Functional Impairment of Von Willebrand Factor in Hypertrophic Cardiomyopathy
Relation to Rest and Exercise Obstruction
Background— Hypertrophic obstructive cardiomyopathy submits blood to conditions of high shear stress. High shear stress impairs von Willebrand factor (VWF) and promotes abnormal bleeding in aortic stenosis. We sought to evaluate VWF impairment and its relationships to baseline or exercise obstruction in hypertrophic cardiomyopathy (HCM).
Methods and Results— Outflow obstruction was evaluated by rest and exercise echocardiography in 62 patients with HCM (age 44±16 years, 40 males). HCM was considered obstructive in 28 patients with rest or exercise peak gradient ≥30 mm Hg. Blood was sampled to assess VWF. History of bleeding was recorded. Baseline median (25th to 75th percentile) peak gradient was 11 (5–62) mm Hg. Shear-induced platelet adhesion was impaired in patients with obstructive HCM. The ratio of VWF–collagen-binding activity to antigen and the percentage of high-molecular-weight multimers of VWF were lower in patients with obstructive HCM than in those with nonobstructive HCM (0.49 [0.43 to 0.59] versus 0.82 [0.73 to 1.03] and 5.0% [3.9% to 7.2%] versus 11.7% [10.8% to 12.5%], respectively; both P<0.0001). Platelet adhesion time, VWF–collagen-binding activity–to-antigen ratio, and the percentage of high-molecular-weight multimers correlated closely and independently with peak gradient (r=0.81, r=−0.68, and r=−0.89, respectively; all P<0.0001). According to receiver operating characteristic curves, a peak gradient threshold of 15 mm Hg at rest and 35 mm Hg during exercise was sufficient to impair VWF. Conversely, VWF function tended to improve with a decrease in peak gradient. Obstructive HCM patients had a trend toward abnormal spontaneous bleeding.
Conclusions— In obstructive HCM, VWF impairment is frequent and is closely and independently related to the magnitude of outflow obstruction. A resting peak gradient of 15 mm Hg is sufficient to impair VWF. VWF abnormalities might favor abnormal bleeding in this setting.
Received April 15, 2008; accepted August 5, 2008.
In patients with hypertrophic cardiomyopathy (HCM), baseline left ventricular (LV) outflow tract obstruction is an important feature and is present in 25% of patients.1 Baseline obstruction, which is associated with heart failure symptoms and poor outcome,2,3 is an important therapeutic goal in these patients.1,4 Obstruction varies substantially with numerous physiological alterations in HCM and can be elicited with exercise in up to 70% of patients.5 The long-term consequences of outflow obstruction are thought to be related to LV wall stress and myocardial ischemia.1 However, beyond its known effect on myocardium, outflow gradient could be involved in other pathophysiological processes through increases in shear stress level.
Clinical Perspective p 1557
Von Willebrand factor (VWF) is a multimeric plasma glycoprotein synthesized by endothelial cells and megakaryocytes that is required for normal hemostasis.6 The largest high-molecular-weight multimers (HMWMs) of VWF are most effective in mediating platelet binding to the extracellular matrix at sites of vascular injury7 and are cleaved rapidly when subjected to high shear stress.8 Impairment of primary hemostasis, characterized by the loss of the largest multimers of VWF, has been reported previously in aortic stenosis (AS).9–12 This loss is related to enhanced proteolysis of VWF under conditions of high shear stress and correlates with transvalvular gradient in AS.7,12,13 VWF impairment is thought to be the link between hemorrhagic syndrome and AS (Heyde’s syndrome).7,12,13 Given the unique shear stress characteristics of VWF, we hypothesized that high outflow tract velocity at rest or during exercise might increase proteolysis of VWF and impair primary hemostasis in the obstructive form of HCM.
Therefore, the purpose of the present study was to evaluate the presence and magnitude of VWF abnormalities in HCM and assess their relationships with baseline or exercise obstruction. We also sought to evaluate the occurrence of spontaneous bleeding in HCM.
Sixty-two consecutive patients (22 women and 40 men) referred for evaluation of HCM were enrolled in the study. Patients had typical echocardiographic features of HCM, with a maximal LV wall thickness ≥15 mm or ≥13 mm in the setting of a familial history in the absence of another cardiac or systemic disease that could produce this hypertrophy.1 Patients with inflammatory disease or significant valve disease other than mitral regurgitation and those receiving antiplatelet treatment at the time of evaluation were excluded from the present study. Warfarin does not influence the results of the primary hemostasis tests used in the present study and was not a criterion for exclusion. Patients with disability or comorbidity that prevented the performance of exercise and those with end-stage heart failure or documented ventricular arrhythmia were not included. A history of spontaneous minor or major bleeding or blood transfusion from the time of first diagnosis of the disease was assessed. Major bleeding was defined as clinically overt bleeding that led to a significant decrease in hemoglobin level (≥2 g/dL) and that required blood transfusion and hospitalization. Written informed consent was obtained from each patient, and the local ethics committee approved the study.
Rest and Exercise Echocardiography
A comprehensive echocardiography examination was performed with a Hewlett-Packard Sonos 5500 (Hewlett-Packard, Palo Alto, Calif) system. Baseline peak LV outflow tract gradient was measured with continuous-wave Doppler. LV wall thickness and LV cavity size were measured. LV ejection fraction was measured by the biplane method of disks from the apical window. A symptom-limited semisupine bicycle exercise test was performed with a standard ergometer protocol (for patients with a baseline gradient ≤50 mm Hg). Blood pressure was measured every 2 minutes. A 12-lead ECG was monitored continuously. Peak gradient was monitored during exercise and for 6 minutes after the end of exercise. Medical therapy was left unchanged for the study. Patients with a baseline peak gradient ≥30 mm Hg (baseline obstruction), which is an independent predictor of poor outcome,2 were identified by rest echocardiography. Latent obstruction was defined as a peak gradient <30 mm Hg at rest and ≥30 mm Hg during exercise. Hence, HCM was further classified as obstructive (baseline or exercise peak gradient ≥30 mm Hg) or nonobstructive after exercise.5
A blood sample was collected for the assessment of VWF, hemoglobin level, and platelet count 126±102 minutes before echocardiography in the entire patient population. Blood samples were collected again within 1 hour after exercise echocardiography and within 24 hours after exercise to assess the effect of an exercise-free night on VWF function. Blood samples were collected during follow-up (mean delay 308±224 days) from 32 patients who returned for a scheduled medical appointment related to the standard follow-up of their disease. There was no scheduled follow-up appointment in the design of the study.
Primary Hemostasis Assessment
Primary hemostasis was tested with a platelet-function analyzer (PFA-100; Dade Behring, Deerfield, Ill) by determining closure time with collagen-epinephrine (PFA-100 epinephrine, normal value <170 seconds) or collagen-ADP (PFA-100 ADP, normal value <114 seconds) cartridges.12 The platelet-function analyzer (PFA) is a high-shear system for in vitro testing of platelet function that simulates primary hemostasis after injury to a small vessel. It is a highly sensitive way to screen patients for VWF defects.14 Plasma VWF antigen (VWF:Ag) was measured by immunoturbidimetry (Liatest; Diagnostica Stago, Asnieres, France). Functional analysis of VWF was performed by measuring its collagen-binding activity with an ELISA (VWF:CB, normal value >50%) as described previously,15 with the use of equine type I collagen (Horm; Nycomed, Oslo, Norway). The ratio between VWF:CB and VWF:Ag was calculated (VWF:CB/Ag, normal value >0.7). The multimeric structure of plasma VWF was analyzed by electrophoresis with 0.1% SDS and 1.5% agarose gel.16 The percentage of the HMWMs (%HMWM, >15 mers) was determined after densitometric scanning, as described previously.12,16–18 A normal plasma was used as reference in each gel electrophoresis. The lower limit of the normal range for the %HMWM is 10.5%, as reported previously.12 The result of a primary hemostasis assay was considered impaired when the value of the assay was beyond the predefined normal value. Interseries variability assessed in plasma controls (n=30) for VWF: Ag (3.3%), VWF:CB (8.9%), and HMWM >15 mers (10.4%) was judged satisfactory. The accuracy of the closure time with collagen-epinephrine or collagen-ADP cartridges is evaluated to 12.4% and 12.7%, respectively, by the manufacturer.
The normality of distribution was tested with the Shapiro-Wilk test for each variable and in each subgroup. Age, hemoglobin level, and platelet count were normally distributed and were expressed as mean±SD. Other variables were expressed as median (25th to 75th percentile) or percentage. Comparisons between groups were performed with χ2 tests, Fisher exact tests, Student’s t tests, Mann–Whitney U, or Wilcoxon signed rank tests, as appropriate. Comparisons between 3 subgroups according to obstruction status were performed by nonparametric ANOVA (Kruskal-Wallis tests). Post hoc comparisons between groups were performed with Mann–Whitney U tests. A repeated-measures ANOVA was used to assess changes in VWF function over time with a 2-way analysis for comparison between groups. Correlations between variables were assessed by linear regression analysis, except for the relationship between primary hemostasis tests and outflow gradient, which was log-linear. Linearity assessment was performed by plotting studentized residuals against each covariable. To identify independent clinical or echocardiographic factors associated with primary hemostasis parameters, all variables that correlated with a probability value less than 0.1 in univariate analysis were submitted to a multivariate regression analysis adjusted for age and gender. The receiver operating characteristic curves were used to evaluate the peak gradient threshold sufficient to impair VWF function (value of each hemostasis assay beyond the normal value). The thresholds were defined when sensitivity and specificity were maximized. A 2-tailed probability value ≤0.05 was considered significant. Statistical analyses were performed with SPSS software version 13 (SPSS, Chicago, Ill).
The authors had full access to the data and take full responsibility for its integrity. All authors have read and agree to the manuscript as written.
Clinical and Echocardiography Data
Baseline clinical and echocardiography data for the entire patient population are shown in Table 1. Mean age was 44±16 years, and 40 patients (65%) were male. The median delay between the time of first diagnosis of HCM and enrollment in the present study was 7.6 (1.3 to 12.7) years (up to 26 years). Thirty-one patients were in New York Heart Association functional class I (50%), 25 in class II (40%), and 6 in class III (10%). Eight patients (13%) were in atrial fibrillation. Septal thickness was 20(16–23) mm, and LV ejection fraction was 66% (63% to 70%). For the overall study group, the baseline peak gradient ranged from 3 to 130 mm Hg, with a median value of 11 (5–62) mm Hg. Baseline peak gradient was ≥30 mm Hg in 21 patients and >50 mm Hg in 16 patients. Characteristics of patients stratified according to baseline peak gradient <30 or ≥30 mm Hg are shown in Table 1.
Exercise was performed under current medications in the 46 patients with a baseline peak gradient ≤50 mm Hg, which elicited a maximal peak gradient that ranged from 4 to 170 mm Hg. Exercise-induced or latent obstruction was found in 7 patients. After exercise testing, HCM was therefore considered as obstructive in 28 patients and nonobstructive in 34. Characteristics of patients stratified according to the result of exercise echocardiography are shown in Table 2.
Baseline Peak Gradient and VWF Function
Platelet count and hemoglobin levels did not differ between patients stratified according to baseline peak gradient. The VWF:Ag level was normal in all cases. Quantitative results of the primary hemostasis assays stratified according to baseline peak gradient are shown in Figure 1. Values of the 5 primary hemostasis assays were significantly impaired in patients with baseline obstruction (all P<0.0001). When the normal values of these hemostasis assays were considered, the closure time determined by the PFA-100 in the presence of collagen and either epinephrine or ADP was significantly prolonged in all but 1 patient (95%) with baseline obstruction. In patients with baseline obstruction, all but 1 patient (95%) had a significant impairment of VWF:CB/Ag, and all patients (100%) had a substantial reduction in %HMWM that ranged from 1.7% to 8.4% (median 4.6% [3.1% to 5.5%]).
Exercise and VWF Function
Results of the primary hemostasis assays in patients without obstruction, with latent obstruction, and with baseline obstruction are shown in Table 3. VWF function was significantly impaired in the presence of baseline or exercise obstruction (all P<0.0001), with an intermediate pattern compared with the 2 other subgroups for patients with exercise obstruction (Table 3). The closure times determined by the PFA-100 with collagen-epinephrine or collagen-ADP cartridges were strongly impaired in obstructive patients (293 [200–300] versus 114 [102–154] seconds and 218 [151–271] versus 97 [87–118] seconds, respectively; both P<0.0001). VWF:CB/Ag and %HMWM were also strongly impaired in obstructive versus nonobstructive patients (0.49 [0.43 to 0.59] versus 0.82 [0.73 to 1.03] and 5.0% [3.9% to 7.2%] versus 11.7% [10.8% to 12.5%], respectively; both P<0.0001). Given the normal values of these hemostasis assays, VWF function was impaired in all but 1 obstructive patient (96%) with a borderline exercise obstruction. In the obstructive group, PFA-100 with epinephrine was prolonged in 93% of patients, as it was in 89% of patients with ADP. All but 2 patients (93%) with obstruction had a decrease in VWF:CB/Ag, and 89% of patients with obstruction had a decrease in %HMWM.
In the 46 patients who underwent exercise echocardiography, repeated hemostasis assays (before and after exercise and after an exercise-free night) demonstrated the stability of measurements overall (P=NS for time effect for each subgroup). The difference between obstructive and nonobstructive patients remained highly significant for each hemostasis assay over time (all P<0.0001).
Relation Between Peak Gradient and VWF Function
Relations between primary hemostasis assays and baseline peak gradient are shown in Figure 2. PFA values correlated closely with baseline peak gradient (epinephrine, r=0.81; ADP, r=0.81; both P<0.0001). VWF:CB, VWF:CB/Ag, and the %HMWM also correlated closely with baseline peak gradient (r=−0.64, r=−0.68, and r=−0.89, respectively; all P<0.0001; Figure 2). When adjusted for age and gender, the association between VWF function (PFA-100, VWF:CB, VWF:CB/Ag, and %HMWM) and peak gradient remained highly significant in a multiple regression analysis (for each hemostasis assay, P<0.0001). In the 46 patients who underwent exercise, the relationships between hemostasis assays and maximal peak gradient during exercise were also highly significant for PFA (epinephrine, P<0.0001; ADP, P=0.006) and for VWF:CB, VWF:CB/Ag, and %HMWM (P=0.001, P<0.0001, and P<0.0001, respectively).
Threshold Value of Peak Gradient Impairing VWF Function
We aimed to evaluate the threshold value of peak gradient that led to an impairment of VWF (value of each hemostasis assay beyond the predefined normal value) in HCM. Receiver operating characteristic curve analyses (Figure 3) demonstrated that a baseline peak gradient that ranged from 13 to 17 mm Hg (mean 15 mm Hg) led to VWF impairment according to the 5 hemostasis assays. When the same analysis was performed in the 46 patients who exercised (Figure 3), the threshold value of maximal peak gradient during exercise that led to VWF impairment ranged from 32 to 37 mm Hg (mean 35 mm Hg) according to the 5 hemostasis assays.
VWF assessment and echocardiographic follow-up were available for 32 patients (52%; obstruction, n=22; no obstruction, n=10) 265 (131–401) days after the first assessment. Relationships between peak gradient and hemostasis assays remained quite similar at the time of follow-up. Outflow gradient changes during follow-up correlated with changes in VWF function, as assessed by VWF:CB, VWF: CB/Ag, and %HMWM (Table 4), which suggests that a decreased obstruction level was associated with an improvement in VWF-related hemostasis. The strongest correlation was found between peak gradient changes and %HMWM changes (r=−0.60, P=0.001). In 8 patients, treatment led to a substantial decrease in peak gradient (≥30 mm Hg; 89 [65–100] to 44 [17–55] mm Hg, P=0.008) associated with a marked increase in VWF:CB (49% [40% to 61%] to 74% [59% to 93%], P=0.008), VWF:CB/Ag (51 [43–56] to 61 [57–71], P=0.008), and %HMWM (5.9% [4.7% to 6.4%] to 8.7% [6.9% to 9.8%], P=0.008).
A past history of blood transfusion was recorded in 1 patient with nonobstructive HCM and in 5 patients with obstructive HCM (P=0.03; spontaneous major bleeding in 3 patients, after abdominal surgery in 1 patient, and after pacemaker implantation in 1 patient). In the obstruction group, 8 patients presented with a history of abnormal spontaneous bleeding without intake of any anticoagulant or antiplatelet drug (5 patients with minor bleeding [epistaxis and ecchymosis, menorrhagia] and 3 with spontaneous major bleeding related to epistaxis [1 episode] and gastrointestinal bleeding [3 episodes]). Two patients without obstruction had a history of minor bleeding, but none experienced spontaneous major bleeding. Hence, the estimated incidence of spontaneous major bleeding was 1.77% per year in patients with obstructive HCM and only 0.66% per year in the overall group of patients. Patients with a history of spontaneous bleeding (n=10) had greater VWF impairment on the basis of the 5 functional tests used (all P<0.05) at the time of the study.
The present study demonstrates that primary hemostasis impairment is common in the obstructive form of HCM. Outflow tract obstruction impairs VWF function through enhanced proteolysis of VWF multimers. The biological effect of obstruction on VWF function is detectable in patients with either baseline or latent (exercise-induced) obstruction. Primary hemostasis assays that assess VWF function impairment strongly correlate with the peak gradient and improve with gradient reduction. Primary hemostasis impairment might favor spontaneous bleeding in patients with an obstructive form of HCM.
Impairment of von Willebrand Function
The effect of high-shear-stress cardiac lesions on VWF was first suggested by Warkentin et al10 in the setting of AS. VWF impairment, defined as an acquired von Willebrand syndrome type 2A, is common in this setting and directly relates to the level of transvalvular gradient.12 Acquired von Willebrand syndrome has also been described in limited series or case reports of patients with other types of high-shear-stress cardiovascular disorders.19–22 Experimental and biochemical data have demonstrated that high-shear forces could induce structural changes in the shape of the VWF molecule, facilitating the action of the specific VWF protease ADAMTS 13, which would lead to the loss of the HMWM of VWF.8 The enhanced proteolysis of the largest VWF multimers is thought to be the underlying link between AS and bleeding syndrome.12,13,19
HCM is a relatively frequent genetic cardiac disorder (≈1:500 in the general population) with a high phenotypic variability that leads to baseline outflow tract obstruction in ≈25% of patients overall1 but in up to 70% of patients with exercise.5 The present study demonstrates that high outflow tract velocity, in addition to its known effect on myocardium,1 also impairs VWF. The normal value of VWF:Ag associated with a marked reduction in the %HMWM suggests that the hemostatic defect is related to direct proteolysis of the largest multimers of VWF, which defines a type 2A von Willebrand syndrome.
Peak Gradient Level Impairing VWF
The percentage of HMWM in the baseline obstructive form of HCM (median 4.6% [3.1% to 5.5%]) appears to be lower than reported previously in AS (median 8.7%).12 It is noteworthy that a resting peak gradient of ≈15 mm Hg was able to impair VWF function in the present study. In the same way, an exercise peak gradient of ≈35 mm Hg was sufficient to impair VWF. These observations suggest that outflow gradient in HCM has a greater effect on VWF than transvalvular aortic gradient. The main explanation for this difference lies in the dynamic pattern of obstruction compared with AS.23 It is well known that outflow gradient in HCM varies substantially with numerous alterations such as a heavy meal, alcohol ingestion, or exercise, which can elicit severe obstruction despite a low (<30 mm Hg) or moderate (30 to 50 mm Hg) baseline gradient.1,5,24
VWF and the Magnitude of Obstruction
Despite the lability and variability of obstruction in HCM,1 we found that overall, there was a tight correlation between VWF function impairment and peak gradient in patients with HCM. In addition, the peak gradient was the strongest independent determinant of VWF impairment. Moreover, the values of the 5 hemostasis assays remained stable over 24 hours, despite the increased gradient during exercise and even after a night of bed rest. Therefore, we can hypothesize that VWF impairment likely reflects a cumulative effect of obstruction episodes over time on circulating VWF not compensated by physiological release. This result is consistent with the rapid cleavage of VWF multimers when subjected to high shear stress and the relatively long half-life of the multimers (12 hours).7,8 Other parameters such as the duration of obstruction during systole, the persistence of obstruction, and the repetition of obstruction episodes may also account for the magnitude of VWF impairment.
Patients with latent obstruction exhibited an intermediate pattern of VWF impairment compared with patients with nonobstructive and baseline obstructive HCM. They tended to have higher gradients at rest than patients with nonobstructive HCM (15 [11–20] versus 6 [5–9] mm Hg, P<0.005), which reflects the logarithmic relationship between gradient and VWF function (Figure 2). Finally, the lower VWF impairment in these patients is consistent with the cumulative effect hypothesis, with VWF impairment reflecting an average effect of daily nonobstructive and obstructive episodes.
In the present study, we also found a good correlation between peak gradient changes and changes in hemostasis assays during follow-up in the overall population. Substantial peak gradient reduction with medical treatment in 8 patients with obstruction was associated with a marked improvement of VWF abnormalities. This observation extends and emphasizes the present results and suggests that VWF impairment can be reversed with gradient reduction in patients with obstructive HCM.
VWF and Bleeding
Acquired von Willebrand syndrome is a recognized risk factor for cutaneous or mucosal bleeding, particularly from gastrointestinal angiodysplasia.12,19–21 It is thus reasonable to hypothesize that acquired VWF syndrome may favor bleeding in the obstructive form of HCM. The occurrence of severe angiodysplasia hemorrhage has been described in HCM,25–28 and VWF impairment has been demonstrated recently in 1 patient.22 In some cases, bleeding cessation was obtained after initiation of β-blocker or calcium antagonist therapy,22,26 after septal myectomy,27 or after percutaneous alcohol septal ablation.28 Although an acquired VWF syndrome has been suspected previously in HCM,19 the present study emphasizes the frequent occurrence of VWF impairment, which can predispose to bleeding syndrome in the presence of bleeding-prone lesions or after invasive intervention.12,20 Nevertheless, a spontaneous major bleed was a rare event in the present study, with an estimated annual incidence of <2% in obstructive patients and ≈0.7% overall in patients with HCM. Given our preliminary results, further studies are required to explore the actual impact of VWF function on bleeding incidence in HCM.
The PFA is a nonspecific but simple, widely used, and highly sensitive way to screen patients for VWF defects that is adapted to routine testing.14 However, the closure time determined by the PFA-100 with collagen-epinephrine cartridges is sensitive to aspirin intake, whereas collagen-ADP cartridges can be influenced by thienopyridine intake. Although the PFA may be a simple way to detect VWF impairment in daily clinical practice, antiplatelet treatment must be checked before its use. Conversely, VWF:Ag, VWF:CB, VWF:CB/Ag, and %HMWM of VWF are insensitive to antiplatelet or warfarin treatment but are technically more complex. Although VWF:Ag and VWF:CB, and therefore VWF:CB/Ag, can be performed with commercial kits, the %HMWM of VWF determination is complex and requires a highly specialized laboratory. VWF:Ag assessment is important to rule out a quantitative defect of VWF; however, VWF:CB and VWF:Ag assays are complementary in assessing VWF function. VWF:CB/Ag is more accurate for the detection of a qualitative defect of VWF in clinical practice than VWF:CB, because it is adjusted to the VWF:Ag level. Finally, %HMWM is very specific and definitively confirms the loss of HMWM and therefore the pathophysiology of VWF impairment in the obstructive form of HCM.
In the present study, the proportion of patients with baseline or exercise-induced obstruction was lower than reported recently5; however, the present study was designed to evaluate VWF function in patients with HCM according to their current status, including medical therapy. Therefore, medical therapy was left unchanged, which likely accounts for the lower proportion of patients with obstructive HCM. Patients with latent obstruction exhibited an intermediate profile of VWF impairment, but their limited number (17%) precluded further analysis of such a subgroup. Further studies are needed to explore the latent group and to perform repeated assessment of VWF and peak gradient to ensure the effect of peak gradient changes and especially peak gradient reduction with treatment on VWF. The main objective of the present study was not to evaluate bleeding frequency in HCM but to assess VWF impairment and its relationship with the magnitude of obstruction. Given the frequent occurrence and extent of VWF function impairment in obstructive HCM, we can reasonably hypothesize that patients with obstructive HCM might be at risk of bleeding in the presence of bleeding-prone lesions or in the setting of invasive intervention.
In conclusion, primary hemostasis impairment through VWF proteolysis (acquired VWF syndrome type 2A) is a common finding in the obstructive form of HCM. Outflow tract obstruction induces VWF alterations that correlate closely with the magnitude of peak gradient. VWF impairment is improved with gradient reduction. Finally, one must keep in mind that the acquired VWF syndrome is a recognized risk factor for bleeding. Further studies with a larger population are warranted to confirm and extend the present results.
Sources of Funding
This work was realized in Institut Fédératif de Recherche 114 and supported by grants from the University of Lille 2 (EA2693), the Conseil Régional Nord-Pas de Calais, and Inserm ESPRI ERI-9 and FEDER R04026EE.
Maron BJ, McKenna WJ, Danielson GK, Kappenberger LJ, Kuhn HJ, Seidman CE, Shah PM, Spencer WH III, Spirito P, Ten Cate FJ, Wigle ED. American College of Cardiology/European Society of Cardiology clinical expert consensus document on hypertrophic cardiomyopathy. J Am Coll Cardiol. 2003; 42: 1687–1713.
Elliott PM, Gimeno JR, Tome MT, Shah J, Ward D, Thaman R, Mogensen J, McKenna WJ. Left ventricular outflow tract obstruction and sudden death risk in patients with hypertrophic cardiomyopathy. Eur Heart J. 2006; 27: 1933–1941.
Ommen SR, Maron BJ, Olivotto I, Maron MS, Cecchi F, Betocchi S, Gersh BJ, Ackerman MJ, McCully RB, Dearani JA, Schaff HV, Danielson GK, Tajik AJ, Nishimura RA. Long-term effects of surgical septal myectomy on survival in patients with obstructive hypertrophic cardiomyopathy. J Am Coll Cardiol. 2005; 46: 470–476.
Maron MS, Olivotto I, Zenovich AG, Link MS, Pandian NG, Kuvin JT, Nistri S, Cecchi F, Udelson JE, Maron BJ. Hypertrophic cardiomyopathy is predominantly a disease of left ventricular outflow tract obstruction. Circulation. 2006; 114: 2232–2239.
Tsai HM, Sussman II, Nagel RL. Shear stress enhances the proteolysis of von Willebrand factor in normal plasma. Blood. 1994; 83: 2171–2179.
Pareti FI, Lattuada A, Bressi C, Zanobini M, Sala A, Steffan A, Ruggeri ZM. Proteolysis of von Willebrand factor and shear stress-induced platelet aggregation in patients with aortic valve stenosis. Circulation. 2000; 102: 1290–1295.
Fressinaud E, Veyradier A, Truchaud F, Martin I, Boyer-Neumann C, Trossaert M, Meyer D. Screening for von Willebrand disease with a new analyzer using high shear stress: a study of 60 cases. Blood. 1998; 91: 1325–1331.
Zimmerman TS, Dent JA, Ruggeri ZM, Nannini LH. Subunit composition of plasma von Willebrand factor: cleavage is present in normal individuals, increased in IIA and IIB von Willebrand disease, but minimal in variants with aberrant structure of individual oligomers (types IIC, IID, and IIE). J Clin Invest. 1986; 77: 947–951.
Gill JC, Wilson AD, Endres-Brooks J, Montgomery RR. Loss of the largest von Willebrand factor multimers from the plasma of patients with congenital cardiac defects. Blood. 1986; 67: 758–761.
Kizilbash AM, Heinle SK, Grayburn PA. Spontaneous variability of left ventricular outflow tract gradient in hypertrophic obstructive cardiomyopathy. Circulation. 1998; 97: 461–466.
Left ventricular outflow tract obstruction plays an important role in the pathophysiology of hypertrophic cardiomyopathy (HCM). Obstruction is present in ≈25% of patients at rest but in up to 70% of patients during exercise. The long-term consequences of obstruction (heart failure and cardiovascular death) are thought to be related to wall-stress increase and myocardial ischemia. Beyond the effect on myocardium, outflow gradient could be involved in other pathophysiological alterations through an increase in the shear stress level. High shear forces have been shown to enhance proteolysis of the largest von Willebrand factor (VWF) multimers. We previously demonstrated that VWF impairment, defined as an acquired von Willebrand syndrome, is common in aortic stenosis and is directly related to the magnitude of obstruction. Moreover, VWF impairment favors abnormal bleeding in aortic stenosis. Accordingly, we hypothesized that outflow gradient could impair VWF in the obstructive form of HCM. Outflow gradient was evaluated by rest and exercise echocardiography in 62 patients. VWF impairment was frequent and was closely and independently associated with the magnitude of obstruction. A resting peak gradient of 15 mm Hg was sufficient to impair VWF. Patients with exercise-induced obstruction exhibited an intermediate pattern of VWF impairment compared with patients with nonobstructive HCM and baseline obstructive HCM. Outflow obstruction reduction was associated with an improvement in VWF function. In addition, patients with obstructive HCM had a trend toward abnormal spontaneous bleeding. Finally, these data support the concept that acquired von Willebrand syndrome, which is a recognized risk factor for bleeding, is common in the obstructive form of HCM.
↵*The first 2 authors contributed equally to this work.