Author + information
- Received October 11, 2010
- Revision received December 2, 2010
- Accepted December 14, 2010
- Published online June 21, 2011.
- Vanessa Roldán, MD, PhD⁎,
- Francisco Marín, MD, PhD†,
- Begoña Muiña, MD⁎,
- Jose Miguel Torregrosa, MD⁎,
- Diana Hernández-Romero, PhD†,
- Mariano Valdés, MD†,
- Vicente Vicente, MD⁎ and
- Gregory Y.H. Lip, MD‡,⁎ ()
- ↵⁎Reprint requests and correspondence
: Prof. Gregory Y. H. Lip, University of Birmingham Centre for Cardiovascular Sciences, City Hospital, Birmingham B18 7QH, United Kingdom
Objectives The purpose of this study was to evaluate the prognostic value of plasma von Willebrand factor (vWF) levels and fibrin d-dimer in a large cohort of anticoagulated permanent atrial fibrillation (AF) patients.
Background In nonanticoagulated AF patients, plasma vWF levels have been related to stroke and vascular events. There are limited data on the prognostic role of biomarkers in anticoagulated AF patients in relation to adverse events (including thromboembolism), mortality, and major bleeding.
Methods We studied 829 patients (50% male; median age 76 years) with permanent AF who were stabilized (for at least 6 months) on oral anticoagulation therapy (international normalized ratio: 2.0 to 3.0). Plasma d-dimer and vWF levels were quantified by enzyme-linked immunosorbent assay. Patients were followed for 2 years, and adverse events (thrombotic and vascular events, mortality, and major bleeding) were recorded.
Results Patients were followed for a median of 828 days (range 18 to 1,085 days). On multivariate analysis, age 75 years and older, previous stroke, heart failure, and high plasma vWF levels (≥221 IU/dl) were associated with future adverse cardiovascular events (all p values <0.05). High plasma vWF levels, elderly patients, diabetes, hypercholesterolemia, and current smoking were associated with mortality (all p values <0.05). High plasma vWF levels were also an independent predictor of major bleeding (hazard ratio: 4.47, 95% confidence interval: 1.86 to 10.75; p < 0.001). High plasma vWF levels were able to refine clinical risk stratification schema for stroke (CHADS2 [Congestive heart failure, Hypertension, Age ≥75, Diabetes mellitus, and prior Stroke or transient ischemic attack (doubled)], CHA2DS2-VASc [Congestive heart failure, Hypertension, Age ≥75 years, Diabetes mellitus, Stroke, Vascular disease, Age 65 to 74 years, Sex category]) and bleeding (HAS-BLED [Hypertension, Abnormal renal/liver function, Stroke, Bleeding history or predisposition, Labile International Normalized Ratio, Elderly, Drugs/alcohol concomitantly]). d-dimer did not show any significant impact on adverse events.
Conclusions High plasma vWF levels (≥221 IU/dl) are an independent risk factor for adverse events in anticoagulated permanent AF patients. This biomarker may potentially be used to refine stroke and bleeding clinical risk stratification in AF.
Atrial fibrillation (AF) is a common cardiac arrhythmia associated with increased morbidity and mortality. Much of the AF-associated morbidity is secondary to a 5- to 6-fold increased risk of stroke (1), but these patients also have a high incidence of other cardiovascular events (mainly acute coronary syndromes and vascular death) (2).
Oral anticoagulation (OAC) is highly effective in reducing the risk of stroke and thromboembolism. To aid decisions for thromboprophylaxis, several risk stratification schemes have been developed using clinical characteristics, the most popular being the CHADS2 score (1 point for Congestive heart failure, Hypertension, Age 75 years and older, and Diabetes, and 2 points for previous Stroke) (3). Given the limitations of the CHADS2 score (4), the CHA2DS2-VASc [Congestive heart failure, Hypertension, Age ≥75 years, Diabetes mellitus, Stroke, Vascular disease, Age 65 to 74 years, Sex category] score has been proposed to complement the CHADS2 score in decision making for OAC (5), reflecting a risk factor–based approach to thromboprophylaxis (6). In the recent European Society of Cardiology guidelines, OAC is recommended in AF patients at moderate to high risk of stroke and thromboembolism (e.g., those with ≥1 stroke risk factors ), whereas in the 2006 American College of Cardiology/American Heart Association/European Society of Cardiology guidelines, OAC is recommended with any high-risk factor (previous stroke) or >1 moderate-risk factor (age 75 years and older, hypertension, diabetes, or heart failure) (7). Whether OAC is useful for the prevention of myocardial infarction and other vascular events has been debated (8).
von Willebrand factor (vWF) has been proposed as a biomarker of endothelial damage/dysfunction because increased plasma levels have been found in inflammatory and atherosclerotic vascular diseases in which the endothelium is likely to be damaged (9). Fibrin d-dimer (DD) is a fibrin degradation product and an established marker of fibrin turnover and thrombogenesis, which has been reported to be increased in AF as part of the hypercoagulable state in this arrhythmia (10). Data from the SPAF (Stroke Prevention in Atrial Fibrillation) III study showed an independent association of plasma vWF level with clinical risk factors for stroke in AF patients (11) and was shown to be predictive of subsequent stroke and vascular events in patients who were treated with aspirin and/or inadequate dosing of warfarin (12). We also previously showed how plasma vWF levels correlated with the CHADS2 score (13), and more recently, inflammatory (14,15) and prothrombotic (16) markers have been related to prognosis in AF patients, even though they are receiving OAC therapy. In addition, DD has prognostic significance in AF, even in patients taking OAC medication (16–18).
Stroke risk is also closely related to bleeding risk, and OAC therapy needs to weigh the benefit from stroke prevention against the bleeding risk. Many thromboembolic risk factors have also been identified as bleeding risk factors (e.g., advanced age, uncontrolled hypertension, history of myocardial infarction or ischemic heart disease, cerebrovascular disease, anemia, a history of bleeding, and the concomitant use of other drugs such as antiplatelet drugs) (19). A simple, novel bleeding risk scoring system (HAS-BLED [Hypertension, Abnormal renal/liver function, Stroke, Bleeding history or predisposition, Labile International Normalized Ratio, Elderly, Drugs/alcohol concomitantly]) has recently been proposed as a practical tool to assess the individual bleeding risk of real-world AF patients (20,21) and has been incorporated in the European Society of Cardiology guidelines (6). However, there is scarce information on biomarkers for bleeding risk prediction in AF.
We hypothesized that plasma vWF and DD levels could be used to refine clinical risk stratification for adverse cardiovascular events (mainly thromboembolism), mortality, and major bleeding in anticoagulated AF patients. To test this hypothesis, we studied the prognostic value of 2 biomarkers (vWF and DD) in a large cohort of AF patients on long-term anticoagulation therapy with regard to adverse events and major bleeding episodes.
We recruited consecutive patients with permanent/paroxysmal AF from our outpatient anticoagulation clinic who were stabilized for at least 6 months with OAC therapy (international normalized ratio: 2.0 to 3.0). We excluded patients with valvular AF or prosthetic heart valves, any acute coronary syndrome, stroke (embolic or ischemic) or hemodynamic instability in the preceding 6 months, as well as any hospital admissions or surgical interventions during the same period.
At study entry, we recorded the medical history, and follow-up was conducted by visits to the anticoagulation clinic. The CHADS2 and CHADS2-VASc2 stroke risk scores were recorded as baseline measures of stroke risk (3,5), whereas the HAS-BLED score (20) was calculated as a measure of baseline bleeding risk. The CHADS2 score is based on a point system in which 2 points are assigned for a history of stroke or transient ischemic attack and 1 point is assigned for age 75 years and older, hypertension, diabetes, or congestive heart failure. The CHADS2-VASc2 stroke risk score assigned 1 point to congestive heart failure, hypertension, diabetes, vascular disease, age 65 to 74 years, and sex category (female) and 2 points for age 75 years and older and stroke. The HAS-BLED score is the result of adding 1 point for hypertension, abnormal renal/liver function (1 point for each one), stroke, bleeding history or predisposition, unstable international normalized ratio, advanced age (65 years and older), and drugs/alcohol concomitantly (1 point for each one).
Adverse cardiovascular endpoints (mainly thromboembolic) were defined as stroke/transient ischemic accident as well as both embolic and ischemic peripheral embolism, acute coronary syndrome, and acute heart failure, and cardiac death. Bleeding events were assessed following 2005 International Society on Thrombosis and Haemostasis criteria, which were recently reviewed (22). Finally, both all-cause and cardiovascular deaths were recorded.
Blood samples and laboratory analysis
Blood samples were drawn atraumatically and without stasis into syringes pre-loaded with trisodium citrate (0.011 mol/l). Platelet-poor plasma fractions were obtained by centrifugation at 4°C for 20 min at 2,200 g. Aliquots were stored at −80°C to allow batch analysis. Both plasma vWF and DD levels were assessed in an automated coagulometer ACL Top 3 G, HemosIL von Willebrand factor and Hemosil D-Dimer HS (Instrumentation Laboratory, Lexington, Massachusetts). The interassay and intra-assay coefficients of variation were 1.4% and 9%, respectively, and the lower limits of detection were 2.2 IU/dl and 21 ng/ml, respectively.
Continuous variables were tested for normal distribution by the Kolmogorov-Smirnov test. Continuous variables are presented as a mean ± SD or median (interquartile range), as appropriate, and categorical variables as a percentage. We used Cox models to determine the association of plasma DD and vWF levels with time to adverse events independently to clinical risk factors. Receiver-operator characteristic curve analyses were generated to test the predictive discrimination of both biomarkers to identify association with adverse events during follow-up. The cut point with the best sensitivity and specificity was chosen for each biomarker, as assessed by receiver-operator characteristic curves (23). The independent effect of variables on prognosis was calculated using a Cox proportional hazards regression model, incorporating in the multivariate model only those values that showed a p value <0.15 in the univariate analysis. A p value <0.05 was accepted as statistically significant. Statistical analysis was performed using SPSS version 15.0 for Windows (SPSS, Inc., Chicago, Illinois).
We studied 829 patients (50% male; median age 76 years (interquartile range: 70 to 80 years) with clinical characteristics summarized in Table 1. The median CHADS2 risk score was 2 (interquartile range: 2 to 3), and 76% of patients had a CHADS2 risk score of ≥2. The median CHADS2-VASc2 score was 4 (range 3 to 5), and 94% had a CHA2DS2-VASc score of ≥2. The median HAS-BLED score was 2 (range 2 to 3), and 32% of patients had a HAS-BLED score of ≥3. The median values of vWF and DD were 171 IU/dl (range 131 to 230 IU/dl) and 257 IU/dl (range 177 to 389 ng/ml), respectively.
The median follow-up was 828 days (range 18 to 1,085 days). During this period, 95 patients experienced an adverse cardiovascular event (rate: 5.0% per year): there were 32 stroke/transient ischemic attack events (rate: 1.7% per year), 36 acute coronary syndrome events (rate: 1.9% per year), and 27 acute heart failure events (rate: 1.5% per year) (Table 2). Sixty-nine patients died during follow-up (rate: 3.7% per year); the deaths in 25 of the patients were cardiovascular (rate: 1.13% per year) (Tables 3 and 4).⇓⇓ There were 68 major bleeding episodes (rate: 3.6% per year) (Table 5).
Univariate and multivariate analyses
For each adverse event (e.g., adverse cardiovascular event, death, and bleeding), we constructed receiver-operator characteristic curves that gave a median cutoff point of 221 IU/dl for plasma vWF levels (area under the curve: 0.74; p < 0.001; sensitivity 0.73; specificity 0.71). We did not find any cutoff point for DD (area under the curve: 0.58; p = 0.051). Univariate and multivariate predictors of adverse cardiovascular events are shown in Table 2, for death and cardiovascular death in Tables 3 and 4, and for bleeding in Table 5.
Multivariate Cox regression analysis showed independent associations for age 75 years and older (hazard ratio [HR]: 2.00, 95% confidence interval [CI]: 1.23 to 3.24; p = 0.005), previous stroke (HR: 1.81; 95% CI: 1.14 to 2.87; p = 0.012), history of heart failure (HR: 1.79; 95% CI: 1.18 to 2.72; p = 0.006), and high plasma vWF levels (≥221 IU/dl) (HR: 2.71; 95% CI: 1.78 to 4.13; p < 0.001) for adverse cardiovascular events (Table 2). The association of a plasma vWF level of ≥221 IU/dl with any of the single components of the composite endpoint gave an HR of 5.04 (95% CI: 2.40 to 10.59; p < 0.001), for stroke, an HR of 3.42 (95% CI: 1.75 to 6.70; p < 0.001) for acute coronary syndrome, and an HR of 1.97 (95% CI: 0.93 to 4.20; p = 0.078) for acute heart failure.
Independent predictors of all-cause mortality were age 75 years and older (HR: 3.26; 95% CI: 1.71 to 6.19; p < 0.001), diabetes (HR: 1.92; 95% CI: 1.17 to 3.15; p = 0.010), current smoking habit (HR: 2.76; 95% CI: 1.56 to 4.89; p < 0.001), hypercholesterolemia (HR: 0.46; 95% CI: 0.25 to 0.85; p = 0.014), and plasma vWF level ≥221 IU/dl (HR: 2.03; 95% CI: 1.24 to 3.32; p = 0.005) (Table 3). Risk factors associated with cardiovascular death were age 75 years and older, heart failure, current smoking habit, hypercholesterolemia, and high plasma vWF levels (Table 4).
Independent predictors of major bleeding were male sex (HR: 0.57; 95% CI: 0.35 to 0.95; p = 0.031), renal impairment (HR: 1.99; 95% CI: 1.08 to 3.67; p = 0.028), previous bleeding episode (HR: 4.69; 95% CI: 2.79 to 7.87; p < 0.001), and high plasma vWF levels (HR: 4.47; 95% CI: 1.86 to 10.75; p = 0.001) (Table 5). Figures 1A to 1D show Kaplan-Meier curves for each adverse event in relation to plasma vWF levels (all log-rank tests: p < 0.001).
Relationship of adverse events to CHADS2 and CHA2DS2-VASc scores
Both CHADS2 and CHA2DS2-VASc scores were predictive of adverse cardiovascular events, mortality, and cardiovascular death, but not major hemorrhage. The CHADS2 score showed an HR of 1.41 (95% CI: 1.20 to 1.65; p < 0.001) for adverse cardiovascular (mainly thrombotic) events, an HR of 1.57 (95% CI: 1.31 to 1.88; p < 0.001) for all-cause mortality, and an HR of 1.67 (95% CI: 1.24 to 2.27; p < 0.001) for cardiovascular death. The CHA2DS2-VASc score showed an HR of 1.30 (95% CI: 1.44 to 1.47; p < 0.001) for adverse cardiovascular (mainly thrombotic) events, an HR of 1.31 (95% CI: 1.13 to 1.51; p < 0.001) for all-cause mortality, and an HR of 1.35 (95% CI: 1.06 to 1.71; p = 0.015) for cardiovascular death.
On multivariate analysis, both the CHADS2 score and a high plasma vWF level (≥221 IU/dl) remained significantly associated with prognosis. The CHADS2 score had an HR of 1.31 (95% CI: 1.12 to 1.54; p < 0.001), and for a high plasma vWF level, the HR was 3.02 (95% CI: 1.98 to 4.58; p < 0.001) for adverse cardiovascular events. For all-cause mortality, the CHADS2 score had an HR of 1.49 (95% CI: 1.23 to 1.79; p < 0.001), and a high plasma vWF level had an HR of 2.41 (95% CI: 1.48 to 3.92; p < 0.001). For cardiovascular death, the CHADS2 score had an HR of 1.56 (95% CI: 1.14 to 2.12; p = 0.005), and for a high plasma vWF level, the HR was 3.67 (95% CI: 1.57 to 8.59; p = 0.003). The CHA2DS2-VASc score also maintained its significance on multivariate analysis when combined with plasma vWF levels, with the exception of cardiovascular death (Table 6).
Relationship of major bleeding events to HAS-BLED score
The HAS-BLED score was predictive of major bleeding events (HR: 1.90; 95% CI: 1.53 to 2.37; p < 0.001), and both the HAS-BLED score and high plasma vWF level remained significant after multivariate analysis. The HAS-BLED score showed an HR of 1.80 (95% CI: 1.44 to 2.25; p < 0.001), and for a high plasma vWF level, the HR was 2.81 (95% CI: 1.71 to 4.61; p < 0.001).
Additive effect of vWF on clinical risk scores
Annualized event rates for the CHADS2, CHA2DS2-VASc, and HAS-BLED risk scores by plasma vWF level are shown in Tables 6 and 7.⇓ The highest risk strata were those with high-risk clinical criteria plus high plasma vWF levels for both thrombotic and bleeding events. Those with moderate risk clinical criteria plus low plasma vWF levels were at lower risk of an event, but not as low as those identified as low risk, but there was clear separation of the annualized event rates for moderate-risk clinical criteria plus low plasma vWF levels versus high-risk clinical criteria plus low plasma vWF levels according to all risk scores.
In this study, we showed for the first time how an increased plasma vWF level, an established marker for endothelial damage/dysfunction, is associated with an adverse prognosis in AF patients, with regard to cardiovascular (mainly thrombotic) events, mortality, and bleeding. We also applied the new CHA2DS2-VASc and HAS-BLED scores and showed that the addition of plasma vWF levels as another biomarker risk factor would help to refine these clinical risk stratification schemes for stroke and bleeding.
AF patients constitute a high-risk population with both cardiovascular and hemorrhagic events. The Framingham Heart Study demonstrated that sex, age, diabetes, hypertension, chronic heart failure, and cardiac valve disease were independent risk factors for AF (24). Such risk factors were also recognized risk factors for stroke and thromboembolism and were included in clinical schemes designed for recommending thromboprophylaxis, but these variables are also established cardiovascular risk factors. Data from the REACH (REduction of Atherothrombosis for Continued Health) study demonstrated that CHADS2 is also a good scoring system for predicting cardiovascular outcomes in patients with and at-risk of atherothrombosis (25). During the follow-up period in the present study, we observed a rate of thrombotic events of 5.0% per year, including 1.7% per year for stroke, 1.9% per year for acute coronary syndrome, and 1.5% for acute heart failure. The incidence of stroke in AF patients with adjusted oral anticoagulation ranges from 1.2% to 2.0% per year (26), whereas the rate of fatal and nonfatal myocardial infarction ranges from 0.5% to 4% per year (27).
In the present study, we found that the strongest risk factors for thrombotic events were age 75 years and older and high plasma vWF levels. This is not surprising in an elderly population with AF in which, despite the use of oral anticoagulants, the incidence of thrombotic events (including stroke, myocardial infarction, and vascular death) was 5.9% per year (28). In our study, 57% of AF patients were older than 75 years of age and 29% were older than 80 years of age. Second, we had a prevalent cardiovascular risk factor population, with >80% with hypertension, 30% with hypercholesterolemia, and 25% with diabetes; moreover, most of our patients (76%) were at moderate to high risk of thromboembolism (CHADS2 risk score ≥2). There is also a high incidence of AF among patients with atherothrombosis, in whom AF was associated with a major increase in cardiovascular events, and the CHADS2 score was predictive not only of stroke but also of cardiovascular death (15,29), even in non-AF patients (25). In this setting, we found that the CHADS2 risk score was predictive of adverse cardiovascular events, being a composite of thrombotic events, global mortality, and cardiovascular death.
Increased plasma vWF levels have been found in inflammatory and atherosclerotic vascular diseases (9). vWF may play an important role in the pathogenesis of atherosclerosis, being associated with subsequent cardiovascular events (30). vWF is increased in AF patients (31), but it remains controversial whether such an increase is due to the AF itself or coexistent cardiovascular risk factors (32). Independent associations between several AF stroke risk factors and vWF have been shown (11) as well as a correlation with the CHADS2 risk score (13). Also, plasma vWF levels were predictive of subsequent stroke and vascular events in AF and had an additive role to clinical factors for risk stratification (12,33). Our study provides evidence that the plasma vWF level consistently predicts cardiovascular events, stroke, coronary syndrome and acute heart failure, major bleeding events, and mortality (both cardiovascular and all cause) in a large cohort of consecutive anticoagulated AF patients, even after adjusting for confounding factors and of having an additive effect on clinical risk stratification schemes. Moreover, the plasma vWF level was also predictive of stroke (HR: 5.04) and acute coronary syndrome (HR: 3.42) (both p < 0.001). As recently reviewed, there is a strong correlation between elevated plasma vWF levels and the incidence and prognosis of acute coronary syndrome (34).
Similar prognostic data have been reported for DD, even in patients receiving OAC therapy (15,16), but in this study, we were not able to find such an association. Inclusion criteria at study entry was the time in therapeutic range (TTR) during the previous 6 months of 100% (international normalized ratio between 2.0 and 3.0), and, thus, the DD values may have been suppressed so much that we did not find any significant impact of plasma DD levels, whereas in previously published studies, the TTR was not reported.
Bleeding events prediction deserves special attention. High plasma vWF levels also predict major bleeding, remaining significant after multivariate analysis. Congenitally decreased plasma vWF levels, as seen in von Willebrand disease, are associated with high bleeding risk, whereas high levels are associated with cardiovascular risk. In our population, high plasma vWF levels probably identify a population at very high risk of both thrombotic and hemorrhagic events. Indeed, risk factors for warfarin-associated bleeding are also indications for warfarin use in AF patients (35).
In the pooled analysis of the first 5 trials with warfarin in AF, the annual rate of major bleeding was 1.0% in control patients compared with 1.3% in patients treated with warfarin, with the main predisposing factors being the intensity of the anticoagulant effect, patient characteristics, concomitant use of drugs that interfere with hemostasis, and the length of therapy (36). Our independent predictors of hemorrhage were female sex, renal impairment, previous hemorrhagic event, and high plasma vWF levels. On univariate analysis, age 75 years and older was predictive of bleeding, but it lost its significance after multivariate analysis. Our rate of major hemorrhage was 3.6% per year, and several studies showed that elderly patients are more prone to bleeding (37–39), with annual rates of major hemorrhages of 1.4% (36) to 7.2% (37). No previous study found female sex as an independent risk factor for bleeding, except for a slightly increased HR of 1.4 of borderline significance (40).
Accordingly, the ACTIVE-W (Atrial Fibrillation Clopidogrel Trial With Irbesartan for Prevention of Vascular Events) showed that patients with a CHADS2 risk score of 1 had a lower major bleeding risk than patients with a CHADS2 score of ≥2 (41). However, we did not find any association between the CHADS2 score and bleeding risk. Although a systematic review identified the following patient characteristics as risk factors for anticoagulation-related bleeding complications—advanced age, uncontrolled hypertension, history of ischemic heart disease, cerebrovascular disease, anemia or a history of bleeding, and the concomitant use of other drugs such as antiplatelet agents (42)—our study only identified a history of bleeding as an independent risk factor. Renal impairment has not been previously reported to be a risk factor for bleeding in AF (42), except for patients undergoing coronary stenting (43). We found that the new HAS-BLED scheme was predictive of subsequent bleeding events, even remaining significant after multivariate analysis of plasma vWF levels. However, a high plasma vWF level showed an additive effect on the HAS-BLED score, for an intermediate-risk category for bleeding (HAS-BLED score of 1 to 2 points), so that high plasma vWF levels changed the annual risk of a hemorrhagic event from 1.2% (based on clinical criteria) to 4.7%.
At entry, the TTR was 100%, but we do not have data on the TTR during follow-up, which could affect both thrombotic and hemorrhagic events. It is also uncertain whether plasma vWF levels bring a new, different physiological factor into stroke prediction in AF patients or simply a measure of severity of previously identified clinical predictors. Of note, the plasma vWF level maintained its statistical significance after adjusting for several risk factors for both thrombotic and hemorrhagic events.
We report in a large cohort of anticoagulated AF patients that plasma vWF levels were independent predictors of thrombotic and bleeding events and death during >2 years of follow-up. This biomarker may potentially be used to refine stroke and bleeding risk stratification in AF.
This work was partially supported by Sociedad Española de Cardiología, RD06/0014/039, (RECAVA) from ISCIII; and PI081531-FEDER from ISCIII. Dr. Marín has received a research grant from Abbott Laboratories and Boston Scientific. Dr. Hernández-Romero has received a research grant from Abbott Laboratories and Boston Scientific, and holds a postdoctoral position funded by the Instituto de Salud Carlos III. Dr. Valdés has received a research grant from Abbott Laboratories and Boston Scientific. Dr. Lip has provided consultancy, lectures, and/or educational symposia for Astellas, Bayer, AstraZeneca, Merck, Pfizer, Bristol-Myers Squibb, Portola, and Biotronic. All other authors have reported that they have no relationships to disclose.
- Abbreviations and Acronyms
- atrial fibrillation
- Congestive heart failure, Hypertension, Age ≥75 years, Diabetes mellitus, Stroke, Vascular disease, Age 65 to 74 years, Sex category
- Congestive heart failure, Hypertension, Age >75, Diabetes mellitus, and prior Stroke or transient ischemic attack (doubled)
- confidence interval
- Hypertension, Abnormal renal/liver function, Stroke, Bleeding history or predisposition, Labile International Normalized Ratio, Elderly, Drugs/alcohol concomitantly
- hazard ratio
- oral anticoagulation
- time in therapeutic range
- von Willebrand factor
- Received October 11, 2010.
- Revision received December 2, 2010.
- Accepted December 14, 2010.
- American College of Cardiology Foundation
- Benjamin E.J.,
- Wolf P.A.,
- D'Agostino R.B.,
- Silberhatz H.,
- Kannel W.B.,
- Levy D.
- Camm A.J.,
- Kirchhof P.,
- Lip G.Y.,
- et al.
- Fuster V.,
- Rydén L.E.,
- Cannom D.S.,
- et al.
- Conway D.S.,
- Chin B.S.,
- Pearce L.A.,
- Hart R.G.,
- Lip G.Y.
- Conway D.S.,
- Pearce L.A.,
- Chin B.S.,
- Hart R.G.,
- Lip G.Y.
- Ferro D.,
- Loffredo L.,
- Polimeni L.,
- et al.
- Sadanaga T.,
- Sadanaga M.,
- Ogawa S.
- Hughes M.,
- Lip G.Y.,
- Guideline Development Group for the NICE national clinical guideline for management of atrial fibrillation in primarysecondary care
- Lip G.Y.H.,
- Frison L.,
- Halperin J.L.,
- Lane D.A.
- Schulman S.,
- Angerås U.,
- Bergqvist D.,
- Eriksson B.,
- Lassen M.R.,
- Fisher W.,
- Subcommittee on Control of Anticoagulation of the ScientificStandardization Committee of the International Society on ThrombosisHaemostasis
- Zweig M.H.,
- Campbell G.
- Barthelemy O.,
- Beygui F.,
- Silvain J.,
- et al.
- Mant J.,
- Hobbs F.D.,
- Fletcher K.,
- et al.,
- BAFTA Investigators,
- Midland Research Practices Network (MidReC)
- Freestone B.,
- Lip G.Y.
- Lip G.Y.,
- Lane D.,
- van Walraben C.,
- Hart R.G.
- Spiel A.O.,
- Gilbert J.C.,
- Jilma B.
- Schulman S.,
- Beyth R.J.,
- Kearon C.,
- Levine M.N.,
- American College of Chest Physicians
- Poli D.,
- Antonucci E.,
- Grifoni E.,
- Abbate R.,
- Gensini G.F.,
- Prisco D.
- Hylek E.M.,
- Evans-Molina C.,
- Shea C.,
- Henault L.E.,
- Regan S.
- Shireman T.I.,
- Howard P.A.,
- Kresowik T.F.,
- Ellerbeck E.F.
- Healey J.S.,
- Hart R.G.,
- Pogue J.,
- et al.
- Hughes M.,
- Lip G.Y.H.