Pregnancy-Induced Shifts in Von Willebrand Factor (VWF) Parameters Support a Prolongation in VWF Survival

Abstract 33

Levels of the hemostatic protein von Willebrand Factor (VWF) are known to increase markedly during pregnancy. The mechanism by which VWF levels are increased has long been thought to be due to estrogen stimulation of vascular endothelial cells to synthesize and release more VWF. We hypothesized an alternate scenario in which prolongation of VWF survival contributes to the overall increase in VWF levels during pregnancy.

To test this hypothesis, we measured VWF parameters in a repository of plasma samples acquired longitudinally over healthy pregnancies. Third trimester and postpartum (≥ 6 weeks) samples were available for 45 pregnancies, while prepregnancy, first, and second trimester samples were also available on a subset of pregnancies. VWF propeptide (VWFpp) and VWF antigen (VWF:Ag) levels were determined by ELISA (GTI Diagnostics), Factor VIII:C (FVIII:C) was performed by a one-stage assay, and ADAMTS13 activity was measured via cleavage of an enzyme-linked peptide substrate. ABO blood group, a known modifier of VWF:Ag levels and VWF survival, was determined by ABO allele-specific PCR (GTI Diagnostics).

We found VWFpp, VWF:Ag, and FVIII:C levels were all significantly increased in the third trimester, as expected, while ADAMTS13 activity remained unchanged (see Table 1). Furthermore, the VWFpp:Ag ratio dropped by nearly half, consistent with a prolongation in VWF survival, while the VWFpp also rose, supporting a coexistent increase in VWF production. Unexpectedly, the decrease in the VWFpp:Ag ratio could be detected early during pregnancy, before the second trimester. Also surprisingly, we observed the VWF:Ag/FVIII:C ratio increase significantly during pregnancy. In a subset analysis of the data categorized by ABO blood group (group O, A, or B), the observed differences in VWF parameters between the third trimester and baseline remained statistically significant for all blood types.

Collectively, these data point to previously unsuspected and biologically relevant qualitative changes in VWF during pregnancy. In a model accounting for the volume expansion of pregnancy and evidence of increased VWF production (and assuming one compartment, single-order kinetics, and a stable VWFpp half-life), we calculated that an increase in the half-life of VWF (1.9 fold) would account for the observed excess in VWF:Ag in the third trimester. Furthermore, the change in the VWF/FVIII ratio suggests a pregnancy-induced alteration in the VWF:FVIII stoichiometry such that pregnancy-associated VWF carries less FVIII. This new model of pregnancy-induced VWF change has direct relevance to the prothrombotic state of pregnancy, and could relate to the increased risk of venous thromboembolism (VTE), the major cause of maternal mortality in the developed world. This fresh perspective on the shifts in VWF and FVIII levels which occur in healthy pregnancy may also lend new insight into the mechanisms underlying the derangements of VWF which accompany other life-threatening pregnancy-associated complications, such as preeclampsia and HELLP (hemolysis, elevated liver enzymes, and low platelet syndrome).