Time in the Therapeutic Range During Warfarin Therapy in Japanese Patients With Non-Valvular Atrial Fibrillation

Transcription

1 Circulation Journal Official Journal of the Japanese Circulation Society ORIGINAL ARTICLE Arrhythmia/Electrophysiology Time in the Therapeutic Range During Warfarin Therapy in Japanese Patients With Non-Valvular Atrial Fibrillation A Multicenter Study of Its Status and Influential Factors Ken Okumura, MD; Takashi Komatsu, MD; Takeshi Yamashita, MD; Yuji Okuyama, MD; Masahiko Harada, MD; Yoshiyuki Konta, MD; Toru Hatayama, MD; Daisuke Horiuchi, MD; Eiki Tsushima, PhD Background: Time in the therapeutic range (TTR) assesses the appropriateness of international normalized ratio of prothrombin time (PT-INR) control during warfarin therapy. We examined the status of and the factors influencing TTR in Japanese patients with non-valvular atrial fibrillation (AF). Methods and Results: We enrolled 501 AF patients (mean age, 70±10 years; males 66%; mean CHADS2 score 2.0±1.2) taking warfarin for 2 years from 5 prefectures. The PT-INR therapeutic range was set up according to the 2008 Japanese Guideline. TTR was 64±25% for all patients and varied from 56% to 74% with the institution. Time below and above TTR was 31±26% and 5±7%, respectively. TTR was not affected by gender or antiplatelet co-administration. TTR in patients <70 and 70 years old was 46±23% and 77±17%, respectively (P<0.0001). TTR in patients with CHADS2 score 1 and 2 was 59±27% and 68±23%, respectively (P<0.0001). TTR in patients with warfarin doses <2.0, , and 5.0 mg/day was 72±22%, 63±25% and 48±24%, respectively (all P<0.001). Multivariate analysis revealed age and warfarin dose as independent predictors of TTR. Conclusions: TTR is generally high in Japan, although it varies with institutions. Most of the time spent out of therapeutic range is below the range. TTR is influenced by age presumably because of the low range recommendation for elderly patients, and by warfarin dose presumably because of physicians anxiety about the hemorrhage risk. (Circ J 2011; 75: ) Key Words: Anticoagulants; Atrial fibrillation; Stroke Based on evidence accumulated mainly in the 1990s, 1 recent guidelines on the management of atrial fibrillation (AF) published in the United States, 2 Europe, 3 and Japan 4 recommend anticoagulation therapy for patients with non-valvular AF and risk factor(s) for ischemic stroke and systemic thromboembolism. Warfarin has been the only drug used for anticoagulation therapy for the past half century until the oral direct thrombin inhibitor, dabigatran, was recently demonstrated to be as effective as or more effective than warfarin in preventing ischemic stroke and thromboembolism in patients with non-valvular AF, 5,6 and subsequently approved for clinical use. A recent focused guideline update recommended dabigatran as a class I drug that is useful as an alternative to warfarin for the prevention of stroke and systemic thromboembolism in AF patients with risk factors for stroke or systemic embolization. 7 This guideline update also noted that because of the twice-daily dosing and greater risk of non-hemorrhagic side effects with dabigatran, patients already taking warfarin with excellent control of the international normalized ratio of prothrombin time (PT-INR) may have little to gain by switching to dabigatran. Editorial p 2048 It is well known that strict control of PT-INR during warfarin therapy is required to ensure not only efficacy but also safety. 8 Japanese guidelines recommend PT-INR control between 2.0 and 3.0 in patients younger than 70 years Received April 1, 2011; revised manuscript received April 15, 2011; accepted April 22, 2011; released online July 8, 2011 Time for primary review: 10 days Department of Cardiology, Hirosaki University Hospital, Hirosaki (K.O., D.H.); Department of Cardiology, Iwate Medical University Hospital, Morioka (T.K.); Department of Cardiovascular Medicine, The Cardiovascular Institute, Tokyo (T.Y.); Department of Cardiology, Osaka General Medical Center, Osaka (Y.O.); Department of Cardiology, Ube Industry Central Hospital, Ube (M.H.); Department of Internal Medicine, Reimeikyo Rehabilitation Hospital, Hirosaki (Y.K.); Hatayama Clinic, Hirosaki (T.H.); and Department of Physical Therapy, Hirosaki University School of Health Science, Hirosaki (E.T.), Japan Mailing address: Ken Okumura, MD, Department of Cardiology, Hirosaki University Graduate School of Medicine, Zaifu-cho 5, Hirosaki , Japan. okumura@cc.hirosaki-u.ac.jp ISSN doi: /circj.CJ All rights are reserved to the Japanese Circulation Society. For permissions, please cj@j-circ.or.jp

2 2088 OKUMURA K et al. Figure 1. Time in the therapeutic range (TTR) calculated for all 501 patients and separately for the patients of each group. and between 1.6 and 2.6 in those equal to or older than 70 years. 4 The PT-INR value represents the level of control at the time of measurement, and 1 measurement at 1 time does not necessarily reflect the appropriateness of long-term warfarin therapy in the individual patient. Recently, analysis of time in the therapeutic range (TTR) has been used to assess the appropriateness of PT-INR. 9 This technique is based on the assumption that anticoagulation intensity during warfarin therapy shifts linearly between any 2 consecutive measurements, 10 and TTR for an individual is calculated from the entire duration of observation. Subanalyses of recent largescale clinical studies revealed that TTR >60 65% is associated with reduced risk for stroke. 11,12 However, the status of TTR in Japan where the recommended therapeutic range for elderly patients is different from that in Western countries is not well known. The factors known to influence PT-INR include patient adherence and age, foods, concomitant drugs, and management by the physician However, the factors influencing TTR in Japanese AF patients undergoing longterm warfarin therapy are not fully understood. We considered that identifying the factors related to a low TTR value (ie, poor PT-INR control and increased risk of stroke) may be useful in the decision of whether to continue with warfarin or switch to dabigatran in Japanese patients with nonvalvular AF undergoing long-term warfarin therapy. Methods This multicenter study was conducted in 6 institutions and 1 clinic from 5 prefectures in Japan. The study protocol was approved by the ethics committee of each institution. written informed consent was given by each patient. Study Patients We enrolled patients with non-valvular AF undergoing warfarin therapy for 2 years in principle who had been followed up by each institution and visited an outpatient clinic between June 1, 2010, and July 31, The maximal number of enrollments from each institution was 100. The indication for prophylactic warfarin therapy, especially in patients with CHADS2 score of 0 or 1, was left to the decision of the attending physician. From a review of the patients charts, we collected the data including the patient profiles, warfarin dose, concomitant antiplatelet drug use, and PT-INR determined monthly or bimonthly in principle for the past 2 years. Warfarin dose during follow-up was adjusted according to TTR proposed by the Japanese Guidelines for Pharmacotherapy of Atrial Fibrillation (JCS2008); that is, between 2.0 and 3.0 for patients <70 years and between 1.6 and 2.6 for those In this study, the PT-INR value measured after at least 3 months after initiation of warfarin was used. Patients with stroke or systemic thromboembolism or major hemorrhage during the study period were excluded from analysis. Study Protocol Using specially programmed software (Medi-Skette Corporation, Tokyo, Japan), we input the successive PT-INR values of each patient and calculated the TTR. This software automatically draws lines successively between any 2 consecutive PT-INR values during the observation period, and calculates the percentage of the total time within the preset therapeutic range (ie, between 2.0 and 3.0 for patients <70 years and between 1.6 and 2.6 for those 70) over the entire period. Because this software was not designed to alter TTR in accordance with an increase in age from 69 to 70 during the 2-year

3 Warfarin Control in Japanese AF Patients 2089 Figure 2. Distribution of the number of the patients with time in the therapeutic range (TTR) from 0 to 100% at every 10% for all patients (Left), for the patients aged <70 years (Middle) and for the patients aged 70 (Right). study period, the range between 2.0 and 3.0 was used throughout the period to calculate TTR in patients who were younger than 70 years at the initiation of PT-INR collection and got closer to 70 during the study period. First, we analyzed the status of PT-INR control for all patients and for the patients from each institution. There were 3 institutions participating from Aomori Prefecture, so to simplify the comparative analysis of institutions from the 5 prefectures, the data for the patients from Aomori were combined and grouped as A. The patient data from Iwate Medical University Hospital, Cardiovascular Institute, Osaka General Medical Center, and Ube Industry Central Hospital were grouped as B, C, D and E, respectively. To examine the chronological change in TTR, we compared the TTR determined in the first year and for the second year for all patients. Also, any seasonal effect on TTR was examined. Second, we analyzed the possible influential factors of TTR, including age, gender, warfarin dose at the last visit, co-administration of antiplatelet drug, and CHADS2 score consisting of congestive heart failure (1 point), hypertension (1 point), age 75 years (1 point), diabetes mellitus (1 point) and previous stroke or transient ischemic attack (2 points). 18 Statistical Analysis All data are expressed as mean ± 1 standard deviation. For comparison of TTR between 2 groups, and for the assessment of TTR change from the first to the second year, a 2-tailed paired t-test was used. For comparison of TTR among 3 or more groups, 1-way analysis of variance (ANOVA) was used followed by post-hoc Bonferroni test. After these univariate analyses, multiple stepwise regression analysis with the use of SPSS Version 19 (SPAA Japan Inc, Tokyo, Japan) was performed to detect the independent predictors of TTR. P<0.05 was considered significant. Results Patients Clinical Characteristics From 7 institutions in the 5 prefectures we enrolled 501 AF patients undergoing warfarin therapy: 134 patients enrolled in Group A (Aomori), 100 in Group B (Iwate), 100 in Group C (Tokyo), 67 in Group D (Osaka) and 100 in Group E (Yamaguchi). Mean follow-up period was 706±58 days (1.9± 0.2 years) for all patients. Mean interval of PT-INR measurements was 45±13 days. There were 331 male and 170 female patients, and their mean age was 70±10 years, ranging from 29 to 92. The mean age of the male and female patients was 69±10 and 73±9 years, respectively (P<0.0001). Mean CHADS2 score was 2.0±1.2 (median 2), ranging from 0 to 6. Mean warfarin dose administered at the last visit was 3.0±1.2 mg/day (median 3.0 mg/day), ranging from 0.5 to 8.5 mg/day. Antiplatelet drugs were co-administered in 172 patients (34%). Status of PT-INR Control and Influential Factors TTR was calculated for all patients and separately for each group. It was 64±25%, ranging from 0% to 100%, for all patients and in each group it varied from of 56% to 74% (P<0.001 by ANOVA) with significant low values in Groups A and D compared with the others (Figure 1). The distribution of the numbers of patients with TTR from 0 to 100% at every 10% for all patients, for patients <70 years and those aged 70 is shown in Figure 2. In the total patient group, 303 (60.5%) had a TTR >60%. The distribution was quite different between the patient groups aged <70 and 70 years. The TTR calculated for the first and second observation years for all patients was 63±29% and 65±28%, respectively (P=0.0737). To examine the effect of seasons, TTR was calculated separately for the periods from April to September

4 2090 OKUMURA K et al. Figure 3. Chronological change in the time in the therapeutic range (TTR). The 2-year observation period was divided into spring and summer periods (from April to September) and autumn and winter periods (from October to March). Figure 4. Comparison of the time in the therapeutic range (TTR) among patient groups with different ages. and from October to May. As shown in Figure 3, no significant seasonal effect was observed, although TTR was slightly increased as the year changed from the first to the second. Time below and above TTR was 31±26% and 5±7%, respectively, for all patients. Groups with low TTR values showed longer times under TTR. TTR was not affected by gender: in male and female patients it was 64±25% and 65±25%, respectively. Also, it was not affected by co-administration of antiplatelet drugs: TTR in patients with and without antiplatelet drugs was 64±25%

5 Warfarin Control in Japanese AF Patients 2091 Figure 5. Comparison of the time in the therapeutic range (TTR) among patient groups with different CHADS2 scores. Figure 6. Comparison of the time in the therapeutic range (TTR) among patient groups with different doses of warfarin. and 65±25%, respectively. Figure 4 shows TTR separately for patients aged 49, between 50 and 59, between 60 and 69, between 70 and 74, and 75 years. TTR values for the groups aged <70 years were all significantly low compared with TTR for the groups aged 70 (all P<0.0001). TTR in the groups aged <70 and 70 years was 46±23% and 77±17%, respectively (P<0.0001), and this significant difference was observed in each of groups A E. Time below and above TTR in the patient group aged <70 years was 51±24% and 2±4%, respectively, while in the group aged 70 years it was 23±16% and 7±9%, respectively. Figure 5 shows TTR separately for the patient groups with CHADS2 scores from 0 to 6. TTR in the group with score 0 was 50±31% and was significantly lower than in the groups with scores of 2 or 3. When TTR was compared between the groups with CHADS2 score 1 and 2, it was 59±27% and 68±23%, respectively (P<0.0001). The average age of the patients with CHADS2 score 1 (n=198) was 65±11 years,

6 2092 OKUMURA K et al. Table. Multiple Stepwise Regression Analysis of the Variables That Were Related to Time in the Therapeutic Range β SEM t P value Age CHADS2 score Warfarin dose whereas that of the patients with a score 2 (n=303) was 73± 8 years (P<0.0001). Of the 198 patients with CHADS2 score 1, 121 (61%) were younger than 70 years, and their TTR was 45±23%. Of the other 303 with a score 2, 87 (29%) were younger than 70 years, and their TTR was 48±22% (P= NS, vs. patients with the score 1). Thus, in patients younger than 70 years, TTR was low, irrespectively of CHADS2 score. Figure 6 shows the effect of warfarin dose on TTR, separately for the patient groups with a dose <2.5 mg/day (mean 1.7±0.4), between 2.5 and 4.9 mg/day (mean 3.3±0.6), and 5.0 mg/day (mean 5.6±0.8). TTR in these 3 groups was 72± 22%, 63±25% and 48±24%, respectively (all P<0.001). The mean dose in patients aged <70 years was 3.5±1.2 mg/day and in the elderly it was 2.7±1.0 mg/day (P<0.0001). Multiple stepwise regression analysis revealed that age and warfarin dose were independent predictors of TTR (Table). Discussion Major Findings By analyzing the TTR of 501 patients enrolled from 5 prefectures in Japan, we found that PT-INR was well-controlled or moderately well-controlled, although it varied significantly with the institutions. Most of the time spent out of TTR was under the range, reflecting a tendency toward relatively weak PT-INR control in Japan. There were slight chronological changes in PT-INR control over the 2-year period. The factors influencing TTR included age, CHADS2 score and warfarin dose. Multivariate analysis indicated that age and warfarin dose were independent predictors of TTR. Status of PT-INR Control in Japan In Japan in the early 1990s, warfarin was used only in <20% of the patients with AF, but usage increased to 50% around year Currently, warfarin is used in approximately 90% of patients that are treated by cardiologists. To secure not only efficacy but also safety during warfarin therapy, Japanese guidelines recommended PT-INR control between 2.0 and 3.0 in patients aged <70 years and between 1.6 and 2.6 in those aged 70 years. 4 TTR is the proportion of time during which PT-INR is supposed to be within the recommended therapeutic range for the observation period, and the higher TTR is, the better the PT-INR control. 9,10 The analysis of the outcome of patients randomized to warfarin therapy in the SPORTIF III and V studies indicated that the risks of death, myocardial infarction, and stroke or systemic embolic event were lower in patients with TTR 60% than in those with TTR <60%. 11 A post-hoc analysis of the ACTIVE W study showed that in patients treated at hospitals with a TTR below the median value (65%), no treatment benefit with warfarin was demonstrated compared with combined antiplatelet therapy, whereas in patients at hospitals with a TTR 65%, warfarin showed a marked benefit, reducing vascular events by >2-fold. 12 The ACTIVE W study estimated the target threshold TTR to be between 58% and 65%, below which there is little benefit of warfarin over antiplatelet therapy. Masaki et al retrospectively analyzed 120 Japanese patients with AF and aged 70 years who were treated with warfarin according to the Japanese Guideline, and showed that TTR >68% was associated with reduced risk of stroke compared with 68%. 20 They also showed that the average PT-INR was not an effective measure for risk stratification. In this study, we calculated the TTR for PT-INR values measured over a 2-year period in 501 AF patients enrolled from 5 prefectures. We found that TTR calculated from the first- and second-year data was unchanged, and furthermore, TTR was not affected by the seasons. Thus, TTR is considered to be a stable and reliable marker of PT-INR control. The study results showed that the mean TTR for the 2-year period was 64% for all patients, varying from 56% to 74% among the institutions. In the ACTIVE W study, 29 different countries participated, and the country mean TTR was shown to vary from 46% to 78%. 12 In the recent RELY study, 44 different countries participated, and the country mean TTR varied from 44% to 77%. 21 The RELY study subanalysis demonstrated that the most important baseline characteristic associated with variability in TTR in the individual patient was the mean TTR of the institution. Although the present results may not necessarily reflect the status of PT-INR control in all Japanese institutions, PT-INR as a whole is considered to be well-controlled or moderately well-controlled in Japan. This study showed that most of the time spent out of TTR was below the range. Importantly, the amount of time below therapeutic range was markedly high (51±24%) in the patient group aged <70 years. This finding reflects a tendency to relatively weak PT-INR control in Japan, especially in the patients aged <70 years, even though the recommended therapeutic range is between 2.0 and 3.0. Because there was little chronological change in PT-INR control during the 2-year period, this weak PT-INR control may be related to the conviction of the attending physicians. A Japan-specific, relatively low therapeutic range ( ) set for patients aged 70 years may have affected the range set for patients aged <70, resulting in an increased proportion of time spent below therapeutic range in the younger patients. Factors Affecting TTR There was no difference in TTR between male and female patients. There was also no difference by co-administration of antiplatelet drugs, although the combined use of warfarin and an antiplatelet drug has been shown to be associated with increased risk of severe bleeding events. 22 Age strongly affected TTR: in patients aged 70 years, TTR was 77±17%, whereas in those aged <70, it was significantly decreased to 46±23%. As clearly shown in Figure 2, most of the patients aged <70 years had a TTR <60%, whereas most of the patients aged 70 had a TTR >60%. It should be emphasized that the difference in TTR by age was observed in all institutions. As discussed before, most of the time spent out of TTR (ie, >50% of the time) was below the range, especially in the patients aged <70 years. It should be noted that many of the patients aged <70 years had a low risk of stroke (CHADS2 score <2), which might have affected the level of PT-INR control maintained by the physicians. However, TTR was also low (48± 22%) in patients aged <70 years and at high risk (CHADS2 score 2). Thus, adherence to the Japanese Guideline seems to be insufficient in many AF patients aged <70 years undergoing warfarin therapy. TTR between 1.6 and 2.6 for patients aged 70 years recommended by the guideline seems to have affected PT-INR control even in patients aged <70 in whom

7 Warfarin Control in Japanese AF Patients 2093 the recommended range is between 2.0 and 3.0. CHADS2 score also significantly affected TTR: in patients with a score 1, TTR was 59±27%, and in those with a score 2, it was significantly increased to 68±23%. In our multivariate analysis, however, CHADS2 score was not an independent predictor of TTR, which may be explained by the fact that the age of the patients with a score 2 was significantly higher than that of the patients with a score 1, and many (72%) of the patients with a score 2 were 70 years old. Warfarin dose significantly affected TTR: in the patient groups with a dose <2.5 mg/day or between 2.5 and 4.9 mg/day, TTR was 72±22% and 63±25%, respectively and in the group with a dose 5.0 mg/day TTR was significantly decreased to 48±24%. Thus, the higher the required dose of warfarin dose, the poorer the PT-INR control. It should be noted that the warfarin dose differed between the patient groups aged <70 and 70 years, and age might have affected the dose administered by the physician. In fact, the mean dose was higher in the younger patients than in the elderly. After multivariate analysis, however, warfarin dose remained an independent predictor of TTR. The dose adjustment needed to obtain optimal PT-INR may be difficult in the patients who require doses 5.0 mg/day. Or increasing the warfarin dose 5.0 mg/day may make physicians anxious about the risk of hemorrhage. Study Limitations We retrospectively analyzed the records of patients taking warfarin for 2 years. A prospective study may be associated with better PT-INR control. However, the data presented here seem to reflect the real life status of PT-INR control in Japan. This study did not include patients with stroke or systemic thromboembolism or major hemorrhage during the study period. The purpose of the study was to examine the status of TTR and identify the influential factors on TTR in AF patients undergoing long-term warfarin therapy, and not the relation of TTR to cardiovascular events, including ischemic and hemorrhagic stroke. A prospective study will be required to show the relation of TTR to the efficacy and safety of warfarin therapy. In this study, multivariate analysis revealed that age and warfarin dose were independent predictors of TTR. Because TTR used for calculating TTR was different between the young (<70 years) and old patient groups ( 70 years), this age-specific protocol of PT-INR control might have affected TTR primarily rather than age itself. Clinical Implications This study revealed the influential factors on TTR in Japanese patients with AF undergoing long-term warfarin therapy. Age and warfarin dose were found to be independent predictors of TTR. PT-INR control was poor in the patients <70 years and taking warfarin 5.0 mg/day. Because this remained unchanged during the 2-year observation period, continuance of appropriate warfarin therapy may be difficult, at least in patients with these clinical characteristics in Japan. The new oral anticoagulant, dabigatran, could be considered in such patients. 5,6 Finally, PT-INR control was excellent in the present patients aged 70 years, largely because TTR is set at a low level compared with the range for patients <70. The efficacy and safety of adjusting the warfarin dose to keep the PT-INR within this low range in patients aged 70 years are now being investigated by the J-RHYHTM registry. 23 A proposal for selecting either warfarin or dabigatran in this high-risk patient group will be made after analysis of the results of J-RHYHTM. Disclosure The authors declare no financial conflicts of interest. References 1. Hart RG, Benavente O, McBride R, Pearce LA. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: A metaanalysis. Ann Intern Med 1999; 131: Fuster V, Rydén LE, Cannom DS, Crijns HJ, Curtis AB, Ellenbogen KA, et al. ACC/AHA/ESC 2006 Guidelines for the Management of Patients with Atrial Fibrillation: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines. Circulation 2006; 114: e257 e Camm AJ, Kirchhof P, Lip GY, Schotten U, Savelieva I, Ernst S, et al. Guidelines for the management of atrial fibrillation: The Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Europace 2010; 12: JCS Joint Working Group. Guidelines for pharmacotherapy of atrial fibrillation (JCS 2008): Digest version. Circ J 2010; 74: Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J, Parekh A, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 2009; 361: Hori M, Connolly SJ, Ezekowitz, MD, Reilly PA, Yusuf S, Wallentin L, et al. Efficacy and safety of dabigatran vs. warfarin in patients with atrial fibrillation: Sub-analysis in Japanese population in RE-LY trial. Circ J 2011; 75: Wann LS, Curtis AB, Ellenbogen KA, Estes NAM 3rd, Ezekowitz MD, Jackman WM, et al ACCF/AHA/HRS focused update on the management of patients with atrial fibrillation (update on dabigatran): A report of the American College of Cardiology Foundation/ American Heart Association Task Force on Practice Guidelines. Circulation 2011; 123: Hylek EM, Go AS, Chang Y, Jensvold NG, Henault LE, Selby JV, et al. Effect of intensity of oral anticoagulation on stroke severity and mortality in atrial fibrillation. N Engl J Med 2003; 349: van Walraven C, Jennings A, Oake N, Fergusson D, Forster AJ. Effect of study setting on anticoagulation control: A systematic review and metaregression. Chest 2006; 129: Rosendaal FR, Cannegieter SC, van der Meer FJ, Briët E. A method to determine the optimal intensity of oral anticoagulant therapy. Thromb Haemost 1993; 69: White HD, Gruber M, Feyzi J, Kaatz S, Tse HF, Husted S, et al. Comparison of outcomes among patients randomized to warfarin therapy according to anticoagulant control: Results from SPORTIF III and V. Arch Intern Med 2007; 167: Connolly SJ, Pogue J, Eikelboom J, Flaker G, Commerford P, Franzosi MG, et al. Benefit of oral anticoagulant over antiplatelet therapy in atrial fibrillation depends on the quality of international normalized ratio control achieved by centers and countries as measured by time in therapeutic range. Circulation 2008; 118: Froom P, Miron E, Barak M. Oral anticoagulants in the elderly. Br J Haematol 2003; 120: Wells PS, Holbrook AM, Crowther NR, Hirsh J. Interactions of warfarin with drugs and food. Ann Intern Med 1994; 121: Wittkowsky AK. Effective anticoagulation therapy: Defining the gap between clinical studies and clinical practice. Am J Manag Care 2004; 10: S297 S Friberg L, Hammar N, Ringh M, Pettersson H, Rosenqvist M. Stroke prophylaxis in atrial fibrillation: Who gets it and who does not? Report from the Stockholm Cohort-study on Atrial Fibrillation (SCAF-study). Eur Heart J 2006; 27: Gattellari M, Worthington J, Zwar N, Middleton S. Barriers to the use of anticoagulation for nonvalvular atrial fibrillation: A representative survey of Australian family physicians. Stroke 2008; 39: Gage BF, Waterman AD, Shannon W, Boechler M, Rich MW, Radford MJ. Validation of clinical classification schemes for predicting stroke: Results from the National Registry of Atrial Fibrillation. JAMA 2001; 285: Inoue H. Antithrombotic therapy: ESC guidelines vs JCS guidelines. Circ J 2010; 74: Masaki N, Suzuki M, Matsumura A, Maruyama Y, Hashimoto Y. Quality of warfarin control affects the incidence of stroke in elderly patients with atrial fibrillation. Intern Med 2010; 49:

8 2094 OKUMURA K et al. 21. Wallentin L, Yusuf S, Ezekowitz MD, Alings M, Flather M, Franzosi MG, 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: Toyoda K, Yasaka M, Iwade K, Nagata K, Koretsune Y, Sakamoto T, et al. Dual antithrombotic therapy increases severe bleeding events in patients with stroke and cardiovascular disease: A prospective, multicenter, observational study. Stroke 2008; 39: Atarashi H, Inoue H, Okumura K, Yamashita T, Origasa H, and for the J-RHYTHM Registry Investigators. Investigation of optimal anticoagulation strategy for stroke prevention in Japanese patients with atrial fibrillation: The J-RHYTHM Registry study design. J Cardiol 2011; 57: