellular retention, defective biosynthesis, or abnormal storage of VWF; these are referred to as type 2A group I [48,57].

Some type 2A variants are located in the A2 region, where the ADAMTS13 cleavage site is located; these act by increasing susceptibility of VWF to proteolysis by ADAMTS13 (referred to as type 2, group II variants) [48,59,60]. Other variants may affect several mechanisms of multimer formation [61]. (See 'Multimer cleavage by ADAMTS13' below.)

Some variants in the propeptide (VWF:pp) that alter multimer formation (previously called type 2C VWD) and the A2801D variant (substitution of aspartic acid for alanine at amino acid 2801), which prevents dimer formation (previously called type 2D VWD), act in a dominant fashion [62-65].

Individuals with type 2A VWD generally present with moderate to moderately severe bleeding. Multimer analysis will show characteristic loss of large multimers. (See "Clinical presentation and diagnosis of von Willebrand disease", section on 'VWF multimer analysis'.)

Type 2A is usually transmitted as an autosomal dominant trait. However, recessive forms of type 2A also exist. Type 2A VWD accounts for approximately 10 to 20 percent of cases of VWD.

●Type 2B (increased binding to platelets) – Type 2B (type 2B) is characterized by loss of high molecular weight multimers and increased binding of VWF to platelets via the platelet receptor glycoprotein Ib (GPIb) [60,66,67]. These variants often affect the A1 domain (a 38 amino acid sequence on opposite side of the A1 domain as type 2M variants) or, infrequently, the D3 domain (figure 1) [40,60,68,69]. The A1 domain variants alter the binding site of VWF to platelet GPIb alpha to increase binding, which causes VWF and bound platelets to be sequestered from the circulation, possibly due to clearance of small platelet aggregates [70-72]. They are almost always (>95 percent) due to missense mutations [48]. Type 2B contrasts with type 2M, which is characterized by reduced binding of VWF to platelet GPIb.

•Montreal platelet syndrome – Patients previously diagnosed as having the Montreal platelet syndrome, characterized by moderately severe thrombocytopenia, mucocutaneous bleeding, giant platelets, and spontaneous in vitro platelet aggregation, have been shown to have type 2B VWD due to the VWF V1316M mutation in exon 28 [73,74].

•Type 2B New York/Malmö – This variant is characterized by a normal pattern of high molecular weight multimers at baseline, with loss of high molecular weight multimers and development of thrombocytopenia under stress situations that lead to increased release of the mutant VWF (eg, after DDAVP) [75]. This is due to mutations affecting the D3 domain [16].

Patients with type 2B VWD generally present with moderately severe bleeding and usually show decreased high molecular weight multimers, a low ratio of VWF:RCo to VWF:Ag, increased ristocetin-induced platelet aggregation, and thrombocytopenia. Thrombocytopenia may be an independent risk factor for bleeding [72,73].

There is some variation in expert opinion regarding whether individuals with type 2B should have a trial of DDAVP, due to the possible risk of exacerbating thrombocytopenia. (See "von Willebrand disease (VWD): Treatment of minor bleeding, use of DDAVP, and routine preventive care", section on 'DDAVP trial'.)

Type 2B is transmitted in an autosomal dominant pattern. It accounts for approximately 5 percent of cases of VWD.

Phenotypic abnormalities similar to VWD type 2B can also be produced by variants in the gene that encodes the GPIb receptor that cause it to bind more avidly to normal VWF; this disorder is called platelet-type or pseudo VWD [76,77]. This is a platelet disease, not an aberration in VWF. (See "Inherited platelet function disorders (IPFDs)", section on 'Specific disorders'.)

●Type 2M (decreased binding to platelets) – Type 2M VWD is characterized by normal multimers and reduced binding of VWF to platelet GPIb, or, less commonly, decreased binding of VWF to collagen [78]. Type 2M variants cause decreased platelet adhesion to sites of injury [60,78]. (See 'Bridging between platelets and vascular subendothelium' below.)

Most type 2M variants are located in the A1 domain (the opposite end of the domain from those that cause type 2B VWD) [60]. Some individuals with decreased binding to collagen are also classified as having type 2M VWD; these variants are located in the A1 and A3 domains [23,79].

Individuals with type 2M typically have significant bleeding symptoms [78,80,81]. Bleeding symptoms are related to decreased interactions of platelets with the vascular endothelium. (See 'Bridging between platelets and vascular subendothelium' below.)

Type 2M is transmitted in an autosomal dominant pattern. It is less common than types 2A or 2B.

●Type 2N (decreased binding to factor VIII) – Type 2N VWD (N stands for Normandy, where one of the first patients was described) is characterized by decreased binding of VWF to factor VIII.

Most type 2N variants are distributed throughout a 91 amino acid region of the amino-terminus of the mature VWF monomer including the D' and D3 domains (figure 1), which include the factor VIII binding site [59,82-88]. Binding of VWF to factor VIII protects circulating factor VIII from proteolysis; Some patients with type 2N also have impaired VWF multimerization [89].

Individuals with type 2N have low factor VIII (usually 5 to 15 percent of normal); their bleeding phenotype is at least partly due to factor VIII deficiency, resulting in joint and soft tissue bleeding. There can be normal VWF antigen (VWF:Ag) and ristocetin cofactor activity (VWF:RCo) and deficient factor VIII; some individuals are misdiagnosed as having hemophilia A. However, in type 2N the transmission pattern is autosomal recessive, with females affected at the same frequency and to the same degree as males, whereas hemophilia A is X-linked [85,86]. (See "Clinical manifestations and diagnosis of hemophilia", section on 'Differential diagnosis'.)

Type 2N inheritance is autosomal recessive trait; homozygosity is generally required for type 2N to manifest. Type 2N can also occur with compound heterozygosity for a type 1 VWD variant on one allele combined with a type 2N variant on the other allele. In a heterozygous carrier for a type 2N variant, the unaffected allele usually produces enough functional VWF to bind and stabilize factor VIII and prevent factor VIII deficiency. Type 2N may also be caused by compound heterozygosity for a type 2N variant and a pathogenic variant in the F8 gene (hemophilia A carrier) or compound heterozygosity for a type 2N variant and a VWF variant that decreases VWF levels enough to eliminate the factor VIII stabilizing function [86,90].

Type 3 (undetectable VWF) — Type 3 VWD is characterized by a marked decrease or absence of detectable VWF [12-14]. This is caused by homozygosity or compound heterozygosity for variants that substantively reduce or abolish VWF production. These variants include nonsense, missense, and splice site mutations and often partial or complete deletions (most common). Nondeletional variants can result in the loss of VWF mRNA expression or impaired secretion and intracellular retention [21,91-95].

Individuals with compound heterozygous variants often have a VWF protein that is not secreted or undergoes rapid clearance [96]. A common variant involves a frameshift mutation in exon 18; this was demonstrated to be the variant present in the initial kindred described by von Willebrand [97,98]. (See 'Initial description' above.)

Type 3 patients have severe bleeding due to decreased or absent VWF-related platelet function (mucocutaneous bleeding) as well as significantly decreased factor VIII function (joint and soft tissue bleeding) (see 'Stabilization of factor VIII' below). Affected individuals may be initially misdiagnosed as having hemophilia A. (See "Clinical manifestations and diagnosis of hemophilia", section on 'Differential diagnosis'.)

Type 3 VWD is transmitted in an autosomal recessive pattern. Children of an individual with type 3 VWD who only inherit one of the variants may have type 1 disease. Type 3 VWD is rare (estimated prevalence 1 in 1 million) [99].

VWF PROTEIN

Domain structure — VWF is a large multimeric glycoprotein [100]. It is composed of a series of homologous domains, with most regions repeated three to six times. These include (in order) three D domains, three A domains, a fourth D domain, and six C domains [101]. The carboxyl terminus of the protein contains a carboxyl-terminal cysteine knot (CTCK) [102].

Each domain has different functional properties (figure 1):

●D' and the contiguous part of D3 contains the binding site for factor VIII [103-105]

●A1 contains binding sites for platelet GPIb, heparin, collagen, and ristocetin [60,66,106]

●A2 contains the cleavage site for ADAMTS13 [60,107]

●A3 contains the primary binding site for collagen [60]

●C4 contains a binding site for the platelet integrin alphaIIb-beta3 (GPIIbIIIa) [102]

●CTCK contains the dimerization interface [102]

Each D domain (referred to as a "D assembly") is subdivided into several modules based on their electron microscopic appearance; they are named VWD C8 (cysteine 8), TIL (trypsin-inhibitor-like), E, and D4N modules. The C domains have also been divided into six C regions based on structural studies and by homology with the known structure of other proteins. Platelet binding via the integrin alphaIIbbeta3 (GPIIbIIIa) is assigned to C4. The CTCK crystal structure has revealed a highly reinforced dimerization interface; this extremely strong binding surface explains how large VWF multimers can assemble and resist the exceptionally high shear forces from blood flow at sites of clot formation [102,108].

Each mature VWF monomer contains binding sites for collagen, platelet integrins, factor VIII, and other VWF monomers. Since the larger VWF multimers contain repeated binding sites, they are well suited to act as bridging molecules between platelets and the vascular subendothelium: The repeated binding sites allow multiple interactions with both platelet receptors and subendothelial structures at sites of vessel injury. (See 'Bridging between platelets and vascular subendothelium' below.)

Synthesis and multimerization — VWF is synthesized in endothelial cells and megakaryocytes (platelet precursor cells) [109,110]. The primary translation product contains a signal peptide of 22 amino acids followed by a propeptide of 741 residues, also known as VWF propeptide [111,112]. This polypeptide undergoes considerable post-translational processing. (See "Basic genetics concepts: DNA regulation and gene expression", section on 'Translation'.)

Post-translational modifications of VWF include [110,113]:

●Cleavage of the propeptide

●Dimerization via disulfide bonds, and multimerization to very large forms

●Targeting of larger multimers to storage granules (Weibel-Palade bodies in endothelial cells, alpha granules in platelets)

Several of these processes occur simultaneously.

Dimerization occurs early during processing of the protein by formation of interchain disulfide bonds between the C-termini of the pro-VWF within the endoplasmic reticulum [114]. N-linked carbohydrate residues are also added in the endoplasmic reticulum and play a critical role in the formation of dimers and the exit from the endoplasmic reticulum to the Golgi [113]. Further glycosylation and sulfation of the pro-VWF dimers occurs in the Golgi [115]

Dimers are assembled into multimers by disulfide bond formation between D3 regions of the N-termini of the pro-dimers. Multimer formation requires the propeptide, a low pH environment, and calcium ions [116-118]. Cleavage of the propeptide sequence by furin occurs at approximately the same time as multimerization, but the propeptide remains to interact with the mature VWF protein to support multimer formation [119]. Multimer sizes are a spectrum, with various numbers of repeating units ranging from tetramers to multimers of >40 to 80 subunits and a size as large as >20 million daltons (>20,000 kilodaltons [kd]) [120,121]. The multimers form flexible 2 nm thick strands in a tangled coil configuration [122]. High shear stress converts the coiled configuration to a linear structure that is more hemostatically active. (See 'Enhanced activity under high shear stress' below.)

Tubule formation, which allows compaction of VWF, occurs during pro-VWF multimerization, and tubule formation drives formation of Weibel-Palade bodies, where VWF is stored as large tubular multimers [123,124]. (See 'Storage and release' below.)

Synthesis of VWF is regulated in part by hormones. As an example, VWF production in endothelial cells is increased by both estrogen and thyroid hormone [125,126]. Increased estrogen levels during pregnancy cause VWF levels to rise during the second and third trimesters [127]; VWF levels decline to nonpregnant values within hours to days after delivery. (See "von Willebrand disease (VWD): Gynecologic and obstetric considerations", section on 'Pregnancy' and "von Willebrand disease (VWD): Gynecologic and obstetric considerations", section on 'Delivery and postpartum care'.)

Storage and release — Large VWF multimers are stored in secretory granules in endothelial cells (Weibel-Palade bodies) and platelets (alpha granules) [128-130].

Weibel-Palade bodies share several characteristics with lysosomal-related organelles, cell-specific specialized organelles that include platelet dense granules, and melanosomes [131]. Secretion and release of VWF from endothelial cells is polarized (the release pathway is directed to a particular extracellular compartment). A constitutive pathway releases smaller VWF multimers via the basolateral side to the subendothelial matrix of the endothelial cells, while the basal release and regulated release pathways secrete larger VWF multimers via the apical side into the blood vessel lumen [132].

Mechanisms of VWF storage and release from Weibel-Palade bodies are complex and depend on many proteins and signaling molecules [131,133]. Regulation of storage and release differs for different size multimers [132,134]. SNARE proteins are important in the fusion of secretory vesicles with the plasma membrane.

●Smaller VWF multimers are constitutively secreted from endothelial cells and megakaryocytes. Abluminal secretion from endothelial cells occurs across the basolateral cell membrane, directed to the subendothelial extracellular matrix that contributes to the binding of platelets to the subendothelium of injured blood vessels.

●Larger, more functional multimers (including "ultralarge" multimers) are targeted to cytoplasmic storage granules for storage (Weibel-Palade bodies in endothelial cells and alpha granules in platelets) [129]. These storage granules contain a variety of other hemostatic factors (P-selectin, tissue plasminogen activator, CD63, osteoprotegerin, angiopoietin-2, interleukin 8, and others [119,135-138]. Coalescence of the Weibel-Palade bodies to form pods occurs before VWF secretion [139]. (See "Megakaryocyte biology and platelet production", section on 'Granules'.)  

●Secretion of the larger, more functional multimers is an active process that occurs by basal (unstimulated) and stimulated secretion pathways.

•Basal (unstimulated) secretion of VWF occurs from the basolateral cell membrane of endothelial cells and delivers VWF into the vessel lumen. This process involving endothelial cell VWF is largely responsible for circulating VWF [130,132].

•Stimulated secretion occurs in response to physiologic agonists, resulting in release of VWF into the vessel lumen, via the apical surface of endothelial cells. Luminal secretion of ultralarge and large multimers enhances VWF binding to platelets (via GPIb) and may also increase the local concentration of factor VIII to enhance clot formation at the site of injury [113]. Agonists include:

-Alpha-adrenergic agonists such as epinephrine

-Thrombin

-Fibrin

-Histamine

-The vasopressin analog DDAVP (used in treatment of VWD) (see "von Willebrand disease (VWD): Treatment of minor bleeding, use of DDAVP, and routine preventive care", section on 'DDAVP')

During activated release of VWF from Weibel-Palade bodies, the pH rises from the more acidic tubule environment to neutral, and this may allow the tubules to unfold in an orderly manner without tangles [119,140]. The pH change may also allow the cleavage of the prosequence by furin, which occurs at approximately the same time [141].

Release of VWF from endothelial cells under shear stress produces long strands of VWF that remain anchored to the endothelial cells by P-selectin [142]. These long strands are assembled under shear using a number of sites in the VWF protein, including unfolded A2 regions that are exposed by shear forces (figure 1) [143]. This allows binding of additional VWF molecules, platelets, and ADAMTS13, the primary enzyme that breaks down VWF multimers in the circulation [142,144,145]. Ristocetin can mimic the changes induced by shear stress-induced self-association [146]. (See 'Enhanced activity under high shear stress' below and 'Multimer cleavage by ADAMTS13' below.)

Shear stress-induced self-association of VWF is influenced by several molecules including high-density lipoprotein (HDL) and low-density lipoprotein (LDL):

●HDL opposes self-association, decreasing VWF strand size and reducing platelet adhesion [147].

●LDL enhances self-association, increasing the length and thickness of VWF fibers produced under shear [148].

●Thrombospondin-1 decreases self-association by acting as a disulfide reductase [149].

●ADAMTS13 proteolytically cleaves VWF. (See 'Multimer cleavage by ADAMTS13' below.)

Clearance and control of plasma VWF levels — Circulating VWF concentrations have a wide reference range. The concentration of circulating (plasma) VWF protein is 500 to 1000 mcg/dL, defined as 50 to 150 percent activity. VWF levels are higher than baseline for approximately six months after birth, making it impossible to exclude VWD in newborns. Platelets contain approximately 15 percent of the quantity of VWF present in plasma [150]. (See "Laboratory test reference ranges in adults", section on 'von Willebrand factor antigen, plasma'.)

The typical half-life of circulating VWF is approximately 8 to 12 hours; this is determined by the balance between the rates of production (see 'Synthesis and multimerization' above) and rates of clearance from the circulation.

The rate of clearance is in part determined by glycosylation of the VWF protein, which is regulated by fucosyltransferases, ABO blood group glycosyltransferases, and beta-galactoside alpha 2,3 sialyltransferase 4.

VWF levels increase with aging [151].

Adults with blood type O have approximately 25 to 30 percent lower VWF levels than individuals with types A, B, or AB. This is due to a portion of the carbohydrate present on VWF that is similar to the blood groups and influences the clearance of VWF [20,42,152-158]. These physiologic differences may not be present during the first year of life, perhaps related to the slow postnatal development of the blood group system [159].

Several different receptors participate in the clearance of VWF from the circulation [160].

●One mechanism involves the loss of terminal sialic acid with aging of the protein (or due to variants in the VWF gene), exposing a penultimate galactose residue [161]. Receptors for this mechanism include:

•The Ashwell-Morrell receptor (which clears hyposialylated N-linked galactose residues) [162,163].

•The macrophage galactose-type lectin receptor (which clears O-linked and N-linked galactose residues) [160,163].

●The LRP1 receptor on macrophages clears VWF via exposure of shear stress-induced interactive sites in VWF [160]. N-linked sites within the VWF A2 domain that are exposed by the unfolding of VWF are able to bind to the LRP1 receptor, which mediates clearance [162].

●CLEC4M receptors, members of the C-type lectin domain family 4, which interact with N-linked glycans, bind to and internalize VWF, decreasing VWF levels [164].

Other genes modifying clearance of VWF include STAB2 (encodes a lectin-like scavenger receptor that is associated with the clearance of VWF and factor VIII), STXBP5, SCARA5, and UFM1 [96,164-168].

Causes of reduced VWF in acquired VWS — Acquired von Willebrand syndrome (AVWS) can be caused by several different mechanisms that affect VWF production, multimerization, or clearance. (See 'Synthesis and multimerization' above and 'Clearance and control of plasma VWF levels' above.)

Underlying disorders are summarized in the table (table 2) and discussed separately. (See "Acquired von Willebrand syndrome", section on 'Pathophysiology'.)

VWF FUNCTIONS — VWF performs three critical functions in hemostasis as well as playing a role in angiogenesis [169,170].

●Two functions are in primary hemostasis. At sites of injury, VWF acts as a bridging molecule between platelets and vascular subendothelium (platelet adhesion); at high shear stress, VWF promotes platelet aggregation. (See 'Bridging between platelets and vascular subendothelium' below and 'Enhanced activity under high shear stress' below.)

●A third function is in fibrin formation. VWF acts as a carrier for factor VIII in the circulation, increasing the factor VIII half-life fivefold and maintaining normal factor VIII levels, which is essential in the clotting cascade that leads to production of a fibrin clot. (See 'Stabilization of factor VIII' below.)

●VWF plays a negative regulatory role in angiogenesis [171]. Gastrointestinal angiodysplasia in von Willegrand disease (VWD) and acquired von Willebrand syndrome (AVWS) are discussed separately. (See "Clinical presentation and diagnosis of von Willebrand disease", section on 'Gastrointestinal bleeding and angiodysplasia' and "Acquired von Willebrand syndrome", section on 'Consequences of reduced VWF'.)

Bridging between platelets and vascular subendothelium — VWF functions in primary hemostasis by forming an adhesive bridge between platelets and vascular subendothelial structures as well as between adjacent platelets at sites of vascular injury [113,172]. The binding of VWF to platelets and subendothelial components is critical for normal platelet adhesion.

High molecular weight VWF multimers are the most active forms of VWF, providing multiple binding sites that can interact with platelets and subendothelial structures [113,130,172]. (See 'Synthesis and multimerization' above.)

The components in the subendothelial connective tissue to which VWF binds are not completely defined. VWF binds to multiple types of collagen (types I, II, III, IV, V, and VI) [173,174]. Type VI collagen, which binds within the A1 domain, appears to be especially important, both in initial platelet binding and in areas of high shear stress [175-177]. (See 'Enhanced activity under high shear stress' below.)

Binding of collagen types I and III may also be important [178]. Free thiols in the C domain of VWF are a requirement for VWF binding to collagen [179]. Collagen binding appears to induce a conformational change within the factor VIII-binding motif of VWF that lowers the affinity of VWF for factor VIII, perhaps releasing factor VIII locally to aid in the formation of the fibrin clot [180].

Enhanced activity under high shear stress — The shear stress that occurs in small arterioles and atherosclerotic arteries (shear rates >1000/second) causes a conformational change in VWF from a globular form to a linear structure in which the A1 domains (figure 1) are exposed and unhindered by self-association [143,181].

The mechanism involves an autoinhibitory module composed of discontinuous sequences located at each end of the A1 domain in VWF, which masks the rest of A1 and prevents the binding of VWF to platelet GPIb. Release of this autoinhibition and exposure of the A1 domain allows platelet GPIb to bind [182]. Ristocetin appears to activate VWF in a similar manner, by exposing A1 for platelet GPIb binding [182].

The linear conformation appears to be the functional form for binding to the platelet receptor glycoprotein Ib (GPIb), which engages the GPIb-IX-V complex on the platelet surface and mediates platelet adhesion and aggregation under high shear [60,66,106,183-187]. This interaction is counteracted by circulating beta2 GPI, which binds VWF and inhibits the binding of activated VWF to GPIb, decreasing platelet adhesion. In some autoimmune settings, antibodies to beta2 GPI remove this inhibition of VWF binding, which may contribute to thrombosis [188,189]. (See "Clinical manifestations of antiphospholipid syndrome".)

A second platelet receptor for VWF, integrin alphaIIb-beta3 (GPIIbIIIa), does not bind VWF unless the platelets are activated [185,187]. With platelet activation, alphaIIb-beta3 undergoes a conformational change and becomes accessible on the platelet surface. Platelets can be activated by GPIb binding to immobilized VWF strands, which transmits a signal within the platelet [190].

The platelet alphaIIb-beta3 interaction with VWF appears to contribute to the final, irreversible binding of platelets to the subendothelium after VWF has bound to GPIb [186]. The alphaIIb-beta3 interaction with VWF may also contribute to platelet aggregation, especially under high shear conditions (under low shear conditions, platelet aggregation is mediated primarily by fibrinogen binding to alphaIIb-beta3) [187,191-193].

Vimentin, a structural protein found in plasma and on the surface of platelets, has also been shown to enhance the binding of VWF to the platelet surface under high shear stress [194].

Stabilization of factor VIII — Factor VIII is an important protein cofactor in the pathway of thrombin generation. (See "Overview of hemostasis".)

VWF normally binds to and protects factor VIII from proteolytic inactivation by activated protein C and its cofactor protein S [104,195,196]. VWF slows the inactivation of factor VIII by activated protein C by 10-fold to 20-fold [195,196]. In the absence of VWF, the half-life of factor VIII is only approximately two hours [169,195,196]. Since the concentration of VWF is much higher than factor VIII in normal plasma, considerably less than one-half of the normal VWF is required for factor VIII stabilization.

VWF domains D' to D3 have been shown to be sufficient to stabilize endogenous factor VIII (figure 2); this has led to interest in their development as an innovative treatment for hemophilia A [197]. The same product should also be sufficient for treatment of VWD type 2N, although this has not been tested. (See "Hemophilia A and B: Routine management including prophylaxis", section on 'Efanesoctocog alfa (factor VIII-VWF fusion)'.)

VWF can bind factor VIII only in the inactive form (when factor VIII has not been cleaved by thrombin) [198]. After activation to factor VIIIa by thrombin cleavage, the VWF dissociates (figure 2) [195].

Multimer cleavage by ADAMTS13 — Ultralarge VWF multimers can be prothrombotic by promoting platelet-platelet and platelet-endothelial interactions [199]. In vitro evidence indicates that the inflammatory cytokine interleukin (IL)-8 may stimulate the release of ultralarge VWF and IL-6 may inhibit cleavage of the ultralarge multimers, suggesting a connection between inflammation and thrombosis [200]. (See 'Thrombosis associated with high VWF levels' below.)

ADAMTS13 is a member of the "a disintegrin and metalloproteinase with thrombospondin type 1 motif" (ADAMTS) family of metalloproteinases. It is synthesized in hepatic stellate cells, megakaryocytes, platelets, and endothelial cells [201-204]. These cells release ADAMTS13 into the circulation where it can bind large VWF multimers that are unfolded by shear stress, particularly when VWF is bound to platelets.

ADAMTS13 cleaves ultralarge VWF multimers at the endothelial cell and platelet surface, converting them to the normal multimer size distribution. It also cleaves smaller (normal size) VWF multimers, leading to a spectrum of multimer sizes. ADAMTS13 contains an active site that cleaves VWF in the A2 domain (figure 1); the fluid shear stress acting on VWF tethered to the endothelium opens the protein configuration and exposes the bond between amino acids tyrosine 1605 and methionine 1606 (Y1605-M1606) for cleavage [107,145,205-211]. This in turn decrease platelet adhesion and renders the VWF fragments subject to clearance [212].

Factor VIII plays a role in accelerating the proteolysis of these ultralarge multimers by ADAMTS13, and factor H, also released by endothelial cells, inhibits cleavage [213-215].

Severe deficiency of ADAMTS13 causes microvascular thrombosis in thrombotic thrombocytopenic purpura (TTP). In animal models of ADAMTS13 inhibition, co-inhibition of the VWF-platelet GPIb interaction was shown to abrogate thrombus formation [216,217]. (See "Pathophysiology of TTP and other primary thrombotic microangiopathies (TMAs)", section on 'Deficient ADAMTS13 activity'.)

Another possible mechanism for processing of the unusually large multimers involves thrombospondin-1, which acts as a protein disulfide reductase and has been shown to reversibly generate new thiol groups in VWF and reduce the size of the multimers [218]. The relative importance of thrombospondin-1 in the determination of the size of plasma VWF multimers is not fully understood. Thrombospondin-1 competes with ADAMTS13 for interaction with the A3 domain of VWF, slowing the proteolysis of VWF by ADAMTS13 [219].

MECHANISM OF CLINICAL FINDINGS

Bleeding — Bleeding can be multifactorial, due to disruption of VWF functions, leading to:

●Decreased association of platelets with sites of vascular injury

●Decreased aggregation of platelets in regions of high shear stress from rapid blood flow

●Reduced factor VIII activity

In most individuals with type 1 VWD, mucocutaneous bleeding predominates, due primarily to decreased platelet adhesion and aggregation (table 3). (See 'Type 1 (reduced VWF)' above.)

Severely reduced factor VIII activity and the associated joint and soft tissue bleeding generally only occurs in the less common type 2N disease. (See 'Type 2 (dysfunctional protein; 2A, 2B, 2M, 2N)' above.)

In type 3 disease, absence of VWF eliminates all of these functions and causes both types of bleeding. (See 'Type 3 (undetectable VWF)' above.)

Gastrointestinal angiodysplasia — Angiodysplasia of the gastrointestinal tract has been reported in individuals with hereditary VWD or acquired von Willebrand syndrome (AVWS). However, it is not clear whether this represents a causal association or an increased likelihood of identifying these lesions due to the underlying bleeding disorder. The combination of aortic stenosis and gastrointestinal angiodysplasia has been referred to as Heyde syndrome; this may also represent increased identification due to AVWS related to aortic valve disease. (See "Angiodysplasia of the gastrointestinal tract", section on 'Conditions associated with angiodysplasia' and "Clinical manifestations and diagnosis of aortic stenosis in adults", section on 'Bleeding tendency'.)

Thrombosis associated with high VWF levels — The ultralarge VWF multimers released from platelets and endothelial cells are even more hemostatically active than the largest forms of VWF in normal plasma, and they are considered prothrombotic, binding spontaneously to platelets in the circulation [220].

The most striking example is in thrombotic thrombocytopenic purpura (TTP), in which deficiency of ADAMTS13 activity secondary to autoantibodies allows ultralarge VWF multimers to accumulate in small vessels and cause platelet microthrombi to form. (See 'Multimer cleavage by ADAMTS13' above and "Pathophysiology of TTP and other primary thrombotic microangiopathies (TMAs)", section on 'Consequences of ADAMTS13 deficiency'.)

Elevated VWF levels are also seen in individuals with deep vein thrombosis (DVT) [221,222]. (See "Overview of the causes of venous thrombosis", section on 'Other plasma components'.)

Increased VWF levels have been documented in individuals with severe coronavirus disease 2019 (COVID-19), which is a highly prothrombotic state. (See "COVID-19: Hypercoagulability", section on 'Coagulation abnormalities'.)

Studies in chimeric pigs in which plasma and platelet VWF are dissociated suggest that plasma but not platelet VWF is critical for the development of arterial thrombosis [223]. Thrombosis occurred as expected in pigs with normal plasma VWF and no platelet VWF, while no thrombosis occurred in pigs with absent plasma VWF but normal platelet VWF.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: von Willebrand disease".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: von Willebrand disease (The Basics)")

●Beyond the Basics topic (see "Patient education: von Willebrand disease (Beyond the Basics)")

PATIENT PERSPECTIVE TOPIC — Patient perspectives are provided for selected disorders to help clinicians better understand the patient experience and patient concerns. These narratives may offer insights into patient values and preferences not included in other UpToDate topics. (See "Patient perspective: von Willebrand disease".)

SUMMARY

●VWF gene – Inherited von Willebrand disease (VWD) was first described in the 1920s; the VWF gene was only identified in the 1980s. (See 'Initial description' above and 'VWF gene' above.)

VWD is classified into three types (table 1). VWF variants are collected in the Leiden Open Variation Database (LOVD) or ClinVar.

•Type 1 – Decreased VWF levels (<30 percent; 30 to 50 percent in individuals with a bleeding phenotype). Numerous pathogenic variants distributed across the gene that alter VWF production, trafficking, stability, or clearance (Type 1C). Autosomal dominant inheritance. Most common type. (See 'Type 1 (reduced VWF)' above.)

•Type 2 – Four subtypes, each with different types of disease variants and different alterations in protein function. Type 2A has reduced large multimers. Type 2B has increased platelet binding and thrombocytopenia. Type 2M has decreased platelet binding. Type 2N lacks factor VIII binding. 2A, 2B, and 2M are autosomal dominant; 2N is autosomal recessive. (See 'Type 2 (dysfunctional protein; 2A, 2B, 2M, 2N)' above.)

•Type 3 – Absent VWF. Autosomal recessive inheritance. Extremely rare. (See 'Type 3 (undetectable VWF)' above.)

●VWF regulation – VWF is a large multimeric glycoprotein (figure 1) synthesized in endothelial cells and megakaryocytes. Dimerization occurs via interchain disulfide bonds. Multimers form a spectrum of sizes from tetramers to >40 to 80 subunits. They are stored in Weibel-Palade bodies in endothelial cells and alpha granules in platelets. Small multimers are constitutively released; larger multimers undergo basal and stimulated secretion in response to various agonists. The plasma half-life is 8 to 12 hours, with a wide reference range (50 to 150 percent). Adults with blood type O have levels approximately 25 to 30 percent lower than other ABO types. Certain disease states decrease VWF production or increase clearance, causing acquired von Willebrand syndrome. (See 'VWF protein' above.)

●VWF functions – At sites of injury, VWF bridges between platelets and vascular subendothelium (platelet adhesion). At high shear stress, VWF elongates and promotes platelet aggregation. VWF also binds to factor VIII and dramatically prolongs its half-life (figure 2). (See 'VWF functions' above.)

●Pathophysiology – Mucocutaneous bleeding in type 1 and most type 2 VWD is due to loss of platelet adhesion and aggregation at sites of vascular injury (table 3). In type 2N VWD, factor VIII is not stabilized, and factor VIII deficiency leads to joint and soft tissue bleeding. In type 3 VWD, all VWF functions are absent; both types of bleeding are seen. Gastric angiodysplasia is often reported; it is not clear whether this is more common in VWD or more likely to be identified due to the bleeding phenotype. High VWF levels, especially of high molecular weight multimers, can cause thrombosis. (See 'Mechanism of clinical findings' above.)

●Diagnosis and treatment – (See "Clinical presentation and diagnosis of von Willebrand disease" and "von Willebrand disease (VWD): Treatment of major bleeding and major surgery" and "von Willebrand disease (VWD): Treatment of minor bleeding, use of DDAVP, and routine preventive care" and "Acquired von Willebrand syndrome".)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Margaret E Rick, MD, who contributed to earlier versions of this topic review.