Validation of Risk Scoring System Excluding Female Sex From CHA₂DS₂-VASc in Japanese Patients With Nonvalvular Atrial Fibrillation　– Subanalysis of the J-RHYTHM Registry –

Abstract

Background: Because the current Japanese guideline recommends CHADS₂ score-based risk stratification in nonvalvular atrial fibrillation (NVAF) patients and does not list female sex as a risk for thromboembolic events, we designed the present study to compare the CHA₂DS₂-VASc and CHA₂DS₂-VA scores in the J-RHYTHM Registry.

Methods and Results: We prospectively assessed the incidence of thromboembolic events for 2 years in 997 NVAF patients without warfarin treatment (age 68±12 years, 294 females). The predictive value of the CHA₂DS₂-VASc and CHA₂DS₂-VA scores for thromboembolic events was evaluated by c-statistic difference and net reclassification improvement (NRI). Thromboembolic events occurred in 7/294 females (1.2%/year) and 23/703 males (1.6%/year) (odds ratio 0.72 for female to male, 95% confidence interval (CI) 0.28–1.62, P=0.44). No sex difference was found in patient groups stratified by CHA₂DS₂-VASc and CHA₂DS₂-VA scores. There were significant c-statistic difference (0.029, Z=2.3, P=0.02) and NRI (0.11, 95% CI 0.01–0.20, P=0.02), with the CHA₂DS₂-VA score being superior to the CHA₂DS₂-VASc score. In patients with CHA₂DS₂-VASc scores 0 and 1 (n=374), there were markedly significant c-statistic difference (0.053, Z=6.6, P<0.0001) and NRI (0.11, 95% CI 0.07–0.14, P<0.0001), again supporting superiority of CHA₂DS₂-VA to CHA₂DS₂-VASc score.

Conclusions: In Japanese NVAF patients, the CHA₂DS₂-VA score, a risk scoring system excluding female sex from CHA₂DS₂-VASc, may be more useful in risk stratification for thromboembolic events than CHA₂DS₂-VASc score, especially in identifying truly low-risk patients.

Numerous studies have shown that female sex is a risk for thromboembolism in Caucasian patients with nonvalvular atrial fibrillation (NVAF).^1–6 Accordingly, recent European and American guidelines recommend the CHA₂DS₂-VASc score for thromboembolic risk stratification, which includes female sex (Sc) as a risk factor.^7,8 However, literature review suggests that female sex as a risk factor is still controversial,^9,10 because some studies demonstrated that the odds ratio (OR) or hazard ratio (HR) of female to male sex for thromboembolism is not significant.^11–15 Indeed, the recent Canadian guideline states that “female sex should not factor into antithrombotic therapy selection because of the low associated stroke risk and the nonsignificant HR compared with men” and does not use female sex as a risk factor.¹⁶ There is also an exception to the use of CHA₂DS₂-VASc score in the European and American guidelines, whereby female patients aged <65 years without other risk factors (still CHA₂DS₂-VASc score 1) do not need anticoagulation therapy.^7,8

Editorial p ????

In the Asian population, there have been few studies showing female sex as a risk for thromboembolism.^17–21 Our recent subanalysis of the J-RHYTHM Registry also demonstrated that female sex was not a risk for thromboembolism in the Japanese cohort.²² In fact, the recent revised Japanese guideline has excluded female sex as a risk factor.²³ However, a risk score excluding female sex from the CHA₂DS₂-VASc score has never been evaluated. The present study aimed to validate the risk scoring system excluding female sex from the CHA₂DS₂-VASc score (ie, the “CHA₂DS₂-VA score”) in Japanese NVAF patients without warfarin treatment enrolled in the J-RHYTHM Registry.

Methods

Study Subjects

The J-RHYTHM Registry is an observational cohort study that aims to determine the current status of anticoagulation therapy with warfarin and the optimal anticoagulation levels for preventing thromboembolic events in Japanese AF patients (mean age 69.7 years, n=7,937).^22,24–30 The details of the study design and baseline characteristics of this registry have been reported elsewhere.^24,25 Briefly, of the 7,937 AF patients enrolled, 421 had mitral stenosis or had undergone mechanical valve replacement, and the remaining 7,516 NVAF patients with and without warfarin treatment (n=6,514 and 1,002, respectively) were prospectively followed up for 2 years or until an endpoint occurred. The endpoints consisted of thromboembolic events, including symptomatic cerebral infarction, transient ischemic attack (TIA), and systemic embolism, major bleeding requiring hospitalization, and death. The diagnostic criteria for cerebral infarction and TIA have been reported previously.^24,25 As 110 patients were lost during follow-up and there were no available data on stroke risk stratification in 22 patients, the remaining 7,384 patients (98.2%) were originally considered for analysis.²⁸ In the present analyses, we focused on the incidences of thromboembolic events for a 2-year period in 997 patients (997/1,002=99.5%) without warfarin treatment enrolled in the J-RHYTHM Registry after excluding 5 patients who were lost to follow-up (2 male and 3 female patients). The study is registered at UMIN Clinical Trials Registry (UMIN000001569).

Thromboembolic Risk Stratification

The CHADS₂ score was determined in each patient at enrollment, as previously described.³¹ The CHA₂DS₂-VASc score was also determined according to the 2012 ESC Guidelines for the management of AF,⁷ in which V was replaced by coronary artery disease, because the J-RHYTHM Registry stratified thromboembolic risk by CHADS₂ score and therefore no data were available regarding peripheral artery disease and aortic plaque. The risk score excluding female sex (Sc) from the CHA₂DS₂-VASc score was defined as the CHA₂DS₂-VA score. Because these risk scores were modified from the original ones, they were defined as the modified CHA₂DS₂-VASc (mCHA₂DS₂-VASc) score and the modified CHA₂DS₂-VA (mCHA₂DS₂-VA) score, respectively (Table 1).

Table 1. Clinical Characteristics Included in the mCHA₂DS₂-VASc Score and mCHA₂DS₂-VA Score

Acronym	Score
mCHA2DS2-VASc score
Congestive heart failure/left ventricular dysfunction	1
Hypertension	1
Age (≥75 years)	2
Diabetes mellitus	1
Stroke/TIA	2
Vascular disease (coronary artery disease)	1
Age (65–74 years)	1
Sexcategory (female)	1
mCHA2DS2-VA score
Congestive heart failure/left ventricular dysfunction	1
Hypertension	1
Age (≥75 years)	2
Diabetes mellitus	1
Stroke/TIA	2
Vascular disease (coronary artery disease)	1
Age (65–74 years)	1

mCHA₂DS₂-VA, modified CHA₂DS₂-VA; mCHA₂DS₂-VASc, modified CHA₂DS₂-VASc; TIA, transient ischemic attack.

Statistical Analysis

Data are expressed as mean±standard deviation or n (%). The unpaired t-test or chi-square test was used to compare differences between 2 groups. The CHADS₂, mCHA₂DS₂-VASc, and mCHA₂DS₂-VA scores were assessed as a categorical variable (0, 1, 2, and ≥3). The OR and 95% confidence interval (CI) for the annual incidence of thromboembolic events were evaluated. The predictive value of the CHADS₂, mCHA₂DS₂-VASc, and mCHA₂DS₂-VA scores for thromboembolic events was compared by c-statistic difference and the net reclassification improvement (NRI). Statistical analyses, including the c-statistic analysis, were performed using JMP 10 software (SAS, Cary, NC, USA). The NRI was calculated using R software. A P-value <0.05 was considered statistically significant.

Results

Distribution of the Patients Stratified by Each Scoring System

As shown in Figure 1A, 37% of the study patients were classified into CHADS₂ score 1, followed by 29% into score 0 and 21% into score 2. Therefore, 66% of the study patients were classified into CHADS₂ scores 0 and 1. On the other hand, only 37% of the patients were classified into mCHA₂DS₂-VASc scores 0 and 1, and 43% into mCHA₂DS₂-VA scores 0 and 1 (Figures 1B,C). The difference in the distribution pattern indicates that the mCHA₂DS₂-VASc and mCHA₂DS₂-VA scores would be more appropriate for identifying truly low-risk patients.

Figure 1.

Distribution of NVAF patients stratified by CHADS₂ (A), mCHA₂DS₂-VASc (B), and mCHA₂DS₂-VA (C) scores. The number (%) is shown on each bar. Notably, 66% of the study patients were classified into CHADS₂ scores of 0 and 1 (A), while 37% and 43% were classified into mCHA₂DS₂-VASc scores of 0 and 1 (B) and into mCHA₂DS₂-VA scores of 0 and 1 (C), respectively. mCHA₂DS₂-VA, modified CHA₂DS₂-VA; mCHA₂DS₂-VASc, modified CHA₂DS₂-VASc; NVAF, nonvalvular atrial fibrillation.

Clinical Profiles of the Study Patients

Sex differences in each of the clinical profiles of the study patients are briefly summarized in Table 2. Mean age was significantly higher in female than in male patients. There was no sex difference in the type of AF. The CHADS₂ and mCHA₂DS₂-VASc scores were both significantly higher in female patients. Although a 1-point decrease was found in the mCHA₂DS₂-VA score in female patients compared with the mCHA₂DS₂-VASc score by definition, the mCHA₂DS₂-VA score was still significantly higher in female than in male patients. Antiplatelet use was significantly higher in male than in female patients. Treatment with warfarin was initiated in 234 (23%) patients during the follow-up period and there was no sex difference in the proportion of patients with warfarin initiation.

Table 2. Sex Differences in the Clinical Profiles of the Study Patients With NVAF

	Total (n=997)	Male (n=703)	Female (n=294)	P value
Age	68±12	66±12	71±10	<0.0001
Type of AF				0.18
Paroxysmal (%)	603 (60%)	414 (59%)	189 (64%)
Persistent (%)	108 (11%)	83 (12%)	25 (9%)
Permanent (%)	286 (29%)	206 (29%)	80 (27%)
CHADS2 score	1.3±1.2	1.2±1.2	1.4±1.2	0.007
mCHA2DS2-VASc score	2.3±1.6	1.9±1.5	3.3±1.5	<0.0001
mCHA2DS2-VA score	2.0±1.5	1.9±1.5	2.3±1.5	0.0005
History of stroke/TIA	67 (7%)	46 (7%)	21 (7%)	0.73
Antiplatelet use	577 (58%)	424 (60%)	153 (52%)	0.016
Warfarin initiation	234 (23%)	168 (24%)	66 (22%)	0.62

Data are shown as mean±standard deviation or n (%). NVAF, nonvalvular atrial fibrillation. Other abbreviations as in Table 1.

Annual Incidence of Thromboembolic Events

Thromboembolic events occurred in 23 of the 703 male (1.6%/year) and 7 of the 294 female patients (1.2%/year), and no sex difference was found (OR 0.72 for female to male, 95% CI 0.28–1.62, P=0.44) (Figure 2). Even when the patients were divided into 2 groups at the age of 75 years, there was no sex difference in the incidence of thromboembolism in either group (<75 and ≥75 years). Also, there was a significant age effect on the incidence of thromboembolism (OR 2.87 for patients aged ≥75 years, 95% CI 1.18–8.20, P=0.02), but no interaction between age and sex (P=0.55).

Figure 2.

Odds ratio of female to male patients in the incidence of thromboembolism among those with nonvalvular atrial fibrillation. Analyses were performed in all study patients and in those divided at the age of 75 years.

Thromboembolic events were analyzed using each risk scoring system. As shown in Figure 3A, there was a significant increase in the annual incidence of thromboembolic events with increasing CHADS₂ score (OR 1.65 by 1-category increase, 1.16–2.37, P=0.006). No sex difference was found in this scoring system (OR 0.65 for female to male, 95% CI 0.25–1.46, P=0.31) (Figure 3B). Furthermore, there was a significant increase in thromboembolic events with increasing mCHA₂DS₂-VASc score (OR 1.43 by 1-category increase, 1.00–2.15, P=0.04) (Figure 4A) and mCHA₂DS₂-VA score (OR 1.56 by 1-category increase, 1.09–2.31, P=0.01) (Figure 4B). No sex differences were found in both scoring systems (OR 0.51 for female to male in mCHA₂DS₂-VASc score, 95% CI 0.20–1.20, P=0.13; OR 0.63 in mCHA₂DS₂-VA score, 95% CI 0.25–1.42, P=0.28) (Figures 4C,D). No thromboembolic events occurred in female patients with mCHA₂DS₂-VASc score 1 or mCHA₂DS₂-VA score 0 (n=39). Also, there were no thromboembolic events in female patients aged <65 years with mCHA₂DS₂-VASc score 2 or mCHA₂DS₂-VA score 1 (n=23).

Figure 3.

Thromboembolic events stratified by CHADS₂ score in all nonvalvular atrial fibrillation patients (A) and in male and female patients separately (B). The thromboembolic event rate (%/year) in each category is shown on each bar. CI, confidence interval; OR, odds ratio.

Figure 4.

Thromboembolic events stratified by mCHA₂DS₂-VASc score in all nonvalvular atrial fibrillation patients (A) and in male and female patients separately (C). Similarly, thromboembolic events stratified by mCHA₂DS₂-VA score in all patients (B) and in male and female patients separately (D). The thromboembolic event rate (%/year) in each category is shown on each bar. CI, confidence interval; mCHA₂DS₂-VA, modified CHA₂DS₂-VA; mCHA₂DS₂-VASc, modified CHA₂DS₂-VASc; OR, odds ratio.

Comparison of the Risk Scoring Systems

The predictive value of the CHADS₂, mCHA₂DS₂-VASc, and mCHA₂DS₂-VA scores for thromboembolic events was compared by c-statistic and the NRI in all patients (n=997). The c-statistic in each risk scoring system was 0.638 (95% CI, 0.534–0.730), 0.595 (95% CI, 0.504–0.680), and 0.624 (95% CI, 0.531–0.709), respectively (Table 3A). Importantly, there was a significant difference in the c-statistic (0.029, Z=2.3, P=0.02) and a significant NRI (0.11, 95% CI 0.01–0.20, P=0.02) between mCHA₂DS₂-VASc and mCHA₂DS₂-VA scores, whereas no differences were found in the other combinations of risk scores (Table 3B). To determine the usefulness of the mCHA₂DS₂-VASc and mCHA₂DS₂-VA scores for identifying truly low-risk patients, patients with CHADS₂ scores 0 and 1 (n=653) were analyzed. The c-statistic in the CHADS₂, mCHA₂DS₂-VASc, and mCHA₂DS₂-VA scores was 0.577 (95% CI, 0.451–0.694), 0.632 (95% CI, 0.479–0.763), and 0.631 (95% CI, 0.491–0.751), respectively (Table S1). Although no significant difference in the c-statistic and no significant NRI were found among these risk scoring systems, c-statistic values of both the mCHA₂DS₂-VASc and mCHA₂DS₂-VA score showed a trend to be higher than that of the CHADS₂ score. When only patients with mCHA₂DS₂-VASc scores 0 and 1 (n=374) were analyzed, there were a markedly significant difference in the c-statistic (0.053, Z=6.6, P<0.0001) and a significant NRI (0.11, 95% CI 0.07–0.14, P<0.0001) between the mCHA₂DS₂-VASc and mCHA₂DS₂-VA scores (Table 4B).

Table 3. Comparison of Risk Scores for Predicting Thromboembolism in All NVAF Patients (n=997)

A
Risk score	c-statistic		SE		95% CI		P value
CHADS2	0.638		0.050		0.534–0.730		0.006
mCHA2DS2-VASc	0.595		0.045		0.504–0.680		0.047
mCHA2DS2-VA	0.624		0.046		0.531–0.709		0.013
B
Score difference	c-statistic analysis				NRI analysis
	c-statistic difference	Z		P value	NRI	95% CI		P value
CHADS2 vs. mCHA2DS2-VASc	0.043	1.514		0.13	−0.10	−0.28 to 0.08		0.29
CHADS2 vs. mCHA2DS2-VA	0.014	0.583		0.56	−0.03	−0.21 to 0.16		0.78
mCHA2DS2-VASc vs. mCHA2DS2-VA	0.029	2.321		0.02	0.11	0.01–0.20		0.02

Each risk score was assessed as a categorical variable (0, 1, 2, and ≥3). (A) C-statistic for predicting thromboembolism in each score. (B) Comparison of risk scores for predicting thromboembolism using the c-statistic and the NRI. CI, confidence interval; NRI, net reclassification improvement; SE, standard error. Other abbreviations as in Tables 1,2.

Table 4. Comparison of Risk Scores for Predicting Thromboembolism in NVAF Patients With mCHA₂DS₂-VASc Scores of 0 and 1 (n=374)

A
Risk score	c-statistic		SE		95% CI		P-value
mCHA2DS2-VASc	0.522		0.106		0.322–0.716		0.82
mCHA2DS2-VA	0.575		0.106		0.367–0.760		0.46
B
Score difference	c-statistic analysis				NRI analysis
	c-statistic difference	Z		P value	NRI	95% CI		P value
mCHA2DS2-VASc vs. mCHA2DS2-VA	0.053	6.600		<0.0001	0.11	0.07–0.14		<0.0001

(A) C-statistic for predicting thromboembolism in each score. (B) Comparison of risk scores for predicting thromboembolism using the c-statistic and the NRI. Abbreviations as in Tables 1–3.

Discussion

Major Findings

The present subanalysis of the J-RHYTHM Registry aimed to validate the mCHA₂DS₂-VA score, a risk scoring system excluding female sex from the mCHA₂DS₂-VASc score, in Japanese NVAF patients without warfarin treatment. We found that there was no sex difference in thromboembolic events observed in patients stratified by all 3 risk scores. Importantly, a significant difference in the c-statistic and a significant NRI were found between the mCHA₂DS₂-VASc and mCHA₂DS₂-VA scores in all study patients. Furthermore, these statistical differences were prominent in patients with mCHA₂DS₂-VASc scores 0 and 1. These findings indicate that the mCHA₂DS₂-VA score may be useful in the risk stratification for thromboembolic events in Japanese NVAF population, especially for identifying truly low-risk patients.

Female Sex Is Not a Risk Factor for Thromboembolism in Japanese NVAF Patients

The ATRIA and Framingham studies in the American population showed female sex as a risk factor for thromboembolism.^1,2 Previous cohort studies in Denmark, Sweden, and Canada also showed that female sex is a risk factor for thromboembolism.^4–6 However, only female patients aged ≥75 years had a higher risk for thromboembolism than age-matched male patients, and those aged <75 years showed no gender difference in those 3 cohorts. The recent Danish Diet, Cancer and Health study also supports no sex difference in a relatively young population.³² These findings suggest that there is a significant interaction between age and sex for thromboembolism; that is, the older age, the higher risk in female patients. However, no sex difference and no such an interaction were found in the present Japanese cohort, although there was a significant age effect on the incidence of thromboembolism.

In contrast, Asian cohort studies in NVAF patients without any antithrombotic therapy in China, Hong Kong, and Taiwan showed that female sex is not a risk factor for thromboembolism (HR 1.92 for female to male in China (n=885), 95% CI 0.66–5.58, P=0.23; HR 1.03 in Hong Kong (n=3882), 95% CI 0.88–1.20, P=0.72; OR 0.94 in Taiwan (n=7920), 95% CI 0.79–1.13, P=0.51).^19–21 Our previous analysis of the combined patients with and without warfarin treatment in the J-RHYTHM Registry (n=7,406) also showed no sex difference after adjustment for warfarin use and other covariates.²² Another small Japanese cohort study further demonstrated that male sex is a higher risk for thromboembolism rather than female.¹⁸ All these cohort studies suggest that there may be a racial difference in the incidence of thromboembolism between Asian and Caucasian NVAF patients. Namely, female sex is a risk in patients aged ≥75 years in the Caucasian population, whereas there is no risk at any age in the female Asian population. Although possible mechanisms for higher risk of female sex, including hormone therapy and menopause, are discussed,³³ there are no studies available with a racial difference in the incidence of thromboembolism. The latest paper by the same research group as for the cohort study in Hong Kong showed female sex as a risk in high-risk patients only (CHA₂DS₂-VASc score ≥2) by multivariate analysis after adjustment for antithrombotic therapy and other covariates (HR 1.16 (n=8,754), 95% CI 1.05–1.29, P=0.003),^21,34 but more clinical evidence regarding this important issue is needed.

Validation of CHA₂DS₂-VA Score in Japanese NVAF Patients

We previously demonstrated that the mCHA₂DS₂-VASc score is useful in risk stratification for thromboembolism in Japanese NVAF patients with and without warfarin treatment.²⁸ However, in the present analysis, more appropriate stepwise thromboembolic event rate per incremental score and higher OR were found in patients stratified by the mCHA₂DS₂-VA score (Figure 4B) compared with those stratified by mCHA₂DS₂-VASc score (Figure 4A). Furthermore, detailed analyses showed that there was a significant difference in the c-statistic and a significant NRI between the mCHA₂DS₂-VASc and mCHA₂DS₂-VA scores in all study patients. Although we previously validated the use of the mCHA₂DS₂-VASc score in Japanese NVAF patients,²⁸ the present analyses further validate that the mCHA₂DS₂-VA score may be more useful in risk stratification for thromboembolism in Japanese NVAF patients. Notably, the c-statistic in the CHADS₂ score showed the highest value among the 3 risk scoring systems in all patients (Table 3A). This finding suggests the usefulness of the CHADS₂ score for identifying patients at risk of thromboembolism, which further supports the current Japanese guideline using a CHADS₂ score-based risk stratification system.²³

The CHA₂DS₂-VASc score was originally recognized as a risk scoring system to identify truly low-risk patients for thromboembolism. Indeed, the c-statistic values of both the mCHA₂DS₂-VASc and mCHA₂DS₂-VA score showed a trend to be higher than that of the CHADS₂ score in patients with CHADS₂ scores 0 and 1 (Table S1). However, the thromboembolism event rate in patients with CHADS₂ score 0 and in those with mCHA₂DS₂-VASc score 0 was the same (0.70%/year and 0.71%/year) in the present study, although the patient population classified into mCHA₂DS₂-VASc scores 0 and 1 was smaller than that into CHADS₂ scores 0 and 1 (37% vs. 66%, Figures 1A,B). This indicates that the CHA₂DS₂-VASc score may not be a risk scoring system for identifying truly low-risk patients. Our scoring system excluding female sex as a risk factor from the mCHA₂DS₂-VASc score (ie, mCHA₂DS₂-VA score) showed that thromboembolic events occurred in 2/180 patients (0.56%/year) with mCHA₂DS₂-VA score 0 (Figure 4B), which was 21% lower than that in patients with mCHA₂DS₂-VASc score 0. We also found that there was a highly significant c-statistic difference and a highly significant NRI between the mCHA₂DS₂-VASc and mCHA₂DS₂-VA scores in patients with mCHA₂DS₂-VASc scores 0 and 1. Furthermore, there were no thromboembolic events in female patients aged <65 years without other risk factors (ie, mCHA₂DS₂-VASc score 1) in the present study. All these findings partly support the European and American guidelines for the management of AF in female patients with CHA₂DS₂-VASc score 1, who do not need to have an anticoagulation therapy, and indicate that in the Japanese population the CHA₂DS₂-VA score may be more useful in identifying truly low-risk patients for thromboembolism rather than the CHA₂DS₂-VASc score.

Study Limitations

First, the number of enrolled patients without warfarin treatment was relatively small and they were at relatively low risk of thromboembolism, because of lower mean CHADS₂ score and younger age, although the patients were prospectively followed up for 2 years. The small number of lost patients (n=5, 0.5%) during follow-up may minimize the effect on the overall results of the study. Also, more detailed analyses using the pooled data of 3 Japanese AF cohorts regarding a low incidence of thromboembolic events have been recently reported.³⁵ Second, clinical characteristics comprising the original CHA₂DS₂-VASc score could not be obtained from the database at enrollment. Therefore, we used a modified CHA₂DS₂-VASc score instead of the original score. Third, only information regarding warfarin treatment at baseline was used in the analyses, and therefore the effects of warfarin initiation observed in 23% of the study patients were not considered. Although there was no sex difference in the warfarin-initiated patients, the incidence of thromboembolic events might be underestimated. Finally, the diagnosis of the endpoints was determined by participating physicians, and not validated by an independent committee.

Conclusions

In Japanese NVAF patients, female sex is not a risk factor for thromboembolism. A modified CHA₂DS₂-VA score, a risk scoring system excluding female sex from a modified CHA₂DS₂-VASc score, may be more useful in risk stratification for thromboembolic events than the modified CHA₂DS₂-VASc score, especially in identifying truly low-risk patients.

Acknowledgments

This study was planned by the Japanese Society of Electrocardiology and supported by a grant from the Japan Heart Foundation.

Disclosures

K.O. reports receiving research fund from Boehringer Ingelheim and Daiichi-Sankyo, and remuneration from Boehringer Ingelheim, Bayer Healthcare, Daiichi-Sankyo, and Pfizer; H.I., receiving research fund from Boehringer Ingelheim and Daiichi-Sankyo, and remuneration from Daiichi-Sankyo, Bayer Healthcare and Boehringer Ingelheim; H.A., receiving research fund from Daiichi-Sankyo and Boehringer Ingelheim, and lecture fees from Bayer Healthcare and Boehringer Ingelheim; T.Y., receiving research fund from Boehringer Ingelheim and Daiichi-Sankyo, and remuneration from Boehringer Ingelheim, Daiichi-Sankyo, Bayer Healthcare, Pfizer, Bristol-Myers Squibb, and Eisai; H.O., receiving lecture fee from Boehringer Ingelheim, and consultancy fees from Daiichi-Sankyo and Bayer Healthcare; H.T. and E.T. have no disclosures.

Supplementary Files

Supplementary File 1

Table S1. Comparison of risk scores for predicting thromboembolism in NVAF patients with CHADS₂ scores of 0 and 1 (n=653)

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-15-0095

References

• 1.    Wang  TJ,  Massaro  JM,  Levy  D,  Vasan  RS,  Wolf  PA,  D’Agostino  RB, et al. A risk score for predicting stroke or death in individuals with new-onset atrial fibrillation in the community: The Framingham Heart Study. JAMA 2003; 290: 1049–1056.
• 2.    Fang  MC,  Singer  DE,  Chang  Y,  Hylek  EM,  Henault  LE,  Jensvold  NG, et al. Gender differences in the risk of ischemic stroke and peripheral embolism in atrial fibrillation: The AnTicoagulation and Risk factors In Atrial fibrillation (ATRIA) study. Circulation 2005; 112: 1687–1691.
• 3.    Lip  GY,  Nieuwlaat  R,  Pisters  R,  Lane  DA,  Crijns  HJ. 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.
• 4.    Tsadok  MA,  Jackevicius  CA,  Rahme  E,  Humphries  KH,  Behlouli  H,  Pilote  L. Sex differences in stroke risk among older patients with recently diagnosed atrial fibrillation. JAMA 2012; 307: 1952–1958.
• 5.    Friberg  L,  Benson  L,  Rosenqvist  M,  Lip  GY. Assessment of female sex as a risk factor in atrial fibrillation in Sweden: Nationwide retrospective cohort study. BMJ 2012; 344: e3522, doi:10.1136/bmj.e3522.
• 6.    Mikkelsen  AP,  Lindhardsen  J,  Lip  GY,  Gislason  GH,  Torp-Pedersen  C,  Olesen  JB. Female sex as a risk factor for stroke in atrial fibrillation: A nationwide cohort study. J Thromb Haemost 2012; 10: 1745–1751.
• 7.    Camm  AJ,  Lip  GY,  De Caterina  R,  Savelieva  I,  Atar  D,  Hohnloser  SH, 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. Eur Heart J 2012; 33: 2719–2747.
• 8.    January  CT,  Wann  LS,  Alpert  JS,  Calkins  H,  Cigarroa  JE,  Cleveland  JC, et al. 2014 AHA/ACC/HRS Guideline 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 Heart Rhythm Society. Circulation 2014; 130: e199–e267, doi:10.1161/CIR.0000000000000041.
• 9.    Hughes  M,  Lip  GY. Stroke and thromboembolism in atrial fibrillation: A systematic review of stroke risk factors, risk stratification schema and cost effectiveness data. Thromb Haemost 2008; 99: 295–304.
• 10.    Pisters  R,  Lane  DA,  Marin  F,  Camm  AJ,  Lip  GY. Stroke and thromboembolism in atrial fibrillation. Circ J 2012; 76: 2289–2304.
• 11.   Atrial Fibrillation Investigators. Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation: Analysis of pooled data from five randomized controlled trials. Arch Intern Med 1994; 154: 1449–1457.
• 12.   Atrial Fibrillation Investigators. Echocardiographic predictors of stroke in patients with atrial fibrillation: A prospective study of 1066 patients from 3 clinical trials. Arch Intern Med 1998; 158: 1316–1320.
• 13.   The SPAF III Writing Committee for the Stroke Prevention in Atrial Fibrillation Investigators. Patients with nonvalvular atrial fibrillation at low risk of stroke during treatment with aspirin: Stroke Prevention in Atrial Fibrillation III Study. JAMA 1998; 279: 1273–1277.
• 14.    Ruigomez  A,  Johansson  S,  Wallander  MA,  Edvardsson  N,  Garcia Rodriguez  LA. Risk of cardiovascular and cerebrovascular events after atrial fibrillation diagnosis. Int J Cardiol 2009; 136: 186–192.
• 15.    Van Staa  TP,  Setakis  E,  Di Tanna  GL,  Lane  DA,  Lip  GY. A comparison of risk stratification schemes for stroke in 79,884 atrial fibrillation patients in general practice. Thromb Haemost 2011; 9: 39–48.
• 16.    Verma  A,  Cairns  JA,  Mitchell  LB,  Macle  L,  Stiell  IG,  Gladstone  D, et al. 2014 focused update of the Canadian Cardiovascular Society Guidelines for the management of atrial fibrillation. Can J Cardiol 2014; 30: 1114–1130.
• 17.    Nakagami  H,  Yamamoto  K,  Ikeda  U,  Mitsuhashi  T,  Goto  T,  Shimada  K. Mitral regurgitation reduces the risk of stroke in patients with nonrheumatic atrial fibrillation. Am Heart J 1998; 136: 528–532.
• 18.    Inoue  H,  Atarashi  H. Risk factors for thromboembolism in patients with paroxysmal atrial fibrillation. Am J Cardiol 2000; 86: 852–855.
• 19.    Lin  LY,  Lee  CH,  Yu  CC,  Tsai  CT,  Lai  LP,  Hwang  JJ, et al. Risk factors and incidence of ischemic stroke in Taiwanese with nonvalvular atrial fibrillation: A nation wide database analysis. Atherosclerosis 2011; 217: 292–295.
• 20.    Guo  Y,  Apostolakis  S,  Blann  AD,  Wang  H,  Zhao  X,  Zhang  Y, et al. Validation of contemporary stroke and bleeding risk stratification scores in non-anticoagulated Chinese patients with atrial fibrillation. Int J Cardiol 2013; 168: 904–909.
• 21.    Siu  CW,  Lip  GY,  Lam  KF,  Tse  HF. Risk of stroke and intracranial hemorrhage in 9727 Chinese with atrial fibrillation in Hong Kong. Heart Rhythm 2014; 11: 1401–1408.
• 22.    Inoue  H,  Atarashi  H,  Okumura  K,  Yamashita  T,  Origasa  H,  Kumagai  N, et al. Impact of gender on the prognosis of patients with nonvalvular atrial fibrillation. Am J Cardiol 2014; 113: 957–962.
• 23.   JCS Joint Working Group. Guidelines for pharmacotherapy of atrial fibrillation (JCS 2013): Digest version. Circ J 2014; 78: 1997–2021.
• 24.    Atarashi  H,  Inoue  H,  Okumura  K,  Yamashita  T,  Origasa  H. Investigation of optimal anticoagulation strategy for stroke prevention in Japanese patients with atrial fibrillation: The J-RHYTHM Registry study design. J Cardiol 2011; 57: 95–99.
• 25.    Atarashi  H,  Inoue  H,  Okumura  K,  Yamashita  T,  Kumagai  N,  Origasa  H. Present status of anticoagulation treatment in Japanese patients with atrial fibrillation: A report from the J-RHYTHM Registry. Circ J 2011; 75: 1328–1333.
• 26.   J-RHYTHM Registry Investigators. Determinants of warfarin use and international normalized ratio levels in atrial fibrillation patients in Japan: Subanalysis of the J-RHYTHM Registry. Circ J 2011; 75: 2357–2362.
• 27.    Inoue  H,  Okumura  K,  Atarashi  H,  Yamashita  T,  Origasa  H,  Kumagai  N, et al. Target international normalized ratio values for preventing thromboembolic and hemorrhagic events in Japanese patients with non-valvular atrial fibrillation: Results of the J-RHYTHM Registry. Circ J 2013; 77: 2264–2270.
• 28.    Okumura  K,  Inoue  H,  Atarashi  H,  Yamashita  T,  Tomita  H,  Origasa  H. Validation of CHA₂DS₂-VASc and HAS-BLED scores in Japanese patients with nonvalvular atrial fibrillation: An analysis of the J-RHYTHM Registry. Circ J 2014; 78: 1593–1599.
• 29.    Yamashita  T,  Inoue  H,  Okumura  K,  Atarashi  H,  Origasa  H; on behalf of the J-RHYTHM Registry Investigators. Warfarin anticoagulation intensity in Japanese nonvalvular atrial fibrillation patients: A J-RHYTHM Registry analysis. J Cardiol 2015; 65: 175–177.
• 30.    Inoue  H,  Atarashi  H,  Okumura  K,  Yamashita  T,  Kumagai  N,  Origasa  H. Thromboembolic events in paroxysmal vs. permanent non-valvular atrial fibrillation: Subanalysis of the J-RHYTHM Registry. Circ J 2014; 78: 2388–2393.
• 31.    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: 2864–2870.
• 32.    Overvad  TF,  Rasmussen  LH,  Skjoth  F,  Overvad  K,  Albertsen  IE,  Lane  DA, et al. Female sex as a risk factor for thromboembolism and death in patients with incident atrial fibrillation: The prospective Danish Diet, Cancer and Health study. Thromb Haemost 2014; 112: 789–795.
• 33.    Cove  CL,  Albert  CM,  Andreotti  F,  Badimon  L,  Van Gelder  IC,  Hylek  EM. Female sex as an independent risk factor for stroke in atrial fibrillation: Possible mechanisms. Thromb Haemost 2014; 111: 385–391.
• 34.    Ho  CW,  Ho  MH,  Chan  PH,  Hai  JJ,  Cheung  E,  Yeung  CY, et al. Ischemic stroke and intracranial hemorrhage with aspirin, dabigatran, and warfarin: Impact of quality of anticoagulation control. Stroke 2015; 46: 23–30.
• 35.    Suzuki  S,  Yamashita  T,  Okumura  K,  Atarashi  H,  Akao  M,  Ogawa  H, et al. Incidence of ischemic stroke in Japanese patients with atrial fibrillation not receiving anticoagulation therapy. Circ J 2015; 79: 432–438.