Atrial Fibrillation (AF) is the most common cardiac arrhythmia of clinical significance.¹ The presence of AF independently increases the risk of mortality and morbidity due to stroke and thromboembolism, congestive heart failure and impaired quality of life, resulting in high health-care cost and a public health burden.² AF is an epidemic disease, affecting 1% to 1.5% of the population in the developed world.³ Approximately 2.3 million people are currently diagnosed with AF in the United States and this number is expected to increase to 15.9 million by 2050.⁴

Risk factors for the development of AF include: increasing age, hypertension, myocardial infarction, heart failure, diastolic dysfunction, valvular heart disease, thyrotoxicosis, alcoholism, obesity and diabetes.^5-7 The term nonvalvular AF (NVAF) is restricted to cases in which the rhythm disturbance occurs in the absence of rheumatic mitral stenosis or a prosthetic heart valve.⁸

The prevalence of AF increases with advancing age.⁹ The SAFE Study (Screening for Atrial Fibrillation in the Elderly) showed a prevalence of AF of 7.2 % in patients over 65 years and in those over 75 years old (10.3%)¹⁰.

AF is classified into paroxysmal, persistent and permanent categories. It is designated paroxysmal AF when the arrhythmia episode terminates spontaneously within 7 days or by electrical/pharmacological cardioversion within 48 hours of its onset; AF is persistent when it is sustained beyond 7 days or terminated by electrical/pharmacological cardioversion after 48 hours of sustenance, it also includes cases of long-standing AF where AF has lasted for 12 months uninterruptedly; finally, the term permanent AF is applied to clinical AF when the attempts of restoration of sinus rhythm are not contemplated.²

AF occurring in the absence of structural heart disease, is called lone AF² and is considered a nosographic entity, when conditions such as hypertension, diabetes, hyperthyroidism, acute infections, recent cardiothoracic or abdominal surgery, and systemic inflammatory diseases, should be excluded.¹¹

AF is associated with a prothrombotic (hypercoagulable) state and with an increased thromboembolic risk, by virtue of the Virchow’s triad for thrombogenesis: endothelial or endocardial damage/dysfunction; abnormal blood stasis; and abnormal haemostasis, increased platelet activity, and fibrinolysis.¹²

The management of AF includes, reducing symptoms through rhythm or rate management, treating underlying medical conditions, concomitant cardiovascular disease and reducing the risk of stroke and thromboembolic events¹³ with oral anticoagulation or aspirin.¹⁴

Atrial fibrillation confers a five-fold increased risk of stroke, and one in five of all strokes are attributed to this arrhythmia.² The mechanisms behind cerebral thrombo-embolism in AF are not completely understood, but it is well documented that AF is associated with a prothrombotic state, demonstrated by higher levels of von Willebrand Factor (vWF), when compared to healthy control subjects.^15,16

von Willebrand factor (vWF) is a multimeric glycoprotein that plays a crucial role in platelet adhesion and aggregation, which are the main initial steps in haemostasis after vascular injury and also under conditions of high shear stress as it happens in lesions in the coronary arteries.¹⁷ Recent studies have implicated vWF as a regulator of angiogenesis, smooth muscle cell proliferation, and interactions in the immune system. It is synthesized in endothelial cells and megakaryocytes as a propeptide. Following synthesis, vWF undergoes dimerization and further multimerization to finally be proteolized by ADAMTS 13 protease, into functional vWF multimers of varying size.^18,19

vWF has two main functions: this glycoprotein carries and protects factor VIII in circulation and mediates the initial platelet adhesion to the subendothelium, by linking to specific platelet membrane receptors: glycoprotein Ib-V-IX complex, facilitating platelet adhesion and glycoprotein αIIbβ3 mediating platelet aggregation at the site of injury, and consequently, activated coagulation.¹⁸

The mean vWF plasma level is 100 IU/dl, but the variability in the population is wide, with 95% of values between 50 IU/dl and 200 IU/dl.^20,21

Plasma vWF concentration is affected by the ABO blood group. O group individuals having plasma levels 25% lower than non-O (A, B and AB) subjects.²² The mechanism by which ABO group determines vWF plasma levels has not been well established.²³

Circulating plasma vWF, is almost exclusively synthesized, stored, and secreted by endothelial cells.²⁴ It has been proposed as a biomarker of endothelial damage/dysfunction as increased plasma levels have been found in inflammatory and atherosclerotic vascular diseases.²⁵

The availability of a useful index of endothelial dysfunction, may therefore have potential value, since the measurement of such a marker can be a non-invasive way of assisting in diagnosis as an indicator of disease progression and prognosis.²⁶

Plasma vWF levels are elevated in patients with NVAF associated with endothelial dysfunction.^27,28 Raised plasma levels of vWF are predictive of stroke and vascular events among patients with AF²⁹ and are associated with a poor prognosis. These levels in AF patients are not altered by warfarin and/or aspirin treatment.³⁰

Multiple studies show increased plasma levels of vWF in patients with AF when compared with a control group; several studies associate this increase to the presence of cardiovascular risk factors. Finally, results have been reported on the levels of vWF according to the type of AF, however, there are no descriptions of the ranges of variation available.

Therefore, our hypothesis is that vWF plasma levels show variability in patients with NVAF and that this variability is influenced by risk factors and by AF type. The aim of the present study is to identify these ranges of variability in patients with AF.

A search of the literature was conducted using electronic databases. Search strategy included: PubMed and Ovid. The keywords used were: “Atrial Fibrillation” and “von Willebrand Factor” (examination term of January 31, 2014).

A separate search for each keyword was realized. With the objective of providing reliability in the selection process of articles, two researchers were given the task of performing the process independently, afterwards their level of agreement was found.

Original articles were reviewed with experimental, quasi-experimental and observational designs, the articles must have been written in English, and followed methodological characteristics like: inclusion of patients with nonvalvular AF (NVAF), with description of associated risk factors, average age of the sample, percentage of male, analysis of plasma vWF levels by ELISA and report vWF concentrations in International Unit/deciliter (IU/dl), with presence or absence of a control group, without acute stroke or acute coronary syndrome, and provide statistical data for analysis and interpretation of results. Review articles and meta-analysis articles with incomplete methodological description or unclear data, were excluded.

Continuous data are expressed as mean and interquartile range, while categorical variables are expressed as a percentage. Statistical analysis was performed using the program IBM® SPSS® Statistics 20 version.

Cardiovascular risk factors associated with AF in the reviewed studies included: hypertension (39%), heart failure (33%), coronary artery disease (24%), diabetes mellitus (15%), prior stroke (12%) and hyperlipidemia (8%).

Four studies described the effects of risk factors in vWF levels in AF patients (Table 2). Kalyonku et al.⁴⁶ evidence the highest levels of vWF in patients with hypertension and diabetes compared with AF patients without the risk factor. Herringa et al.⁵¹ and Lip et al.⁵² show the greatest vWF levels by effect of hypertension and heart failure, respectively. Finally, Conway et al.²⁷ describe the effect of 6 risk factors: heart failure, hypertension, diabetes, prior stroke, age and female sex. The influence of age and female sex in vWF levels were: patient >75 years 151 IU/dl (123-179) and ≤75 years 145 IU/dl (113-177); women 150 IU/dl (119-181) and men 145 IU/dl (114-176).

Differences of vWF concentrations in patients with and without warfarin treatment are showed in Table 3. Two studies had a longitudinal design and show vWF levels in AF patients before and after warfarin treatment: Freestone et al. 2007⁴⁹ and Li-Saw-Hee et al. 2000⁵⁸ showed the same variation of 4I U/dl in the mean value, after 4 and 8 weeks of treatment with warfarin, respectively.

Three studies show differences in vWF by comparing two groups of patients: AF with and without warfarin. Barber et al. in 2004⁵⁴ showed a difference of 25 IU/dl between their means: AF without warfarin group (n = 92, age 72 years, 52% men) and warfarin group (n = 166, Age 70 years, 50% men). Lip et al. 1995⁵⁸ described a difference of 17 IU/dl between the 2 groups: without warfarin (n = 37, age 73 years, 51% men) and warfarin (n = 31, age = 60, 39% male). And Finally, Freestone et al. 2008,⁴⁷ presented high levels of vWF in patients with warfarin therapy, but they don´t specify the differences between their groups.

The general characteristics and vWF plasma levels in Lone AF patients are shown in Table 4. In this group, 193 subjects with AF were included; 2 types of Lone AF were found: Lone AF (Fu et al. 2011) with patients <60 years of age (mean 49 years) where variability in plasma vWF concentrations ranged between 79 and 147 IU/dl, with a mean value of 114 IU/dl. More studies^49,56,58,59 with Lone AF in patients >60 years of age (mean 69 years) had vWF levels ranged between 84 and 254 IU/dl, with an average of 148 IU/dl.

Table 5 summarizes the variability of vWF plasma levels in AF studies categorized into 3 types. These studies included a total of 658 participants. In Paroxysmal AF (47.6%) concentrations of vWF ranged between 92 to 264 IU/dl, with a mean of 141 IU/dl in patients 62 years old and 70% men; Persistent AF (30.7%) varied between 76 and 234 IU/dl with a mean of 137 IU/dl, 62 years and 70% men; finally in Permanent AF (21.7%) values rangedfrom 91 to 247 IU/dl with a mean of 162 IU/dl, mean age of 66 years and 61% of participants were male.

Studies that involved a control group are shown in Table 6; 2 types of subjects were found: healthy subjects (no AF and no risk factors) and patients with sinus rhythm matched (by age and gender, in terms of smoking status, body mass index, and systolic and diastolic blood pressure). Healthy patients (n=233) showed vWF plasma levels between 55 and 136 IU/dl with a mean of 95 IU/dl, age of 60 years and 70% men. And 371 participants with sinus rhythm matched had values of 54 to 176 IU/dl with a mean of 114 IU/dl, mean age of 64 years and 58% men.

Most of these studies were conducted in Europe (82%), mainly in the UK (73%) and the rest in Asia (18%).

Table 7 summarizes the vWF plasma concentrations, shows the variability of these based on the different types of atrial fibrillation. The highest values were found in patients with Permanent AF (162 IU/dl) and the lowest (114 IU/dl) were found in Lone AF patients younger than 60 years. Healthy subjects showed the lowest vWF plasma levels (95 IU/dl).