von Willebrand Disease

An inherited bleeding disorder caused by quantitative or qualitative deficiency of von Willebrand factor (VWF), predominantly involving mucocutaneous bleeding, encompassing types 1, 2A, 2B, 2M, 2N and 3, with autosomal dominant or recessive inheritance depending on type and treatment tailored to each type's mechanism.

Literature and regulatory information updated: 6 October 2026.

0. Initial Reference Table

1. Introduction

In 1926, the Finnish physician Erik von Willebrand described a new bleeding disorder in a family from the Åland Islands in the Baltic Sea. The index case was a young girl who died from exsanguination during her fourth menstrual period, and several other family members had excessive bleeding. The original report called the disease “pseudohaemophilia”; the term “von Willebrand disease” was adopted later [1]. The original article, published in Swedish in Finska Läkaresällskapets Handlingar, was translated into English and republished in Haemophilia in 1999 [2].

The disease results from deficiency or dysfunction of VWF, a plasma multimeric glycoprotein with two haemostatic functions: it mediates platelet adhesion and aggregation at sites of vascular injury and transports and protects FVIII from proteolytic degradation in the circulation [1, 3]. The International Society on Thrombosis and Haemostasis (ISTH) classification recognises three types: type 1, partial quantitative deficiency; type 2, qualitative abnormalities, subdivided into 2A (reduced or absent high-molecular-weight multimers), 2B (gain of function with increased platelet affinity), 2M (reduced interaction with platelets or collagen without multimer loss) and 2N (reduced FVIII binding); and type 3, virtual absence of protein with very low FVIII. The 2021 guidelines added type 1C, defined by accelerated VWF clearance, because of its therapeutic relevance [1].

VWD is the most common inherited bleeding disorder in humans and is transmitted equally to men and women; women are more likely to seek medical attention because of gynaecological and obstetric bleeding [1]. Diagnosis is often delayed: the diagnosis guidelines cite delays of 15 years or more between symptom onset and diagnosis, identifying difficulty distinguishing normal from abnormal bleeding, lack of clarity regarding the diagnostic approach and limited availability of specialised laboratories as barriers [1].

In 2021, the American Society of Hematology (ASH), ISTH, National Hemophilia Foundation (NHF; now the National Bleeding Disorders Foundation) and World Federation of Hemophilia (WFH) jointly published two guidelines developed using GRADE methodology (Grading of Recommendations Assessment, Development and Evaluation): one on diagnosis with 11 recommendations and another on management with 12 recommendations, both including patient representatives on the panel [1, 6]. Reviews published between 2025 and 2026 continue to regard them as the international reference guidelines [19, 20]. The GeneReviews chapter on the disease underwent a full update on 14 November 2024 [3]. In France, the national reference centre for the disease drafted three national protocols for the HAS: the general VWD PNDS (French National Diagnosis and Care Protocol; text dated February 2021, published on 9 March 2021), the type 3 PNDS (text dated 3 November 2021, published on 18 January 2022), which is more extensive and includes doses, laboratory targets and care pathways, and the PNDS on heavy uterine bleeding in young women with a constitutional or acquired bleeding disorder (November 2022, published on 12 December 2022), coordinated by the Toulouse reference centre for rare gynaecological conditions [7, 16, 17]. These protocols follow the 2012 HAS method for rare diseases, rest mainly on expert consensus and observational series, with explicit grading only for isolated recommendations, and describe French practice; divergences from the international guidelines are gathered in section 7.11.

The Oxford clinical genetics manual highlights the geneticist's role in these families: clarifying the nature of the familial bleeding disorder, determining severity and estimating the counsellee's risk. In the authors' experience, only approximately half of families referred with a family history of “haemophilia” actually had a relative with FVIII or factor IX deficiency; other diagnoses included VWD and factor XI deficiency [21].

2. Molecular Genetics and Pathophysiology

2.1. The gene and protein

The VWF gene is located at 12p13.31, contains 52 exons and encodes a protein of 2813 amino acids [1, 3]; the 2025 therapeutic review gives a genomic span of approximately 178 kb, contributing to the difficulties of gene therapy [19]. The precursor comprises a 22-amino-acid signal peptide, a 741-amino-acid propeptide and a mature protein of 2050 amino acids [3]. The MANE Select reference transcript is NM_000552.5, encoding protein NP_000543.3 [5].

VWF is synthesised in two cell types, endothelial cells and megakaryocytes. During synthesis, dimers form through “tail-to-tail” disulphide bonds between C-terminal cystine knot (CK) domains, followed by “head-to-head” oligomers that can reach 40 dimers [3]. The propeptide (VWFpp) is cleaved by furin in the Golgi apparatus, is essential for multimerisation, remains noncovalently associated with VWF during transport and storage (in endothelial Weibel–Palade bodies, in a 1:1 ratio), and is secreted with it [1, 3]. The VWFpp/VWF:Ag ratio estimates the relative half-life of mature VWF, although it correlates poorly with rapid clearance when VWF is between 30 and 50 IU/dL [3]. In the circulation, high-molecular-weight multimers are cleaved by ADAMTS13 between amino acids 1605 and 1606, and the pattern of proteolysis products on agarose-SDS gel electrophoresis helps infer the pathogenic mechanism [3].

Normal plasma contains VWF complexes ranging from dimers to very large multimers (500 kD to more than 10,000 kD), classified as low (1–5 bands), intermediate (6–9 bands) and high molecular weight (≥10 bands) [3]. The A1 domain binds platelet glycoprotein Ibα (GPIbα) and collagens IV and VI; the A3 domain binds collagens I and III; the D′D3 region interacts with FVIII, and this is precisely the region fused to FVIII in efanesoctocog alfa to prolong its half-life [3, 19]. Loss of high-molecular-weight multimers, shared by type 2A mechanisms, impairs clot formation, explaining why bleeding can occur despite only a modest reduction in protein quantity [3].

2.2. Mechanisms by type

Table sources: [1, 3].

In type 1, partial quantitative deficiency is associated mainly with missense variants [3]. Type 1C was first described in “Vicenza-type” VWD, caused by p.Arg1205His, and currently accounts for approximately 15–20% of type 1 cases [1, 3]. In type 2A, four mechanisms converge on loss of high-molecular-weight multimers and less effective clot formation [3]. In type 2B, inappropriate VWF binding to GPIbα removes the platelet–VWF complex from the circulation and causes mild to moderate thrombocytopenia in up to 50% of patients; platelet binding also increases ADAMTS13 susceptibility, contributing to loss of larger multimers [3]. In type 2N, FVIII loses its stabilising interaction with VWF, and without this binding its half-life is approximately sixfold shorter [20]; the laboratory and clinical phenotype resembles mild haemophilia A, with recessive inheritance [1, 3]. In type 3, absence of VWF deprives FVIII of its carrier, and FVIII falls below 10 IU/dL [3].

2.3. Genotype–phenotype correlation and penetrance

The three phenotypes reflect partial quantitative deficiency (type 1), complete quantitative deficiency (type 3) or qualitative defects (type 2) [3]. In type 1, variants associated with plasma VWF below 30 IU/dL tend to be more penetrant; in types 2A, 2B and 2M, variants are often fully penetrant, particularly in types 2A and 2B [3]. The diagnosis guidelines note that type 1 inheritance is autosomal dominant in families with VWF below 0.30 IU/mL, whereas incomplete penetrance and variable expressivity complicate transmission in families with VWF between 0.30 and 0.50 IU/mL; bleeding tends to be complex, with contributions from genes outside VWF, and a concomitant bleeding disorder, such as platelet dysfunction, should be considered [1]. The Oxford manual expresses the same reservation: inheritance in mild type 1 is more variable than classical dominant inheritance, and one school of thought links it to genetic or environmental influences independent of VWF, such as blood group [21]. In a multiethnic study of 600 US children, seven of eight children meeting VWD criteria had blood group O, and group O children had significantly lower VWF [22]; however, the 2021 guidelines do not require ABO-specific reference ranges [1].

In type 2B, Federici and colleagues studied 67 patients from 38 families with characterised variants: thrombocytopenia was present in 30% at baseline and 57% after stressors (pregnancy, infection, surgery), but platelet counts remained normal in 16 patients from five families carrying P1266L/Q or R1308L substitutions (the article's protein nomenclature, numbered from the initiating methionine; = NM_000552.5:c.3797C>T p.(Pro1266Leu), NM_000552.5:c.3797C>A p.(Pro1266Gln) and NM_000552.5:c.3923G>T p.(Arg1308Leu)) [23]. Large VWF rearrangements confer the highest risk of anti-VWF alloantibodies in type 3, although null alleles are also associated with this complication [3, 24].

2.4. Reference variants and nomenclature

Variants from GeneReviews Table 9 [3]; c., p. and g. descriptions confirmed in VariantValidator using the MANE Select transcript [5]; ClinVar classifications consulted on 6 October 2026 [25]. Older VWF literature uses the same protein numbering from the initiating methionine as HGVS (for example, R1205H corresponds to p.Arg1205His), but often describes variants only at protein level; a complete report requires c. notation with transcript and version.

Gene conversion involving the VWFP1 pseudogene, which reproduces exons 23–34, can alter those exons; it is usually recognised by two or more sequential variants identical to the pseudogene sequence. The most frequent conversions span 6 to 335 bp and have been reported in types 1, 2B, 2M and 3 [3]. Maternal uniparental disomy of chromosome 12 has been reported as a cause of type 3 [3].

2.5. Tissue and systemic pathophysiology

Failure of VWF-mediated primary haemostasis mainly causes mucocutaneous bleeding. When FVIII transport also fails, as in types 2N and 3, a secondary haemostatic defect resembling haemophilia is added, with joint and muscle bleeding [3]. VWF is an acute-phase protein that increases with bleeding, trauma, pregnancy, age and comorbidities [1]; this rise can mask diagnosis and explains the recommendation for baseline testing. Shear-induced VWF conformational changes, with relative loss of high-molecular-weight multimers, also underlie acquired von Willebrand syndrome associated with aortic stenosis (Heyde syndrome), ventricular septal defects and left ventricular assist devices [3, 6]. Gastrointestinal angiodysplasia occurs in all inherited VWD types, most frequently in adults with type 2A [3].

3. Clinical Spectrum and Manifestations by System

3.1. General pattern

VWD causes mucocutaneous bleeding and excessive bleeding following trauma and procedures; the most severe forms also carry a risk of musculoskeletal bleeding [3]. Mucocutaneous bleeding includes easy bruising, prolonged bleeding from small wounds, epistaxis, oral cavity bleeding, heavy menstrual bleeding, gastrointestinal bleeding and bleeding during haemostatic challenges such as dental treatment, childbirth and surgery [1, 3]. Severity varies considerably, even within families; in some individuals, the phenotype emerges only during a haemostatic challenge and becomes apparent only after repeated exposure, whereas others have frequent spontaneous bleeding [3]. Without trauma, musculoskeletal bleeding is rare unless FVIII is low, as in types 2N and 3 [3].

3.2. Manifestations by type

Table sources: [1, 3, 23, 26].

In type 1, heavy menstrual bleeding is the commonest finding in menstruating women [3]. The diagnosis guidelines emphasise that studies correlating VWF levels with symptoms found a similar bleeding phenotype throughout the level range, without more severe bleeding in patients with VWF below 0.30 IU/mL; among 93 patients with VWF 0.30–0.50 IU/mL, 70 had been investigated after a bleeding episode: mucocutaneous bleeding in 35, postoperative bleeding in 25 and bleeding after dental procedures in 10 [1].

In type 2B, the Italian cohort of 67 patients showed that thrombocytopenia independently predicted bleeding (adjusted hazard ratio 4.57; 95% confidence interval [95% CI] 1.17–17.90), as did the highest bleeding-score tertile (5.66; 95% CI 1.03–31.07) [23]. Patients with type 2B generally have a more severe bleeding phenotype than those with other type 2 forms or type 1, and bleeding risk correlates with the degree of thrombocytopenia [1].

In an Italian prospective study lasting 24 months, 46 patients with type 2A (17 families) and 61 with type 2M (15 families) had incidences of spontaneous bleeding requiring treatment of 107 and 40 per 100 patient-years, respectively; the difference was not explained by FVIII or VWF levels and was mainly attributable to more frequent gastrointestinal bleeding in type 2A [26].

In type 3, the international 3WINTERS-IPS study compared 223 unrelated patients with type 1 patients from the European MCMDM-1VWD study: median bleeding score was 15 versus 6, and the number of clinically relevant bleeding symptoms 5 versus 3; intracranial bleeding, oral cavity bleeding, haemarthroses and deep haematomas were at least five times overrepresented in type 3 [8]. Heterozygotes (parents and siblings) of type 3 patients may have normal VWF and be asymptomatic, but 15% to 50% have mild bleeding symptoms and may receive a type 1 diagnosis [3].

The French type 3 PNDS brings together data from four cohorts (Iranian, 385 patients; Italian, 66; North American, 150; and the international 3WINTERS-IPS) and stresses that 3WINTERS-IPS found no difference from type 1 in post-surgical or post-dental-extraction bleeding, unlike deep, oral and central nervous system sites [8, 16]. Symptoms begin early: in an Italian series of 128 consecutive patients with severe forms, median age at first symptoms was 1 year in type 3, versus 7 and 8 years in types 1 and 2, and 25 patients (two with type 3) had been diagnosed through family history or preoperative testing; in the Dutch WiN paediatric cohort, median age at first signs was 1.5 years, at diagnosis approximately 2 years, and the interval between them was 0.1 year in type 3 [16]. The ISTH-BAT score in type 3 is high from childhood, with medians of 19 (range 1–35; 26 patients; de Wee 2012), 15 (8–20; 223 patients; 3WINTERS-IPS) and 17 (11.5–25.5; nine children; Sanders 2015) [8, 16]. Because inheritance is recessive, the absence of a family history does not exclude the diagnosis, and an isolated prolonged aPTT on preoperative testing, due to FVIII deficiency, should raise the possibility [16].

Source: table 12 of the type 3 PNDS, which compiles Lak, Federici, Tosetto and Metjian; epistaxis, menstrual bleeding, haemarthroses and gastrointestinal bleeding are excluded from the table; “—” indicates data not reported [16]. The PNDS gives 233 patients for 3WINTERS-IPS in this table, but the original article includes 223 [8].

In the same protocol, epistaxis is the most frequent manifestation of type 3, present in 77% of children, versus 20% of children with severe haemophilia A in the Iranian series; heavy menstrual bleeding is the second symptom in women, with frequencies of 61% to 81%, similar to those of other types but more often severe (transfusion, concentrate, therapeutic amenorrhoea and sometimes haemostatic hysterectomy); gastrointestinal bleeding affects 24% of men and 29% of women; haemarthroses are the second cause of bleeding in men and the fourth in women; and muscle haematomas occur in 28% to 50% of patients, five times more often than in type 1 and less often than in severe haemophilia [8, 16]. Bruising affects 81% of children with type 3 and requires no specific treatment in 75% of cases; oral bleeding (53% to 70%) is the first bleeding site in 54% of children [16]. Haematomas may cause neurovascular compression and compartment syndrome requiring fasciotomy, acute anaemia (psoas haematoma, rarely described in VWD) or airway compromise (tongue base and neck); small-bowel wall haematomas may mimic subocclusion or gastrointestinal lymphoma on imaging; myometrial haematomas present with pelvic pain; and pseudotumours, described in fewer than 2% of patients with severe haemophilia, are exceptional in type 3 [16].

3.3. Female reproductive system

Female reproductive tract bleeding includes heavy menstrual bleeding, postpartum haemorrhage and ovarian bleeding [3]. In the VWDMin trial, the authors state that heavy menstrual bleeding occurs in 80% of women with VWD and is associated with iron deficiency and inadequate response to current treatments [27]. Monthly menstrual bleeding affects quality of life and overall health, and many women develop iron deficiency and anaemia from unrecognised or inadequately treated bleeding [6]. Postpartum bleeding complications are more frequent in women with VWD [6]; obstetric data are presented in section 8.

The French PNDS on heavy uterine bleeding (HUB) defines it as loss exceeding 80 mL per cycle and/or duration exceeding 7 days, or as bleeding that impairs physical, emotional, social or material quality of life; normal duration is 3 to 6 days and mean loss approximately 30 mL, with only 5% of menstrual periods lasting more than 7 days and 0.5% more than 10 [17]. HUB affects 3% to 30% of women of reproductive age, and constitutional or acquired bleeding disorders account for 10% to 20% of cases; among adolescents with HUB and no known disorder, a bleeding disorder is found in 10% to 30%, and VWD is the leading cause, with a prevalence ten times higher than in the general population [17]. Diagnosis is late: on average 4 years after the first bleeding episodes in adolescents and, in women of reproductive age, at around 35 years; median age at diagnosis of constitutional bleeding disorders in Europe is around 16 years, well after menarche (11–12 years) [17]. In the first two years after menarche, 55% to 82% of cycles are anovulatory (30% to 55% between 2 and 4 years and 20% between 4 and 5 years), which aggravates HUB in adolescents with VWD [17]. HUB is associated with regular school absenteeism (more than one day per month in 50% of adolescents in the series cited), restriction of outdoor activities (65%) and of physical education during menstruation (80%), iron-deficiency-related fatigue and reduced sexual activity [17]. Young women with a bleeding disorder also have more haemorrhagic ovarian cysts, sometimes with haemoperitoneum, and mid-cycle ovulatory pain [17].

3.4. Gastrointestinal system: angiodysplasia

Gastrointestinal angiodysplasia occurs in all types, most commonly in adults with type 2A, followed by types 3 and 1; it affects the entire gastrointestinal tract, particularly the colon, small bowel and stomach [3]. Bleeding may begin insidiously, be recognised only through iron deficiency, anaemia or endoscopy, and worsen as lesions progress to life-threatening episodes [3].

The type 3 PNDS cites a 1993 international survey of 297 haemophilia treatment centres, which predates capsule endoscopy and is therefore probably an underestimate: angiodysplasia in 0 of 3593 patients with type 1, 13 of 661 (2%) with type 2, 9 of 198 (4.5%) with type 3 and 7 of 51 (13.7%) with the acquired syndrome, with median ages of 50, 51 and 69 years in the affected groups [16]. This survey diverges from GeneReviews, which places the highest frequency in adult type 2A [3]; in a retrospective study of the VWD International Prophylaxis Study (NCT00557908; 48 patients from 18 centres in 10 countries with unexplained or angiodysplasia-associated gastrointestinal bleeding), type 3 was the most represented subgroup (37.7%) [16]. Lesions predominate in the caecum and ascending colon but occur in any colonic segment, the small bowel and the stomach; they arise mainly after the age of 40, although there are paediatric cases in type 3; and the proposed pathophysiological model is dysregulated angiogenesis due to lack of plasma and intraendothelial VWF, more marked in complete deficiency [16]. Gastrointestinal bleeding may also reveal lesions common in the general population (tumours, ulcers, diverticula) [16].

3.5. Musculoskeletal and central nervous systems

In types 2N and 3, joint and muscle bleeding creates a presentation resembling haemophilia [3, 6]. In a long-term follow-up study cited by the management guidelines, patients with severe VWD developed significant joint damage, affecting quality of life and limiting physical activity through bleeding or fear of bleeding [6]. Intracranial bleeding risk is documented in type 3 [3, 8].

In type 3, FVIII usually lies between 1 and 5 IU/dL, and haemarthroses begin in childhood, mainly in the ankles, knees and elbows; arthropathy affects 30% to 40% of patients (probably an underestimate), generally involves one or two joints, with clinical and radiological scores close to those of moderate haemophilia A and lower than those of severe haemophilia A, in which arthropathy exceeds 50% without early prophylaxis; more than two-thirds of patients with type 3 and chronic arthropathy have a substantial impairment of quality of life, similar to that in severe haemophilia A [16]. A target joint is defined by at least three spontaneous bleeds in 6 consecutive months [16]. Intracranial haemorrhage occurs in 2% to 9% of patients with type 3, five times more often than in type 1, spontaneous or post-traumatic and sometimes fatal; an Iranian review from 2000 found 2%, and at the Toronto centre, among 153 children with VWD (17 with type 3), 24 sustained 33 head injuries, 15 underwent computed tomography and six had intracranial haemorrhage, three of them with type 3, all post-traumatic and followed by long-term prophylaxis [16]. The literature cited contains no cases of spontaneous neonatal haemorrhage, except one secondary to venous sinus thrombosis [16].

3.6. Treatment complications

Development of anti-VWF alloantibodies (inhibitors) is a rare but serious complication, estimated in 5% to 10% of type 3 patients and exceptional in other types [3]. It presents as reduced or absent response to VWF replacement and, rarely, anaphylaxis; risk is higher with large VWF deletions and multiple replacement exposure days [3, 24]. Patients exposed to blood products or plasma-derived concentrates before 1985 should be screened for hepatitis B, hepatitis C and HIV [3]; the Oxford manual recalls that concentrate production from pooled plasma in the 1980s transmitted these infections to many patients with bleeding disorders [21].

The type 3 PNDS devotes a section to thrombotic risk. The risk of arterial or venous thromboembolic events in VWD appears lower than in the general population but higher than in haemophilia; in a study of 47 patients with type 3 and 84 healthy volunteers, intima–media thickness and the proportion of atherosclerotic plaques (19% versus 17%) did not differ, and concentrates did not alter these parameters, so complete VWF deficiency does not protect against atherosclerosis [16]. The cases of venous thromboembolism described in type 3 arose in the setting of repeated concentrates at short intervals and elevated FVIII: a deep vein thrombosis discovered on day 13 after haemorrhoid surgery with FVIII of 248 IU/dL in a plasma-derived concentrate study, an asymptomatic femoral thrombosis on day 4 after total hip replacement under dabigatran with FVIII of 158 IU/dL in a vonicog alfa study, and pharmacovigilance reports of deep vein thrombosis with pulmonary embolism and of central retinal vein thrombosis; in the remaining plasma-derived and recombinant concentrate studies, covering hundreds of bleeding episodes and dozens of surgical procedures, there was no thromboembolism [16]. Hypersensitivity reactions are rare with all concentrates; in type 3, an anaphylactic reaction or loss of efficacy requires testing for anti-VWF alloantibodies, sometimes accompanied by anti-FVIII antibodies [16].

4. Clinical Recognition and Warning Signs

GeneReviews lists findings that should prompt suspicion: excessive bruising, particularly without recognised trauma; prolonged bleeding from skin wounds; prolonged, recurrent or severe epistaxis; gingival bleeding after brushing or flossing, particularly if lifelong or without periodontal disease, and prolonged bleeding after dental cleaning or extraction; excessive female reproductive tract bleeding (heavy menstrual bleeding, postpartum haemorrhage, ovarian bleeding); prolonged bleeding after invasive procedures, surgery or trauma; excessive gastrointestinal bleeding or bleeding associated with angiodysplasia; and, in more severe forms, haemarthroses and arthropathy [3].

Routine investigations are frequently normal. The full blood count may show microcytic iron-deficiency anaemia or thrombocytopenia in type 2B, accompanied by platelet clumps on the blood film. Activated partial thromboplastin time (aPTT) is often normal but becomes prolonged when FVIII falls below approximately 20–30 IU/dL, occurring in some type 1 and 2 patients and all type 2N and 3 patients. Prothrombin time is normal [3]. The Oxford manual emphasises that normal coagulation screening does not exclude VWD [21].

History-taking should be structured. For patients with a low pretest probability (for example, in primary care), the guidelines recommend a validated bleeding assessment tool (BAT) as initial screening to determine who requires specific blood tests (strong recommendation; moderate certainty) [1]. Most BATs derive from the 2005 Vicenza bleeding questionnaire and overlap considerably; ISTH published a clinician-administered consensus BAT (ISTH-BAT), translated into several languages, and a self-administered version (Self-BAT) is available in English and French [1]. For genetics consultations, the Oxford manual recommends a three-generation pedigree with targeted questions about joint bleeding, post-traumatic bleeding and bleeding after surgery, dental extraction or tonsillectomy [21].

Particular warning signs include women with heavy menstrual bleeding, iron deficiency or postpartum haemorrhage [3, 6]; children and men, for whom BAT performance data are more limited and who may not yet have manifested symptoms, meaning that outside primary care BAT should not be the sole criterion for requesting tests [1]; a patient's first-degree relative, whose pretest probability is estimated at approximately 50% and who should be tested regardless of bleeding score [1]; thrombocytopenia worsened by stress or desmopressin, suggesting type 2B or platelet-type VWD [3]; a “mild haemophiliac” with a family history consistent with autosomal recessive inheritance, who may have type 2N [1, 3]; and patients with gastrointestinal bleeding from angiodysplasia or an unexpectedly poor VWF replacement response, raising suspicion of alloantibodies in type 3 [3].

Patients never exposed to haemostatic challenges may lack a suggestive history; an old VWD diagnosis without recent laboratory confirmation should not underpin therapeutic planning for an invasive procedure [3]. All invasive procedures, even those perceived as minor (circumcision, dental and dermatological treatment, piercings), require prior haematology consultation [3].

The French National Diagnosis and Care Protocols (PNDS) of the Haute Autorité de Santé (HAS) add specific signs. The general PNDS calls for testing for VWD in unexplained chronic thrombocytopenia, for review of anticoagulants and antiplatelet agents that may aggravate bleeding, and recalls that incidental diagnosis on preoperative testing is typical of patients with few or no symptoms and moderate deficiency; once a severe form is diagnosed, family investigation begins with systematic testing of first-degree relatives [7]. The attending physician is responsible for recognising the personal and family history, requesting FVIII:C, VWF:Ag and VWF:Act assays with a platelet count, checking that the patient is followed at a competence centre, renewing adjunctive treatments, treating iron deficiency, ensuring vaccination, preventing invasive procedures without a haemostatic plan and avoiding antiplatelet agents and non-steroidal anti-inflammatory drugs (NSAIDs) [7, 16].

In type 3, manifestations appear during periods of motor acquisition (rolling over, creeping, crawling, walking), and the age of walking is a period of particular risk, with bruising, haematomas and post-traumatic haemarthroses, mouth wounds and gingival bleeding at eruption or loss of primary teeth [16]. The type 3 PNDS lists the paediatric signs to look for, with the frequencies from the series it cites: cephalohaematoma from birth (favoured by vacuum extraction but also described after caesarean section), cheek haematoma from sucking (22%), prolonged bleeding after venepuncture (44%), after circumcision (7% and 11% in two series), haematoma after intramuscular vaccination (33% and 5%), petechiae and bruising (20%) and oral bleeding (50% and 54%), as well as conjunctival haemorrhage; bleeding at the fetal scalp electrode site, marked bruising from birth and prolonged bleeding after the newborn heel-prick test (Guthrie test) or after vaccinations are not scored by the ISTH-BAT but are associated with earlier diagnosis when sought [16]. The severity of epistaxis may be underestimated in children because the blood is swallowed and may appear as bloody vomiting or as blood-stained tears [16].

For the ISTH-BAT, the type 3 PNDS regards a score above 2 as abnormal in children, whereas the PNDS on heavy uterine bleeding (HUB) regards scores above 6 in women and above 3 in children as highly suggestive [16, 17]. The HUB PNDS thresholds lie one point above those derived from the normal population: in the reference study (1040 healthy adults and 328 healthy children), the normal range was 0–3 in men, 0–5 in women and 0–2 in children, so the score is abnormal from 4 in men, 6 in women and 3 in children [28]; these are the reference thresholds, and that of the type 3 PNDS (>2 in children) coincides with them. In severe or clinically probable forms, including those without a family history, the type 3 PNDS uses the score mainly for monitoring symptoms rather than for deciding on testing, in line with recommendations 2 and 3 of the 2021 guidelines [1, 16].

HUB is often the first manifestation in girls, and its recognition is difficult: in a series cited by the PNDS, 14% of women with mild to moderate losses considered them heavy and 40% of women with heavy losses considered them normal [17]. The protocol proposes the following signs, which adolescents should learn to recognise before menarche.

Source: [17]. The Higham score covers only 7 days and does not include menstrual cups or period underwear.

The following increase the probability of a bleeding disorder: heavy bleeding since menarche and throughout menstruation, not only during the first 3 days; mid-cycle ovulatory pain, haemorrhagic ovarian cysts or haemoperitoneum; iron deficiency with marked fatigue; physical, emotional, social or material repercussions; extragynaecological bleeding (bilateral epistaxis lasting more than 10 minutes, spontaneous haematomas once or twice a month, bleeding from a small wound for more than 5 minutes, oral or gastrointestinal bleeding without a lesion, unexpected bleeding after dental extraction or surgery, haemarthroses, previous transfusion); and a family history, which may paradoxically delay diagnosis when the adolescent compares her losses with those of an equally affected mother or sister [17].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Principles

Diagnosis is established in a proband with excessive bleeding by demonstrating quantitative VWF deficiency, a qualitative defect in an activity assay or multimer gel, or causative VWF variant(s) [3]. The 2021 guidelines prioritise avoiding missed diagnoses in affected patients to ensure access to care, and considered major bleeding, transfusion and treatment, gastrointestinal bleeding, blood loss, symptom severity, minor bleeding, mortality and unnecessary testing critical outcomes [1]. The overall guideline algorithm (James et al., Figure 1, figure link) begins with BAT, proceeds through full blood count, prothrombin time and aPTT (optional fibrinogen and thrombin time), then VWF:Ag, platelet-dependent VWF activity and FVIII:C, and branches according to VWF levels (>50, 30–50 and <30 IU/dL), activity/antigen ratio (≥0.7 or <0.7) and subtyping tests [1].

The general French PNDS bases laboratory diagnosis on three inseparable assays — FVIII:C, VWF:Act (generic term for VWF:RCo, VWF:GPIbM and VWF:GPIbR) and VWF:Ag — with systematic calculation of the VWF:Act/VWF:Ag and FVIII:C/VWF:Ag ratios and a platelet count, because some forms are accompanied by thrombocytopenia [7]. A VWF:Act and/or VWF:Ag <50 IU/dL, a VWF:Act/VWF:Ag ratio <0.7 or an FVIII:C/VWF:Ag ratio <0.7 justify referral to the competence centre for confirmation and typing; a normal aPTT does not exclude the diagnosis; inflammation and pregnancy raise VWF and may mask the disease, so tests are repeated when suspicion is strong; and the bleeding time is no longer used [7]. The type 3 PNDS requires the three assays to be performed on at least two separate samples and, in line with the 2021 guidelines, recommends VWF:GPIbM or VWF:GPIbR rather than VWF:RCo [1, 16].

5.2. ASH/ISTH/NHF/WFH 2021 diagnosis recommendations

Source: [1].

Associated remarks are integral to the recommendations. Recommendation 1 applies mainly to adult women, the group with the strongest data; across seven cohorts, pooled BAT sensitivity was 0.75 (95% CI 0.66–0.83; 112 patients) and pooled specificity 0.54 (95% CI 0.29–0.77; 863 patients) [1]. Specific tests comprise VWF:Ag, platelet-dependent VWF activity (for example, VWF:GPIbM) and FVIII:C. In recommendation 6, “VWF level” means VWF:Ag and/or platelet-dependent activity; the lower limit of the local reference range should be used if below 0.50 IU/mL; ABO-specific reference ranges are unnecessary; and, because VWF is an acute-phase reactant increasing with bleeding, trauma and pregnancy, investigation should take place at baseline health [1]. The strong recommendation despite low certainty was justified by the need to ensure access to care and international uniformity of criteria [1].

Data supporting the thresholds show that patients with VWF below 0.30 IU/mL had detectable VWF variants in 75% to 82% of cases, compared with 44% to 60% in those with 0.30–0.50 IU/mL; likelihood-ratio analysis yielded 0.73 (0.41–1.30) for the 0.41–0.50 IU/mL range and 0.33 (0.18–0.62) for 0.51–0.60 IU/mL [1]. In recommendation 8, pooled sensitivity of the <0.70 threshold was 0.90 (95% CI 0.83–0.94), compared with 0.58–0.79 for <0.5, with 65 additional false negatives per 1000 patients at the lower threshold in a population with 30% prevalence [1]. In recommendation 9, for types 2A and 2B, the VWF:CB/VWF:Ag ratio had sensitivity 0.90 and specificity 0.95, and multimers sensitivity 0.90 and specificity 0.97; for type 2M, VWF:CB/VWF:Ag had sensitivity 0.98 and specificity 0.99, and multimers 0.86 and 0.97, with the caveat that type 2M is defined by normal multimers [1]. For VWF:RCo, sensitivities and specificities ranged from 0.83–1.00 and 0.87–0.95; for VWF:GPIbR, 0.80–1.00 and 0.81–0.97; for VWF:GPIbM, 0.62–0.82 and 0.90–0.97 [1].

5.3. Laboratory assays and classification

Adapted from GeneReviews Table 1; “low” means <50 IU/dL for VWF:Act, VWF:Ag and VWF:CB, or below the laboratory reference range for FVIII [3].

VWF:Ag is measured by ELISA or latex immunoassay. Platelet-dependent activity assays include VWF:GPIbR (ristocetin-induced VWF binding to GPIb), VWF:GPIbM (binding to a GPIb fragment carrying a gain-of-function variant, without ristocetin) and classical VWF:RCo, with a high coefficient of variation [3]. Ristocetin-dependent assays yield artificially low results in carriers of benign variants interfering with the ristocetin–VWF interaction, such as p.Asp1472His; the guidelines noted that accuracy studies included few patients of African ancestry, a population in which this variant creates a risk of VWF:RCo-based overdiagnosis [1, 3]. FVIII is measured by a one-stage or chromogenic assay. Subtyping uses agarose-SDS multimer analysis, whose protocols capable of showing satellite bands (“triplets”) are not available in all laboratories; ristocetin-induced platelet aggregation (RIPA), in which low-dose (~0.5–0.7 mg/mL) or spontaneous agglutination indicates increased platelet binding (type 2B or platelet-type VWD); VWF:FVIIIB, reduced in 2N; and VWF:CB, whose VWF:CB/VWF:Ag ratio is reduced in types 2A, 2B and 2M—most laboratories use collagen I and/or III, while collagen IV and VI assays, required for some 2M cases, are available only in reference laboratories [1, 3].

The NHF Medical and Scientific Advisory Council highlighted preanalytical variables and falsely low results in improperly processed or transported samples, a documented source of incorrect VWD diagnosis in a cohort of women investigated for bleeding [1]. When the proband's diagnosis is uncertain, the Oxford manual recommends full blood count, coagulation studies, VWF:Ag, VWF:RCo and specific factor measurements, and DNA banking for future studies [21].

The decision tree of the general PNDS (appendix 3, adapted from de Jong and Eikenboom, 2016) starts from three elements — bleeding symptoms, reduced VWF and family history — and branches as follows: undetectable VWF:Ag with absent propeptide (VWFpp) indicates type 3, and with VWFpp present severe type 1; with a VWF:Act/VWF:Ag ratio >0.7, an FVIII/VWF:Ag ratio >0.7 points to type 1 or to “collagen” type 2M (VWF:CB/VWF:Ag <0.7 with high-molecular-weight multimers present), and a ratio <0.7 leads to VWF:FVIIIB, reduced in type 2N and normal in mild or moderate haemophilia A; with a VWF:Act/VWF:Ag ratio <0.7, increased RIPA indicates type 2B or platelet-type VWD and, without it, a VWF:CB/VWF:Ag ratio <0.7 with absent high-molecular-weight multimers indicates type 2A and a ratio >0.7 with multimers present indicates type 2M; molecular biology confirms the type [7]. Second-line tests (VWFpp, multimers, VWF:CB, VWF:FVIIIB and VWF binding to recombinant GPIb) and VWF analysis are performed at the competence centre [7].

In type 3, patients generally have VWF:Ag and VWF:Act <5 IU/dL and FVIII:C <10 IU/dL, often <2–3 IU/dL and <5 IU/dL; because VWF may also fall below the method's detection limit in severe type 1, an FVIII:C <10 IU/dL points towards type 3, and FVIII:C correlates with severity and with age at first symptoms [16]. The method's lower limit of detection determines whether an “undetectable” value corresponds to absence of VWF or merely to a value below the threshold.

Source: table 2 and text of the type 3 PNDS [16]. The protocol expresses coefficients of variation in IU/dL. The table assigns ELISA a limit of 5–15 IU/dL, but the text states that ELISA and chemiluminescence have quantification thresholds below 1 IU/dL and are the only methods that confirm type 3.

5.4. Type 1C and the desmopressin trial

A desmopressin trial with samples before, approximately 30–60 minutes after and approximately 4 hours after administration detects accelerated clearance; a fall exceeding 30% from the peak at 4 hours identifies these patients [1, 6]. The simpler VWFpp/VWF:Ag ratio is unavailable in most laboratories and may be normal in patients with rapid clearance; it may be useful when a trial is inadvisable (very young or very old patients) [1]. Response definition and trial technique are described in section 7.

The type 3 PNDS regards VWFpp measurement as the most discriminating test between type 3 and severe type 1. VWF and its propeptide are secreted in a 1:1 ratio, but VWFpp has a half-life of 2–3 hours, versus 10–14 hours for mature VWF; in type 3 VWFpp is undetectable, whereas in severe type 1 with VWF:Ag <10 IU/dL it remains above that value [16]. The VWFpp/VWF:Ag ratio is generally <3 in healthy individuals, >3 in type 1C and >5 in “Vicenza” type [16]. On multimer electrophoresis, type 3 shows no bands at all, whereas severe type 1 retains detectable multimers even with very low VWF:Ag; sampling should take place more than 5 days after any replacement; and VWF:CB is <5 IU/dL in type 3, with ELISA or chemiluminescence assays with a limit below 1 IU/dL [16]. The PNDS uses the VWFpp/VWF:Ag ratio to identify type 1C, whereas the 2021 guidelines prefer the desmopressin trial with sampling at 4 hours (recommendation 7) [1, 16].

5.5. Subtypes 2B and 2N: specific algorithms

For type 2B, the guidelines prefer targeted genetic testing of exon 28, which contains the known variants (pooled sensitivity 1.00 for genetics and 0.99 for RIPA, with no specificity available) [1]. The Figure 2 algorithm (figure link) specifies: positive genetic testing for 2B variants—diagnose 2B; negative testing—perform RIPA; positive RIPA—investigate platelet GPIb variants or other bleeding disorders; negative RIPA—type 2A [1]. RIPA requires a fresh sample that cannot be shipped, and its methodology is not standardised [1].

For type 2N, the Figure 3 algorithm (figure link) allows either VWF:FVIIIB or genetics first: abnormal VWF:FVIIIB leads to testing for 2N variants; normal VWF:FVIIIB with low FVIII leads to F8 testing; a novel VWF variant requires functional confirmation by VWF:FVIIIB [1]. Genetics is more informative for counselling because patients may be homozygous for a 2N variant, compound heterozygous for two 2N variants, or carry a 2N variant with an unidentified null allele—in which case a child may be an asymptomatic heterozygote or have type 1 [1].

5.6. Genetic testing: indications, yield and limitations

Molecular testing helps establish diagnosis when quantitative and qualitative tests are ambiguous or contradictory and helps classify subtype [3]. Identification of a heterozygous (or rarely biallelic) variant establishes diagnosis in types 1C, 2A, 2B and 2M, and biallelic variants in types 2N and 3; in type 1, the proportion of patients with an identified variant is lower, but testing remains informative [3].

Proportion of causative variants detected by each method, according to GeneReviews Table 2 [3]. These figures describe the proportion of detectable variants in cohorts with an established phenotype, rather than clinical test sensitivity in suspected cases without laboratory confirmation.

Sequencing detects missense, nonsense and splicing variants and small insertions/deletions, and should include regulatory regions and the promoter; large deletions, duplications, structural variants, pseudogene conversions and deep intronic variants are often missed, so a convincing phenotype with negative sequencing is followed by rearrangement analysis (quantitative or long-range PCR, next-generation data analysis, MLPA, comparative genomic hybridisation or targeted microarray) [3]. Analysis of exons 23–34 is complicated by the VWFP1 pseudogene [3]. A panel including VWF, F8 and GP1BA helps distinguish VWD from haemophilia A and platelet-type VWD, but panels usually do not detect recurrent F8 intron 1 and 22 inversions [3]. Some laboratories may not report the common benign p.Asp1472His variant [3]. Testing should be performed in a laboratory experienced in VWD, and the report should specify gene, versioned transcript (NM_000552.5), HGVS nomenclature, zygosity, ACMG/AMP classification and methodological limitations. A variant of uncertain significance does not become pathogenic through treatment response or partial segregation; familial segregation and functional assays are tools for investigating such variants. DNA banking should be considered when molecular diagnosis is unconfirmed or incompletely explains the phenotype [3].

The type 3 PNDS regards VWF analysis as necessary in all patients, with five aims: confirming the diagnosis through genotype–phenotype correlation, demonstrating that the parents are heterozygous, enabling prenatal diagnosis (recurrence risk of 25%), testing relatives and estimating the risk of alloantibody, which is greater when there is a gene deletion [16]. Variants are distributed throughout the gene and give rise to null alleles in approximately 80% of cases (splicing and nonsense variants, small and large deletions or insertions, gene conversions) and missense variants in the remaining 20%; GeneReviews estimates approximately 25% missense alleles [3, 16]. In approximately 15% of patients only one or no variant is identified [16]. Next-generation sequencing has replaced targeted Sanger sequencing, which is used to confirm point variants; large rearrangements are confirmed by MLPA and/or quantitative PCR; parental testing confirms that the variants lie on different alleles and may suggest uniparental disomy; and microsatellite and VNTR markers within VWF are used to confirm disomy or to establish the parental origin of the variants in prenatal diagnosis [16].

5.7. Reassessment with age and historical diagnoses

Six observational studies indicate that approximately 43% of patients with VWD have VWF levels that normalise with age; a cross-sectional donor study found an annual increase of 0.01 IU/mL between ages 20 and 60; no study longitudinally assessed whether bleeding phenotype improves [1]. Recommendation 5 therefore proposes reconsidering the diagnosis (for example, investigating concomitant platelet dysfunction) rather than removing it, through shared decision-making [1].

5.8. Differences between guidelines

The 2008 National Heart, Lung, and Blood Institute (NHLBI) guidelines used a 30 IU/dL threshold for definitive diagnosis without excluding diagnosis at 30–50 IU/dL with clinical or familial evidence; the 2014 UK Haemophilia Centre Doctors' Organisation (UKHCDO) guidelines propose regarding patients with bleeding and VWF 0.30–0.50 IU/mL as having “low VWF”, a risk factor rather than VWD [1]. The 2021 panel recommended type 1 diagnosis in this range when abnormal bleeding is present, concerned that the “low VWF” designation could impede treatment, particularly in countries without universal coverage; it debated and rejected diagnosis based solely on positive family history in the same range because of inaccurate diagnosis risk [1].

5.9. Heavy uterine bleeding in young women: investigation

The HUB PNDS distinguishes acute bleeding, which requires immediate assessment of severity and possible admission, from chronic bleeding, which may lead to severe anaemia without active bleeding; a bleeding disorder is found in 20% to 30% of adolescents with HUB [17]. First-line assessment comprises a full blood count with platelets and reticulocytes, ferritin, prothrombin time, aPTT, fibrinogen, VWF antigen and activity, FVIII, blood group and C-reactive protein (to interpret VWF), βhCG, TSH and prolactin; blood pressure and heart and respiratory rates are measured, and signs of poorly tolerated anaemia and iron deficiency are sought (pallor, fatigue, dyspnoea, tachycardia, dizziness, headache, hair loss, brittle nails, restless legs) [17]. Because stress, severe bleeding and high-dose oestrogens alter VWF, the assay is repeated outside the episode when it is abnormal or when the phenotype is severe (Higham >150 per cycle, ISTH-BAT >6) [17].

Even with normal first-line tests, the young woman should be referred to a haemostasis specialist if she has had at least one acute episode (emergency attendance, acute anaemia, worsening of chronic anaemia, transfusion), a family history of HUB or of a bleeding disorder, or extragynaecological bleeding; any coagulation abnormality also justifies referral [17]. Second-line testing, performed in a specialised laboratory, confirms or refutes the abnormalities, characterises the disorder and, if the phenotype and family justify it, looks for rarer disorders: platelet function, fibrinolysis, prothrombin consumption and chromogenic FXIII [17]. Iron deficiency, the leading cause of anaemia in women, is always sought with haemoglobin, ferritin and transferrin saturation (useful when ferritin is falsely normal because of inflammation), and dietary assessment looks for deficiencies of iron, vitamin C and folate, frequent with selective, vegetarian or vegan diets [17].

Pelvic ultrasound excludes structural causes in the PALM-COEIN classification (uterine malformation, polyps, fibroids) and measures endometrial thickness, distinguishing hyperplasia due to relative hyperoestrogenism from atrophy; in girls who are not sexually active it is performed suprapubically with a full bladder by a trained operator, and magnetic resonance imaging is reserved for uncertain cases (fibroid versus adenomyosis, the latter exceptional in adolescents) [17]. An intrauterine blood collection may sustain bleeding after the acute episode [17]. In adolescents, ovulatory dysfunction due to immaturity of the hypothalamic–pituitary–ovarian axis explains most cases of HUB, and coagulopathy is the leading non-structural cause [17].

6. Differential Diagnosis

Table sources: [3, 7, 16, 17, 21, 29–35].

Distinguishing type 2N from mild haemophilia A is particularly relevant to geneticists: low FVIII:C and normal to low VWF occur in both, and only inheritance (autosomal recessive versus X-linked), VWF:FVIIIB or genetics differentiates them [1, 3]. Haemophilia A patients receive FVIII concentrate; type 2N patients require VWF replacement to prevent or treat severe bleeding, and FVIII alone has a short half-life without VWF binding [1, 20].

Platelet-type VWD is a “non-identical twin” of type 2B: the GP1BA variant confers excessive GPIbα affinity for VWF, with clearance of platelets and high-molecular-weight multimers; nine causative variants are known [30]. Distinction changes treatment—platelet transfusion in platelet-type disease and VWF/FVIII concentrate in type 2B—although VWF has a shortened half-life in platelet-type disease and VWF replacement may also be necessary [3, 30]. In an international study of 110 samples from eight countries with suspected 2B or platelet-type disease, 48 had exon 28 variants consistent with 2B, 17 had GP1BA variants, three had other types (2A and 2M), and 26 had no variant in either gene; the authors estimated a 15% misdiagnosis rate for platelet-type disease [29]. ISTH published specific guidance on platelet-type disease, cited by the 2021 guidelines [1].

Acquired von Willebrand syndrome denotes deficiencies in VWF concentration, structure or function that are neither inherited nor caused by VWF variants; anti-VWF antibodies have been detected in some lymphoproliferative disorders [3]. In aortic stenosis, shear-induced multimer loss is associated with angiodysplasia bleeding (Heyde syndrome) [6].

The Oxford manual includes factor VII deficiency (autosomal recessive, approximately 1/500,000), factor X deficiency (autosomal recessive, approximately 1/100,000) and factor XI deficiency in the differential of inherited bleeding disorders; factor XI deficiency is particularly frequent in Ashkenazi Jews, with severe deficiency in 1/190 and approximately 8% carriers [21]. For Noonan and cardiofaciocutaneous syndromes, Smith's book describes a bleeding diathesis including VWD [32, 33]; in glycogen storage disease type I, some patients have reduced or dysfunctional VWF and coagulopathy may improve with metabolic control, particularly relevant before surgery [34]. The Brazilian book on genetic syndromes lists VWD among causes of aPTT prolongation to distinguish from haemophilia A [35].

The general PNDS summarises the differential diagnosis in three entities: acquired von Willebrand syndrome (associated mainly with monoclonal gammopathy or advanced valvular disease), mild or moderate haemophilia A when the deficiency predominantly affects FVIII, and platelet-type VWD [7]. The HUB PNDS adds myeloproliferative and lymphoproliferative syndromes, nephroblastoma, hypothyroidism, severe aortic stenosis and circulatory support devices (extracorporeal membrane oxygenation, Impella) to the causes of the acquired syndrome, regards it as exceptional in young women and recalls that mucocutaneous bleeding without any personal or family history should prompt a search for an acquired cause first [17]. The same protocol includes Quebec platelet disorder, due to platelet release of urokinase-type plasminogen activator, among the hereditary fibrinolytic abnormalities that may cause menorrhagia [17].

7. Treatments and Risk Management Strategies

7.1. Principles and organisation of care

Treatment aims to prevent and manage bleeding and enable a full, active life, ideally through a multidisciplinary bleeding disorders programme [3]. Options comprise three classes: measures increasing endogenous VWF (desmopressin), replacement with exogenous VWF concentrates and general haemostatic adjuncts (antifibrinolytics, hormone therapy and local measures) [19]. Most patients require only on-demand treatment; prophylaxis is reserved for severe, frequent bleeding [6, 19]. Treatment depends on type: applying another type's treatment can be dangerous, as with desmopressin in type 2B or isolated FVIII concentrate in type 2N [1, 3].

Table sources: [3, 6].

In France, diagnosis and treatment are coordinated by the haemostasis physicians of the 27 resource and competence centres for constitutional bleeding disorders (CRC-MHC) and by the Centre de Référence de la Maladie de Willebrand (CRMW; coordinating site in Lille and constituent sites in Paris-Lariboisière and Caen), associated with the reference centres for haemophilia and platelet disorders in the MHémo rare disease health network, with haemophilia treatment centres integrated into the network; difficult type 3 cases may be discussed at a national multidisciplinary consultation meeting organised by the network several times a year [7, 16]. At the disclosure consultation, held in person and never by telephone by the centre physician, ideally with the nurse and, in type 3, with the psychologist, the patient receives the Ministry of Health care and emergency card, stating the VWD type, the appropriate treatments and emergency contacts, and a booklet in which to record the dates, times, reasons, products and batch numbers of injections (affixing the labels); the first application for recognition as a long-term condition (ALD 30), which exempts from co-payment, is then made [7, 16]. The general PNDS defines three levels of strategy according to severity: local and adjunctive measures; desmopressin or replacement for spontaneous bleeds that exceed local measures and for invasive procedures; and regular prophylaxis for recurrent and severe bleeding (epistaxis, gastrointestinal bleeding, haemarthroses) [7]. Treatments outside the marketing authorisation require informing the patient, a statement on the prescription and justification in the clinical record, and the French medicines agency (ANSM) may draw on the PNDS for a temporary recommendation for use [7].

7.2. Desmopressin

Desmopressin (1-deamino-8-D-arginine vasopressin, DDAVP), a synthetic vasopressin analogue, releases endothelial stores of VWF and FVIII and increases their levels three- to fourfold [3, 19]. For patients in whom it is a valid option (mainly type 1) with baseline VWF below 0.30 IU/mL, the guidelines suggest undertaking a desmopressin trial and treating according to its results rather than empirically with tranexamic acid or concentrate (recommendation 2a), and avoiding desmopressin treatment without trial results (2b), both conditional with very low certainty [6]. Adults with type 1 and baseline VWF ≥0.30 IU/mL may be presumed responsive, with post-administration measurement reasonably confirming response [6]. Desmopressin should never be sole therapy for major surgery, including sites where minor bleeding can damage a critical organ, such as the central nervous system [6].

The US label indication for both injectable DDAVP® and nasal Stimate® is restricted to mild to moderate type I VWD with FVIII >5%, excluding severe type I and forms with a molecular abnormality of FVIII antigen; the injectable label requires FVIII testing and exclusion of severe forms before treatment, and monitoring of sodium, FVIII, VWF:RCo and VWF:Ag [13, 14]. Injectable DDAVP® carries a boxed warning for hyponatraemia, which can cause seizures, coma, respiratory arrest or death, and specifies sodium measurement within 7 days and approximately 1 month after starting therapy, then periodically [13]. Use in types 2A, 2M and 2N and in children under 2 falls outside the label wording or conflicts with guidelines; the injectable label considers safety established from 3 months and the Stimate® label from 11 months, whereas guidelines and GeneReviews advise against desmopressin before age 2 because fluid restriction is difficult [3, 6, 13, 14]. GeneReviews also lists contraindication above age 70 and in arteriovascular disease [3]. In 2020, the nasal spray was withdrawn from the market because of manufacturing problems, and compounding pharmacies in the USA and Netherlands maintained limited supplies with short shelf lives [19]; the Stimate® label remains published (22 July 2026 version), but the consulted sources do not establish current commercial availability [14].

In France, the general PNDS refers to intravenous desmopressin (Minirin®) and the nasal spray (Octim®): after administration, VWF and FVIII rise two- to fivefold within 30 to 60 minutes and return to baseline within 6 to 9 hours; the response is reproducible in the same individual but variable between individuals, including within the same family, and may be short-lived in some forms; and the effect wanes with repeated administration over a short period [7]. The therapeutic trial is performed after diagnosis and before first clinical use, in hospital, at a centre with a haemostasis laboratory; the nasal spray may be used at home if the patient follows the centre's instructions and fluid restriction; desmopressin is indicated only from the age of 2 years and is ineffective in type 3 [7, 16]. In women who respond, it may be used on the first day of menstruation [7].

In series underpinning the recommendation, the trial-and-result-guided treatment strategy achieved excellent, good or effective haemostasis in 94% of 211 operations and 97% of 29 bleeding episodes; postoperative bleeding occurred in 6% of 199 operations; in one series, 10 of 41 patients vomited (five hospitalised) and one developed hyponatraemia; in a series of 37 children, all developed some hyponatraemia, two severe and one with seizures [6]. With empirical desmopressin, hyponatraemia was reported in 4% to 72% of patients, using varying definitions, and mild to moderate adverse effects (headache, flushing, nausea, dizziness, asthenia, vomiting, peripheral oedema) in 43% with mild type 1 and 14% with moderate type 1 [6].

7.3. VWF concentrates

Plasma-derived or recombinant VWF replacement is the mainstay of treatment [3]. Plasma-derived concentrates are prepared from pooled donations and undergo pathogen inactivation such as pasteurisation and detergent treatment; VWF/FVIII ratios differ: approximately 2:1 for Humate-P®, 1:1 for Wilate®, 1.58:1 for Fanhdi®, and 50:1 for Wilfactin®, available only in Europe [3, 19]. Voncento® has an FVIII:C/VWF:RCo ratio of approximately 1:2.4 [36]. Vonicog alfa (recombinant VWF; Vonvendi® in the USA, Veyvondi® in the EU) contains no FVIII; because it has not been exposed to ADAMTS13, it is enriched in high-molecular-weight multimers; plasma-derived concentrates have a half-life of approximately 10–12 hours, while recombinant VWF approaches 20 hours, with substantial individual variability; because it is not derived from human blood, it is often acceptable to people declining blood products, such as Jehovah's Witnesses [19]. For all these products, the management guidelines give a typical intravenous dose of 40–80 VWF:RCo units/kg, noting that FVIII-free products may require additional FVIII during emergency treatment when baseline FVIII is low [6]. In types 2N and 3, a product containing VWF and FVIII is initially preferred [3].

Table sources: [9–12, 15, 36–39]. The consulted EMA medicine registry lists only Veyvondi® and Voncento® among centrally authorised medicines indicated for VWD; other concentrates available in Europe depend on national authorisations, whose summaries of product characteristics were not consulted for this article [15]. Authorisation does not guarantee marketing or reimbursement in each country.

Each Vonvendi® vial is labelled in VWF:RCo IU; endogenous FVIII rises above 40% approximately 6 hours after a single infusion, depending on baseline level; for elective surgery, a dose may be given 12–24 hours beforehand to raise endogenous FVIII to at least 30 IU/dL (minor surgery) or 60 IU/dL (major surgery); peak targets are VWF:RCo 50–60 IU/dL and FVIII 40–50 IU/dL for minor surgery, and VWF:RCo 100 IU/dL and FVIII 80–100 IU/dL for major surgery, with VWF:RCo troughs >50 IU/dL up to 72 hours and >30 IU/dL thereafter in major surgery, for at least 72 hours [9]. The European label uses preoperative FVIII targets of at least 0.4 IU/mL for minor and oral surgery and 0.8 IU/mL for major surgery [38]. The Veyvondi® SmPC states that endogenous FVIII remains normal or near normal while VWF administration continues [38].

The most common vonicog alfa adverse effects in trials (≥2%) were headache, vomiting, nausea, dizziness and generalised pruritus; labels warn about thromboembolic events (particularly with persistently elevated FVIII following repeated recombinant FVIII coadministration and in patients with risk factors, including low ADAMTS13), hypersensitivity reactions (the product contains traces of mouse and hamster proteins), and neutralising antibodies against VWF or FVIII, to be investigated when expected levels are not achieved or bleeding is uncontrolled [9, 38]. For plasma-derived concentrates, the commonest adverse reactions were hypersensitivity (urticaria, chest discomfort, pruritus, oedema) and, during surgery, wound and injection-site bleeding [10, 11]. FVIII-containing concentrates can excessively elevate FVIII:C during prolonged treatment, so dose reduction, interval extension or use of a low-FVIII product should be considered after 24–48 hours [36].

The type 3 PNDS describes the medicinal products available in France in 2021, all with a marketing authorisation, and groups them by FVIII content, which determines correction of the secondary FVIII deficiency and the risk of accumulation: without FVIII (Wilfactin® and Veyvondi®), with intermediate concentration (Voncento® and Wilstart®) and with high concentration (Eqwilate®, the French name for Wilate®) [16].

Source: tables 5 and 6 of the type 3 PNDS, which summarise the French summaries of product characteristics (SmPCs) [16]. The current European indications of Veyvondi® and Voncento® appear in the preceding table; since 2021, Veyvondi® has also covered treatment of bleeding in children [38]. According to the SmPCs, 1 IU/kg of VWF raises VWF:RCo by approximately 2 IU/dL [16].

All concentrates available in France contain high-molecular-weight multimers, in variable quantity and quality; the plasma-derived products lack the ultra-high-molecular-weight multimers of the recombinant product, which has not been exposed to ADAMTS13 — a feature that might confer additional efficacy, for example in angiodysplasia, but also a greater thrombotic risk, with no data confirming either hypothesis [16]. Wilfactin® and Eqwilate® contain very little ADAMTS13 and retain the triplet structure of normal VWF, whereas Voncento® contains approximately 0.9 IU/mL of ADAMTS13 and, like Haemate P®, an excess of fast-migrating bands [16]. In the pivotal studies compiled by the PNDS, good or excellent efficacy in treating bleeds was 89% with Wilfactin® (20 patients, 35% with type 3; median dose 41.8 IU/kg; median of 3 injections), 98% with Voncento® (22 patients, 59% with type 3; 74 IU/kg; 2 injections), 96% with Eqwilate® (44 patients, 55% with type 3; 29 IU/kg; 1.93 injections) and 100% with Veyvondi® (39 patients, 29 of 37 evaluated with type 3; 46.5 IU/kg; 1 injection); in surgery, efficacy was 96% to 100%, with type 3 accounting for 20% to 59% of procedures and one thrombotic event in 100 procedures with Voncento® and in 15 with Veyvondi® [16]. In elective surgery with FVIII-free concentrates, injections 12–24 hours beforehand raise endogenous FVIII at the time of the procedure [16].

The PNDS stresses the risk of FVIII accumulation: after a VWF injection, endogenous FVIII is stabilised by the injected VWF and, with FVIII-containing concentrates, exogenous FVIII is added, which with repeated injections leads to high values [16]. FVIII:C values >150 IU/dL are found in 25% of patients with venous thrombosis and in 11% of the general population; in the MEGA study, in the surgical setting, the relative risk of venous thrombosis tripled with FVIII of 109–138 IU/dL and increased fivefold with 139–173 IU/dL [16]. The protocol recommends FVIII:C measurements to avoid values >150 IU/dL, preference for the concentrate with less FVIII and particular care in older people, patients with cancer or cardiovascular disease, and in pregnancy and the postpartum period [16].

Anti-VWF alloantibodies arise in 5% to 10% of patients with type 3, mainly in those homozygous for complete gene deletion, practically never in other forms, and may be associated with anti-FVIII antibodies; an intolerance or anaphylactic reaction or loss of efficacy requires prompt inhibitor testing [7, 16]. The possible treatments are off-label uses, to be decided case by case with the reference centre's opinion recorded in the clinical record [16]:

Source: [16]. When bleeding persists despite adequate circulating VWF levels, adding platelet concentrates, a source of VWF rich in high-molecular-weight multimers, has been proposed [16].

7.4. Prophylaxis

Recommendation 1 suggests long-term prophylaxis in patients with a history of severe, frequent bleeding (conditional; low certainty), with periodic reassessment of need [6]. In the only randomised trial (PRO.WILL; 19 patients; Fanhdi®/Alphanate®), prophylaxis reduced bleeding episodes (rate ratio 0.24; 95% CI 0.17–0.35), prolonged time to first bleeding by 31.4 days and reduced epistaxis (0.38; 0.21–0.67); the prophylaxis group had more bleeding lasting over 2 days and more gastrointestinal bleeding, concentrated mainly in one patient [6, 40]. In observational studies with explicit comparisons, prophylaxis reduced bleeding episodes (0.34; 0.25–0.46), hospitalisations (0.64; 0.44–0.93) and heavy menstrual bleeding [6]. For the review, prophylaxis was defined as at least 6 months of VWF replacement at least once weekly; later proposed definitions accept at least 3–6 months and, for heavy menstrual bleeding, at least one administration per cycle [6, 19].

The general French PNDS defines long-term prophylaxis as the administration of VWF at home, by a nurse or by self-treatment, at least once a week for at least 45 weeks a year, or intermittently during menstruation; the more recent type 3 PNDS adopts the definition of the 2021 guidelines (at least 6 months, at least once a week, or once per cycle in HUB) [7, 16]. In France, prophylaxis is indicated in children with type 3 from the first haemarthrosis or first few haemarthroses, because of the risk of arthropathy, and in other patients according to the 2015 VWD Prophylaxis Network criteria; severe recurrent epistaxis, recurrent gastrointestinal bleeding, angiodysplasia and HUB in adolescents may also justify it [7, 16]. Regimens use 30–50 IU/kg one to three times a week, usually starting with one weekly infusion and increasing the frequency stepwise in the event of further bleeding; an implantable port may be required when venous access is difficult; and efficacy, tolerance and benefit are reassessed at least every 6 months, with an outpatient blood count at the start [7, 16]. Prophylaxis reduces bleeding frequency to an extent broadly comparable with haemophilia but is less effective for gastrointestinal bleeding, which often requires higher doses and injections every 2–3 days [7, 16]. Primary prophylaxis, before any haemarthrosis and before the age of 2 years, is not usual in type 3, because some patients bleed little at first, although Swedish experience shows benefit of early prophylaxis in children with haemarthroses [16]. The expected risks in children are anti-VWF or anti-FVIII alloantibodies, allergic reactions, poor tolerance of injections and difficult venous access, and initiation is accompanied by a therapeutic education programme to make the patient autonomous [16].

For severe epistaxis, the type 3 PNDS cites a stepwise regimen from a North American series of 11 patients with severe VWD (five with type 3), in whom epistaxis prompted prophylaxis in six (two with type 3): 50 IU/kg once a week, moving to twice and then three times a week whenever epistaxis persisted with need for iron, transfusion, packing or admission [16]. In severe HUB or in type 3, replacement may be limited to the days of heaviest loss, generally the first to the third day; HUB prompts prophylaxis in approximately 5% of patients, without relevant change in the number of bleeds (4% before and 3% after in type 3), which reflects the efficacy of non-specific treatments [16, 17].

For haemarthrosis in type 3, the PNDS recommends VWF 40–80 IU/kg combined with FVIII 20–40 IU/kg at the first injection, followed by VWF (with or without FVIII) daily or every 48 hours according to progress; secondary prophylaxis, indicated in repeated haemarthroses and particularly with a target joint, reduces the number of haemarthroses by more than 80%, with 25–40 IU/kg one to three times a week adjusted to efficacy and adherence [16]. In disabling chronic synovitis, chemical, radioisotopic or surgical synovectomies may be offered; in advanced arthropathy, synoviortheses, viscosupplementation with hyaluronic acid and, in advanced cases, arthroplasty; there are no data on platelet-rich plasma or mesenchymal stem cells in this population [16]. The physiotherapist detects the first musculoskeletal signs, applies the RICE protocol in the acute phase (rest for 24–72 hours, ice in periods of 15–20 minutes several times a day, compression, elevation above the heart), immobilises and offloads lower-limb haemarthroses, restores range of motion, strength and proprioception in the subacute phase, treats postural abnormalities (flat foot, genu valgum) and corrects the sequelae of chronic arthropathy; assessment uses the Hemophilia Joint Health Score (HJHS; elbows, knees and ankles, more than 2 weeks after a bleed; validated in VWD), the Haemophilia Activity List (HAL) and joint ultrasound, which shows effusion, synovial thickening and complete resorption of blood [16].

7.5. Surgery and invasive procedures

For major surgery, the guidelines suggest maintaining FVIII and VWF activity ≥0.50 IU/mL for at least 3 days (recommendation 4a) and advise against targeting only FVIII ≥0.50 IU/mL (4b), with targets individualised to patient, procedure and laboratory availability [6]. Major surgery included procedures entering large cavities, those with severe bleeding risk, joint surgery, third-molar extraction and life-threatening interventions [6]. The 2008 NHLBI guidelines recommended maintaining VWF and FVIII above 0.50 IU/mL for 7–10 days, while the 2014 UK guidelines discussed evidence for 6 days and 7–10 days without specifying a duration [6]. The 2025 review adds that, in major bleeding, aiming for FVIII and VWF peaks of 0.80–1.00 IU/mL and maintaining troughs ≥0.50 IU/mL until haemostasis is reasonable [19]. Where real-time laboratory monitoring is unavailable, high-risk elective procedures may require transfer, or a pharmacokinetic study of the planned dose can be performed beforehand [6].

For minor surgery or minor invasive procedures, the guidelines suggest raising VWF to ≥0.50 IU/mL with desmopressin or concentrate and adding tranexamic acid (recommendation 5a): raising VWF alone was associated with greater postoperative bleeding risk (risk ratio 6.29; 95% CI 2.12–18.65) than combined treatment in two randomised haemophilia trials used as indirect evidence [6]. For type 1 with baseline VWF >0.30 IU/mL and a mild phenotype undergoing a minor mucosal procedure, tranexamic acid alone is suggested (5b) [6]. Remarks add that type 3 requires concentrate, many type 2 patients (including 2B) require concentrate rather than desmopressin, prolonged combined VWF and FVIII >1.50 IU/mL with prolonged tranexamic acid should be avoided in those with greater thrombotic risk, and gelatin sponges, fibrin glue or tranexamic acid rinses can be used in dentistry [6]. ε-Aminocaproic acid is a reasonable alternative, particularly as a dental rinse or paediatric liquid formulation [6].

In type 3, the PNDS requires multidisciplinary assessment of any invasive procedure, at a centre with a haemostasis laboratory available 24 hours a day, and a written protocol drawn up by the haemostasis physician with the operator and the anaesthetist; some authors propose a prior study of recovery and pharmacokinetics of the concentrate, which also allows an alloantibody to be detected, and testing for one is mandatory before surgery in the event of poor efficacy or poor recovery [16]. With FVIII-free concentrates (Wilfactin®, Veyvondi®), FVIII is corrected with a single FVIII dose of 30–50 IU/kg, according to baseline level, together with the first VWF injection, or with a VWF injection 12–18 hours before surgery, which stabilises endogenous FVIII, followed by a further injection immediately beforehand; in patients on prophylaxis, preoperative FVIII is checked to decide on a supplement [16]. Oral tranexamic acid 1 g four times a day (20 mg/kg/day in 2–3 doses in children) is added for 7 days, and antiplatelet agents and NSAIDs are contraindicated [16].

Source of the compilation: [6, 16]. The PNDS attributes to the 2021 guidelines VWF:Act targets of >50 IU/dL in minor surgery and 100 IU/dL in major surgery, understood as peak and minimum trough; the published management guidelines state the major surgery target as ≥0.50 IU/mL for at least 3 days.

In high-risk surgery, VWF:Act and FVIII are measured twice on the first day and then once a day before the injection, without waiting for the result to inject, and adjustment applies to subsequent doses; postoperative VWF:RCo values >200 IU/dL and FVIII:C values >200–250 IU/dL are avoided [16]. Peripheral regional anaesthesia may be performed, after weighing the benefit, by an experienced anaesthetist under ultrasound guidance to avoid vascular puncture, without requiring supratherapeutic levels; neuraxial anaesthesia is considered only with normalised VWF and FVIII, case by case, and there is no published experience in type 3 [16]. Venous thromboprophylaxis follows the 2018 European Society of Anaesthesiology recommendation (monitor FVIII and prescribe thromboprophylaxis whenever there is another patient-related or surgical risk factor; grade 2C), detailed by the Groupe d'Intérêt en Hémostase Périopératoire, and, in bariatric or orthopaedic surgery, the Société Française d'Anesthésie et de Réanimation dosage identical to that for patients without VWD, because the replaced patient is postulated to have the thrombotic risk of the general population [16]. Low-molecular-weight heparin is used; direct oral anticoagulants are not recommended owing to lack of experience in constitutional bleeding disorders, fondaparinux is discouraged because of its long half-life, and intermittent pneumatic compression is added when pharmacological prophylaxis is unfavourable or the risk is very high [16]. The PNDS lists the risk factors to be weighed (previous thromboembolism, major thrombophilia, age >70 years, body mass index >30 kg/m², active cancer or chemotherapy, heart or respiratory failure, hormonal or oestrogen therapy, recent neurological deficit, postpartum up to 2 months, immobility, acute illness, inflammatory bowel disease, nephrotic syndrome, myeloproliferative syndrome, smoking, varicose veins, central venous catheter) and classifies surgery as low thrombotic risk (arthroscopy, foot surgery, appendicectomy, non-inflammatory cholecystectomy, proctology, breast surgery, laparoscopy <60 min) or high thrombotic risk (hip or knee arthroplasty, femoral neck or tibial plateau fractures, major abdominal surgery, hysterectomy by laparotomy, bariatric surgery, nephrectomy, neurosurgery, duration >120 min) [16]. In general practice, the 2021 PNDS requires high-risk procedures to be performed at a centre with 24-hour clinical and laboratory monitoring and access to blood products, and allows prolonged admission in severe patients [7].

7.6. Cardiovascular disease and antithrombotics

For patients with VWD and cardiovascular disease requiring antiplatelet agents or anticoagulants, the guidelines suggest providing the necessary therapy (recommendation 3; conditional, low certainty), with ongoing bleeding-risk reassessment, multidisciplinary decisions involving cardiology and possible VWF concentrate prophylaxis for severe phenotypes; desmopressin is generally contraindicated in cardiovascular disease or increased thrombotic risk [6]. Limiting antiplatelet treatment duration should be considered, for example using bare-metal stents [6]. Extrapolating from haemophilia without specific data, the 2025 review considers maintaining VWF activity ≥0.05–0.10 IU/mL for single antiplatelet therapy and FVIII ≥0.20–0.30 IU/mL for dual antiplatelet therapy or anticoagulation reasonable [19].

The type 3 PNDS stresses that, unlike moderate type 1, severe deficiency does not improve with age, and that the comorbidities of ageing require decisions shared with the centre [16]. The protective effect of severe deficiency on arterial thrombosis is debated: mouse and pig models suggest less atherosclerosis, but the intima–media thickness study in type 3 showed no difference, and the studies with a lower prevalence of arterial thrombosis included few severe patients; an arterial event is treated as in the general population, including percutaneous procedures, under prophylaxis with VWF concentrate at adjusted dose and frequency, choosing with the cardiologist the type of stent and the shortest possible antithrombotic duration and intensity; statins do not alter the haemostatic balance and can be used without bleeding risk [16]. In cancer, there are no data on a different incidence in type 3 (experimental data suggest that lack of VWF might reduce metastatic spread, without clinical confirmation); VWD does not change screening; all invasive procedures are covered with VWF; radiotherapy is accompanied by VWF prophylaxis; chemotherapy may aggravate the phenotype through thrombocytopenia; cancer surgery requires thromboprophylaxis; the disease should be presented at the oncology multidisciplinary meeting; and patients with HIV or HCV infection have a higher risk of lymphoma and hepatocellular carcinoma [16]. Venous thromboembolism is treated according to general population guidelines, discussing regular VWF prophylaxis for as long as anticoagulation lasts; arthropathy, falls and renal failure, which adds platelet dysfunction, are systematically sought during follow-up [16].

7.7. Heavy menstrual bleeding

The guidelines suggest hormone therapy (combined hormonal contraception or a levonorgestrel-releasing intrauterine system) or tranexamic acid rather than desmopressin for women not wishing to conceive (recommendation 6a), and tranexamic acid rather than desmopressin for those wishing to conceive (6b), both conditional with very low certainty [6]. Options may be combined if control is inadequate; desmopressin is ineffective in type 3 and many type 2 forms and contraindicated in 2B; additional treatments may be needed in the first cycles following intrauterine system insertion [6]. Good practice includes joint gynaecology and haematology clinics, standard gynaecological assessment to exclude fibroids and polyps, attention to endometrial hyperplasia risk (age >35, polycystic ovaries, obesity, diabetes, hypertension) and regular iron-deficiency assessment and treatment [6]. The 2025 review proposes a ferritin target ≥50 ng/mL [19].

In Kouides' crossover trial (116 women with heavy menstrual bleeding and abnormal laboratory haemostasis), tranexamic acid reduced the pictorial blood assessment chart (PBAC) score by 105.7 points and intranasal desmopressin by 64.1 points, a 41.6-point difference favouring tranexamic acid [41]. Levonorgestrel intrauterine system series reported expulsion in 15% and malposition in 10% [6]. VWDMin (phase 3, randomised crossover; 39 women aged 13–45 with mild or moderate VWD) compared recombinant VWF 40 IU/kg on day 1 with oral tranexamic acid 1300 mg three times daily on days 1–5; it stopped because of slow recruitment and failed to meet its primary objective. At interim analysis, median PBAC was lower with tranexamic acid (146 versus 213; adjusted difference 46; 95% CI 2–90), without serious adverse events; neither treatment normalised PBAC [27].

The US tranexamic acid tablet label indicates, in women and post-menarchal adolescents of reproductive age (not indicated before menarche), 1300 mg three times daily (3900 mg/day, maximum daily dose) for a maximum of 5 days per menstrual period for cyclic heavy menstrual bleeding, with renal dose reduction, and contraindicates concomitant combined hormonal contraception and active thromboembolic disease or intrinsic thrombosis risk [42]. This contraindication differs from the guideline remark that options may be combined: combining tranexamic acid with combined contraception conflicts with the US label, and the decision requires individual thrombotic-risk assessment [6, 42].

The French HUB PNDS, coordinated by the Toulouse reference centre for rare gynaecological conditions with the MHémo network, is the only national guidance dedicated to adolescents and young adult women with a bleeding disorder, and rests on expert consensus and small observational studies, mostly of women with VWD [17]. It requires a gynaecological cause to be excluded by ultrasound and combines, stepwise, tranexamic acid, hormonal therapy and, when necessary, disease-specific treatment, with systematic treatment of iron deficiency and contraindication of NSAIDs and acetylsalicylic acid, including those taken as self-medication; antidepressants, herbal medicines and supplements that aggravate bleeding should also be sought [16, 17].

Sources: [16, 17].

In severe acute bleeding, the PNDS emergency flowchart immediately starts tranexamic acid (20 mg/kg/day in 2–3 doses in children, 2–4 g/day in adults, or 0.5–1 g by slow intravenous injection 2–3 times a day), corrects haemodynamics with transfusion according to haemoglobin and tolerance (transfusions can be avoided in most patients with haemoglobin above 7–8 g/dL), seeks the haematologist's opinion on specific treatment and starts hormonal therapy: without contraindication, one tablet of ethinylestradiol 30 µg with levonorgestrel 150 µg as soon as possible, every 8 hours if bleeding persists 8 hours after the first and, once controlled, one tablet every 8 hours for 2 days, every 12 hours for 3 days and one a day continuously; with a contraindication, chlormadinone acetate 10 mg or medrogestone continuously [17]. The protocol text allows, in the same situation, one 30 µg tablet every 6–8 hours with tapering over 2 days. Discharge includes continuous hormonal therapy until the consultation, tranexamic acid for at least 5 days and to be resumed at each recurrence, oral or intravenous iron, and a consultation with full blood count and ferritin 2 to 3 months later [17]. In an unstable adolescent in whom medical treatment fails, intrauterine balloon tamponade (double-lumen, allowing irrigation with tranexamic acid) is the procedure that least threatens fertility, followed by curettage or hysteroscopic endometrectomy, uterine artery embolisation and vascular ligation [17].

Menarche is prepared for with a gynaecology consultation in the year following the onset of puberty (in France, Tanner breast stage 2 on average at 10.5 years and menarche at 12.5 years), assessment of pubertal stage and, from stage 3, an indicative suprapubic ultrasound (pubertal uterus with height >35 mm; bleeding expected when it exceeds 45–50 mm with endometrium >4 mm) [17]. Before the first menstrual periods a checklist is provided: prescriptions for tranexamic acid (20–40 mg/kg/day in 2–4 doses in the flowchart, to be started on the first day), hormonal treatment and a non-NSAID analgesic, a request for full blood count and ferritin on day 2 (48 hours) if bleeding is heavy, contact details of the team available 24 hours a day, written instructions, an individualised school care plan (PAI) and, if indicated, specific treatment at home; if the Higham score exceeds 100 or HUB persists at 48 hours, hormonal treatment is started without stopping tranexamic acid, and the assessment consultation takes place 3 to 6 months after menarche [16, 17]. In the Dutch WiN study, the 26 girls with type 3 had haemorrhagic first menstrual periods and approximately 80% required treatment, ranging from antifibrinolytics to admission with transfusion [16]. In the first two years, anovulatory cycles increase flow, and a menarche without HUB does not obviate monitoring of subsequent cycles [17].

For dysmenorrhoea, paracetamol (1 g three times a day in adults; 60 mg/kg/day in adolescents), antispasmodics, codeine or, in type 3, COX-2 inhibitors, and hormonal therapy are used; persistent disabling dysmenorrhoea prompts a search for endometriosis; high-frequency transcutaneous electrical nerve stimulation, acupuncture and cognitive behavioural therapies have limited evidence [16, 17]. Haemorrhagic ovarian cysts are managed conservatively, with analgesia, tranexamic acid or specific treatment and close monitoring, and are prevented with antigonadotropic contraception, preferably continuous combined oestrogen–progestogen [17]. The risk of bleeding at first sexual intercourse is discussed beforehand, with cover by tranexamic acid or specific treatment in severe forms; in a series cited, sexual activity was 25.7% in adolescents with HUB, versus 51.4% at 15–17 years and 91.8% at 20–24 years without HUB [17]. Before COC, blood pressure is measured, the only clinical examination required, and 3 to 6 months later fasting glucose, cholesterol and triglycerides are measured, to be repeated every 5 years if normal; COC is stopped if blood pressure rises or new migraines or headaches appear; in France, 2nd-generation contraception is free of charge up to the age of 26 years [17].

7.8. Antifibrinolytics, local measures and specific situations

Antifibrinolytics (tranexamic acid and ε-aminocaproic acid) inhibit plasmin action and delay fibrinolysis; they can be used alone for minor bleeding or alongside VWF-raising treatments, and are relatively contraindicated in gross haematuria [3, 19]. Intravenous tranexamic acid (Cyklokapron®) has a US indication limited to dental extraction in haemophilia (10 mg/kg beforehand and 10 mg/kg 3–4 times daily for 2–8 days, adjusted for renal function) and a boxed warning against inadvertent neuraxial administration; VWD use is extrapolation beyond the label wording [43]. Epistaxis management includes nasal care (humidification, gels, saline, treatment of allergies and infection), nasal clips, packing and cauterisation, with haemostatic treatment before cauterisation and VWF prophylaxis for refractory cases [3]. Gastrointestinal angiodysplasia is managed with endoscopic intervention, VWF replacement and tranexamic acid, with prophylaxis in severe cases [3]; vascular endothelial growth factor-targeted treatments and high-dose statins have been described in series and cases [19]. Recombinant FVIIa is an option for refractory bleeding and patients with inhibitors [19]. Hepatitis B vaccination is recommended [3].

Activities with high trauma risk and contact sports with head-injury risk should be avoided; antiplatelet agents (acetylsalicylic acid, clopidogrel) should be used only with a clear indication and monitoring; nonsteroidal anti-inflammatory drugs should be used cautiously and briefly; and supplements that may impair haemostasis, such as fish oil and turmeric, should be avoided [3].

The PNDS summarise the adjunctive measures available in France, to be weighed case by case and combined with specific treatment when necessary [7, 16].

Source: [7, 16, 17]. The PNDS do not set a paediatric maximum for tranexamic acid in mg/kg; the adult oral range ends at 4 g/day, type 3 epistaxis allows 3–4 g/day for 10 days and the menarche flowchart up to 40 mg/kg/day.

The contraindications to tranexamic acid in the HUB PNDS are hypersensitivity, active thromboembolic disease, fibrinolytic states of consumption coagulopathy (except predominant hyperfibrinolysis with severe bleeding), severe renal failure and a history of seizures; the ANSM has warned of renal cortical necrosis with doses above 2 g in postpartum haemorrhage in patients with pre-eclampsia; and, outside an acute thromboembolic episode, combination with combined oestrogen–progestogen contraception does not appear to increase risk, although there are no data from large cohorts with bleeding disorders [17]. French nutritional references for iron are 13 mg/day in girls aged 12–17 years with heavy menstrual losses and 16 mg/day from 18 years, with preference for haem iron (absorption of 20% to 30%, versus 2% to 5% for non-haem iron), tea away from meals and vitamin D and calcium supplementation [17].

In severe epistaxis, severity is assessed by haemodynamics (heart rate, blood pressure, capillary refill time in children) and by full blood count with reticulocytes, ferritin, ABO, RhD and Kell grouping and an irregular antibody screen (grade C) [16]. Acute treatment has four simultaneous lines: correcting shock (transfusion if anaemia is poorly tolerated), correcting haemostasis without delay (VWF with or without FVIII in type 3 and severe forms; intravenous desmopressin in type 1 and some type 2), ear, nose and throat treatment and prevention of recurrence, with oral tranexamic acid (3–4 g/day in adults, 20 mg/kg/day in children, in 3–4 doses) for 10 days to cover clot detachment [7, 16]. Local measures are gentle nose-blowing to remove clots, the sitting position with the head tilted slightly forward and bidigital compression of both nostrils for 20 minutes; alginate wicks (from 30 minutes to 24 hours), HEC® ointment or gelatin sponges; then, if necessary, anterior packing, preferably absorbable (oxidised cellulose, polyurethane, gel wick) after topical anaesthesia with a vasoconstrictor, or fibrin or gelatin glues, avoiding absorbable materials and glues in young children because of the risk of migration and inhalation; posterior packing with a double-balloon catheter for 48–72 hours; sphenopalatine artery embolisation by an experienced neuroradiologist (the ethmoidal arteries, branches of the ophthalmic artery, are not embolised because of the risk of blindness); endoscopic ligation of the sphenopalatine arteries and, as a last resort, approach to the ethmoidal arteries via the medial canthus, always preceded by haemostatic treatment [16]. Recurrence due to eschar detachment occurs 7–10 days later, sometimes up to 21 days; prevention rests on air humidification, saline sprays, petroleum jelly or oily nasal solution, treatment of allergic rhinitis and hypertension, and cauterisation of the vascular spot away from the episode [7, 16]. Tonsillar bleeding, spontaneous or triggered by infection, is an emergency requiring replacement, tranexamic acid and sometimes cauterisation [7, 16].

In oral bleeding, a cold semi-liquid diet and tranexamic acid as a mouthwash and orally for 3–5 days may not suffice, and if bleeding persists for more than 24 hours concentrate may be required; oral hygiene and regular follow-up by the dentist are essential [16].

In gastrointestinal bleeding, an emergency when overt, treatment with VWF concentrate (with or without FVIII) begins before endoscopy and laboratory tests, with transfusion support, and may be continued until the eschars detach (1–2 weeks), with oral tranexamic acid for 8–10 days [16]. After the first episode, upper endoscopy and ileocolonoscopy are performed, repeatable if preparation or active bleeding limit them; capsule endoscopy is the examination of choice when standard endoscopy is normal and requires no cover; enteroscopy allows argon plasma coagulation and biopsies; percutaneous embolisation guided by CT angiography or arteriography treats accessible lesions after endoscopic failure; and surgical resection is a last resort [7, 16]. In type 3, all endoscopic procedures carry the same bleeding risk during the procedure, but the European Society of Gastrointestinal Endoscopy classification helps set the duration of cover: at least 2–3 days with VWF for low-risk procedures (endoscopy and colonoscopy with biopsy, endoscopic ultrasound without puncture, polypectomy <1 cm, dilatations, stents, retrograde cholangiopancreatography without sphincterotomy, argon plasma) and 5–8 days for high-risk procedures (mucosectomy, submucosal dissection, ampullectomy, sphincterotomy, polypectomy >1 cm, puncture of cystic lesions, gastrostomy, variceal ligation), with clips or chemical haemostatic agents to prevent secondary bleeding [16]. Recurrent bleeding leads to prophylaxis or its intensification, after reviewing the quality of the endoscopic investigation with the gastroenterologist; in refractory forms, particularly with inaccessible angiodysplasias, octreotide, thalidomide, lenalidomide, high-dose atorvastatin, danazol, tamoxifen or recombinant FVIIa have been described in case reports and small series and are used only as a last resort, off-label, after a multidisciplinary decision with the patient [7, 16]. Small-bowel wall haematoma should be considered in the presence of atypical gastrointestinal symptoms in a severe form [7].

Vaccination follows the national schedule and includes hepatitis A and B; the PNDS accept the intramuscular route, with a 25G needle in the outer aspect of the arm or in the thigh and compression without rubbing for at least 10 minutes, given immediately after a prophylaxis dose when the child receives one; in severe haemophilia, post-vaccination intramuscular haematoma occurred in 1 of 549 injections, and there are no studies in type 3, although the risk is considered to depend mainly on FVIII deficiency [16, 17]. Human papillomavirus vaccination (9-valent) is given in France between 11 and 15 years in two doses and, after that age, in three catch-up doses [17]. Regular physical activity (at least 60 minutes a day above 5 years, according to the World Health Organization) is recommended; the type of sport and competition are decided with the physician, contact sports and some amusement park activities are avoided in severe forms, and sport is practised as close as possible to a prophylaxis infusion [7, 16].

7.9. Pivotal trials and interpretation

Table sources: [6, 10, 27, 39–41, 45, 46].

Except for PRO.WILL, concentrate trials are open-label, lack a concurrent control group and compare treatment with historical data or a preceding on-demand period, favouring regression-to-the-mean effects and greater adherence during trials. Populations are dominated by type 3 and severe forms, so findings cannot be extrapolated to mild type 1. Relative reductions should be read alongside residual absolute rates (for example, 5.2 bleeds per year in WIL-31). Products cannot legitimately be ranked through indirect comparisons between trials with different definitions, durations and populations. Two menstrual bleeding trials (Kouides and VWDMin) showed that tranexamic acid is not inferior and may be superior to more costly therapies, and VWDMin is a relevant negative trial for VWF replacement in this indication.

7.10. Regulatory developments and investigational treatments (2025–2026)

On 5 September 2025, the FDA approved extension of Vonvendi® prophylaxis to adults with types 1 and 2 (previously limited to severe type 3, approved 28 January 2022), and paediatric use for on-demand treatment and perioperative management, based on NCT02932618, NCT03879135 and NCT02973087 [37]. On 2 July 2026, the FDA extended Wilate® prophylaxis to children under 6, based on WIL-33 (NCT04953884); prophylaxis from age 6 had been approved on 1 December 2023 [39]. In the EU, the current Veyvondi® SmPC includes bleeding treatment in children under 18 [38]. A phase 3 vonicog alfa prophylaxis trial in children with severe VWD is ongoing (NCT05582993; 24 planned patients; estimated primary completion in 2030) [47].

Table sources: [19, 20, 48–54].

The evidence hierarchy differs for each row: efanesoctocog alfa and rondoraptivon pegol have phase 1 or 2 results published in peer-reviewed journals [51, 53]; VGA039 and HMB-002 had only conference data cited by third parties at the time of the consulted reviews [20]; emicizumab in VWD is supported by case reports and a recruiting phase 3 trial [20, 49]. None of these molecules has an approved VWD indication in the USA or EU, and use outside a trial is off-label. The 2026 review also describes preclinical approaches: the single-domain antibody KB-V13A12 (linking VWF to albumin), platelet-inspired synthetic nanoparticles, small interfering RNA targeting plasminogen, selective allele silencing and CRISPR editing in endothelial cells from type 2A and 2B patients, and gene therapy limited by VWF complementary DNA size (>8 kb) relative to AAV vector capacity (~4.5 kb) [20].

7.11. Divergences between the French PNDS, the international guidelines and the labels

Sources: [1, 3, 6, 7, 9, 10, 16, 17, 19, 42]. The international guidelines, based on GRADE methodology, are the reference for diagnosis and for the main decisions, and the labels serve as a comparison for doses; the PNDS, based on expert consensus, add operational detail (adjunct doses, durations, care pathways) and reflect the range of medicinal products available in France.

8. Reproductive Options and Family Counselling

8.1. Transmission and risks

Types 1, 2A and 2M are usually caused by a heterozygous variant and show autosomal dominant inheritance, rarely involving biallelic variants; type 2B is autosomal dominant; types 2N and 3 result from biallelic variants and are autosomal recessive [3]. In dominant forms, many probands have an affected parent; rare de novo variants have been reported in types 1 and 2, and sibling risk when a proband's variant is absent from parental leukocyte DNA is slightly above population risk because of possible gonadal mosaicism [3]. An apparently negative family history may reflect nonrecognition, reduced penetrance, variable expressivity, early death of an informative relative or absence of haemostatic challenges, and can be confirmed only through parental laboratory and/or molecular testing [3]. Each child of a patient with a dominant form has a 50% chance of inheriting the variant; if the partner is also a carrier, offspring may inherit biallelic variants with a more severe or different-type phenotype [3].

In recessive forms, parents are usually heterozygous, and each sibling has a 25% chance of being affected, 50% of being heterozygous and 25% of inheriting neither variant; children of an affected individual are obligate heterozygotes unless the partner is a carrier [3]. Type 2N heterozygotes are generally asymptomatic, but a small proportion have mild bleeding and lower FVIII; 15% to 50% of type 3 heterozygotes have symptoms and may be diagnosed as type 1 [3]. If only one parent is a carrier, a de novo variant, postzygotic parental mosaicism or uniparental disomy should be considered, the latter reported in type 3 [3]. In type 2N with only one identified variant, coexistence of a null allele is assumed, changing offspring risk [1].

Source: [3].

8.2. Family testing, prenatal and preimplantation diagnosis

At-risk relatives should be assessed even if asymptomatic, using sensitive and specific haemostatic assays regardless of whether the familial type or variant is known, and molecular testing when the variant is known [3]. Counselling should be offered to young adult carriers, ideally before pregnancy, and testing considered for reproductive partners, particularly consanguineous couples or those sharing ancestral origin, such as Romani and Amish communities [3]. Once familial variants are identified, prenatal diagnosis and preimplantation genetic diagnosis are possible; views on their use vary among professionals and families, and the decision is personal [3]. The diagnosis guidelines note that some groups, such as certain First Nations communities, have reservations about genetic testing [1].

The French PNDS specify the care pathway. Counselling on the mode of transmission is part of the haemostasis consultation; types 1, 2A, 2B and 2M are generally dominant, with sometimes incomplete penetrance, and types 2N and 3 recessive; and a clinical genetics consultation should be offered in the most severe forms, particularly type 3, in which preimplantation genetic diagnosis or termination of pregnancy for medical reasons are available to couples who wish it [7]. In type 3, the risk is 25% in each pregnancy of a heterozygous couple and fetal sex is of no relevance; all first-degree relatives (and, where there is consanguinity, other relatives) should have FVIII and VWF measured, to detect other affected individuals and heterozygotes with low VWF and bleeding risk; the patient and both parents undergo genetic testing, and at-risk relatives can then be tested for the identified variants [16]. Genetic testing requires informed consent and a consultation certificate, and the patient is obliged to inform the family, with support if necessary [7, 16].

Source: [16]. Heterozygous carriers with low VWF need haemostatic cover for oocyte retrieval, biopsy, amniocentesis or termination; a prior consultation at the haemostasis centre informs the couple about the treatment and outlook of type 3; and sessions with a psychologist help in dealing with anxiety, guilt or grief [16]. A woman with type 3 should know that her child will be an obligate heterozygote, asymptomatic or with a type 1 profile, unless the father is also a carrier [16].

8.3. Pregnancy

VWF levels rise throughout pregnancy, peaking approximately 4 hours after delivery, but pregnant women with VWD remain at increased risk of bleeding complications and should be managed in centres experienced in perinatal bleeding disorders [3]. In a US prospective study of 32 women with VWD and 40 without VWD, VWF reached 250% of baseline 4 hours after delivery in women with VWD and 12 hours in the remainder, then fell rapidly, approached baseline within one week and reached it at 3 weeks; except immediately after delivery, levels in treated women remained below those in women without VWD, suggesting persistent bleeding risk despite treatment [55]. In a Swedish series of 59 deliveries in 34 women (1995–2012), primary postpartum haemorrhage occurred in 44%, severe primary haemorrhage in 20% and secondary haemorrhage in 12%; type 3 was associated with more severe primary bleeding, and secondary bleeding was more frequent when diagnosis was unknown at delivery [56]. In a systematic review of 811 deliveries, cohorts reported primary postpartum haemorrhage in 32% and secondary haemorrhage in 13%, with similar primary incidence with and without prophylaxis, and neonatal bleeding in 4.6% of deliveries, within a low-quality evidence base [57].

For women considered suitable for neuraxial anaesthesia, the management guidelines suggest a VWF activity target of 0.50–1.50 IU/mL rather than above 1.50 IU/mL to permit the procedure (recommendation 7; conditional, very low certainty) [6]. Remarks recommend a third-trimester consultation with VWF and FVIII measurement and a prospective anaesthesia and delivery plan, maintaining VWF >0.50 IU/mL while the epidural catheter remains in place and for at least 6 hours after removal, final decision-making by the obstetric anaesthetist, and postpartum thrombotic-risk assessment with thromboprophylaxis when indicated [6]. This recommendation focused on anaesthesia rather than preventing postpartum haemorrhage, for which levels above 1.50 IU/mL may be advisable in some situations; in the panel's collective experience (110 women with 0.50–1.50 and 34 with >1.50 IU/mL), the lower-target group had more major bleeding (2.7% versus 0%), postpartum haemorrhage (17% versus 5.9%) and transfusion (10% versus 0%), unpublished data considered insufficient to recommend the higher target [6].

GeneReviews notes that the 2021 guidelines do not define a delivery VWF target: some centres treat to maintain VWF above 100 IU/dL, and almost all agree on correcting values below 50 IU/dL before delivery [3]. The 2025 review proposes measuring VWF and FVIII at 30–32 weeks and aiming for minimum VWF activity ≥0.80 IU/mL to permit neuraxial anaesthesia and reduce postpartum bleeding risk; administering tranexamic acid 1000 mg intravenously at delivery (usually at cord clamping), then oral 1000–1500 mg every 8 hours for at least 10–14 days or while lochia remains heavy; and, if concentrate is required, continuing for at least 72 hours after vaginal delivery and 5 days after caesarean section, including at least 24 hours after epidural catheter removal, according to the Royal College of Obstetricians and Gynaecologists guideline [19]. These proposals represent author opinion and a national guideline rather than international panel recommendations.

Delivery should follow obstetric indications, minimising instrumentation (ventouse, forceps) and avoiding episiotomy whenever possible; fetal scalp electrodes and scalp blood sampling should be avoided in most forms because the fetus may have VWD [3]. Desmopressin has been used safely in pregnancy in women with bleeding disorders and diabetes insipidus, but should be avoided in pre-eclampsia and cardiovascular disease, and intravenous fluids and oxytocic agents during labour increase hyponatraemia risk; desmopressin trials should be avoided during pregnancy [6]. The DDAVP® label identifies no drug-associated risk of malformations or pregnancy loss and reports poor placental transfer in vitro [13]. No pregnancy or breastfeeding experience exists for vonicog alfa, and the European SmPC reserves it for a clear indication [9, 38].

The general PNDS recommends following the woman from the wish to conceive, at a level 2 or 3 maternity unit in severe forms, with FVIII and VWF measured at the first antenatal visit, at 28 weeks, before any invasive procedure and at 34 weeks, and with a multidisciplinary delivery protocol; it cites the international target of FVIII and VWF >50% for at least 3 days after vaginal delivery and 5 to 7 days after caesarean section; and it allows epidural analgesia in type 1 if levels are >50% at 34 weeks, does not recommend it in type 2 and contraindicates it in type 3 [7]. The mode of delivery is decided on obstetric criteria [7].

In type 3, the PNDS adds the following particulars [16]:

• Before conception: make the woman, already on hormonal therapy for HUB, aware early of the preconception strategy; when pregnancy is desired, stop hormonal therapy and start prophylaxis, generally on the first days of menstruation; offer ovulation tests to shorten the period of spontaneous cycles.
• During pregnancy: VWF and FVIII levels do not correct at any point, so they need not be measured; chorionic villus sampling, amniocentesis or cerclage require replacement to VWF:Act and FVIII:C >50 IU/dL; intermittent bleeding requires replacement; iron deficiency is corrected; the miscarriage rate does not appear higher than in the general population, but evacuation is surgical under replacement; medical elective termination of pregnancy is contraindicated because of the risk of unpredictable bleeding at home and is performed by aspiration with at least 3 days of replacement and tranexamic acid.
• Delivery in a level 3 maternity unit, planned by obstetrician, anaesthetist, haematologist, pharmacist and paediatrician; vaginal delivery permitted without traumatic manoeuvres; episiotomy possible under replacement, with monitoring of healing; neuraxial anaesthesia discussed case by case.
• Induction: VWF 12 hours beforehand to stabilise endogenous FVIII, a second dose immediately before delivery and FVIII added according to the level measured 12 hours after onset; at delivery, FVIII and VWF:Act ≥100 IU/dL, up to 150 IU/dL; replacement at therapeutic doses for at least 5 days (vaginal) or 7 days (caesarean), with troughs of approximately 50 IU/dL, followed by prophylaxis until menstruation returns, with early hormonal therapy, even during breastfeeding, and tranexamic acid.
• Spontaneous labour: VWF of at least 60–80 IU/kg with FVIII of at least 30–40 IU/kg immediately, then following the induction regimen.
• Tranexamic acid in all cases for 7 to 14 days, up to day 14 after caesarean section; low-molecular-weight heparin discussed after caesarean section and stopped if there is bleeding.

The risk of postpartum haemorrhage, greater in VWD than in the general population, is higher still in type 3, even with replacement according to the recommendations [16].

8.4. Postpartum period and breastfeeding

The guidelines suggest postpartum tranexamic acid for women with type 1 or low VWF, potentially also applying to types 2 and 3 (recommendation 8; conditional, low certainty) [6]. In two retrospective studies, tranexamic acid reduced secondary postpartum haemorrhage (risk ratio 0.42; 95% CI 0.20–0.91), with imprecise estimates for primary haemorrhage, transfusion and vaginal haematoma [6]. Good practice specifies oral or intravenous administration, with an oral dose of 25 mg/kg (usually 1000–1300 mg) three times daily for 10–14 days, or longer if blood loss remains heavy (no maximum dose defined in the source), and education about tranexamic acid safety during breastfeeding [6]. This regimen derives from the guideline rather than the US tablet label, which is limited to cyclic menstrual bleeding for up to 5 days; it constitutes off-label use [6, 42]. VWF rapidly returns to prepregnancy values after delivery [3, 55].

The general PNDS calls for anticipation of delayed postpartum haemorrhage, which may arise around day 15 and up to 6 weeks: discharge treatment may combine tranexamic acid, hormonal therapy, nasal desmopressin or replacement with VWF, with or without FVIII, according to type, and in type 3 replacement continues until menstruation returns; NSAIDs increase bleeding risk [7]. In type 3, the puerperal woman is clinically reassessed on day 15 and taught to attend hospital for any delayed bleeding [16].

8.5. At-risk newborn

The mother should be informed during pregnancy, and the obstetrician, neonatologist and paediatric haematologist should know the risk before birth [3]. In newborns at risk of more severe type 1, 2 or 3 forms, cord blood should be collected for VWF and FVIII, and platelets if type 2B is a possibility; delivery stress, sample handling and immature neonatal haemostasis affect results, and definitive diagnosis should be established later [3]. Cord-blood molecular testing is more informative when the familial variant is known; without it, a negative result does not exclude VWD [3].

The French PNDS recommend avoiding invasive fetal monitoring (scalp electrodes) and traumatic manoeuvres, particularly vacuum extraction, and screening for type 3 in cord blood (after double clamping) when there is a family history, with later confirmation at a centre consultation [7]. When the newborn has a potential or confirmed risk of type 3, delivery takes place in a level 2B or 3 maternity unit with neonatal intensive care; there is no indication for routine caesarean section, but difficult progress requires rapid conversion; vacuum extraction, scalp punctures and intramuscular injections are contraindicated; venepunctures, including the heel-prick test, are followed by manual compression for approximately 10 minutes; and, if cord sampling fails, peripheral puncture is performed with 10 minutes of compression and a moderately compressive dressing until the following day [16]. Routine preventive replacement of VWF and FVIII is not indicated, except after traumatic delivery (particularly difficult instrumental extraction) with a confirmed laboratory diagnosis; the newborn has a daily clinical examination, including neurological examination, for 5 days before discharge, and suspicion of intracranial haemorrhage leads to transfer to neonatal intensive care, cross-sectional imaging and immediate replacement, because transfontanellar ultrasound is not sufficient to exclude it [16]. At discharge, the child health record notes the bleeding risk and an appointment is made at the competence centre [16].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Prevalence and what each estimate measures

The variation reflects the definition used: laboratory prevalence under broad criteria, symptomatic disease prompting care, or patients actually followed in specialised centres. US haemophilia centre data show treated type 1 prevalence twice as high in women and treated type 1 incidence almost three times higher in women born between 1995 and 1999 (26.2 versus 9.9 per 100,000 live births), differences the authors attribute mainly to differing diagnosis rates, because inheritance is sex-independent [58]. A treatment-centre series measures access to care rather than population prevalence. The French PNDS note that symptomatic forms are more frequent in women, because of the haemostatic challenge of menstruation, pregnancy and childbirth, although autosomal transmission affects both sexes equally [7].

9.2. Type 3 in Europe and Israel, including Portugal

A 1984 registry of patients with VWF:Ag below 1 IU/dL, confirmed by immunoassay at four reference centres, showed severe VWD prevalence in Scandinavian countries approximately tenfold higher than elsewhere in Western Europe, with Israel intermediate [59]. Values derived from this registry and validated by Orphanet, per million inhabitants, are: Sweden 3.12, Norway 2.5, Finland 2.4, Israel 1.6, Italy 0.55, Greece 0.44 (at birth), Austria and Netherlands 0.37, UK 0.27, Portugal 0.22, France 0.16 and Spain 0.11; Orphanet estimates worldwide type 3 prevalence at approximately 1.9 per million [4, 59]. These are historical data dependent on reporting and each country's diagnostic capacity in the 1980s. In the 2025 review, the WFH global registry had already gathered data on more than 1000 people with VWD, including over 500 with type 3 [19]. No Brazil-specific prevalence data were found in the consulted sources.

9.3. Founder effects and particular populations

The p.Gln1311Ter variant (Q1311X in the original article's nomenclature) was identified homozygously in four Spanish type 3 patients from two apparently unrelated Romani families; a shared haplotype suggests a single origin approximately 2300 years ago [3, 60]. The p.Arg1374Cys variant is a founder variant in Amish populations in Indiana and Ohio; in a cross-sectional study of 779 Amish individuals from Indiana and Wisconsin, 17% had VWD by VWF:RCo, with highly variable bleeding phenotype and generally low bleeding scores despite the shared variant [3, 61]. The family described by von Willebrand lived in the Baltic Åland Islands [1].

The benign p.Asp1472His variant, discussed by the guidelines in relation to African ancestry, artificially lowers VWF:RCo without affecting VWF function or bleeding risk; the guidelines also recommend ristocetin-free assays for equity reasons, because accuracy studies included few patients of African ancestry and identified false-positive results in African Americans [1, 3]. Women with VWD more often receive medical attention because of gynaecological and obstetric bleeding [1]; for transgender individuals, some hormonal treatments for menstrual bleeding may be less acceptable or contraindicated within gender-affirming care plans [6].

10. Illustrative Clinical Case

In 2019, Weyand and colleagues described a boy with severe type 3 VWD who began plasma-derived VWF concentrate prophylaxis at age 5 (approximately 45 units/kg three times weekly) because haemarthrosis-induced left ankle synovitis caused recurrent bleeding [62]. At age 7, after approximately 270 exposure days, he developed more bruising, haematomas and joint swelling. Pre- and post-infusion testing showed FVIII 2%, ristocetin cofactor activity below 10% and VWF:Ag below 15%; thirty minutes after concentrate, VWF activity and antigen remained undetectable, with FVIII 0.64% and an FVIII inhibitor of 17.5 Bethesda units. Serial plasma dilutions demonstrated inhibition of normal plasma VWF activity, consistent with an anti-VWF alloantibody; FVIII inhibitor positivity was attributed to steric interference by the anti-VWF antibody. Sequencing and comparative genomic hybridisation showed a heterozygous partial VWF deletion encompassing exons 17 and 18, spanning at least 8.7 kb, and c.2435del, described as causing absent protein expression [62]. Article Table 1 summarises laboratory evolution before and after inhibitor development (table link); the article has no figures.

Prophylaxis was changed to off-label recombinant FVIIa (270 µg/kg three times weekly), but target-ankle haemarthroses persisted, requiring hospitalisation and alternation between recombinant FVIIa and activated prothrombin complex concentrate (aPCC). Daily aPCC prophylaxis followed (approximately 100 units/kg), with central venous catheter placement because of difficult access and radioisotopic synoviorthesis, after which aPCC could be spaced to alternate days and then three times weekly, with rare spontaneous haemarthroses and a high treatment burden persisting [62]. The authors then started off-label emicizumab: loading 3 mg/kg weekly for 4 weeks and 1.5 mg/kg weekly for approximately 9 months without any bleeding, followed by 3 mg/kg every two weeks; recombinant FVIIa (90–120 µg/kg) was needed only for one traumatic soft-tissue haematoma [62].

The case illustrates several lessons of this article: alloantibody risk in type 3 with large deletions and repeated exposure; suspicion prompted by loss of response to replacement; the need to distinguish anti-VWF from FVIII inhibitors; and the contribution of FVIII deficiency to joint bleeding in type 3 [3, 62]. The authors mention the theoretical risk of thrombosis and microangiopathy with the combination of emicizumab and recombinant FVIIa [62]. This is a single case with less than one year of follow-up; emicizumab has no approved indication in VWD and its efficacy is being assessed in a phase 3 trial in type 3 [20, 49]. The reported management should not be read as a routine recommendation.

11. Prognosis and Long-Term Follow-Up

Prognosis depends on type and access to specialised care. Most type 1 patients have mucocutaneous bleeding controllable with on-demand treatment, but heavy menstrual bleeding and iron deficiency impair quality of life [6, 19]. Severe forms, particularly type 3, carry risks of haemarthroses with joint damage, intracranial bleeding and treatment complications such as alloantibodies [3, 6, 8]. Gastrointestinal angiodysplasia tends to worsen over time and can cause life-threatening bleeding [3]. Age increases VWF in some type 1 patients without proven parallel reduction in bleeding risk [1].

The first five rows reproduce GeneReviews Table 8 [3]; the remaining rows come from the French PNDS.

At initial assessment, GeneReviews recommends ensuring a modern, complete evaluation before invasive procedures, avoiding reliance on old diagnoses without recent laboratory confirmation, testing for inhibitors in type 3, assessing iron and ferritin, joints in forms with low FVIII, gynaecological causes of bleeding, angiodysplasia when gastrointestinal bleeding is suspected and serology in those exposed before 1985, and referring for genetic counselling [3]. The management guidelines recommend periodically reassessing symptoms and prophylaxis requirements and regularly evaluating and treating iron deficiency in women with heavy menstrual bleeding [6]. The 2025 review suggests patients carry a card or information on their phone specifying VWD type, emergency treatment and centre contact details, and elective surgery be planned with a multidisciplinary team and pharmacy in a centre offering FVIII and VWF assays, including at weekends [19].

The French PNDS make therapeutic education part of every consultation, with an educational diagnosis, a personalised programme, individual or group workshops and evaluation, following the HAS guide; the The3P group of the MHémo network and the Association Française des Hémophiles (AFH) proposed in 2015 a competency framework for VWD: understanding the disease and its inheritance, recognising and assessing a bleed, using simple measures, identifying an emergency that requires contacting the centre, knowing when and how to inject concentrate and learning self-infusion in severe forms, managing menstruation, informing third parties, reducing stress and asserting life with the disease [16]. There are a national therapeutic education weekend for severe VWD, organised by the AFH with trained patients and parents (“resource patients”), a distance-learning platform of the reference centre and, for HUB, programmes for adolescents that teach the cycle, the Higham score, treatment adjustment and an iron-rich diet, as well as prototype menstrual tracking apps [16, 17]. Consultations check coordination with the attending physician, home nursing, the gynaecologist, the gastroenterologist, the dentist and the physiotherapist, update schooling, occupation and sports, offer psychological support to the patient and family and provide information on clinical trials [16].

At school, the individualised school care plan (PAI; interministerial circular no. 2003-135 of 8 September 2003), drawn up with the centre physician, states the warning signs, the measures to take in case of accident, the contacts and the information for emergency services, and contraindicated sports; for girls with HUB, it provides permission to leave class, change protection or take analgesics, exemption from sport or swimming during menstruation, spare clothing available in the school infirmary and calling the emergency service in the presence of signs of severity (malaise, nausea, dizziness, pallor, dyspnoea), with the girl lying down with legs raised until help arrives [16, 17]. Transition from paediatric to adult care is prepared in consultation, progressively transferring responsibility from the parents to the young person and clinical information to the adult team; in adolescent girls, the heaviness of losses, their impact, the adaptation of treatment to school and social life, sexuality, prevention of sexually transmitted infections and contraception before first intercourse are reassessed [16, 17]. Before stopping contraception to conceive, women with severe forms have tranexamic acid at home and checked haemoglobin and ferritin and, according to the HUB PNDS, undergo a fertility assessment (with semen analysis of the partner), to shorten the period of poorly tolerated spontaneous cycles [17].

The AFH, an association recognised as being of public utility and accredited to represent users, has a committee of patients with VWD and a women's committee, 24 regional committees linked to the centres, a biennial national congress and a biennial holiday camp [7, 16, 17]. French centres liaise with the medico-social sector (Maison Départementale des Personnes Handicapées) and systematically offer inclusion in the FranceCoag and BAMARA registries [7, 16].

In Europe, the relevant reference network is ERN-EuroBloodNet, which includes a subnetwork for rare bleeding and coagulation disorders [18]. In France, the centres of the MHémo network (the CRMW, the reference centres for haemophilia and platelet disorders and the CRC-MHC) and the Toulouse reference centre for rare gynaecological conditions provide the care described in the PNDS [7, 17]. GeneReviews lists patient resources including the National Bleeding Disorders Foundation (USA), Canadian Hemophilia Society, Haemophilia Society (UK) and World Federation of Hemophilia, alongside Centers for Disease Control and Prevention and MedlinePlus pages [3]. The VWD Connect Foundation maintains a severe VWD patient registry (NCT05437536) [63].

12. Summary and Final Considerations

VWD results from quantitative (types 1 and 3) or qualitative (types 2A, 2B, 2M and 2N) defects in VWF, encoded by VWF at 12p13.31. Inheritance is autosomal dominant in types 1, 2A, 2B and 2M and recessive in types 2N and 3; penetrance is incomplete in type 1 with VWF 30–50 IU/dL, where factors outside VWF contribute. The clinical picture is dominated by mucocutaneous bleeding, with heavy menstrual bleeding and postpartum haemorrhage particularly significant in women; types 2N and 3 also cause joint and muscle bleeding, and type 3 adds intracranial bleeding and alloantibody risks.

Diagnosis follows the 2021 ASH/ISTH/NHF/WFH guidelines: validated BAT in low-probability settings; VWF:Ag, platelet-dependent activity using ristocetin-free assays and FVIII; thresholds of 0.30 IU/mL regardless of bleeding and 0.50 IU/mL with bleeding; activity/antigen ratio <0.7 for type 2; multimers or VWF:CB for 2A, 2B and 2M; genetics for 2B and VWF:FVIIIB or genetics for 2N; and a desmopressin trial with a 4-hour sample for type 1C. Molecular tests detect the causative variant in 80–90% of severe forms and 40–60% of type 1 with VWF 30–50 IU/dL, require rearrangement analysis and attention to the pseudogene, and reports should use transcript NM_000552.5.

Treatment is type-specific: desmopressin in responders (intravenous 0.3 µg/kg, maximum 20 µg), contraindicated in 2B and ineffective in 3; plasma-derived or recombinant VWF concentrates, typically 40–80 IU VWF:RCo/kg; FVIII and VWF ≥0.50 IU/mL for at least 3 days after major surgery; adjunctive tranexamic acid for minor procedures and postpartum; and hormone therapy or tranexamic acid for heavy menstrual bleeding, where tranexamic acid outperformed desmopressin and was not inferior to recombinant VWF. In the USA, vonicog alfa prophylaxis was extended to adults with types 1 and 2 in 2025 and paediatric on-demand and perioperative use approved, while Wilate® prophylaxis was extended to children under 6 in 2026. Emicizumab, VGA039, HMB-002, rondoraptivon pegol and efanesoctocog alfa are in trials and lack an approved VWD indication.

For geneticists, critical issues are distinguishing type 2N from mild haemophilia A and 2B from platelet-type VWD, which change inheritance and treatment; assessing first-degree relatives regardless of symptoms; counselling consanguineous couples or communities with founder variants; and multidisciplinary pregnancy and delivery planning, protecting the at-risk newborn.

The three HAS PNDS complement the international guidelines with operational detail: referral criteria from VWF <50 IU/dL or ratios <0.7; propeptide and methods with a low detection limit for type 3; prophylaxis from the first haemarthrosis in type 3; higher and longer surgical and obstetric targets in type 3 (peak ≥100 IU/dL); monitoring of accumulated FVIII (<150 IU/dL); decision pathways for alloantibodies; and a structured approach to HUB in adolescents, from menarche to the emergency setting, with the Higham score, tranexamic acid and hormonal therapy [7, 16, 17]. The main divergences — neuraxial anaesthesia in types 2 and 3, duration of surgical cover, schedule of measurements in pregnancy and combination of tranexamic acid with combined contraception — reflect differences in method and population and are decided case by case at the competence centre.

13. Bibliographic References

1. James PD, Connell NT, Ameer B, Di Paola J, Eikenboom J, Giraud N, et al. ASH ISTH NHF WFH 2021 guidelines on the diagnosis of von Willebrand disease. Blood Adv. 2021;5(1):280–300. doi:10.1182/bloodadvances.2020003265. PMID:33570651; PMCID:PMC7805340.
2. Von Willebrand EA. Hereditary pseudohaemophilia. Haemophilia. 1999;5(3):223–31 (translation of the original published in 1926 in Finska Läkaresällskapets Handlingar). doi:10.1046/j.1365-2516.1999.00302.x. PMID:10444294.
3. Johnsen J. Von Willebrand Disease. In: Adam MP, et al., editors. GeneReviews® [Internet]. Seattle: University of Washington; 1993–2026. Published 4 June 2009; fully updated 14 November 2024. NBK7014.
4. Orphanet. Von Willebrand disease. ORPHA:903 (subtypes ORPHA:166078, 166081, 166096; pseudo-VWD ORPHA:52530) [Internet]. Disease entry. Definitions and ICD-10 D68.0 / ICD-11 3B12 mapping from the OrphaCode API (2 July 2026 version); Orphadata prevalence data (23 June 2026). Accessed 6 October 2026.
5. VariantValidator [Internet]. Manchester: University of Manchester. Validation of NM_000552.5:c.3614G>A, c.3931C>T, c.4120C>T and c.4414G>C in GRCh38 with the MANE Select transcript. [https://variantvalidator.org](https://variantvalidator.org). Accessed 6 October 2026.
6. Connell NT, Flood VH, Brignardello-Petersen R, Abdul-Kadir R, Arapshian A, Couper S, et al. ASH ISTH NHF WFH 2021 guidelines on the management of von Willebrand disease. Blood Adv. 2021;5(1):301–25. doi:10.1182/bloodadvances.2020003264. PMID:33570647; PMCID:PMC7805326.
7. Haute Autorité de Santé. Protocole National de Diagnostic et de Soins (PNDS): Maladie de Willebrand. Saint-Denis La Plaine: HAS; 9 Mar 2021 (text by the Centre de Référence de la Maladie de Willebrand, February 2021). [https://www.has-sante.fr/jcms/p_3236792](https://www.has-sante.fr/jcms/p_3236792). Accessed 6 November 2026.
8. Tosetto A, Badiee Z, Baghaipour MR, Baronciani L, Battle J, Berntorp E, et al. Bleeding symptoms in patients diagnosed as type 3 von Willebrand disease: results from 3WINTERS-IPS, an international and collaborative cross-sectional study. J Thromb Haemost. 2020;18(9):2145–54. doi:10.1111/jth.14886. PMID:32379400.
9. Takeda Pharmaceuticals America. VONVENDI [von Willebrand factor (recombinant)] — prescribing information. Revised July 2026. DailyMed set ID 53b6d198-6175-4a26-8606-dc0d2a0f12d6. DailyMed. Accessed 6 October 2026.
10. Octapharma USA. WILATE (von Willebrand factor/coagulation factor VIII complex [human]) — prescribing information. Revised July 2026. DailyMed set ID a8ef06b8-b041-054e-e58d-fd8a994a8071. DailyMed. Accessed 6 October 2026.
11. CSL Behring. HUMATE-P (antihemophilic factor/von Willebrand factor complex [human]) — prescribing information. Version published 26 August 2020. DailyMed set ID c5f4b63b-8d8b-45b3-b530-2f1c32a1b852. DailyMed. Accessed 6 October 2026.
12. Grifols. ALPHANATE (antihemophilic factor/von Willebrand factor complex [human]) — prescribing information. Revised July 2026 (approval letter dated 31 July 2026, labelling changes). DailyMed set ID 5a7aea94-654c-4113-9014-c3137de0a931. DailyMed. Accessed 6 October 2026.
13. Ferring Pharmaceuticals. DDAVP Injection (desmopressin acetate) — prescribing information. Version dated 28 September 2022. Set ID 651f6fee-a2c7-431b-8d5d-58b156c72244. openFDA. Accessed 6 October 2026.
14. Ferring Pharmaceuticals. STIMATE (desmopressin acetate) nasal spray — prescribing information. Version dated 22 July 2026. Set ID f5ada39c-20ed-4500-9dce-1e4ab642a358. openFDA. Accessed 6 October 2026.
15. European Medicines Agency. Medicines output — list of human medicines (JSON file), Veyvondi (authorised 31 August 2018) and Voncento (authorised 12 August 2013) entries. EMA. Accessed 6 October 2026.
16. Haute Autorité de Santé. Protocole National de Diagnostic et de Soins (PNDS): Maladie de Willebrand type 3. Saint-Denis La Plaine: HAS; 18 Jan 2022 (text and supporting argument by the Centre de Référence de la Maladie de Willebrand, 3 November 2021). [https://www.has-sante.fr/jcms/p_3309664](https://www.has-sante.fr/jcms/p_3309664). Accessed 6 November 2026.
17. Haute Autorité de Santé. Protocole National de Diagnostic et de Soins (PNDS): Saignements utérins abondants (SUA) chez la jeune femme atteinte de maladies hémorragiques rares constitutionnelles ou acquises (MHCA). Saint-Denis La Plaine: HAS; 12 Dec 2022 (text by the Centre de Référence des Pathologies Gynécologiques Rares, Toulouse, November 2022). [https://www.has-sante.fr/jcms/p_3389732](https://www.has-sante.fr/jcms/p_3389732). Accessed 6 November 2026.
18. ERN-EuroBloodNet. Subnetworks — Rare bleeding-coagulation disorders and related diseases [Internet]. [https://eurobloodnet.eu/eurobloodnet/subnetworks/](https://eurobloodnet.eu/eurobloodnet/subnetworks/). Accessed 6 October 2026.
19. Connell NT. Treatment of von Willebrand disease. Blood Adv. 2026;10(3):794–801. doi:10.1182/bloodadvances.2025016487. PMID:41201390; PMCID:PMC12887782.
20. Denis CV, Dich H, Casari C. Novel therapies for von Willebrand disease. Blood Adv. 2026;10(15):5253–63. doi:10.1182/bloodadvances.2025016488. PMID:42241704; PMCID:PMC13449370.
21. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Haemophilia and other inherited coagulation disorders, p. 442–444.
22. Werner EJ, Broxson EH, Tucker EL, Giroux DS, Shults J, Abshire TC. Prevalence of von Willebrand disease in children: a multiethnic study. J Pediatr. 1993;123(6):893–8. doi:10.1016/s0022-3476(05)80384-1. PMID:8229521.
23. Federici AB, Mannucci PM, Castaman G, Baronciani L, Bucciarelli P, Canciani MT, et al. Clinical and molecular predictors of thrombocytopenia and risk of bleeding in patients with von Willebrand disease type 2B: a cohort study of 67 patients. Blood. 2009;113(3):526–34. doi:10.1182/blood-2008-04-152280. PMID:18805962.
24. Franchini M, Mannucci PM. Alloantibodies in von Willebrand disease. Semin Thromb Hemost. 2018;44(6):590–4. doi:10.1055/s-0037-1607440. PMID:29165738.
25. National Center for Biotechnology Information. ClinVar: VCV000000308 (c.3614G>A), VCV000100307 (c.3931C>T), VCV000100329 (c.4120C>T), VCV000256679 (c.4414G>C) [Internet]. [https://www.ncbi.nlm.nih.gov/clinvar/](https://www.ncbi.nlm.nih.gov/clinvar/). Accessed 6 October 2026.
26. Castaman G, Federici AB, Tosetto A, La Marca S, Stufano F, Mannucci PM, et al. Different bleeding risk in type 2A and 2M von Willebrand disease: a 2-year prospective study in 107 patients. J Thromb Haemost. 2012;10(4):632–8. doi:10.1111/j.1538-7836.2012.04661.x. PMID:22329792.
27. Ragni MV, Rothenberger SD, Feldman R, Nance D, Leavitt AD, Malec L, et al. Recombinant von Willebrand factor and tranexamic acid for heavy menstrual bleeding in patients with mild and moderate von Willebrand disease in the USA (VWDMin): a phase 3, open-label, randomised, crossover trial. Lancet Haematol. 2023;10(8):e612–23. doi:10.1016/S2352-3026(23)00119-9. PMID:37385272.
28. Elbatarny M, Mollah S, Grabell J, Bae S, Deforest M, Tuttle A, et al. Normal range of bleeding scores for the ISTH-BAT: adult and pediatric data from the merging project. Haemophilia. 2014;20(6):831-835. doi:10.1111/hae.12503. PMID: 25196510
29. Hamilton A, Ozelo M, Leggo J, Notley C, Brown H, Frontroth JP, et al. Frequency of platelet type versus type 2B von Willebrand disease. An international registry-based study. Thromb Haemost. 2011;105(3):501–8. doi:10.1160/TH10-08-0523. PMID:21301777.
30. Fu A, Kazmirchuk TDD, Bradbury-Jost C, Golshani A, Othman M. Platelet-type von Willebrand disease: complex pathophysiology and insights on novel therapeutic and diagnostic strategies. Semin Thromb Hemost. 2025;51(2):219–26. doi:10.1055/s-0044-1789183. PMID:39191406.
31. Savoia A, Kunishima S, De Rocco D, Zieger B, Rand ML, Pujol-Moix N, et al. Spectrum of the mutations in Bernard-Soulier syndrome. Hum Mutat. 2014;35(9):1033–45. doi:10.1002/humu.22607. PMID:24934643.
32. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. Noonan syndrome, p. 148.
33. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. Cardio-facio-cutaneous syndrome, p. 156.
34. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 5.1, Hepatic glycogenoses, section 5.1.2 Glycogen storage disease type I — Coagulopathy.
35. Ribeiro EM, coordinator. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. Chapter 5.2, Hemofilias — Differential diagnosis, p. 109.
36. CSL Behring GmbH. Voncento — summary of product characteristics (EPAR, product information; text revised 11/2021). EMA. Accessed 6 October 2026.
37. U.S. Food and Drug Administration, CBER. VONVENDI (STN 125577): supplement approval letters dated 28 January 2022 (BL 125577/412) and 5 September 2025 (BL 125577/691). Product page. Accessed 6 October 2026.
38. Baxalta Innovations GmbH. Veyvondi (vonicog alfa) — summary of product characteristics (EPAR, product information) and post-authorisation procedural steps. EMA. Accessed 6 October 2026.
39. U.S. Food and Drug Administration, CBER. WILATE (BL 125251): supplement approval letters dated 1 December 2023 (BL 125251/382) and 2 July 2026 (BL 125251/454). Product page. Accessed 6 October 2026.
40. Peyvandi F, Castaman G, Gresele P, De Cristofaro R, Schinco P, Bertomoro A, et al. A phase III study comparing secondary long-term prophylaxis versus on-demand treatment with vWF/FVIII concentrates in severe inherited von Willebrand disease. Blood Transfus. 2019;17(5):391–8. doi:10.2450/2019.0183-18. PMID:30747707.
41. Kouides PA, Byams VR, Philipp CS, Stein SF, Heit JA, Lukes AS, et al. Multisite management study of menorrhagia with abnormal laboratory haemostasis: a prospective crossover study of intranasal desmopressin and oral tranexamic acid. Br J Haematol. 2009;145(2):212–20. doi:10.1111/j.1365-2141.2009.07610.x. PMID:19236375.
42. Nordic Pharma. Tranexamic acid tablets USP — prescribing information. Version dated 29 September 2025. Set ID 031bf7c2-3b07-469d-bc5a-31fc6238e25f. openFDA. Accessed 6 October 2026.
43. Pfizer. CYKLOKAPRON (tranexamic acid injection) — prescribing information. Version dated 3 September 2025. Set ID 6e89a7d9-4da4-42aa-b7f8-c602c24eefe5. openFDA. Accessed 6 October 2026.
44. College ter Beoordeling van Geneesmiddelen (CBG-MEB). Ferinject 50 mg ijzer/ml, dispersie voor injectie/infusie (ferric carboxymaltose): samenvatting van de productkenmerken, rubriek 4.2 [Internet]. Geneesmiddeleninformatiebank, RVG 33865 (European decentralised procedure). [https://www.geneesmiddeleninformatiebank.nl/smpc/h33865_smpc.pdf](https://www.geneesmiddeleninformatiebank.nl/smpc/h33865_smpc.pdf). Accessed 7 November 2026.
45. Leebeek FWG, Peyvandi F, Escobar M, Tiede A, Castaman G, Wang M, et al. Recombinant von Willebrand factor prophylaxis in patients with severe von Willebrand disease: phase 3 study results. Blood. 2022;140(2):89–98. doi:10.1182/blood.2021014810. PMID:35439298.
46. Sidonio RF Jr, Boban A, Dubey L, Inati A, Kiss C, Boda Z, et al. von Willebrand factor/factor VIII concentrate (Wilate) prophylaxis in children and adults with von Willebrand disease. Blood Adv. 2024;8(6):1405–14. doi:10.1182/bloodadvances.2023011742. PMID:38237075.
47. Takeda. A phase 3, prospective, open-label, uncontrolled, multicenter study on efficacy and safety of prophylaxis with vonicog alfa (rVWF) in children diagnosed with severe von Willebrand disease. ClinicalTrials.gov NCT05582993. Updated 9 September 2026; accessed 6 October 2026.
48. Vega Therapeutics. VIVID (NCT05776069, phase 1/2) and VIVID-6 (NCT07115004, phase 3): VGA039 in patients with VWD. ClinicalTrials.gov NCT05776069; NCT07115004. Accessed 6 October 2026.
49. Hoffmann-La Roche. A phase III study of emicizumab prophylaxis in patients with type 3 von Willebrand disease (NCT06998524); Bleeding and Clotting Disorders Institute. Emicizumab for severe VWD and VWD/hemophilia A (NCT05500807). ClinicalTrials.gov NCT06998524; NCT05500807. Accessed 6 October 2026.
50. Hemab ApS. A phase 1/2 study of HMB-002 in participants with von Willebrand disease (Velora Pioneer). ClinicalTrials.gov NCT06754852. Accessed 6 October 2026.
51. Ay C, Pabinger I, Kovacevic KD, Gelbenegger G, Schörgenhofer C, Quehenberger P, et al. The VWF binding aptamer rondoraptivon pegol increases platelet counts and VWF/FVIII in type 2B von Willebrand disease. Blood Adv. 2022;6(18):5467–76. doi:10.1182/bloodadvances.2022007805. PMID:35772170.
52. Medical University of Vienna. Efficacy and safety of BT200 (rondoraptivon pegol) in patients with type 2B von Willebrand disease. ClinicalTrials.gov NCT07273721. Accessed 6 October 2026.
53. Susen S, Trossaërt M, Zilberstein M, Mamikonian L, Santagostino E, Yarramaneni A, et al. A phase 1, open-label study to assess the pharmacokinetics, safety, and tolerability of a single intravenous injection of efanesoctocog alfa in adults with type 2N or type 3 von Willebrand disease. J Thromb Haemost. 2026;24(1):79–87. doi:10.1016/j.jtha.2025.07.013. PMID:40685140.
54. Alnylam Pharmaceuticals. HMBeacon: a phase 2 study to evaluate ALN-6400 in adult and adolescent female patients with VWD and heavy menstrual bleeding. ClinicalTrials.gov NCT07575308. Accessed 6 October 2026.
55. James AH, Konkle BA, Kouides P, Ragni MV, Thames B, Gupta S, et al. Postpartum von Willebrand factor levels in women with and without von Willebrand disease and implications for prophylaxis. Haemophilia. 2015;21(1):81–7. doi:10.1111/hae.12568. PMID:25333737.
56. Govorov I, Löfgren S, Chaireti R, Holmström M, Bremme K, Mints M. Postpartum hemorrhage in women with von Willebrand disease — a retrospective observational study. PLoS One. 2016;11(10):e0164683 (correction: PLoS One. 2017;12(2):e0172185). doi:10.1371/journal.pone.0164683. PMID:27780267.
57. Punt MC, Waning ML, Mauser-Bunschoten EP, Kruip MJHA, Eikenboom J, Nieuwenhuizen L, et al. Maternal and neonatal bleeding complications in relation to peripartum management in women with von Willebrand disease: a systematic review. Blood Rev. 2020;39:100633. doi:10.1016/j.blre.2019.100633. PMID:31718817.
58. Soucie JM, Miller CH, Byams VR, Payne AB, Abe K, Sidonio RF, et al. Occurrence rates of von Willebrand disease among people receiving care in specialized treatment centres in the United States. Haemophilia. 2021;27(3):445–53. doi:10.1111/hae.14263. PMID:33780098.
59. Mannucci PM, Bloom AL, Larrieu MJ, Nilsson IM, West RR. Atherosclerosis and von Willebrand factor. I. Prevalence of severe von Willebrand's disease in western Europe and Israel. Br J Haematol. 1984;57(1):163–9. doi:10.1111/j.1365-2141.1984.tb02876.x. PMID:6609712.
60. Casaña P, Martínez F, Haya S, Lorenzo JI, Espinós C, Aznar JA. Q1311X: a novel nonsense mutation of putative ancient origin in the von Willebrand factor gene. Br J Haematol. 2000;111(2):552–5. doi:10.1046/j.1365-2141.2000.02410.x. PMID:11122100.
61. Gupta S, Heiman M, Duncan N, Hinckley J, Di Paola J, Shapiro AD. Variable bleeding phenotype in an Amish pedigree with von Willebrand disease. Am J Hematol. 2016;91(10):E431–5. doi:10.1002/ajh.24474. PMID:27414491.
62. Weyand AC, Flood VH, Shavit JA, Pipe SW. Efficacy of emicizumab in a pediatric patient with type 3 von Willebrand disease and alloantibodies. Blood Adv. 2019;3(18):2748–50. doi:10.1182/bloodadvances.2019000656. PMID:31540901; PMCID:PMC6759735.
63. VWD Connect Foundation. The Severe von Willebrand Disease (sVWD) Patient Registry. ClinicalTrials.gov NCT05437536. Accessed 6 October 2026.