Hollestelle MJ, Loots CM, Squizzato A, et al. . J Thromb Haemost. 2011;9:953-958.

Excess bleeding is noted in up to 25 percent of patients with severe aortic stenosis. The associated platelet dysfunction, detected by prolonged platelet function analyzer (PFA-100) closure times, which measures hemostasis under shear, resolves after valve replacement surgery. Previous investigators hypothesized that bleeding in patients with aortic stenosis (AS) was related to the high shear force of a stenotic aortic valve.1  The shear associated with a high aortic valve gradient of severe AS leads to stretching and unfolding of von Willebrand factor (vWF), facilitating cleavage by ADAMTS13 within the A2 domain, which leads to acquired von Willebrand disease (vWD) with loss of the largest vWF multimers. Although several studies have confirmed the inverse relationship between the aortic valvular gradient and vWF multimer size, the problem remains that some patients with severe AS and bleeding have no decrease in large vWF multimers.

Further evidence suggests that the conformational changes required for vWF binding to platelets are found not only in high-molecular-weight multimers, but also in lower-molecular-weight multimers, indicating that the loss of large vWF multimers might not be the sole explanation for this process. Moreover, platelet activation correlates with valvular pressure gradient even in patients with mild or moderate AS,2  suggesting the potential importance of platelet activation in bleeding in AS patients.

Hollestelle and colleagues from Utrecht in the Netherlands measured vWF activation (i.e., platelet-bound vWF activity) in the plasma of patients with various thrombocytopenias and/or thrombotic disorders3  and proposed an alternate explanation. They reasoned that increased shear in severe AS decreases vWF multimer size, and, since high-molecular-weight-vWF multimers preferentially bind the platelet glycoprotein Ibα (GPIbα) receptor, there is less available “active” platelet-bound vWF. This accounts for the platelet dysfunction and potential bleeding in AS patients.

To determine whether “active” vWF (vWF in platelet-binding GPIbα conformation) is decreased in AS, the investigators tested plasma samples from 62 adult patients using a recombinant antibody fragment (AU/vWF-11)-based ELISA. Among severe AS, defined as mean valve area of 0.6 cm2 and mean gradient 60 mmHg, they found that vWF activation was reduced. In fact, there was a strong negative correlation between vWF activation and aortic gradient: vWF activation decreased as valvular gradient increased. The proportion of vWF able to bind the GPIbα in patients with severe AS was significantly lower than in those with less severe AS.

Table

Table
Aortic StenosisMildModerateSevereCorrelation and p-values
with AS gradientwith PFA-100
No. Patients N=6 N=28 N=28     
AS valve area 1.6 cm2 0.9 cm2 0.6 cm2     
AS valve gradient (mean) 10.5 mmHg 32.0 mmHg 60.5 mmHg     
PFA-100 93 sec 128 sec 208 sec     
Bleeding score 1.5 1.0 1.0     
vWF activation factor* 1.56 1.27 1.01 r=0.47, p=0.01 r=-.38, p=0.01 
vWF Ag 199 U/dl 173 U/dl 147 U/dl NS NS 
vWF propeptide 8.6 nM 8.4 nM 6.7 nM NS NS 
ADAMTS-13 77% 94% 102% r=0.34, p=0.01 NS 
Multimers 98% 96% 96% NS NS 
Aortic StenosisMildModerateSevereCorrelation and p-values
with AS gradientwith PFA-100
No. Patients N=6 N=28 N=28     
AS valve area 1.6 cm2 0.9 cm2 0.6 cm2     
AS valve gradient (mean) 10.5 mmHg 32.0 mmHg 60.5 mmHg     
PFA-100 93 sec 128 sec 208 sec     
Bleeding score 1.5 1.0 1.0     
vWF activation factor* 1.56 1.27 1.01 r=0.47, p=0.01 r=-.38, p=0.01 
vWF Ag 199 U/dl 173 U/dl 147 U/dl NS NS 
vWF propeptide 8.6 nM 8.4 nM 6.7 nM NS NS 
ADAMTS-13 77% 94% 102% r=0.34, p=0.01 NS 
Multimers 98% 96% 96% NS NS 

*vWF activation factor is the proportion of circulating vWF that bind to the platelet GP1bα receptor.

vWF activation in these patients was also negatively correlated with PFA-100. Specifically, vWF activation levels decreased with increasing prolongation of PFA-100 closure times. Closure times were also strongly correlated with valve gradient, with longer closure times noted at higher aortic valve gradients (Table).

By contrast, while there was no correlation between vWF parameters (antigen or propeptide) and valve gradient or PFA-100, vWF parameters did correlate with β-thromboglobulin, a marker of in vivo platelet activation. The authors concluded that the observed decreased vWF activation (i.e., ability to bind GP1bα) with increasing severity of aortic stenosis was likely related to depletion of vWF stores in the vascular endothelium.

So, what are the implications for the bleeding patient with AS? While the novel findings of Hollestelle et al. suggest clinical bleeding with AS arises from a decrease in vWF A1 binding to platelet GPIbα receptor, not a decrease in vWF multimers, management of such patients would still require vWF concentrates to reduce clinical bleeding. With demonstrated depletion of vWF stores, neither DDAVP nor platelets would be expected to help. In the home “stretch,” that is, before valve surgical repair, one wonders whether a pharmacologic agent, such as an aptamer, that can alternately turn on (promote) or off (block) vWF-platelet interaction might provide a novel approach.

1.
Zhang X, Halvorsen K, Zhang CZ, et al. Mechanoenzymatic cleavage of the ultralarge vascular protein von Willebrand factor. Science. 2009;324:1330-1334.
2.
Groot E, de Groot PG, Fijnheer R, et al. The presence of active von Willebrand factor under various pathological conditions. Curr Opin Hematol. 2007;14:284-289.
3.
Hulstein JJ, de Groot PG, Silence K, et al. A novel nanobody that detects the gain-of-function phenotype of von Willebrand factor in ADAMTS13 deficiency and von Willebrand disease type 2B. Blood. ASH Annual Meeting Abstract. 2005;106:3035-3042.

Competing Interests

Dr. Ragni indicated no relevant conflicts of interest.