Advertisement

SAMe-TT2R2 Score, Time in Therapeutic Range, and Outcomes in Anticoagulated Patients with Atrial Fibrillation

      Abstract

      Background

      Oral anticoagulation is highly effective in preventing stroke and mortality in nonvalvular atrial fibrillation patients. However, the efficacy and safety of vitamin K antagonists (the main oral anticoagulation drug used) strongly depends upon the quantity of anticoagulation control, as reflected by the average percentage of the time in therapeutic range of international normalized ratio 2.0-3.0. An easy, simple prediction of which atrial fibrillation patients are likely to do well on vitamin K antagonists (with good average time in therapeutic range) could guide decision-making between using vitamin K antagonists (eg, warfarin) and non-vitamin K antagonist oral anticoagulants.

      Methods and Results

      In a consecutive cohort of nonvalvular atrial fibrillation patients attending our anticoagulation clinic, we tested the hypothesis that the new Sex, Race, Medical history, Tobacco use, Race (SAMe-TT2R2) score was a predictor for good average time in therapeutic range, and second, this would translate into adverse events in a “real world” cohort of patients with nonvalvular atrial fibrillation. The incidence of bleeding, adverse cardiovascular events (including stroke/thromboembolism), and mortality during the follow-up was higher with increasing SAMe-TT2R2 score. The SAMe-TT2R2 score was predictive for the composite of all adverse events (hazard ratio 1.32 [95% Confidence Interval 1.17-1.50]; P <.001), adverse cardiovascular events (1.52 [1.28-1.83]; P <.001), and all-cause mortality (1.41 [1.16-1.67]; P = .001). A trend was also observed for major bleeding events (1.23 [0.99-1.53]; P = .059).

      Conclusion

      In a “real world” cohort of consecutive patients with nonvalvular atrial fibrillation, a high SAMe-TT2R2 score (reflecting poor anticoagulation control with poor time in therapeutic range) was associated with more bleeding, adverse cardiovascular events, and mortality during follow-up.

      Keywords

      Clinical Significance
      • In a “real world” cohort of consecutive patients with nonvalvular atrial fibrillation, a high SAMe-TT2R2 score (reflecting poor anticoagulation control with poor time-in-therapeutic range [TTR]) was associated with more bleeding, adverse cardiovascular events, and mortality during follow-up.
      • The SAMe-TT2R2 score, an easy, simple prediction of which atrial fibrillation patients are likely to do well on vitamin K antagonists (VKA) (with good average TTR), could guide decision-making between using VKAs and non-VKA oral anticoagulants.
      Oral anticoagulation is highly effective in preventing stroke and mortality in nonvalvular atrial fibrillation patients.
      • Hart R.G.
      • Pearce L.A.
      • Aguilar M.I.
      Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation.
      However, the efficacy and safety of vitamin K antagonists depends upon the quality of anticoagulant control, as reflected by the average percentage of the time in therapeutic range of international normalized ratio 2.0-3.0. Various studies have shown how a high time in therapeutic range translates into a lower risk of stroke and bleeding while on oral anticoagulation.
      • Gallagher A.M.
      • Setakis E.
      • Plumb J.M.
      • Clemens A.
      • van Staa T.P.
      Risks of stroke and mortality associated with suboptimal anticoagulation in atrial fibrillation patients.
      • Morgan C.L.
      • McEwan P.
      • Tukiendorf A.
      • Robinson P.A.
      • Clemens A.
      • Plumb J.M.
      Warfarin treatment in patients with atrial fibrillation: observing outcomes associated with varying levels of INR control.
      • Wan Y.
      • Heneghan C.
      • Perera R.
      • et al.
      Anticoagulation control and prediction of adverse events in patients with atrial fibrillation: a systematic review.
      A recent European consensus document recommends that an average individual time in therapeutic range should be >70% for optimal efficacy and safety outcomes whilst on a vitamin K antagonist, and this is also recommended in international guidelines.
      • Camm A.J.
      • Lip G.Y.
      • De Caterina R.
      • et al.
      2012 focused update of the ESC Guidelines for the management of atrial fibrillation: an update of the 2010 ESC Guidelines for the management of atrial fibrillation—developed with the special contribution of the European Heart Rhythm Association.
      The vitamin K antagonists were the main oral anticoagulant drug used until the introduction of the non-vitamin K antagonist oral anticoagulants (NOACs, previously referred to as novel or new oral anticoagulants).
      The difficulties in achieving a high time in therapeutic range, as well as the inconvenience of regular anticoagulation monitoring and the various food/drug restrictions associated with the vitamin K antagonists, have led to the recent introduction of the NOACs, which offer improved efficacy, safety, and convenience compared with the vitamin K antagonists.
      • Camm A.J.
      • Lip G.Y.
      • De Caterina R.
      • et al.
      2012 focused update of the ESC Guidelines for the management of atrial fibrillation: an update of the 2010 ESC Guidelines for the management of atrial fibrillation—developed with the special contribution of the European Heart Rhythm Association.
      Even in clinical trials, patients in centers with high average time in therapeutic range have less marked difference in efficacy outcomes between NOACs and warfarin.
      • Wallentin L.
      • Yusuf S.
      • Ezekowitz M.D.
      • et al.
      Efficacy and safety of dabigatran compared with warfarin at different levels of international normalised ratio control for stroke prevention in atrial fibrillation: an analysis of the RE-LY trial.
      An easy, simple prediction of which atrial fibrillation patients are likely to do well on vitamin K antagonists (with good average time in therapeutic range) could guide decision-making between using vitamin K antagonists and NOACs. Recently, Apostolakis et al
      • Apostolakis S.
      • Sullivan R.M.
      • Olshansky B.
      • Lip G.Y.
      Factors affecting quality of anticoagulation control amongst atrial fibrillation patients on warfarin: the same-TT2R2 score.
      proposed and validated the SAMe-TT2R2 score (Sex, Age [<60 years], Medical history [at least 2 of the following: hypertension, diabetes, coronary artery disease/myocardial infarction, peripheral arterial disease, congestive heart failure, previous stroke, pulmonary disease, hepatic or renal disease], and Treatment [interacting drugs, eg, amiodarone for rhythm control] [all 1 point]; as well as current Tobacco use [2 points] and Race [non-Caucasian; 2 points]) (see Table 1). This simple score (SAMe-TT2R2) could help decision-making by identifying those atrial fibrillation patients that would do well on vitamin K antagonists (with a high average time in therapeutic range, with SAMe-TT2R2 score = 0-1), or conversely, those likely to have anticoagulation control (ie, poor average time in therapeutic range, with SAMe-TT2R2 score ≥2) who require additional interventions to achieve acceptable anticoagulation control or be candidates for NOACs. This score was derived from the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) trial population and externally validated in a small “real world” cohort of anticoagulated nonvalvular atrial fibrillation patients.
      • Apostolakis S.
      • Sullivan R.M.
      • Olshansky B.
      • Lip G.Y.
      Factors affecting quality of anticoagulation control amongst atrial fibrillation patients on warfarin: the same-TT2R2 score.
      Further validations in large “real world” cohorts are necessary, also to assess whether this score (which reflects time in therapeutic range) translates into adverse outcomes in anticoagulated nonvalvular atrial fibrillation patients.
      Table 1The SAME-TT2R2 Score
      SSex (female)1
      AAge (<60 years)1
      MMedical history
      Defined as more than 2 of the following: hypertension, diabetes, coronary artery disease/myocardial infarction, peripheral arterial disease, congestive heart failure, previous stroke, pulmonary disease, hepatic or renal disease.
      1
      e
      TTreatment (rhythm control strategy)1
      TTobacco use (within 2 years)2
      RRace (non-Caucasian)2
      Defined as more than 2 of the following: hypertension, diabetes, coronary artery disease/myocardial infarction, peripheral arterial disease, congestive heart failure, previous stroke, pulmonary disease, hepatic or renal disease.
      In a consecutive cohort of nonvalvular atrial fibrillation patients attending our anticoagulation clinic, we tested the hypothesis that new SAMe-TT2R2 score is a predictor for average time in therapeutic range. Second, we investigated whether the latter translated into adverse events (major bleeding, adverse cardiovascular events, mortality) in our “real world” cohort of patients with nonvalvular atrial fibrillation.

      Methods

      We recruited consecutive patients with permanent or paroxysmal atrial fibrillation on a vitamin K antagonist from our outpatient anticoagulation clinic. In order to homogenize the baseline cohort of patients, only patients who had an international normalized ratio between 2.0 and 3.0 during the previous 6 months were included. Patients with prosthetic heart valves or valvular atrial fibrillation were excluded from the study, as well as those presenting with acute coronary syndrome, stroke (ischemic or embolic), any hemodynamic instability, or hospital admission or surgical intervention in the preceding 6 months. A history of malignancy was allowed if the patient's expected survival duration was more than 6 months and they were not receiving chemotherapy or radiotherapy at study entry. A complete medical history was recorded at inclusion. Follow-up was performed through visits to the anticoagulation clinic, the hospital electronic medical records system, or, when unavailable or there were persisting doubts, by telephone interview.
      Baseline stroke risk was assessed using the CHA2DS2-VASc (Cardiac failure or dysfunction, Hypertension, Age over 75 years [Doubled], Diabetes, Stroke [Doubled] – Vascular disease, Age between 65 and 74 years, and Sex category [Female]) score, as described in recent guidelines.
      • Camm A.J.
      • Lip G.Y.
      • De Caterina R.
      • et al.
      2012 focused update of the ESC Guidelines for the management of atrial fibrillation: an update of the 2010 ESC Guidelines for the management of atrial fibrillation—developed with the special contribution of the European Heart Rhythm Association.
      • Lip G.Y.
      • Nieuwlaat R.
      • Pisters R.
      • Lane D.A.
      • Crijns H.J.
      Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation.
      The HAS-BLED bleeding risk score was calculated as a measure of baseline bleeding risk, as the result of adding one point to Hypertension, Abnormal renal/liver function (one point each), Stroke, Bleeding history or predisposition, Labile international normalized ratio, Elderly (age over 65 years), and Drugs/alcohol concomitantly (one point for each one).
      • Pisters R.
      • Lane D.A.
      • Nieuwlaat R.
      • de Vos C.B.
      • Crijns H.J.
      • Lip G.Y.
      A novel user-friendly score (HAS-BLED) to assess 1-year risk of major bleeding in patients with atrial fibrillation: the Euro Heart Survey.
      Adverse cardiovascular endpoints (mainly thromboembolic) were defined as stroke/transient ischemic attack, peripheral embolism, acute coronary syndrome, acute heart failure, and cardiac death. Bleeding events were assessed by the 2005 International Society on Thrombosis and Haemostasis criteria, including fatal bleeding or symptomatic bleeding in a critical area or organ, such as intracranial, intraspinal, intraocular, retroperitoneal, intra-articular or pericardial, or intramuscular with compartment syndrome or bleeding causing a fall in hemoglobin level of 20 g/L (1.24 mmol/L) or more, or leading to transfusion of 2 or more units of whole blood or red cells.
      • Lip G.Y.
      • Andreotti F.
      • Fauchier L.
      • et al.
      Bleeding risk assessment and management in atrial fibrillation patients. Executive Summary of a Position Document from the European Heart Rhythm Association [EHRA], endorsed by the European Society of Cardiology [ESC] Working Group on Thrombosis.
      Finally, we recorded all-cause mortality, and whether the cause of death was secondary to a cardiovascular event (stroke/transient ischemic attack, peripheral embolism, acute coronary syndrome, acute heart failure, or cardiac death) or a hemorrhagic one. At 6 months after inclusion, we calculated the percentage of international normalized ratio measures within the therapeutic range during the previous 6 months, as an estimation of the time in therapeutic range.
      The protocol study was approved by the Ethical Committee from University Hospital Morales Meseguer, and patients gave informed consent to participation in the study.

      Statistical Analysis

      Continuous variables were tested for normality by the Kolmogorov-Smirnov test. Continuous variables are presented as a mean ± SD or median (interquartile range [IQR]), as appropriate, and categorical variables as a percentage. Cox models were used to determine the association between clinical risk factors and bleeding, as well as cardiovascular events and mortality. The independent effect of clinical variables on prognosis was calculated using a Cox proportional hazard regression model, and analysis of variance models were used to compare means between different groups. The c-statistic was calculated using receiver operating characteristic curves. A P <.05 was accepted as statistically significant. Statistical analyses were performed using SPSS 15.0 for Windows (SPSS, Inc., Chicago, Ill).

      Results

      We studied 972 patients (49% male, median age 76 years, IQR 70-80) (Table 2). The median CHA2DS2-VASc score was 4 (IQR 3-5), and 93% had a CHA2DS2-VASc score ≥2. The median HAS-BLED score was 2 (IQR 2-3).
      Table 2Baseline Clinical Characteristics
      Baseline Characteristicsn = 972
      Male sex478 (49%)
      Age, median (IQR)76 (70-80)
       Age <60 years66 (7%)
      Hypertension796 (82%)
      Diabetes mellitus249 (26%)
      Heart failure350 (36%)
      History of stroke or TIA182 (19%)
      Hepatic impairment11 (1%)
      Renal impairment94 (10%)
      Coronary artery disease182 (19%)
      Hypercholesterolemia303 (31%)
      Current smoking habit136 (14%)
      Current alcoholic consumption24 (2.5%)
      Previous bleeding episode79 (8%)
      Concomitant malignant disease58 (6%)
      Concomitant treatment
       Antiplatelet therapy156 (16%)
       Angiotensin-converting enzyme inhibitors244 (25%)
       Angiotensin-renin blockers214 (22%)
       Calcium antagonist208 (21%)
       Beta-blockers293 (30%)
       Statins202 (21%)
       Digoxin179 (18%)
       Diuretics401 (41%)
      CHA2DS2-VASc score, median (IQR)4 (3-5)
      HAS-BLED score, median (IQR)2 (2-3)
      SAME-TT2R2 score, median (IQR)2 (1-2)
      CHA2DS2-VASc = Cardiac failure or dysfunction, Hypertension, Age ≥75 [Doubled], Diabetes, Stroke [Doubled] − Vascular disease, Age 65-74 and Sex category [Female]; HAS-BLED = Hypertension, Abnormal Renal/Liver Function, Stroke, Bleeding History or Predisposition, Labile international normalized ratio, Elderly, Drugs/Alcohol Concomitantly; IQR = interquartile range; TIA = transient ischemic attack.
      Median follow-up was 952 (784-1078) days, and during this period, 107 patients had an adverse cardiovascular event (4.22%/year): of these, 35 were strokes (1.38%/year), 42 were acute coronary syndrome events (1.65%/year), and 31 acute heart failure events (1.22%/year). Of the cohort, 77 patients presented with major bleeding (3.04%/year) and 91 patients died (3.59%/year). In the following 6 months after inclusion, mean time in therapeutic range was 78.0% ± 19.98%.

      Predictive Value for Anticoagulation Control

      In our anticoagulated patients, increasing baseline SAMe-TT2R2 score was associated with significantly lower mean time in therapeutic range at 6 months after inclusion. This ranged from 79.7% for patients with SAMe-TT2R2 score of 0 points to 72.3% for patients with 5 points (P = .043) (Table 3).
      Table 3Baseline SAMe-TT2R2
      SAMe-TT2R2 ScoreScoreTTR at 6-Month Follow-up Mean (± SD)n = 972
      • Low0-179.67(± 19.46)431 (44%)
      • Borderline278.40 (± 20.28)332 (34%)
      • High≥374.25 (± 20.24)208 (22%)
      TTR = time in therapeutic range.

      Predictive Value of the SAMe-TT2R2 Score for Bleeding, Cardiovascular Events, and Death

      The SAMe-TT2R2 score had a c-statistic for adverse cardiovascular events of 0.62 (95% confidence interval [CI], 0.57-0.68; P <.001); for bleeding, the c-statistic was 0.55 (95% CI, 0.49-0.62; P = .117). For all-cause mortality, the c-statistic was 0.62 (95% CI, 0.55-0.68; P <.001).

      Major Adverse Events

      The incidence of both adverse cardiovascular events (including stroke/thromboembolism) and bleeding and mortality during the follow-up was higher with increasing SAMe-TT2R2 score at baseline (Figure).
      Figure thumbnail gr1
      FigureAdverse events according to SAMe-TT2R2 score at baseline.
      On unadjusted (crude) analyses, the SAMe-TT2R2 score was predictive for the composite of all adverse events (hazard ratio 1.32 [95% confidence interval 1.17-1.50]; P <.001), adverse cardiovascular events (hazard ratio 1.52 [95% confidence interval 1.28-1.83]; P <.001], and for all-cause mortality (1.41 [95% confidence interval 1.16-1.71]; P = .001). Only a nonsignificant trend was seen regarding major bleeding events (1.23 [0.99-1.53]; P = .059).

      Discussion

      In a “real world” cohort of consecutive patients with nonvalvular atrial fibrillation, we have shown that a low SAMe-TT2R2 score was a significant predictor for good anticoagulation control (as reflected by time in therapeutic range). A high SAMe-TT2R2 score (reflecting poor anticoagulation control with poor time in therapeutic range) translated into more bleeding, adverse cardiovascular events (including stroke/thromboembolism) and mortality during follow-up. To the best of our knowledge, we have also validated the SAMe-TT2R2 score for the first time in consecutive atrial fibrillation patients taking acenocoumarol treatment as their oral anticoagulant (rather than warfarin).
      Nonvalvular atrial fibrillation patients are at high risk for cardiovascular events and mortality. Our population was an elderly cohort with nonvalvular atrial fibrillation where over 50% were aged >75 years. Moreover, our population had an intermediate-high thrombotic risk and multiple prevalent comorbidities. Even in patients with good drug compliance, various acute or clinical decompensate events may lead to unstable international normalized ratio control. Therefore, good prior international normalized ratio control does not necessarily imply good control in the future. Also, the largest randomized trial of genotype-guided (pharmacogenetically based) dosing of warfarin did not result in improved anticoagulation control.
      • Kimmel S.E.
      • French B.
      • Kasner S.E.
      • et al.
      A pharmacogenetic versus a clinical algorithm for warfarin dosing.
      Thus, the ability to have a simple, practical way to predict who is likely to do well on vitamin K antagonists, or those less likely to have good anticoagulation control where NOACs would benefit, could help decision-making in everyday clinical practice.
      • De Caterina R.
      • Husted S.
      • Wallentin L.
      • et al.
      Vitamin K antagonists in heart disease: current status and perspectives (Section III). Position Paper of the ESC Working Group on Thrombosis—Task Force on Anticoagulants in Heart Disease.
      • De Caterina R.
      • Husted S.
      • Wallentin L.
      • et al.
      European Society of Cardiology Working Group on Thrombosis Task Force on Anticoagulants in Heart Disease
      General mechanisms of coagulation and targets of anticoagulants (Section I). Position Paper of the ESC Working Group on Thrombosis—Task Force on Anticoagulants in Heart Disease.
      International normalized ratio control and quality of anticoagulation is most often assessed by the time in therapeutic range. The average individual time in therapeutic range is known to increase over time, as more international normalized ratio fluctuations occur during the first 3 months following initiation of vitamin K antagonists.
      • Garcia D.A.
      • Lopes R.D.
      • Hylek E.M.
      New-onset atrial fibrillation and warfarin initiation: high risk periods and implications for new antithrombotic drugs.
      Good adherence to the therapy, as well as minimizing drug or diet interactions, influences time in therapeutic range outcomes. In the present study, we selected experienced anticoagulated patients with excellent anticoagulation control at baseline to ensure some degree of patient homogeneity, and thus, adherence to the therapy was expected to be high and not affect subsequent measurements. However, we found that the SAMe-TT2R2 score was still able to predict which patients were prone to subsequent instability and poor international normalized ratio control, even among previously anticoagulated patients, with good time in therapeutic range at inclusion. Slight differences were observed, among the different range of SAMe-TT2R2 scores, due to our inclusion criteria, given that all patients had shown previously excellent international normalized ratio control.
      The SAMe-TT2R2 score was found to be predictive for both adverse cardiovascular events and mortality, but a trend was observed only regarding major bleeding events. It should be borne in mind that the anticoagulation control in our cohort was optimal, and therefore, only few major bleeding events occurred during the follow-up (with a rate of 3%/year).
      Although the SAMe-TT2R2 score at baseline is related to outcomes (mainly adverse cardiovascular events and mortality, but also to major bleeding), the low c-statistics show a moderate discriminatory ability of the survival model. This might be related to low outcome event rates (mainly few major bleeding events during the follow-up, with an annual rate of 3%/year), due to the good control of anticoagulation therapy in our (selected) cohort at baseline. Given the modest discriminatory ability (as reflected by c-statistics) of the SAMe-TT2R2 score in our well-controlled anticoagulated clinic population, it is unclear how the score would perform in a less meticulously cared-for group of patients.
      Of note, we studied a white population, without any prevalence of other ethnic races, which is one of the risk factors for poor time in therapeutic range, based on the SAMe-TT2R2 score.
      • Apostolakis S.
      • Sullivan R.M.
      • Olshansky B.
      • Lip G.Y.
      Factors affecting quality of anticoagulation control amongst atrial fibrillation patients on warfarin: the same-TT2R2 score.
      Thus, none of the patients could reach a SAMe-TT2R2 score of >6 points. Moreover, high scores were poorly represented in our cohort, as only 6 patients had a score of 5 and none had a score of 6, perhaps reflecting our selection criteria, as stable international normalized ratio was required at baseline. Indeed, patients with the highest SAMe-TT2R2 scores did not achieve good anticoagulation control and therefore were probably excluded from the present cohort.

      Limitations

      The possibility of some selection bias cannot be excluded as our patients were clinically stable at study entry, so unstable patients who are more prone to have adverse events were excluded, unlike other clinical studies where patients recruited would present with a time in therapeutic range ranging from 60% to 75%. Similarly, we selected for our cohort “anticoagulation experienced” patients with proven adherence and good anticoagulation control during the 6 months before inclusion. Thus, patients with erratic anticoagulation (who would probably have higher SAMe-TT2R2 score values) were excluded from the analysis. This specific selection of patients (stable on acenocoumarol with high initial time in therapeutic range at entry) would perhaps limit the generalizability of our findings to less well-controlled populations. Furthermore, we used acenocoumarol (the most widely used vitamin K antagonists in Spain), which differs from warfarin, with a shorter half-life, which is perceived to lead to more unstable anticoagulation, although no results comparing the different vitamin K antagonists have been published. We also presumed that patients with high SAMe-TT2R2 score would be “suitable” to be treated with NOAC, but we have not studied such a population. Finally, we studied a white population, without any other races being studied, and thus, further studies in ethnically diverse populations would be required. Ongoing prospective studies will address all these aspects.
      In conclusion, a low SAMe-TT2R2 score was a significant predictor for good anticoagulation control (as reflected by time in therapeutic range), whilst a high SAMe-TT2R2 score (reflecting poor anticoagulation control) was associated with more bleeding, adverse cardiovascular events, and mortality during follow-up. Thus, the simple SAMe-TT2R2 score may aid decision-making by identifying those atrial fibrillation patients likely to do well on vitamin K antagonist (score 0-2) or those likely to have poor(er) anticoagulation control (score >2) who could be better off being started on NOACs as initial therapy, or be “flagged up” for more aggressive efforts to improve anticoagulation control.

      References

        • Hart R.G.
        • Pearce L.A.
        • Aguilar M.I.
        Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation.
        Ann Intern Med. 2007; 146: 857-867
        • Gallagher A.M.
        • Setakis E.
        • Plumb J.M.
        • Clemens A.
        • van Staa T.P.
        Risks of stroke and mortality associated with suboptimal anticoagulation in atrial fibrillation patients.
        Thromb Haemost. 2011; 106: 968-977
        • Morgan C.L.
        • McEwan P.
        • Tukiendorf A.
        • Robinson P.A.
        • Clemens A.
        • Plumb J.M.
        Warfarin treatment in patients with atrial fibrillation: observing outcomes associated with varying levels of INR control.
        Thromb Res. 2009; 124: 37-41
        • Wan Y.
        • Heneghan C.
        • Perera R.
        • et al.
        Anticoagulation control and prediction of adverse events in patients with atrial fibrillation: a systematic review.
        Circ Cardiovasc Qual Outcomes. 2008; 1: 84-91
        • Camm A.J.
        • Lip G.Y.
        • De Caterina R.
        • et al.
        2012 focused update of the ESC Guidelines for the management of atrial fibrillation: an update of the 2010 ESC Guidelines for the management of atrial fibrillation—developed with the special contribution of the European Heart Rhythm Association.
        Europace. 2012; 14: 1385-1413
        • Wallentin L.
        • Yusuf S.
        • Ezekowitz M.D.
        • et al.
        Efficacy and safety of dabigatran compared with warfarin at different levels of international normalised ratio control for stroke prevention in atrial fibrillation: an analysis of the RE-LY trial.
        Lancet. 2010; 376: 975-983
        • Apostolakis S.
        • Sullivan R.M.
        • Olshansky B.
        • Lip G.Y.
        Factors affecting quality of anticoagulation control amongst atrial fibrillation patients on warfarin: the same-TT2R2 score.
        Chest. 2013; 144: 1555-1563
        • Lip G.Y.
        • Nieuwlaat R.
        • Pisters R.
        • Lane D.A.
        • Crijns H.J.
        Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation.
        Chest. 2010; 137: 263-272
        • Pisters R.
        • Lane D.A.
        • Nieuwlaat R.
        • de Vos C.B.
        • Crijns H.J.
        • Lip G.Y.
        A novel user-friendly score (HAS-BLED) to assess 1-year risk of major bleeding in patients with atrial fibrillation: the Euro Heart Survey.
        Chest. 2010; 138: 1093-1100
        • Lip G.Y.
        • Andreotti F.
        • Fauchier L.
        • et al.
        Bleeding risk assessment and management in atrial fibrillation patients. Executive Summary of a Position Document from the European Heart Rhythm Association [EHRA], endorsed by the European Society of Cardiology [ESC] Working Group on Thrombosis.
        Thromb Haemost. 2011; 106: 997-1011
        • Kimmel S.E.
        • French B.
        • Kasner S.E.
        • et al.
        A pharmacogenetic versus a clinical algorithm for warfarin dosing.
        N Engl J Med. 2013; 369: 2283-2293
        • De Caterina R.
        • Husted S.
        • Wallentin L.
        • et al.
        Vitamin K antagonists in heart disease: current status and perspectives (Section III). Position Paper of the ESC Working Group on Thrombosis—Task Force on Anticoagulants in Heart Disease.
        Thromb Haemost. 2013; 110: 1087-1107
        • De Caterina R.
        • Husted S.
        • Wallentin L.
        • et al.
        • European Society of Cardiology Working Group on Thrombosis Task Force on Anticoagulants in Heart Disease
        General mechanisms of coagulation and targets of anticoagulants (Section I). Position Paper of the ESC Working Group on Thrombosis—Task Force on Anticoagulants in Heart Disease.
        Thromb Haemost. 2013; 109: 569-579
        • Garcia D.A.
        • Lopes R.D.
        • Hylek E.M.
        New-onset atrial fibrillation and warfarin initiation: high risk periods and implications for new antithrombotic drugs.
        Thromb Haemost. 2010; 104: 1099-1105