Predictive value of von Willebrand factor for adverse clinical outcome in hypertensive patients with mild-to-moderate aortic regurgitation

Abstract

Plasma levels of von Willebrand factor (vWF), a marker of endothelial dysfunction/damage, are elevated in high-risk hypertensive patients and in patients with severe aortic regurgitation (AR). Patients with mild-to-moderate AR, frequently detected in hypertensive elderly, have additional left ventricular morphological and functional dysfunctions. We hypothesized that hypertensive patients with mild-to-moderate AR may have enhanced endothelial and/or left ventricular dysfunctions that may lead to a deteriorated prognosis. We measured vWF, prothrombin F1+2 (F1+2) as a marker of thrombin generation, brain natriuretic peptide (BNP) in 104 hypertensive patients with mild-to-moderate AR and 66 hypertensive patients without AR. The left ventricular diameter at systole (LVDs) and left ventricular posterior wall thickness (LVWT) were determined by echocardiography and indexed by body surface area (LVDs/BSA and LVWT/BSA). VWF (median, interquartile range (IQR) 154, 120–196%) and BNP (34.7 pg ml−1, 15–65%) levels were greater in patients with AR than in those without AR (135, 98–175% and 20, 10.3–49 pg ml−1). All patients were prospectively followed up for cardiac events during the period of median 43 months (IQR 31–81). Patients with AR had an increased risk of cardiac events (regression ratio (RR) 1.87, 95% confidence interval 1.28–2.87) when compared to those without AR. A multivariate Cox hazard analysis indicated that log vWF (RR 4.93) and log BNP (RR 1.9) were independent predictors in patients with AR. VWF was an independent predictor of clinical outcome in hypertensive patients with mild-to-moderate AR.

Introduction

Patients with hypertension are at increased risk of thromboembolic events and associated with raised levels of various haemostatic markers. Plasma levels of von Willebrand factor (vWF), a marker of endothelial dysfunction/damage,1 are elevated in hypertensive patients.2 VWF levels increased in the elderly with mild renal impairment,3 in severe aortic regurgitation (AR)4 and in heart failure.5, 6 VWF also has been shown to be a prognostic factor for stroke,2 coronary artery diseases2 and heart failure.5

These findings suggest that vWF may be useful in identifying hypertensive patients with high-risk profiles. Nevertheless, the predictive values of vWF for cardiorenal dysfunction or cardiovascular events in high-risk hypertensive patients have not been fully investigated.

Hypertensive patients with mild-to-moderate AR may be regarded as a high-risk group and clinically more important than we assumed. Mild-to-moderate AR is prevalent in hypertensive patients7 and in the elderly.8 Hypertensive patients with mild-to-moderate AR had additional left ventricular structural and functional changes that may lead to deteriorated prognosis.9 Therefore, we hypothesized that hypertensive patients with mild-to-moderate AR may have enhanced endothelial and/or left ventricular dysfunctions, and therefore, would be predisposed to thromboembolic events or worsening of cardiorenal functions. However, the natural history of such patients has not been the focus of study.

To test this hypothesis, we assessed the predictive powers of vWF for future cardiovascular events in 104 hypertensive patients with mild-to-moderate AR. We also simultaneously measured brain natriuretic peptide (BNP) and prothrombin F1+2 (F1+2) as these are clinically important markers of heart failure and thrombin generation, respectively. Prothrombin F1+2, a marker of thrombin generation, was elevated in hypertensive patients with target organ damage or renal impairment,10, 11 and in patients with heart failure of various aetiologies.12

Methods

Study population

We studied hypertensive patients with mild-to-moderate AR (n=104) who were referred for echocardiography between April 1999 and March 2002. Hypertensive patients without any valvular or ischaemic heart diseases (n=66) served as controls during the same enrolment period.

All study subjects fulfilled the following criteria: (1) medically treated; (2) no clinical or laboratory evidence of ischaemic heart diseases, stroke, significant mitral valve regurgitation or peripheral vascular disease; (3) no clinical evidence or history of atrial fibrillation, malignant disorders or inflammatory diseases; (4) no history of hormone replacement therapy and (5) serum creatinine level <1.5 mg ml−1.

The study design was approved by the ethical committee of our hospital. This study was performed in accordance with the Declaration of Helsinki.

Echocardiography

All the participants underwent transthoracic echocardiography using a commercially available instrument (Sonos 5500, Philips, Best, The Netherlands). Two-dimensional and M-mode measurements were made according to the recommendations of the American Society of Echocardiography. Left ventricular posterior wall thickness (LVWT) at end diastole, left ventricular diameter at diastole (LVDd) and left ventricular diameter at systole (LVDs) were measured and indexed by body surface area (LVWT/BSA, LVDd/BSA and LVDs/BSA). AR was determined by continuous wave and colour Doppler echocardiography. The severity of regurgitation was assessed using regurgitant jet width ratio to left ventricular outflow tract width (AR ratio); an AR ratio <30% was defined as mild AR, 30–60% as moderate and >60% as severe. Out of 104 hypertensive patients with AR, 64 patients were graded as mild AR and 40 patients were graded as moderate AR. We excluded patients with an aortic root diameter >4.0 cm to avoid enrolling patients with systemic aortic diseases. Patients with AR due to aortic valve sclerosis were enrolled in this study. Aortic valve sclerosis was defined as thickening of aortic valve leaflets with preserved leaflet mobility. All measurements were analysed by two independent cardiologists unaware of the characteristics of the patients.

Assessment of haemostatic and neurohormonal markers and estimation of renal function

Blood samples were collected on the same day as echocardiography to measure the plasma levels of vWF, prothrombin fragment (F1+2) and BNP. VWF levels were determined with the ristocetin cofactor activity assay, using a commercially available kit (BC von Willebrand Reagent, Dade Behring, Marburg, Germany). The intra-assay CV and the inter-assay CV of high vWF levels (mean 110) were below 5% and the upper limit of this assay was 200%. F1+2 was measured using an enzyme immunoassay (Enzygnost F1+2, Behringwerke AG, Marburg, Germany). BNP were measured using specific immunoradiometric assays for BNP (Shionoria, Shionogi, Osaka, Japan). Glomerular filtration rate (GFR) was estimated by the Cockcroft–Gault formula (140−age) multiplied by 1.23 (body weight/creatinine), amplified by 0.85, if patient was female. Assay results were expressed as mean±s.d.

Follow-up

All patients were followed prospectively for admission due to cardiovascular events: new onset or decompensation of congestive heart failure, new onset of ischaemic heart diseases, stroke or transient ischaemic attack and cardiac death. Congestive heart failure was confirmed by the presence of exertional dyspnoea and/or radiographic evidence of pulmonary oedema, accompanied by acute elevation of BNP concentration.

For hypertensive patients with AR, follow-up echocardiography was performed at 6 months after the enrolment.

Statistical analysis

Comparisons between the patients and the control group were determined by unpaired t-test or the Mann–Whitney test as appropriate for continuous variables, and the χ2-analysis for categorical variables. Paired comparisons were made using the paired t-test.

The distributions of vWF, F 1+2 and BNP were rightly skewed, so these variables were expressed as median and interquartile range (IQR), and log-transformed for survival analysis.

To investigate the predictive value of vWF in patients with AR, patients were assigned to the high-risk group if they had vWF level >154%. This cut-off level was derived from the median vWF level in hypertensive patients with AR. Comparisons of cardiac event-free survivals between hypertensive patients with AR and those without AR, and comparisons between high- and low-risk groups in AR patients were performed with the Kaplan–Meier curve followed by the log-rank test. Predictive values of the markers for haemostasis and left ventricular dysfunction were evaluated using a multivariate Cox proportional hazard model for patients with AR. The explanatory variables included in the Cox model were age, gender, log vWF, log F1+2, log BNP, LVWT/BSA, LVDs/BSA and estimated GFR. JMP version 5 was used for statistical analysis.

Results

Patient characteristics

The clinical characteristics that might influence vWF levels did not differ between patients with AR and those without AR, except for higher diastolic blood pressure in hypertensive patients without AR (Table 1). The contents of medication such as angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers and lipid-lowering statins were similar between the two groups (Table 1). All hypertensive patients with AR had aortic valve sclerosis.

Table 1 Clinical characteristics of hypertensive patients with or without AR

Von Willebrand factor and parameters of left ventricular dysfunction

Plasma levels of vWF and BNP were greater, and the estimated GFR values were lower in patients with AR than in those of patients without AR (Table 2). Echocardiographic parameter of left ventricular dilatation (LVDd/BSA) was greater in hypertensive patients with AR than in those without AR. F1+2 levels and LVWT/BSA and LVDs/BSA did not differ between the two groups.

Table 2 Haemostatic factors, BNP and left ventricular hypertrophy in hypertensive patients with or without AR

Survival analysis

All patients were on medical treatments and prospectively followed up for median 43 months (IQR 31–81). Out of 104 hypertensive patients with AR, 36 patients experienced cardiac events. Cardiac death occurred in four, heart failure in eighteen, angina in eight and stroke in six (lacunar-type, five; embolic infarction, one). Cardiac death occurred exclusively in the very elderly (mean age 83±4 years and BNP median 120, IQR 74–186 pg ml−1). Out of four cardiac deaths, two died because of decompensation of heart failure, and two experienced sudden death due to acute myocardial infarction. Decompensation of heart failure was experienced by elderly population (mean age 72±11 years and mean BNP level=median 73, IQR 26–133 pg ml−1, vWF levels 183, 154–196%, GFR 47±18 ml min−1). As for sixty-six hypertensive patients without AR, cardiac events occurred in nine (cardiac sudden death in one, angina in five and lacunar-type stroke in three).

Event-free survival analysis indicated that hypertensive patients with AR had a worse prognosis than hypertensive patients without AR (Figure 1). The presence of AR in hypertensive patients had an increased risk of cardiac events (regression ratio (RR) 1.87, 95% confidence interval 1.28–2.87) even after the adjustment for age, gender, BNP and estimated GFR.

Figure 1
figure1

Kaplan–Meier plot demonstrating event-free survival of hypertensive patients with mild-to-moderate aortic regurgitation (AR+) and those without mild-to-moderate aortic regurgitation (AR−).

Hypertensive patients with AR and cardiovascular events had higher mean age, more prevalence of female gender, and higher vWF, BNP and F1+2 levels than those AR patients without the events (Table 3). Out of 104 hypertensive patients with AR, 95 patients underwent follow-up echocardiography at 6 months after the enrolment. LVWT/BSA decreased significantly in AR patients without cardiovascular events, but did not change in those with the events (Table 4). However, the difference of changes of LVWT/BSA was not significant between the two patient groups. The severity of AR did not change in all the patients.

Table 3 Comparisons of the parameters of haemostasis and cardiorenal functions between AR patients with and without event
Table 4 Echocardiographic parameters at initial and follow-up assessments in hypertensive patients with AR

Higher levels of vWF were associated with poor clinical outcome in hypertensive patients with AR. Event-free survival in hypertensive patients with AR was significantly higher in the low vWF group than in the high vWF group (Figure 2). A multivariate Cox hazard analysis revealed that vWF and BNP were the independent predictors of cardiovascular events in hypertensive patients with AR (Table 5).

Figure 2
figure2

Kaplan–Meier plot demonstrating event-free survival of high-von Willebrand factor (vWF) group (vWF154) and low-vWF group (vWF <154) in hypertensive patients with mild-to-moderate aortic regurgitation (AR). Hypertensive patients with AR were stratified according to the median level of vWF.

Table 5 Relative risks of clinical outcome in hypertensive patients with AR

Discussion

The present study indicated two major findings (1) the presence of mild-to-moderate AR in hypertensive patients was associated with a poor clinical outcome; (2) vWF remained an independent prognostic factor after the adjustment for age, gender and the other parameters in hypertensive patients with mild-to-moderate AR.

AR, mostly mild, was found in up to 29% of the elderly population.8 Although frequently detected in routine practice, mild AR rarely causes a haemodynamically significant lesion and its natural history has not been a focus of study.13 However, AR may be more important as a subclinical factor in the hypertensive subjects than previously thought. Lending support to our proposal, the present study showed that the presence of AR was associated with an increased risk of cardiovascular events even after adjustment for age, gender, BNP and estimated GFR. The cause of a poorer prognosis in hypertensive patients with AR may be attributable to the accumulation of endothelial and cardiorenal dysfunctions, as was indicated by higher levels of vWF and BNP and more impaired GFR. Renal dysfunction has emerged as an important prognostic factor of chronic heart failure.14 Insufficient control of fluid volume that could be caused by impaired renal function or neurohormonal activation15 may affect worsening of left ventricular dysfunction in heart failure patients with preserved systolic function.

VWF levels were elevated in patients with severe AR who underwent valve replacement surgery,4 and in patients with non-valvular atrial fibrillation accompanied by left ventricular dysfunction.16 Even though our patient population had less haemodynamically significant lesions than those of the previous studies, the present study indicated that vWF levels were greater in hypertensive patients with mild-to-moderate AR than those without AR, in spite of a lower diastolic blood pressure in hypertensive patients with AR. Although raised blood pressure levels were associated with higher vWF levels,2 hypertension may play a less important role in raising vWF levels compared to pathological properties such as turbulent regurgitant flow and LVD in patients with AR, in the same way as recently indicated by Varughese et al.17 that hypertension or blood pressure levels do not seem to have an additive effect on the prothrombotic state in atrial fibrillation.

One possible mechanism of raised vWF levels in hypertensive patients with AR is that the aetiology of regurgitation in our study is aortic valve sclerosis that has similar risk factors with those of atherosclerosis.18 Aortic valve sclerosis has been regarded as a manifestation of the atherosclerotic process,19 and is related to systemic endothelial dysfunction since flow-mediated dilation (FMD) is predictive of aortic valve sclerosis. VWF levels were also associated with FMD.6 Therefore, we speculate that higher vWF levels may be a manifestation of atherosclerosis that can be related to aortic valve sclerosis at least in our study.

In the present study, vWF and BNP were the independent predictors of cardiovascular events on multivariate analysis, in spite of the finding that left ventricular morphologies did not change in AR patients with the events. Thus, we speculate that endothelial dysfunction and neurohormonal activation may play an important role in the development of cardiovascular events in this clinical setting. However, it is possible that AR patients with the events did not have progression of left ventricular size because of a relatively short follow-up period to repeat echocardiography. Patients with AR but without the events had regression of left ventricular hypertrophy to a certain extent, probably because these patients had less severe cardiorenal dysfunctions and were likely to respond well to medical treatments.

BNP was associated with the onset of symptoms in AR patients. Gerber et al.20 reported that in patients with chronic moderate or severe AR and preserved systolic function BNP levels were higher in symptomatic patients than in asymptomatic patients, even though echocardiographic measurements of left ventricular size were not significantly different between the two groups. The result of Gerber et al.20 is compatible with the finding in the present study that BNP was the independent predictor of the poor outcome after being adjusted for echocardiographic measurements in AR patients.

The present study is the first report, as far as we know, that vWF predicts the future cardiovascular events in hypertensive patients with mild-to-moderate valvular heart diseases with preserved left ventricular systolic function. This result is in accordance with a previous report by Chong et al.5 that high vWF levels were predictive of adverse outcome in patients with heart failure, however, their patient population was severe systolic heart failure with median ejection fraction of 20%. In addition, Varughese et al.21 have reported that vWF levels were not associated with composite cardiovascular events in high-risk hypertensive patients. Compared to our study, their study did not include decompensation of heart failure in end point and their study population was younger than that of ours.

Stam et al.3 revealed that renal function was mildly impaired and inversely correlated with vWF in an elderly general population. They also revealed that mild impairment of renal function was associated with cardiovascular morbidity, and, therefore, suggested that endothelial dysfunction may be an important mechanism linking mildly impaired renal function to cardiovascular diseases. However, the present study revealed that vWF remained an independent prognostic factor in hypertensive patients with AR even after being adjusted for estimated GFR. Renal and endothelial dysfunctions are likely to be coexisting comorbidities leading to a poor prognosis in our patient population, but they may be independent of each other.

F1+2 did not predict the future cardiovascular events in the present study. This could be partly explained by the biological availability of F1+2 in the blood circulation. Because of the short half-life of F1+2 (90 min), it is not clear whether elevated F1+2 represents the tail of acute burst of thrombin generation or a persistent low-grade activation of thrombin.22 Predictive value of F1+2 for future cardiovascular events has not been fully investigated, and still been controversial.23, 24

As this study was an exploratory study, we did not estimate the target number of patients to be recruited. We enrolled all the patients who met the inclusion criteria during the enrolment period. Therefore, we calculated a statistical power (1–β) from the result of this study to secure reliability of this study. Its power was calculated as 0.89, it was confirmed that the number of patients enrolled in this study was reasonable.

In conclusion, vWF was associated with adverse clinical outcome in hypertensive patients with mild-to-moderate AR. These patients can be characterized by a coexistence of several target organ damages caused by hypertension, where endothelial dysfunction seems to play a central role to cause an adverse effect. Further clinical studies involving a large number of patients are required to evaluate if vWF is a useful prognostic marker for cardiovascular events in the hypertensive general population.

Study limitations

AR may not entirely be a manifestation of target organ damage by hypertension, as its prevalence was not solely influenced by the severity of hypertension.25 Therefore, the presence of this valvular disease may be a reflection of other metabolic disorders such as diabetes mellitus or high lipid profiles. However, these clinical characteristics did not differ between hypertensive patients with AR and those without AR.

To properly evaluate the severity of AR, it is required to integrate all of the echocardiographic information. Therefore, our study may be less accurate in the assessment of the severity of AR, since we did not provide supportive evidences such as pressure half time of the rate of deceleration of AR velocity. However, classification of AR by the ratio of regurgitant jet width/LVOT width we employed in this study has been widely regarded as a specific sign of AR severity.26

Von Willebrand factor activity was assessed by ristocetin cofactor assay in our study. This assay may be less reliable than vWF antigen assay, however, intra- and inter-assay CVs of this assay when measuring high vWF level were below 5%. In addition, the assay has been traditionally and widely used for daily clinical practice.

References

  1. 1

    Lip GY, Foster W, Blann AD . Plasma von Willebrand factor levels and surrogates of atherosclerosis. J Thromb Haemost 2005; 3: 659–661.

  2. 2

    Blann AD . Plasma von Willebrand factor, thrombosis, and the endothelium: the first 30 years. Thromb Haemost 2006; 95: 49–55.

  3. 3

    Stam F, van Guldener C, Becker A, Dekker JM, Heine RJ, Bouter LM et al. Endothelial dysfunction contributes to renal function-associated cardiovascular mortality in a population with mild renal insufficiency: the Hoorn study. J Am Soc Nephrol 2006; 17: 537–545.

  4. 4

    Goldsmith IR, Blann AD, Patel RL, Lip GY . Effect of aortic valve replacement on plasma soluble P-selectin, von Willebrand factor, and fibrinogen. Am J Cardiol 2001; 87: 107–110, A9.

  5. 5

    Chong AY, Freestone B, Patel J, Lim HS, Hughes E, Blann AD et al. Endothelial activation, dysfunction, and damage in congestive heart failure and the relation to brain natriuretic peptide and outcomes. Am J Cardiol 2006; 97: 671–675.

  6. 6

    Chong AY, Blann AD, Patel J, Freestone B, Hughes E, Lip GY . Endothelial dysfunction and damage in congestive heart failure: relation of flow-mediated dilation to circulating endothelial cells, plasma indexes of endothelial damage, and brain natriuretic peptide. Circulation 2004; 110: 1794–1798.

  7. 7

    Lebowitz NE, Bella JN, Roman MJ, Liu JE, Fishman DP, Paranicas M et al. Prevalence and correlates of aortic regurgitation in American Indians: the Strong Heart Study. J Am Coll Cardiol 2000; 36: 461–467.

  8. 8

    Lindroos M, Kupari M, Heikkila J, Tilvis R . Prevalence of aortic valve abnormalities in the elderly: an echocardiographic study of a random population sample. J Am Coll Cardiol 1993; 21: 1220–1225.

  9. 9

    Kontos J, Papademetriou V, Wachtell K, Palmieri V, Liu JE, Gerdts E et al. Impact of valvular regurgitation on left ventricular geometry and function in hypertensive patients with left ventricular hypertrophy: the LIFE study. J Hum Hypertens 2004; 18: 431–436.

  10. 10

    Sechi LA, Zingaro L, Catena C, Casaccio D, De Marchi S . Relationship of fibrinogen levels and hemostatic abnormalities with organ damage in hypertension. Hypertension 2000; 36: 978–985.

  11. 11

    Sechi LA, Zingaro L, Catena C, De Marchi S . Increased fibrinogen levels and hemostatic abnormalities in patients with arteriolar nephrosclerosis: association with cardiovascular events. Thromb Haemost 2000; 84: 565–570.

  12. 12

    Jafri SM, Mammen EF, Masura J, Goldstein S . Effects of warfarin on markers of hypercoagulability in patients with heart failure. Am Heart J 1997; 134: 27–36.

  13. 13

    Otto CM . Aortic regurgitation.In: Otto CM (ed). Valvular Heart Disease, 2nd edn. Saunders: Philadelphia, 2004, pp 303–335.

  14. 14

    Hillege HL, Girbes AR, de Kam PJ, Boomsma F, de Zeeuw D, Charlesworth A et al. Renal function, neurohormonal activation, and survival in patients with chronic heart failure. Circulation 2000; 102: 203–210.

  15. 15

    Zile MR, Brutsaert DL . New concepts in diastolic dysfunction and diastolic heart failure: part II: causal mechanisms and treatment. Circulation 2002; 105: 1503–1508.

  16. 16

    Lip GY, Pearce LA, Chin BS, Conway DS, Hart RG . Effects of congestive heart failure on plasma von Willebrand factor and soluble P-selectin concentrations in patients with non-valvar atrial fibrillation. Heart 2005; 91: 759–763.

  17. 17

    Varughese GI, Patel JV, Tomson J, Lip GY . Effects of blood pressure on the prothrombotic risk in 1235 patients with non-valvular atrial fibrillation. Heart 2007; 93: 495–499.

  18. 18

    Stewart BF, Siscovick D, Lind BK, Gardin JM, Gottdiener JS, Smith VE et al. Clinical factors associated with calcific aortic valve disease. Cardiovascular Health Study. J Am Coll Cardiol 1997; 29: 630–634.

  19. 19

    Agmon Y, Khandheria BK, Meissner I, Sicks JR, O'Fallon WM, Wiebers DO et al. Aortic valve sclerosis and aortic atherosclerosis: different manifestations of the same disease? Insights from a population-based study. J Am Coll Cardiol 2001; 38: 827–834.

  20. 20

    Gerber IL, Stewart RA, French JK, Legget ME, Greaves SC, West TM et al. Associations between plasma natriuretic peptide levels, symptoms, and left ventricular function in patients with chronic aortic regurgitation. Am J Cardiol 2003; 92: 755–758.

  21. 21

    Varughese GI, Patel JV, Tomson J, Blann AD, Hughes EA, Lip GY . Prognostic value of plasma soluble P-selectin and von Willebrand factor as indices of platelet activation and endothelial damage/dysfunction in high-risk patients with hypertension: a sub-study of the Anglo-Scandinavian Cardiac Outcomes Trial. J Intern Med 2007; 261: 384–391.

  22. 22

    Biasucci LM, Liuzzo G, Caligiuri G, Quaranta G, Andreotti F, Sperti G et al. Temporal relation between ischemic episodes and activation of the coagulation system in unstable angina. Circulation 1996; 93: 2121–2127.

  23. 23

    Agewall S, Wikstrand J . Prothrombin fragment 1+2 is a risk factor for myocardial infarction in treated hypertensive men. J Hypertens 1998; 16: 1557–1559.

  24. 24

    Lowe GD, Rumley A, Sweetnam PM, Yarnell JW, Rumley J . Fibrin D-dimer, markers of coagulation activation and the risk of major ischaemic heart disease in the Caerphilly Study. Thromb Haemost 2001; 86: 822–827.

  25. 25

    Palmieri V, Bella JN, Arnett DK, Roman MJ, Oberman A, Kitzman DW et al. Aortic root dilatation at sinuses of valsalva and aortic regurgitation in hypertensive and normotensive subjects: the Hypertension Genetic Epidemiology Network Study. Hypertension 2001; 37: 1229–1235.

  26. 26

    Bekeredjian R, Grayburn PA . Valvular heart disease: aortic regurgitation. Circulation 2005; 112: 125–134.

Download references

Acknowledgements

We thank Mr Christopher Wood for checking our writing.

Author information

Correspondence to M Iida.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Iida, M., Nihei, M., Yamazaki, M. et al. Predictive value of von Willebrand factor for adverse clinical outcome in hypertensive patients with mild-to-moderate aortic regurgitation. J Hum Hypertens 22, 275–281 (2008) doi:10.1038/sj.jhh.1002321

Download citation

Keywords

  • vWF
  • BNP
  • aortic regurgitation
  • F1+2
  • endothelial dysfunction

Further reading