Ask the Hematologists: Von Willebrand Disease and Pregnancy

A 37-year-old woman 28 weeks into her first pregnancy is referred to hematology clinic given a history of von Willebrand disease (vWD) diagnosed during her teenage years, after presenting with excessive menstrual bleeding requiring oral iron supplementation. She had taken oral contraceptives until six months prior to this pregnancy with improvement in her bleeding symptoms. She has no family history of vWD and has never had prior surgery or dental extractions. Lab results obtained during her initial evaluation show a platelet count of 135 × 10⁹ /L, normal partial thromboplastin time and prothrombin time, ristocetin cofactor activity (vWF:RCo) 47 IU/dL (normal, 44-165 IU/dL), von Willebrand Factor (vWF) antigen (vWF Ag) 99 IU/dL (normal, 43-176 IU/dL), and FVIII activity 84 percent (normal, 50%-150%).

What is your approach to the diagnosis and management of vWD during pregnancy?

vWD is the most common inherited bleeding disorder, affecting up to 1 percent of the general population.¹  vWF is a large, multimeric protein needed for two crucial roles in hemostasis. First, it serves as a bridging molecule between platelets and subendothelial collagen exposed during vascular injury. Second, it acts as a chaperone protein for factor VIII, preventing premature degradation and increasing availability at sites of active thrombus formation.²  Quantitative (type 1 and 3) and qualitative (type 2, with 4 different subtypes) abnormalities in vWF protein may result in bleeding diatheses. The vast majority of cases are type 1, resulting in mild to moderate reduction of functionally normal vWF. Bleeding severity, however, is variable, and only a fraction of individuals with the disorder present with symptoms of bleeding. The gene encoding vWF is located on chromosome 12, spanning 178 kb over 52 coding exons. The qualitative mutations observed in type 2 vWD can result in 1) a defect in the intracellular assembly and transport of normal vWF multimers (type 2A); 2) an abnormal cleavage site in vWF causing increased susceptibility to proteolysis by ADAMTS13 (type 2A); 3) an abnormal interaction of vWF and platelets (type 2B, 2M, and platelet-type); or 4) impaired binding to FVIII resulting in low FVIII coagulant activity (type 2N).

Consequences of vWD During Pregnancy

During pregnancy, hormonal influences lead to an increase in vWF and clotting factors VII, VIII, and X while anticoagulant factors (such as protein S) decrease, shifting hemostasis to a procoagulant state to compensate for anticipated hemorrhage during parturition.³  Although vWF and FVIII levels rise and peak during the third trimester, women with vWD remain at risk of early pregnancy bleeding, as well as postpartum hemorrhage (PPH), immediate and delayed.⁴  This can be explained by the rapid fall of vWF after delivery.⁵  Other specific considerations include the need to address analgesia options, mode of delivery, and management of obstetric bleeding. This is best conducted with a multidisciplinary team for safest care of the mother and baby.

**<30 IU/dL is designated as the level for a definitive diagnosis of VWD; some patients with type 1 or type 2 VWD have levels of vWF:RCo and/or vWF:Ag of 30-50 IU/dL.

Adapted from the ASH Quick Reference Pocket Guide: 2012 Clinical Practice Guideline on the Evaluation and Management of von Willebrand Disease.

Laboratory Diagnosis of vWD in Pregnancy

Given the hormonal changes that occur during pregnancy, the diagnosis can be obscured and is ideally made prior to conception. For patients without a previous diagnosis, a history of excessive menstrual or mucocutaneous bleeding, prior PPH, bleeding after prior surgical/dental procedures, or family history may prompt diagnostic testing. In addition to a complete blood count (to evaluate platelet count) and standard coagulation studies (PT and PTT), recommended screening tests include plasma vWF:Ag, vWF activity (with ristocetin cofactor activity being most commonly performed, and the collagen binding assay less readily available), and FVIII activity. vWF:Ag determination informs the total amount of vWF in the plasma, but the assay does not distinguish between active or inactive protein. vWF:RCo is a functional assay, testing the ability of patient vWF to bind to GPIb in the presence of ristocetin (an antibiotic that binds both vWF and GPIb), causing agglutination. In patients with type 2A, 2B, and 2M vWD, the qualitative protein defect causes greater decline in activity level than antigen concentration, resulting in a vWF:RCo-vWF:Ag ratio less than 0.5 to 0.7 (Table 1).

Abnormal vWF levels (<30-40%) and/or a low vWF:RCo-vWF:Ag ratio should prompt additional testing, such as vWF multimer pattern using gel electrophoresis, and ristocetin-induced platelet aggregation (RIPA). In type 1 vWD, multimers are normal or mildly decreased with normal distribution. There is loss of high-molecular-weight multimers (HMWM) in types 2A and 2B, whereas all multimers are present in types 2M and 2N. Type 2N affects the FVIII binding site on the vWF protein, resulting in low-circulating FVIII and normal vWF protein. Type 2B variant is characterized by gain of function mutations in the GPIb binding site of the vWF protein, causing a conformational change leading to enhanced affinity for GPIb. This can be demonstrated by performing RIPA, a type of platelet aggregation study using platelet-rich plasma (PRP) to screen for the ability of the patient vWF to bind to platelet GPIb in the presence of a low concentration of ristocetin (0.5-0.6 mg/dL). With normal vWF protein, aggregation does not occur at ristocetin concentrations below 0.6 to 0.7 mg/mL. In contrast, type 2B variant is associated with hyper-responsiveness to a low ristocetin concentration. vWF gene analysis is useful to confirm type 2 and type 3 diseases as the mutations are localized within discrete regions of the vWF gene.⁶  For instance, the majority of 2A, 2B, and 2M vWD have sequence variations in exon 28, and genetic analysis usually begins in this region. Analytic sensitivity is reported to be more than 99 percent for reported mutations, with results typically available within three weeks.⁷  In contrast, genetic testing for type 1 vWD has not been shown to be clinically useful in general.

Thrombocytopenia is frequently seen in patients with Type 2B vWD, occurring in about 40 to 50 percent of cases, and it has been shown to be an independent risk factor for bleeding in such patients.⁸  Thrombocytopenia may develop or worsen during pregnancy.⁸  Differentiation between type 2B vWD and the rarer platelet-type vWD, which produces a similar phenotype due to gain of function mutation in the platelet receptor GPIb, can be challenging. One way to try to distinguish these possibilities is by performing RIPA mixing studies. This RIPA study is performed by mixing a patient’s platelets with normal vWF, and conversely, patient plasma vWF with normal platelets using low concentrations of ristocetin to help determine where the defect exists.⁹  Confirmatory molecular analysis should be performed to verify the diagnosis.

Treatment of vWD During Pregnancy

Antepartum hemorrhage is uncommon, but PPH has been observed in up to 37.5 percent of women with type 2 vWD not receiving adequate prophylaxis.¹⁰  It is therefore recommended that women have their vWF/FVIII levels monitored during the third trimester to facilitate planning for delivery. The need for therapeutic intervention depends on level of vWF and FVIII, thrombocytopenia, and type of vWD. With the rise of FVIII and vWF during pregnancy, women with mild type 1 vWD rarely need treatment, while type 3 disease (where vWF and FVIII do not increase with pregnancy) will require factor replacement. When FVIII or vWF:RCo levels are less than 50 IU/dL, prophylaxis is recommended prior to vaginal or surgical delivery.¹¹  DDAVP, a synthetic analogue of vasopressin that increases vWF and FVIII, is an option for certain patients; however, there is the potential risk of hyponatremia at the time of delivery. In a 2011 Argentinean cohort of 54 pregnant women with vWF levels lower than 50 IU/dL, none treated with DDAVP preprocedurally or prior to epidural developed complications. The agent therefore, may be a safe option in pregnancy.¹²  In type 2 vWD, less benefit is seen with DDAVP because of the qualitative defect. Furthermore, in type 2B, use of DDAVP can exacerbate thrombocytopenia, leading to an additional risk of bleeding; this agent is contraindicated if a persistent decrease in platelet count has been documented. Replacement products are needed for type 2B and 3 vWD. Plasma-derived vWF containing concentrates licensed in the United States for the treatment of vWD include Humate P (CSL Behring), Wilate (Octapharma), and Alphanate (Grifols). More recently, recombinant vWF (rvWF; Vonvendi) has become commercially available. In phase III studies, use of rvWF was highly effective in restoring hemostasis, with hemostatic levels achieved within six hours and sustained for up to 72 hours following infusion.¹³  Platelet transfusions may be needed for type 2B, with transfusion thresholds of 50 × 10⁹ /L frequently used. Other adjunctive options include antifibrinolytic agents such as tranexamic acid and aminocaproic acid in the postpartum setting. Although they have not been studied specifically in the management of PPH, retrospective studies have shown reduction in rates of PPH when used for up to three weeks following discharge.¹⁴  Finally, while there is no consensus on factor levels needed for safe regional anesthesia during labor and delivery, it can be considered if FVIII and vWF:RCo are above 50 IU/dL.⁷ 

Patient Follow-Up

This patient’s baseline vWF activity at 28 weeks was lower than would be expected for the stage of pregnancy. Type 2 vWD was considered in this case given the discrepancy between vWF:RCo and vWF:Ag, with a ratio of 0.47 and relatively low vWF:RCo levels. Later in the third trimester, she developed worsening thrombocytopenia with a platelet count of 70 × 10⁹ /L. vWF activity improved during the latter part of the pregnancy (vWF Activity 55%; vWF Ag 145%), but it is expected to decline following delivery. Multimer analysis showed loss of high-molecular weight multimers. RIPA testing was performed (Table 2), showing typical aggregation findings observed in type 2B vWD, including the hyper-responsiveness to a low ristocetin concentration (see Table 2 results in red type for patient PRP aggregation percentage with ristocetin concentrations of 0.5-0.7 mg/mL).

Sequence analysis of exon 28 showed a R578Q mutation in the GPIb binding site of the vWF protein, confirming the diagnosis of type 2B vWD. Given the type 2B variant and persistent thrombocytopenia, DDAVP is contraindicated. The delivery plan includes administration of vWF concentrates at time of delivery, possible platelet transfusion, and oral tranexamic acid after discharge.