The diagnosis of von Willebrand disease: a guideline from the UK Haemophilia Centre Doctors’ Organization


Prof K. J. Pasi, Chairman of the UKHCDO von Willebrand's Disease Working Party, Department of Haematology, Barts and The London, Queen Mary's School of Medicine and Dentistry, Turner Street, London E1 2AD, UK.
Tel.: +44 20 7377 7000 (x3152); fax: +44 709 222 7649;


Summary.  von Willebrand disease (VWD) is the commonest inherited bleeding disorder. However, despite an increasing understanding of the pathophysiology of VWD, the diagnosis of VWD is frequently difficult because of uncertainty regarding the relationship between laboratory assays and function in vivo. The objective of this guideline is to provide contemporary advice on a rational approach to the diagnosis of VWD. This is the second edition of this UK Haemophilia Centre Doctors’ Organisation (UKHCDO) guideline and supersedes the previous edition which was published in 1997.


von Willebrand factor (VWF) is a large plasma glycoprotein (GP) that exists as a series of multimers of molecular weight 800–20 000 kDa. VWF has two main functions: as a carrier protein for coagulation factor VIII (FVIII) and as an adhesive protein involved in vessel wall–platelet interaction. Its function as an adhesive protein is most important in situations of high shear stress. Inherited defects in VWF may, therefore, cause bleeding by impairing either platelet adhesion or fibrin clot formation.

von Willebrand disease (VWD) is defined as a deficiency of VWF function causing impaired haemostasis. VWD is the commonest of the inherited bleeding disorders. However, uncertainty regarding the relationship between laboratory assays and function in vivo means that diagnosis is frequently difficult. Despite an increasing understanding of the pathophysiology of VWD, the investigation and diagnosis of patients with possible VWD vary widely.

Aims of the guideline

The aim of the guideline is to provide contemporary advice on a rational approach to the diagnosis of VWD. Guidelines on the management of VWD are given in an accompanying paper (Pasi et al., p 218–231). Target users of the guideline include clinical and laboratory staff involved in the diagnosis and care of patients and families with VWD. Users of the guideline should be aware that individual professional judgement is not suspended on the basis of advice contained herein. This guideline will be reviewed annually by the UK Haemophilia Centre Doctors’ Organization (UKHCDO) and updated when required.


The current guideline document was produced by the UKHCDO VWD Working Party. This guideline writing group included UK-based medical and scientific experts in the diagnosis and treatment of inherited bleeding disorders, including VWD. Relevant scientific papers were identified by systematic review from Medline using the index terms VWF, VWD, FVIII, treatment and management. Further publications were obtained from the references cited and from reviews known to the members of the working party. Reports in the literature were used to establish the evidence upon which recommendations have been made. Recommendations are based on reports with the highest level of evidence available and the graded recommendations presented in these guidelines are in accordance with the US Agency for Health Care Policy and Research (see Appendix).

It has been clear in the production of this guideline that there are a number of areas where currently there is no absolute recommendation that could be given from the evidence available: in these situations a consensus within the Working Party has been established to attempt to provide guidance. Drafts were circulated to the Advisory Committee of the UKHCDO for consultation and editorial comment.

Synthesis and function of VWF

VWF is synthesized in vascular endothelial cells and bone marrow megakaryocytes. The 2813 amino acid primary translation product undergoes processing in the endoplasmic reticulum to remove a signal peptide and to form pro-VWF dimers. Subsequently, during passage through the Golgi apparatus, the VWF propeptide (previously also known as VWF AgII) mediates the assembly of the 500 kDa pro-VWF dimers into a range of multimeric structures with molecular weights of up to 20 000 kDa, by means of N-terminal disulphide bond formation [1]. Extensive post-translational modification of VWF occurs, including glycosylation, sulphation and cleavage of the propeptide. The mature VWF protein is secreted directly into plasma or the subendothelial matrix, or is stored in endothelial cell Weibel–Palade bodies and platelet α-granules.

The largest multimers, recognized to be the most haemostatically effective, are present in endothelial cell Weibel–Palade bodies and platelet α-granules, but are detected only transiently in normal blood following their acute release in response to vascular injury or stimulation. This partitioning effect may serve to restrict the potentially thrombogenic highest molecular weight VWF multimers to sites of tissue damage, where they are required to provide maximal haemostatic effect. High-molecular weight (HMW) multimers released into plasma from endothelial cells are broken down by the action of thrombospondin and a specific protease called ADAMTS 13 [2,3].

VWF has an essential function in primary haemostasis to facilitate platelet adhesion and aggregation in the presence of high fluid shear stress forces, such as occur in the microvasculature. HMW VWF multimers are of particular importance for the mediation of interactions between subendothelium exposed at sites of vascular injury and the platelet membrane GPIb and IIb–IIIa. The exact nature of the interaction of VWF with exposed subendothelium is uncertain, however, a binding site for fibrillar collagens is located within the VWF A3 domain. The interaction of VWF with subendothelium and the effect of shear forces is thought to induce a conformational change in VWF, exposing the platelet GpIb-binding site within the VWF A1 domain. The initial reaction of VWF with platelet GPIb is rapid but of low affinity. The resultant slowing of platelet travel along the vessel wall enables the less rapid but higher affinity binding of VWF to GPIIb–IIIa on the surface of activated platelets to take place, via the RGD (Arg-Gly-Asp) sequence located within the VWF C1 domain [4]. This latter binding interaction induces platelet plug formation, and is particularly important under conditions of high shear stress.

VWF also has an essential role to stabilize FVIII in the circulation. The VWF-binding site for FVIII is located in the D′/D3 domains, within the amino-terminal 272 amino acid residues of the mature VWF subunit. This high affinity-binding site has a Kd of 0.2–0.4 nm [5]. In the presence of VWF the half-life of plasma FVIII is c. 8–12 h and in its absence has been estimated to be <1 h [6]. It has been shown to be <3 h in severe VWD [7].

Factors affecting VWF level

The plasma level of VWF in normal individuals varies over a sixfold range; from c. 40–240 IU dL−1 [8]. Several genetic and environmental factors influence the plasma VWF concentration. It is likely that the presence of VWD and its severity are frequently determined by a combination of these factors, which may also obscure the diagnosis in some situations.

Genetic defects

The VWF gene

The human VWF gene is located on the short arm of chromosome 12 at 12p13.2. It comprises 52 exons spanning 178 kbp and gives rise to a 9 kbp mRNA. Analysis of the VWF gene is complicated by the presence of an unprocessed partial pseudogene located at chromosome 22q11.2. The pseudogene corresponds to exons 23–34 of the VWF gene and exhibits 97% sequence homology [9].

Knowledge of regulatory mechanisms at the VWF locus remains limited. However, a recent report identified polymorphic variation in the promoter region of the VWF gene [10], which alters the level of gene expression.

According to current definition, VWD is considered to arise from mutations in the VWF gene [1]. Many mutations in the coding region of the gene, altering expression, processing and function, have been identified and are discussed below according to the type of VWD produced. However, in light of the large number of modifying factors, it is likely that some cases of VWD will segregate with other gene loci. Support for this concept has been provided by a recent study [11], which showed that a personal and family bleeding history and persistently low VWF ristocetin cofactor activity (i.e. consistent with the accepted criteria for VWD type 1) need not co-segregate with genetic markers of the VWF gene locus.

ABO blood group

Plasma VWF level varies according to ABO blood group. People with blood group O have a lower VWF level than people with non-O blood group. In 1101 blood donors, the mean VWF antigen (VWF:Ag) level for people with blood group O was 75 IU dL−1 (range 36–157), blood group A 106 IU dL−1 (range 48–234), blood group B 117 IU dL−1 (range 57–241) and blood group AB 123 IU dL−1 (range 64–238) [12]. Furthermore, a dosage effect is detectable, such that genotype AO has a lower level than AA [13]. Whether or not ABO group has any effect on the specific activity of VWF is unresolved [12,14,15].

Two other glycosyltransferases, responsible for the Lewis and Secretor serotypes, are closely related to ABO blood groups. A recent study showed that homozygous secretors (SeSe) have higher levels of VWF than heterozygotes or non-secretors. Lewis blood group had no effect on VWF [16].

A recent genome-wide linkage analysis confirmed the major influence of the ABO locus on plasma VWF levels and indicated that regions on several other chromosomes may be implicated [17]. Interestingly, the VWF gene itself was reported to have very little influence on normal plasma levels.


Ethnic differences in VWF levels are likely to be largely because of ABO group, although in one study 19 black women had higher levels than the remaining 123 women in the study group even after controlling for this effect [18]. In another study, the VWF:Ag but not VWF ristocetin cofactor activity (VWF:RCo) was higher in African-American than in caucasian American women (mean 123 vs. 103 IU dL−1) [19].

Physiological and environmental


Neonates have a raised VWF level which falls to baseline by 6 months. The mean ± 1 SD on day 1 is 153 ± 67 IU dL−1, day 5 140 ± 57 IU dL−1, day 30 128 ± 59 IU dL−1, day 90 118 ± 44 IU dL−1 and day 180 107 ± 45 IU dL−1 [20]. This wide variation makes the diagnosis or exclusion of type 1 VWD unreliable in children <6 months old. Several reports have shown that VWF level increases slowly throughout adult life [12,18], at a rate of c. 10 IU dL−1 per decade.


In children under 6 months of age VWF levels need to be assessed in relation to age-specific normal ranges and should be repeated when older (grade B, level IIb).

Menstrual cycle

There are conflicting data regarding changes in VWF levels during the menstrual cycle. In a longitudinal study of 95 women not taking oral contraception tested on days 4–7, 11–15 and 21–28, no change was found for VWF:Ag, VWF:RCo or VIII:C [21]. A similar study of 39 women found lower levels of VWF:Ag and VWF:RCo in the follicular phase and peak levels in the luteal phase [18]. Finally, a larger, but cross-sectional, study reported significantly lower levels of VWF:Ag in the menses compared with the follicular phase (days 8–11) [22]. The magnitude of the reported changes was c. 30%.

These data relate to women who do not have VWD. There are no data available for changes in VWF during the menstrual cycle for women with VWD.


At the present time there are insufficient data to recommend testing for VWD during a specific part of the menstrual cycle (grade B, level III).

Combined oral contraceptive pill

Combined oral contraceptive pill (COCP) use causes a small rise in the level of VWF:Ag and VWF:RCo. In 36 women who did not have VWD the VWF:Ag and VWF:RCo rose from 88 ± 29 (IU dL−1, mean ± SD) and 103 ± 31 to 92 ± 38 and 107 ± 46 respectively after 6 months treatment with low dose COCP [23].

No data are available for women with VWD.


The level of VWF increases three- to fivefold during pregnancy in women without VWD [24,25] and in most but not all women with VWD type 1. Although published data most frequently compare baseline and the third trimester [26,27] the level of VWF is often raised earlier in pregnancy.

There is no increase in VWF in women with type 3 VWD. The changes in VWD types 2A and 2M are variable with increases in VWF:Ag and VIII:C not necessarily paralleled by increases in VWF:RCo [28].


Women on the COCP with borderline VWF levels should be retested if the opportunity arises when off the COCP and not pregnant (grade C, level IV).

Hormone replacement therapy

Hormone replacement therapy (HRT) has no effect on the levels of VWF in women without VWD [29]. This has been demonstrated for both oral and transdermal HRT [23,30]. No data are available for women with VWD.


Testing for VWD is often complicated by the concern that the VWF levels may be elevated by stress or exercise. In an uncontrolled study of males without VWD, mental stress (e.g. arithmetic) for 20 min resulted in a significant increase of VWF:Ag from 95 ± 28 IVdL−1 to 123 ± 56 IVdL−1 (mean ± SEM) and FVIII from 125 ± 54 IVdL−1 to 217 ± 170 IVdL−1 [31]. A similar, but less marked effect was subsequently observed in females [32].

Physical exertion

Moderate or strenuous exercise in people without VWD reproducibly leads to a rapid and significant increase in VWF:Ag and FVIII:C. For example, short or long runs at close to maximal capacity altered levels of VWF:Ag to 194 ± 42 IVdL−1 (1.7 km), 96 ± 21 IVdL−1 (4.8 km) and 144 ± 22% (10.5 km) above pre-exercise levels [33]. The VWF:Ag levels remained above baseline at 10 h postexercise.

Lesser degrees of exercise do not result in elevation of VWF. In another study, FVIII:C and VWF:Ag became raised only after 80% maximal activity had been reached and exercising at <50% maximum for 30 min did not result in increased levels of VWF:Ag or FVIII:C [34]. Walking at 3.2 km h−1 on a 10% incline produced only a modest 20% rise in VWF:Ag [35].

Despite these acute effects, prolonged training programmes do not have any long-term effects on FVIII-VWF [36].

These data suggest that near maximal exercise even for short periods of a few minutes leads to rapid increases in levels of FVIII and VWF. However, the moderate exercise related to attending hospital is unlikely to lead to a significant elevation.


The practice of giving 15–30 min rest before venepuncture will be insufficient to allow the VWF and FVIII levels return to baseline if they have been elevated by strenuous exercise and is therefore not recommended (grade B, level IIb).


It is very likely that venepuncture performed under stressful circumstances results in increased FVIII and VWF. In children, stress may be compounded by strenuous physical effort (e.g. crying) and a further increase in FVIII and VWF could occur. It is the experience of most haematologists that stressful venepuncture, particularly in young children, appears to result in raised levels of FVIII and VWF and this should be taken into account when results are interpreted. Fainting after venepuncture has a similar effect.


In children with borderline levels of VWF, the child should be retested with as stress-free venepuncture as possible (grade C, level IV). Samples taken when children are clearly stressed should be assessed in relation to the possible increase in VWF that may have been induced by the stress and this fact recorded in their notes.

There is no evidence that venepuncture alone results in raised levels of VWF although venous occlusion can have this effect. Whether this is a result of changes in haemoconcentration or release of VWF [37,38], is disputed [39,40] but the practical consequence is the same and venous occlusion should be minimized.

Inflammation, malignancy, renal disease, diabetes, liver disease, infection

von Willebrand factor is an acute phase reactant and is raised in a number of inflammatory, malignant, infective and vasculitic diseases [41–43]. Malignancy is associated with raised levels of VWF and in some cases this has been shown to correlate with the stage of the disease [44–47]. In these circumstances VWF may reach very high levels of several times normal.

Thyroid function

The VWF antigen and function are decreased in hypothyroidism and raised in hyperthyroidism. These changes resolve when the patient is treated and has become euthyroid [48,49].

von Willebrand disease

von Willebrand disease (VWD) is the commonest of the inherited bleeding disorders. Up to 1% of the population have VWD as defined by reduced levels of VWD but only about 125 per million have a clinically significant bleeding disorder [50]. The number of patients with VWD registered with the UKHCDO database in 2001 was 6294, of whom c. 10% required treatment in the calendar year. Uncertainty regarding the relationship between laboratory assays and function in vivo means that diagnosis is frequently difficult.

VWD commonly presents as a mild to moderate bleeding disorder, typically with easy bruising or bleeding from mucosal surfaces. However, when there is complete deficiency of VWF, the bleeding symptoms are severe. Consequently the most severe forms of VWD usually present in childhood whilst the mild forms may not present until after a significant haemostatic challenge, often in adulthood. In the absence of such a significant challenge, some patients remain asymptomatic and undiagnosed. In the mild forms there is considerable overlap with normality.

VWD has been defined as an inherited bleeding disorder caused by a quantitative or qualitative defect of VWF secondary to a mutation in the VWF gene [1]. However, it is now clear that many other genetic loci exert quantitative and qualitative influences over plasma VWF so that VWD is a genetically and clinically heterogeneous disorder with variable penetrance.

Historically the classification of VWD has been complex and a major source of confusion. The old classification system had many different subtypes and subclassifications of limited clinical relevance. Today a simpler classification is used [1]. VWD is now subclassified into three major categories (Table 1), one of which, (type 2) is subclassified into four variants dependent upon the type of functional defect present (Table 2). The six categories correspond to distinct pathophysiological mechanisms and are important in determining therapy.

Table 1.  Primary classification of von Willebrand disease (VWD).
SubclassificationType of von Willebrand factor (VWF) deficiencyVWF protein function
Type 1Quantitative partial deficiencyNormal
Type 2Qualitative functional deficiencyAbnormal
Type 3Quantitative complete deficiencyUndetectable
Table 2.  Secondary classification of type 2 von Willebrand disease (VWD).
SubtypePlatelet-associated functionFactor VIII-binding capacityHigh-molecular weight von Willebrand factor (HMW VWF) multimers
2BIncreased affinity for glycoprotein (GP)IbNormalUsually reduced/absent
2MDecreasedNormalNormal and occasionally ultra-large forms
2NNormalMarkedly reducedNormal

According to the International Society on Thrombosis and Haemostasis (ISTH) classification, the diagnosis of VWD hinges on defining quantitative and qualitative deficiencies of VWF [1].

Approach to the diagnosis of VWD

History and examination

History taking is a key part of the assessment of a possible bleeding problem. The pattern of bleeding should be noted. Bleeding manifestations typical of VWD are easy bruising, epistaxis, oral cavity bleeding and in women, menorrhagia. It is particularly important to ascertain the response to haemostatic challenges such as operations, trauma and dental extractions. Women should be asked about postpartum bleeding, although it should be noted that symptoms of VWD may improve in pregnancy.

It is important to establish whether bleeding has been life-long or whether it is of recent onset, indicating an acquired abnormality. There should be a careful review of other medical problems, particularly for conditions known to be associated with acquired VWD such as lymphoproliferative and myeloproliferative disorders. A detailed drug history should be taken since aspirin and NSAIDs are the commonest cause of platelet dysfunction. Examination assesses the type of bleeding if any is present at the time of consultation but often the main purpose is to exclude an underlying disease. Scars from previous trauma or surgery should be examined in case the defect in primary haemostasis is due to a collagen disorder.

Criteria for bleeding history

Bleeding histories are subjective and there is overlap between symptoms suffered by people with VWD and the normal population. In one study of children undergoing tonsillectomy, a history of easy bruising was present in 24% of those who did not bleed excessively at operation and in 67% of those who did [51]. The ISTH suggested stringent criteria to define a positive mucocutaneous bleeding history (1996 Annual Report of the SSC/ISTH Subcommittee on VWF) [52]. In practice, however, these criteria have been reported to be difficult to apply and other, less stringent, criteria have been suggested which appear to result in fewer indeterminate results [52]. It has also been shown that the discriminatory power of the bleeding history is greater in a screening situation than when patients have already been referred for investigation [53].

In general, a history of bleeding is more significant the more symptoms that are present, the more severe these symptoms are and the more frequently they occur. The requirement for a blood transfusion as a result of bleeding, medical intervention to stop the bleeding or the development of anaemia adds weight to the bleeding history. A significant bleeding tendency is less likely if a person has undergone invasive procedures without bleeding and similarly abnormal bleeding after surgery may indicate a bleeding disorder. It should be recognized that the bleeding history of patients with VWD is very variable and evolves throughout their lifetime.

Bleeding events that may suggest VWD include (adapted from ISTH criteria and [52]):

  • 1Prolonged epistaxis without a history of trauma that is not stopped within 20 min by compression, or leads to anaemia or which requires blood transfusion. Epistaxis that required control by medical intervention such as packing or has recurred after cauterization may also be more significant.
  • 2Cutaneous haemorrhage and bruising with minimal or no apparent trauma, as a presenting symptom or requiring medical treatment.
  • 3Prolonged bleeding from trivial wounds, lasting ≥15 min, requiring medical attention to control or recurring spontaneously during the 7 days after wounding.
  • 4Oral cavity bleeding, such as gingival bleeding, or bleeding with tooth eruption or bites to lips and tongue that requires medical attention or recurs over the next 7 days.
  • 5Spontaneous gastrointestinal bleeding requiring medical attention, or resulting in acute or chronic anaemia, unexplained by a local lesion.
  • 6Heavy, prolonged, or recurrent bleeding after tooth extraction or surgery such as tonsillectomy and adenoidectomy, requiring medical attention.
  • 7Menorrhagia not associated with structural lesions of the uterus. Menorrhagia that has been present from the menarche or has led to anaemia or required medical treatment increases the likely significance of this symptom.
  • 8Prolonged bleeding from other skin or mucous membrane surfaces requiring medical treatment.

Family history

A positive family history compatible with the dominant forms of VWD requires that a first degree relative or two second degree relatives have a personal history of significant mucocutaneous bleeding and laboratory tests compatible with VWD. A complete dominant pattern is often not seen because of incomplete penetrance. When available, the identity of VWF mutations or genetic markers linked to the VWF locus may permit linkage of the phenotype to more distant relatives.

Preliminary investigations for a suspected bleeding disorder

These should include full blood count, prothrombin time, activated partial thromboplastin time (APTT), thrombin time or fibrinogen, bleeding time or PFA-100, platelet aggregation studies, FVIII:C and VWF antigen and function.

Full blood count

The platelet count should always be performed when investigating patients with a possible bleeding disorder. The platelet count will be within the normal range in patients with all types of VWD except for those with type 2B VWD where a moderate thrombocytopenia may be present. The identification of anaemia is clearly significant.

Activated partial thromboplastin time

The APTT is a frequently used screening test. Each laboratory should determine its own reference range and be aware of the variable sensitivity of the APTT to reduced levels of clotting factors. Although often prolonged in VWD, a normal APTT does not exclude its diagnosis as the FVIII:C is frequently in the normal range. Further, specific tests should always be performed if the diagnosis is suspected.

Prothrombin time, fibrinogen and thrombin time

These tests are all normal in VWD.

Bleeding time

The bleeding time is defined as the time taken for a standardized skin incision to stop bleeding. The modified Ivy method is the most frequently used approach. In this method, a longitudinal incision is made on the volar aspect of the forearm using a standardized disposable device, keeping a blood pressure cuff inflated to 40 mmHg. Each laboratory should establish its own reference range; which will usually be c. <10.5 min. The bleeding time is also prolonged by a low haematocrit (<0.30), low-platelet count (<100 × 109 L−1), functional platelet disorders (including aspirin or NSAID ingestion) and collagen disorders.

The bleeding time has in the past been considered a screening test for VWD. Its clinical utility, however, is limited because of insensitivity and lack of reproducibility. The bleeding time is frequently normal or only minimally prolonged in milder forms of VWD.


The bleeding time does not have a role as a screening test for VWD. However, it is of value in the composite assessment of haemostasis (grade B, Level III).


The PFA-100 allows a rapid and simple determination of platelet-VWF function. It simulates primary haemostasis in the high shear stress environment that occurs after small vessel injury. Whole blood is aspirated through an aperture in a membrane coated with collagen impregnated with adrenaline or ADP. Platelets adhere, aggregate and occlude the aperture, defining the ‘closure time’.

The closure time in the PFA-100 has recently been proposed as superior to the bleeding time as it is reproducible, simple and predictive; detecting VWD with a sensitivity of >95% compared with 50% for the bleeding time [54]. Like the bleeding time it is not specific for VWD and it may be prolonged by other defects of primary haemostasis except collagen disorders. The PFA-100 appears to offer a useful VWD screening test because of its high sensitivity [55,56]. It will reliably detect moderate or severe disease. However, it cannot avoid the natural fluctuations that make assessment of VWF levels so difficult. For example, in one study the sensitivity for mild type 1 VWD was reported as 83% but four cases with normal PFA-100 closure times all had normal VWF:RCo on the day of testing [52].

The PFA may function as an alternative to performing a bleeding time in the initial investigation of patients with a suspected bleeding disorder but may require repeating if suspicion is high. Repeated normal results make VWD extremely unlikely. Some mild platelet disorders are not detected but it is not clear that these have a significant bleeding tendency [57].

A local normal range must be established.

Tests used for the primary diagnosis of VWD

Factor VIII assay

Factor VIII (FVIII:C) is usually measured using a modified APTT-based one-stage clotting assay or chromogenic assay. A two-stage assay is also sometimes used. Although FVIII half-life is regulated by VWF and is frequently reduced in VWD, FVIII:C levels do not always parallel the levels of plasma VWF, and may be normal in the presence of VWD. A normal FVIII:C does not therefore exclude VWD.

von Willebrand factor antigen

Plasma VWF:Ag levels are measured by immunological methods. Electrophoretic immunoassays have now largely been superseded by immunoradiometric (IRMA) or enzyme-linked immunosorbant (ELISA) assays. ELISA remains the reference method. Latex particle based assays are now becoming available using immunoturbidometric methods which can be automated.


The VWF:Ag should be measured using an assay whose limit of detection is <1 IU dL−1 (grade C, level IV).

von Willebrand factor ristocetin cofactor activity

The VWF:RCo is a functional measure of the ability of VWF to bind GpIb in the presence of ristocetin (a glycopeptide synthesized by Nocardia lurida). This activity assay is performed by measuring the agglutination of normal fixed platelets in dilutions of test plasma containing an excess of ristocetin. Fresh platelets can also be used. Ristocetin dimers bind to both VWF and platelet GpIb leading to the cross-linking of platelets. A proposed ristocetin-binding site on VWF is Glu1239-Pro-Gly-Gly1242, which is in the GpIb-binding region [58].

The patient's VWF:RCo is determined by reference to a plasma standard. The result depends on the presence of HMW multimers and on an intact GpIb-binding site. The assay has high inter-assay and inter-laboratory variability [59].

In recent years, the VWF platelet-dependent function was frequently assessed in the UK by an ELISA incorporating an antibody, which recognizes a functional epitope on VWF associated with GpIb-binding. Data from this ELISA in its original form suggested it performed well across a range of different types of VWD and VWF activity levels [60]. However, data from NEQAS and other surveys [61], in which most participants used a commercial kit, show the median VWF activity estimated by this ELISA to be consistently higher than that obtained by VWF:RCo and analysis of a molecularly defined qualitative variant (type 2A) revealed a highly significant difference between the two assays.

These results showed the inability of the ELISA to identify with certainty type 2A VWD and led to a re-evaluation of its utility by the ISTH VWD Scientific Subcommittee. By consensus, the standard and most discriminatory assay of VWF function remains the platelet-based assay of VWF:RCo.


A platelet-based assay for VWF:RCo should be used to assess VWF:RCo. An ELISA assay cannot presently be recommended as being sufficiently sensitive to detect variant type 2 VWD (grade B, level IIa).

von Willebrand factor collagen binding activity

The use of VWF collagen binding activity (VWF:CB) as a measure of VWF activity is becoming widespread and complements the VWF:RCo assay [62]. The current method is an ELISA in which the plate is coated with collagen and the amount of VWF which binds is measured. The patient's VWF:CB is determined with reference to a plasma standard. The binding is very dependent on the presence of HMW multimers and an intact collagen-binding site. The type of collagen used is an important variable in the performance of the assay [63].

Using a standardized VWF:CB, there is less inter-assay variability than with the VWF:RCo. This may prove useful in cases of diagnostic difficulty where results are close to the lower limit of the reference range and also in more clearly distinguishing type 1 from type 2 VWD [61,63,64]. Furthermore, in type 2A disease treated with DDAVP the VWF:RCo may normalize whilst the VWF:CB remains low reflecting the continuing absence of the larger multimers [64]. A practical advantage of the VWF:CB in the laboratory is that it uses similar methodology to the VWF:Ag, which allows the two tests to be performed in parallel.

Thus, although VWF:CB and VWF:RCo assess different aspects of VWF function, both are sensitive to the loss of HMW multimers. However, the VWF:CB might not detect a type 2 defect because of a mutation in the Ib-binding region of VWF(type 2M), and the VWF:RCo might not detect a type 2 defect because of a mutation in the collagen-binding region of VWF [65]. The use of both assays should improve the ability to detect type 2 variants. The VWF:CB will be more accurate than the VWF:RCo in differentiating mild type 1 VWD from normal (due to the better CV) and in accurately assessing the level in severe types 1 or 3 disease (as the lower limit of detection is much better). When comparing activity to antigen ratios it is also better at differentiating type 2A from type 1 disease [61].


It is recommended that both VWF:RCo and VWF:CB are used to assess VWF:activity to improve the ability to detect type 2 variants and to more clearly define type 1 VWD (grade B, level III).

Ristocetin-induced platelet aggregation

Addition of ristocetin to normal platelet-rich plasma will promote the binding of VWF to platelet GpIb. In the ristocetin-induced platelet aggregation (RIPA) assay, ristocetin is added to patient's platelet-rich plasma at several concentrations (from 0.2 to 1.5 mg mL−1 final concentration) to assess the affinity of VWF for platelets. Thus, the lowest ristocetin concentration that can induce aggregation can be determined. Aggregation at concentrations ≤0.5 mg mL−1 indicates VWF-platelet hyper-reactivity and is an essential diagnostic criterion for the detection of type 2B VWD and platelet type pseudo-VWD. RIPA is absent in severe forms of VWD, but is frequently normal in those with VWF:RCo >30 U dL−1.


The RIPA should be assessed at least at final ristocetin concentrations of both 0.5 and 1.25 mg mL−1(grade C, level IV).

Tests used to subtype VWD

Multimeric analysis of von Willebrand factor

When a diagnosis of VWD has been established according to clinical presentation, family history and the results of appropriate laboratory investigations, VWF multimer analysis is important for classification and subtyping.


The VWF multimer analysis has a role in subclassification of VWD. It is therefore, inappropriate to perform VWF multimer analysis until a diagnosis of VWD has been confirmed (grade C, level IV).

The VWF multimer analysis is carried out by electrophoresis of test plasma samples using non-reducing agarose gels in the presence of sodium dodecyl sulphate (SDS). Higher gel concentrations are useful to provide enhanced discrimination of smaller multimers and individual multimer sub-bands. A number of different methods have been described for the visualization of VWF multimers following electrophoresis. Methods using 125I-labelled VWF, although sensitive, have generally been superseded as the use of radioisotopes is potentially hazardous and many laboratories do not have the appropriate facilities. These methods are also time-consuming and expensive. Non-radioactive methods now used by many laboratories entail the use of enzyme-labelled antibodies and subsequent visualization of VWF multimers using enzyme substrates or enhanced chemiluminescence followed by autoradiography [66].

In normal plasma, VWF multimers appear as a series of bands separated by the mass of 2 subunits (i.e. 1 × 500 kDa dimer). Higher resolution gels reveal the presence of three bands comprising each multimer (a major band with leading and trailing sub-bands), the result of proteolytic cleavage in the circulating blood.

VWF multimer characteristics in VWD

Standard resolution gels (1.0–1.5% agarose) are generally adequate to distinguish the presence (e.g. as in ‘normal’ individuals or in VWD type 1) or absence (e.g. as in VWD type 2A) of HMW VWF multimers. 1.5% gels will also often allow at least partial discrimination of multimer triplet sub-bands.

Table 3 and Fig. 1 illustrate the plasma VWF multimer characteristics associated with different VWD subtypes, classified according to currently generally accepted classification criteria [66].

Table 3.  von Willebrand disease (VWD) multimer interpretation in VWD subtypes.
VWD subtypevon Willebrand factor (VWF) multimer interpretation
Type 1Normal multimer distribution
Type 2AHigh and intermediate molecular weight multimers absent. Frequent abnormal triplet sub-bands
Type 2BVariable loss of high-molecular weight multimers. Occasionally multimer distribution is normal with all high-molecular weight multimers present. Frequent abnormal triplet sub-bands
Type 2MThe full range of VWF multimers is present. Occasionally ultra-large multimers are present
Type 2NNormal multimer distribution and appearance
Type 3No multimers visualized. May be faint low-molecular weight bands
Figure 1.

Example of von Willebrand factor (VWF) multimer analysis using a standard resolution 1.3%agarose gel followed by Western blotting and visualization of multimers by enhanced chemiluminescence [linkRIDb6666]. The high molecular weight multimers are to the top of the Figure. Lanes 2 and 3 show a normal distribution of VWF multimers. Lane 1 shows the multimer pattern in a patient with type 2A von Willebrand disease (VWD), and lane 6 represents a patient with type 1 VWD. In lanes 4 and 5, there is a suggestion that the largest multimers may be reduced compared with normal and these patients would require further analysis using a lower concentration (for example 1%) gel.

VWF-FVIII binding assay


In a typical procedure VWF is isolated from test plasma by antibody capture onto a microtitre plate. Endogenous VWF-bound FVIII is removed by washing with a high concentration solution of calcium chloride. Purified normal FVIII is then added and allowed to interact with the immobilized VWF. VWF:Ag and bound exogenous FVIII are separately quantified by ELISA and chromogenic assay, respectively. A graph is plotted of FVIII-bound vs. VWF:Ag-bound [67].


Markedly reduced binding of FVIII is diagnostic of type 2N VWD. There is a large area of overlap between the binding characteristics of some ‘normal’ individuals and the intermediate binding profile often seen in heterozygous carriers of VWD type 2N mutations (these individuals have normal levels of FVIII). Intermediate VWF-FVIII-binding seen in patients with reduced FVIII:VWF:Ag ratio should be referred for further specialist investigation.

Molecular analyses

In some circumstances the identification of a molecular defect in the VWF gene can help with diagnosis and classification. The types of mutations are characteristic of each VWD subtype and are considered with each subtype below under diagnostic criteria.

Platelet von Willebrand factor

Analysis of platelet VWF:Ag and multimers may be of value to diagnose VWD subtypes identified in the old subclassification [68]. Platelets must be processed to remove any trace of contaminating plasma VWF prior to preparation of platelet lysate solutions. These analyses are sometimes useful when the bleeding tendency appears discordant with the plasma VWF level.

Practical guidelines for diagnosis

General considerations

By definition VWD is a mucocutaneous bleeding disorder caused by a qualitative or quantitative VWF deficiency resulting from a VWF gene defect.

Therefore, to make a definitive diagnosis of VWD a person must have:

  • 1a personal history of mucosal bleeding;
  • 2decreased functional VWF levels; and
  • 3a mutation in the VWF gene or family history of bleeding that segregates with (2).

In practice, each of these criteria pose a number of practical problems:

  • 1A personal history of bleeding may be missing because no significant haemostatic challenge has been encountered or some form of prophylaxis has already been introduced.
  • 2It is not clear that the in vitro assessments of VWF function available adequately reflect VWF function in vivo. This may be due to interactions with other genetic factors. The result is that an increased tendency to bleeding cannot be confidently predicted when these measures fall within a large ‘grey area’ at the lower end of the normal range. This is not resolved by the practice of using ABO group-specific normal ranges [69].
  • 3The VWF gene is large and difficult to analyse so in many cases a genetic defect is not identified. This is particularly true of VWD type 1. Furthermore, the significance of many identified mutations is uncertain. Finally, it is now recognized that some patients with VWD do not have lesions of the VWD gene at all [11]. As already discussed, the plasma VWF level is modulated by numerous other genetic, physiological and environmental factors, which can result in incomplete penetrance and an obscured family history.

Nonetheless, in many cases the diagnosis is straightforward and the diagnostic difficulties are largely restricted to those who are suspected to have mild VWD type 1. For many patients it is often not possible to fulfil all the required diagnostic criteria with certainty. These problems of classification will not concern the patient for whom the important issues are:

  • 1are they at risk of bleeding, either spontaneously or following injury or invasive procedures?
  • 2are other family members at risk of the same disorder?
  • 3what treatment, if any, is required?

It is, therefore, more pragmatic in a clinical setting to consider VWD as a bleeding disorder caused by decreased functional levels of VWF rather than to follow the strict definition of VWD. This approach will include some patients with normal VWF genes and who lack a clear family or personal history of mucosal bleeding but who may nonetheless require treatment for bleeding episodes or to cover invasive procedures.

This approach is supported by a study that demonstrates that people with VWF levels between 35 and 50 IU dL−1 have similar bleeding risks and requirement for treatment whether their low VWF level is due to a clearly defined VWD or blood group O [69]. Similarly, a study using the PFA-100 found that the closure times were significantly longer in those with blood group O, indicating that the lower VWF levels are of functional consequence [70]. Clearly, there is a continuum of VWF levels in the population and a continuum of bleeding risk associated with VWF level. The point at which bleeding or the risk of bleeding becomes unacceptable is necessarily arbitrary but has to be determined for each patient. It is now suggested that a diagnosis of ‘possible VWD’ could be made where either a personal or family history of bleeding is accepted in combination with decreased functional activity of VWF.


To minimize the risk of misdiagnosis, VWF:Ag and function must be measured in samples obtained on at least two occasions, with consistent results (grade C, level IV).

Type 1 VWD


The VWD type 1 is defined as an inherited bleeding disorder because of quantitative deficiency of VWF. Within the ISTH criteria type 1 VWD is due to a defect in the VWF gene.

Minimum criteria for diagnosis

  • 1Significant mucocutaneous bleeding (see above).
  • 2Laboratory tests compatible with VWD type 1 (see below).
  • 3Either a positive family history for VWD type 1 (see above) or an appropriate VWF mutation.
  • 4Threshold for RIPA not reduced.

Frequently a patient's history or laboratory results are consistent with a diagnosis of type 1 VWD but they do not meet these criteria in full. Thus, a category of ‘possible VWD type 1’ is now defined. This includes all patients with laboratory tests compatible with VWD type 1 and either significant mucocutaneous bleeding or a positive family history for VWD type 1.


Refer to general considerations above. Patients have a history of mucosal or trauma-related bleeding and inheritance is usually autosomal dominant.

Laboratory findings

Laboratory tests results are compatible with VWD type 1 if the levels of both VWF:RCo and VWF:Ag are <50 U dL−1 on at least two determinations.

The coagulation screen is normal or may have an isolated prolonged APTT. This will depend upon the residual VIII:C activity.

The VWF:RCo, VWF:CB and VWF:Ag will all be reduced to an approximately equal degree. A function to antigen ratio <0.7 raises the possibility of a type 2 variant [71,72].

The RIPA may be normal or impaired but does not show increased sensitivity to ristocetin.

Platelet count is normal.

Multimeric analysis will show a normal and full distribution of multimers with a normal triplet structure.

Molecular studies

The molecular mechanism in the large majority of type 1 VWD cases is unknown [73] (a database of reported mutations is available at: A small number of missense mutations have been reported, while some cases may represent the heterozygous form of VWD type 3 – the result of null mutations. Recently a dominant VWF gene mutation (Y1584C) was reported in 10 of 70 families with type 1 VWD [74]. This is the first report of a mutation which is present in a significant proportion of type 1 VWD patients. In classical type 1 VWD, inheritance is considered to be autosomal dominant and linked to the VWF gene. Penetrance is incomplete however, and circulating VWF levels and hence clinical presentation are clearly influenced by other factors, both genetic and environmental. Genetic influences include ABO blood group, compound heterozygosity for VWF gene mutations and the influence of the other ‘normal’ VWF allele. A recent report has indicated that circulating levels of VWF may be influenced by polymorphic variation in the promoter region of the VWF gene [10]. There is also evidence that a proportion of human type 1 VWD arises from genetic defects not associated with the VWF gene [10,75], as has previously been demonstrated in a mouse model of dominant type 1 VWD [76].

Type 2A VWD


A qualitative variant of VWF with decreased platelet-dependent function that is associated with the absence of HMW multimers.

Minimum criteria for diagnosis

  • 1A personal or family history of bleeding.
  • 2VWF:RCo below 50 U dL−1 and <0.7 of the VWF:Ag level.
  • 3Absent high molecular weight multimers.
  • 4Threshold for RIPA not reduced.


Patients have a history of mucosal or trauma-related bleeding and inheritance is usually autosomal dominant.

Laboratory findings

The platelet count will be normal.

The coagulation screen is normal or may have an isolated prolonged APTT.

The FVIII:C level may be normal or low.

The VWF:Ag may be low or normal. The VWF:RCo or VWF:CB are below 50 U dL−1, may be very low, and are <0.7 of the VWF:Ag.

The bleeding time or PFA-100 closure time is prolonged in proportion to the function defect in VWF/platelet interaction.

The RIPA is nearly always significantly impaired.

The HMW multimers are absent and the triplet pattern may be abnormal.

Molecular studies

Inheritance is generally autosomal dominant although some cases may be autosomal recessive. VWD type 2A arises by two alternative molecular mechanisms [77,78]. Group 1 mutations are thought to cause defective intracellular transport of VWF, resulting in the retention of large multimeric forms in the endoplasmic reticulum. Group 2 mutations appear to cause enhanced susceptibility of VWF to proteolysis in plasma. This results in the degradation of the larger multimers. The majority of the missense mutations associated with VWD type 2A are clustered within a 134 amino acid segment of the VWF A2 domain, which is encoded by exon 28 of the VWF gene.

A number of rare type 2 VWD variants, previously classified as types IIC, IID, IIE, IIF, IIG, IIH, and II-I, are now included within the VWD type 2A category. The recessively inherited former type IIC VWD results from missense mutations or small deletions or insertions in the VWFD2 propeptide domain [79]. These mutations presumably interfere with multimer processing and assembly, causing a lack of large VWF multimers and an increase in the quantity of VWF dimer. The former type IID VWD subtype results from impaired VWF dimer formation because of defective disulphide bonding of C-terminal domains. Several mutations which abolish C-terminal cysteine residues, thus accounting for the type IID phenotype, have been described [80]. Mutations associated with VWD types IIE to II-I have not been described. The VWD variants previously defined as types Ib and I ‘platelet discordant’, which are also associated with a reduction of HMW VWF multimers, have similarly been reclassified as type 2A (

Type 2B VWD


A qualitative gain of function variant of VWF with increased affinity for platelet GpIb.

Minimum criteria for diagnosis

  • 1A personal or family history of mucosal bleeding.
  • 2Increased sensitivity to RIPA at low-dose ristocetin (≤0.75 mg mL−1).
  • 3Exclusion of pseudo(platelet type) –VWD.


Patients have a history of mucosal or trauma-related bleeding. Inheritance is usually autosomal dominant although some apparently recessive cases have been described. There is usually in positive family history.

Laboratory findings

The platelet count is usually slightly low (75–100 × 109 L−1) but may be normal in mild forms. The thrombocytopenia may be exacerbated in pregnancy and following infusion of DDAVP [81].

The coagulation screen may be normal or have an isolated prolonged APTT.

Antigenic measurements of VWF may be low or normal. The VWF:RCo is usually <0.7 of the VWF:Ag.

The FVIII:C level may be normal or low.

The VWF:RCo is usually decreased. Rare cases exist where VWF:RCo is >50 U dL−1.

The VWF:CB is low and tests of platelet function under high shear are abnormal.

The RIPA is increased, with full aggregation occurring at <0.75 mg mL−1 final concentration of ristocetin.

The bleeding time or PFA-100 closure time is prolonged in proportion to the functional defect in VWF-platelet interaction.

The HMW multimers are usually absent but the deficiency is less marked than in type 2A and intermediate size multimers are present. In mild forms HMW multimers may be present.

Molecular studies

The VWD type 2B results from increased binding of VWF to platelet GpIb and associated removal from the circulation (usually) of the largest VWF multimers. Inheritance is autosomal dominant. Nucleotide substitutions associated with this ‘gain of function’ variant almost all occur within a short segment of the VWF A1 domain, the functional domain containing the platelet GpIb-binding site. Of reported mutations, four account for c. 90% of cases of VWD type 2B [82] (

Platelet-type (pseudo-) VWD

Platelet-type pseudo-VWD is a rare disorder [83], resulting from mutations in the platelet GpIb/IX receptor complex, which cause increased binding between VWF and platelets. Inheritance is autosomal dominant. Clinical presentation is variable, including epistaxis and excessive bleeding associated with surgery but some patients have few bleeding manifestations. Laboratory presentation is markedly similar to that of type 2B VWD, with a variable degree of usually mild thrombocytopenia, enhanced RIPA, and decreased HMW VWF multimers. Plasma/platelet mixing studies are required to distinguish platelet-type pseudo-VWD and type 2B VWD. The addition to normal platelets of plasma from a patient with type 2B VWD, but not from a patient with platelet-type pseudo-VWD, will confer enhanced RIPA. Alternatively, the addition of normal cryoprecipitate (containing a high concentration of normal VWF) will cause platelets in platelet-rich plasma from a patient with platelet-type pseudo-VWD, but not type 2B VWD, to aggregate spontaneously.

Type 2M VWD


A qualitative variant of VWF with decreased platelet-dependent function that is not caused by the absence of HMW multimers.

Minimum criteria for diagnosis

  • 1A personal or family history of mucosal bleeding.
  • 2A VWF:RCo below 50 U dL−1 and <0.7 of the VWF:Ag level.
  • 3Presence of HMW multimers.
  • 4Threshold for RIPA not reduced.


Patients have a history of mucosal or trauma-related bleeding. Inheritance is autosomal dominant and there is usually a positive family history.

Laboratory findings

The platelet count is normal.

The coagulation screen may be normal or have an isolated prolonged APTT.

The FVIII:C level may be normal or low.

The VWF:Ag is low or normal. VWF:RCo is decreased and may be very low. VWF:CB may be low or normal and tests of platelet function under high shear are prolonged. The VWF:RCo is <0.7 of the VWF:Ag.

The bleeding time or PFA-100 closure time is prolonged in proportion to the functional defect in VWF–platelet interaction.

The HMW multimers are present and in some cases unusually large multimers may be present.

The RIPA is nearly always significantly impaired.

Molecular studies

The VWD type 2M phenotype can be caused by VWF gene mutations affecting VWF binding to platelets. Unlike VWD type 2A, the resulting loss of function is not associated with the absence of the higher molecular weight VWF multimers and in some cases unusually large multimers have been reported. To date, VWD type 2M has been reported to be associated with missense mutations and small in-frame deletions in the VWF A1 domain which contains the GpIb-binding site. A number of mutations and candidate mutations have recently been described. Inheritance is autosomal dominant. Recent studies have suggested that, due to the presence of normal VWF multimers, VWD type 2M may be often misclassified as VWD type 1 [84]. The localization of reported mutations to exon 28 of the VWF gene should facilitate rapid genetic diagnosis by direct sequencing in many cases of VWD type 2M [73] (

Type 2N VWD


A qualitative variant of VWF with decreased affinity for factor VIII.

Minimum criteria for diagnosis

  • 1Decreased FVIII:C level.
  • 2Decreased binding of control factor VIII to patient VWF.


Patients may have a history of trauma or surgery-related bleeding. Inheritance is autosomal recessive. Heterozygotes have normal VIII:C and are asymptomatic.

Laboratory findings

The FBC is normal.

The coagulation screen has an isolated prolonged APTT.

The FVIII:C level is low and is often in the range 5–30 IU dL−1. It is important to differentiate from mild and moderate haemophilia A or haemophilia A heterozygosity.

Binding of normal FVIII to VWF is markedly decreased.

The VWF:Ag and VWF:RCo are normal.

The bleeding time is normal.

Multimer analysis is normal.

The RIPA is normal.

Molecular studies

The VWD type 2N is a recessively inherited disorder. Patients are homozygous or compound heterozygous for mutations usually in the N-terminal FVIII-binding region of the VWF gene. The type 2N phenotype can also be produced by inheritance of one 2N allele and one null VWF allele. The recognition of VWD type 2N and exclusion of haemophilia A is important to ensure appropriate treatment of bleeding episodes, valid genetic counselling, and accurate carrier or prenatal diagnosis. VWD type 2N should be considered as a possible diagnosis in patients with congenital FVIII deficiency which is not clearly X-linked [82] (

Unclassified type 2 VWD

A qualitative variant of VWF with decreased collagen-dependent function that is not caused by the absence of HMW multimers has been described [85]. It is detected by VWF:CB but not by VWF:RCo assay. Although this variant clearly belongs within the type 2 category of VWD it has not been formally classified by the ISTH.

Type 3 VWD


An absence or virtual absence of VWF

Minimum criteria for diagnosis

VWF:Ag is below the limit of detection using an assay whose limit of detection is ≤1 IU dL−1.


Patients will usually present as children with a history of severe mucocutaneous haemorrhage and bleeding after minor trauma. Inheritance is autosomal recessive and often there may be no family history of bleeding.

Laboratory findings

The FBC is normal.

The coagulation screen has an isolated prolonged APTT.

The FVIII:C level is markedly reduced.

The VWF:Ag is absent or below the limit of detection.

The VWF:RCo and VWF:CB are below the limit of detection.

The bleeding time is prolonged.

Multimer analysis shows absence of VWF protein. In some cases faintly detected low-molecular bands may be observed.

The RIPA is absent.

Molecular studies

Inheritance is autosomal recessive. Molecular defects associated with VWD type 3 include deletions of large sections of the VWF gene and other null mutations including nonsense and frameshift mutations ( A single cytosine deletion in exon 18, resulting in a frameshift which causes a translational stop codon, is common in Scandinavian VWD type 3 patients. This mutation was responsible for VWD in the original family described by von Willebrand [86,87].

Genetic and family studies in VWD

A full genetic counselling service should be available to patients and families with VWD. Counselling should be undertaken before any genetic studies are carried out in order to ensure that the individuals concerned have an understanding of the inheritance of VWD in their family, and what information may be obtained by any proposed investigations. Further counselling should subsequently be carried out to explain the results of genetic tests when they become available. Genetic counselling must always be non-directive and should be undertaken by an individual(s) who has significant experience in the clinical management in VWD, together with a good understanding of the molecular basis of the disorder.

The demand for molecular genetic family studies in VWD is limited. When required, precise molecular diagnosis may be achieved in the qualitative VWD variants (types 2A, 2B, 2M and 2N) by screening for common mutations. The causative mutation is unknown in the majority of cases of VWD, however, and gene linkage analysis using restriction fragment length polymorphisms (RFLPs) or variable number of tandem repeat (VNTR) sequences in the VWF gene may be useful.

The cloning of VWF cDNA led to the identification of many RFLPs that have been used in genetic studies in VWD. The information from VWF gene RFLPs analysis is limited by their biallelic nature, and the interpretation of RFLP data in autosomally inherited VWD is more complex than in X-linked disorders. In order to determine the RFLP haplotype associated with VWD it may be necessary to carry out multiple restriction fragment analyses. The presence of the VWF pseudogene on chromosome 22 is a further complicating factor. Intron 40 of the VWF gene contains several very informative VNTR sequences and these are useful for PCR-based gene linkage studies in VWD.

In summary, gene linkage analysis has limited applications in family studies in VWD, its main use being in prenatal diagnosis for type 3 VWD. The potential for variations in the VWD phenotype, resulting from variable penetration, should always be taken into account when interpreting the results of genetic studies and when carrying out genetic counselling. Furthermore, unless linkage of VWD with the VWF gene can be clearly demonstrated the results of genetic family studies should be interpreted with caution.

Acquired von Willebrand syndrome


Acquired von Willebrand syndrome (AVWS) is due to an acquired defect in VWF that is associated with a variety of underlying disorders or pharmaceutical agents, and results in clinical symptoms similar to VWD.


The clinical presentation of AVWS is extremely variable, reflecting the diverse underlying conditions (Table 4) and pathological mechanisms (decreased synthesis, autoantibody, increased proteolytic degradation, adsorption of VWF onto cells and mechanical degradation by high shear stress). Patients may present with a sudden onset of bleeding symptoms of varying severity or only laboratory abnormalities. A search for an underlying condition should be made, although occasional idiopathic cases have been described [88].

Table 4.  Diseases associated with acquired von Willebrand syndrome [86].
 Including MGUS
Autoimmune disease
 Structural cardiac defects
 Including drugs

Criteria for diagnosis

The AVWS should be considered in all patients with

  • 1recent onset of bleeding symptoms;
  • 2no family history of VWD;
  • 3lack of previous bleeding symptoms especially in association with previous haemostatic challenges;
  • 4abnormality of VWF parameter(s) on laboratory testing.

Laboratory findings

Bleeding time – may be prolonged or normal.

The APTT – may be prolonged or normal. In some cases the prolonged APTT is not associated with a decreased FVIII:C level (e.g. hypothyroidism).

The platelet count may be normal, reduced or increased.

The FVIII level may be normal or reduced.

The VWF:Ag and VWF:RCo levels are usually both reduced. In some cases, associated with cardiac defects, the VWF:Ag and VWF:RCo levels are both normal, but the ratio of VWF:RCo to VWF:Ag is reduced and the VWF multimer pattern is abnormal (reduced HMW multimers).

The VWF:CB level is affected similarly to the VWF:RCo level.

The RIPA may be normal, decreased or absent.

The VWF multimers may be normal, abnormal with loss of HMW multimers or reduced.

There is no reliable method for detecting VWF inhibitors and not all cases of AVWS are because of the production of a VWF autoantibody. Inhibitors, if present, can be detected by mixing studies using either the RIPA or VWF:RCo assay, but these methods will only detect antibodies that interfere with the VWF–GPIb interaction. Non-inhibitory antibodies have been detected by using an ELISA-based method.

Additional investigations/therapeutic trials

The diagnosis of AVWS may be clarified by therapeutic trials:

  • 1accelerated clearance of VWF post-DDAVP or after infusion of VWF containing concentrates (dependent on the pathogenic mechanism);
  • 2a trial of intravenous immunoglobulin (IVIg) may be helpful. Some individuals with AVWS associated with an IgG monoclonal gammopathy show a temporary improvement or normalization of VWF parameters following an infusion of IVIg.

The VWF propeptide: normal or increased levels are seen in AVWS.

Investigations for an underlying disorder associated with AVWS should be made [89,90].

Note on nomenclature

The nomenclature used for VWF and its properties is that recommended by the von Willebrand Factor Subcommittee of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis (

AttributeRecommended abbreviations
Mature proteinVWF
Ristocetin cofactor activityVWF:RCo
Collagen-binding capacityVWF:CB
Factor VIII-binding capacityVWF:FVIIIB
Factor VIII coagulant activityFVIII:C


The information and advice contained within this guideline are believed to be true and accurate at the time of going to press. Neither the authors nor the publishers can accept any legal responsibility for any errors or omissions that may have occurred.

Declarations of interest

All members of the Executive of the UKHCDO and UKHCDO Working Party members are obliged to present a declaration of interests to the Chairman of the UKHCDO annually. None of the authors has any shareholding in any pharmaceutical company, None of the authors is acting as an advisor or consultant for any of the manufacturers in relation to products currently used for the treatment of VWD.



Note on recommendations

Level of evidence

LevelType of evidence
IaEvidence obtained from meta-analysis of randomized-controlled trials
IbEvidence obtained from at least one randomized-controlled trial
IIaEvidence obtained from at least one well-designed controlled study without randomization
IIbEvidence obtained from at least one other type of well-designed quasi-experimental study
IIIEvidence obtained from well-designed non-experimental descriptive studies, such as comparative studies, correlation studies and case–control studies
IVEvidence obtained from expert committee reports or opinions and/or clinical experiences of respected authorities

Grade of recommendation

GradeEvidenceRecommendation level
  1. Derived from the US Agency for Health Care Policy and Research, Acute Pain Management: Operative or Medical Procedures and Trauma. Agency of Health Care Policy and Research Publications. United States.

  2. Department of Health and Human Services, Washington, 1992.

AIa, IbRequired – at least one randomized-controlled trial as part of the body of literature of overall good quality and consistency addressing specific recommendation
BIIa, IIb, IIIRequired – availability of well-conducted clinical studies but no randomized clinical trials on the topic of recommendation
CIVRequired – evidence obtained from expert committee reports or opinions and/or clinical experiences of respected authorities. Indicates absence of directly applicable clinical studies of good quality