Beneficial Effect of Non-Vitamin K Antagonist Oral Anticoagulants in Patients With Nonvalvular Atrial Fibrillation　– Results of the J-RHYTHM Registry 2 –

Eitaro Kodani; Hirotsugu Atarashi; Hiroshi Inoue; Ken Okumura; Takeshi Yamashita; Hideki Origasa; on behalf of the J-RHYTHM Registry Investigators

doi:10.1253/circj.CJ-16-0066

Abstract

Background: The J-RHYTHM Registry 2 was a multicenter, prospective observational study that extended the follow-up period of the J-RHYTHM Registry in order to investigate long-term outcomes and effects of non-vitamin K antagonist oral anticoagulants (NOACs) in Japanese patients with atrial fibrillation (AF).

Methods and Results: Among 6,616 patients with nonvalvular AF (NVAF) (men 71.0%, 69.7±9.9 years, CHADS₂ score 1.7±1.2), event rates were compared among patients receiving warfarin (n=3,964), NOACs (n=923), and no anticoagulation therapy (No-OAC, n=753) at the end of follow-up, except for 976 patients lacking anticoagulant data. During the 5-year follow-up period, thromboembolism occurred in 196 (4.9%), 19 (2.1%), and 45 (6.0%) patients, respectively; major hemorrhage in 233 (5.9%), 22 (2.4%), and 36 (4.8%); all-cause death in 230 (5.8%), 13 (1.4%), and 105 (13.9%), (P<0.001 for each). After adjusting for the components of the CHA₂DS₂-VASc score and antiplatelet drug use, the odds ratio (OR) in the Warfarin group was significantly lower for all-cause death compared with that in the No-OAC group (OR 0.30, 95% confidence interval [CI] 0.23–0.39, P<0.001), whereas ORs in the NOACs group were significantly lower for all events (OR 0.42, 95% CI 0.24–0.74, P=0.003 for thromboembolism; OR 0.53, 95% CI 0.31–0.93, P=0.027 for major hemorrhage; and OR 0.10, 95% CI 0.06–0.18, P<0.001 for all-cause death, respectively).

Conclusions: NOACs could be beneficial for reducing event rates of all types in Japanese NVAF patients.

Atrial fibrillation (AF) is a potent risk factor for cardiogenic embolism.¹^,² Because oral anticoagulation (OAC) therapy can reduce the risk of AF-related ischemic stroke by 60–70%,³^,⁴ vitamin K antagonists, mainly warfarin, have been used worldwide for more than 50 years. In order to investigate appropriate warfarin-based anticoagulation therapy for prevention of stroke in Japanese patients with AF, we previously conducted the J-RHYTHM Registry, which included more than 7,000 patients with AF.⁵ The status of anticoagulation therapy and the 2-year incidence of events in Japanese AF patients have been reported.⁶^–⁸ We demonstrated that more than 85% of AF patients were taking warfarin⁶ and the controlled international normalized ratio of prothrombin time (PT-INR) was slightly lower than the 2.0–3.0 recommended for management of AF in Western countries.⁹^,¹⁰ PT-INR of 1.6–2.59 may be optimal for preventing thromboembolism without increasing major hemorrhage in Japanese AF patients,⁷ even those with valvular AF,⁸ prior stroke or transient ischemic attack (TIA),¹¹ and in patients aged ≥85 years.¹²

After terminating the J-RHYTHM Registry in 2011, several non-vitamin K antagonist oral anticoagulants (NOACs) such as dabigatran,¹³^,¹⁴ rivaroxaban,¹⁵^,¹⁶ apixaban,¹⁷ and edoxaban¹⁸ have been approved for prevention of ischemic stroke and systemic embolism in patients with nonvalvular AF (NVAF). Although the use of NOACs is increasing worldwide,¹⁹^,²⁰ information on event rates in patients receiving NOACs in Japan is lacking.²¹^,²² Therefore, we extended the follow-up period of the J-RHYTHM Registry to conduct the J-RHYTHM Registry 2 in order to investigate the long-term outcomes of warfarin therapy compared with NOACs in Japanese patients with NVAF.

Methods

J-RHYTHM Registry 2 Study Design

The J-RHYTHM Registry 2 was a multicenter, prospective observational study that extended the J-RHYTHM Registry follow-up period for 3 years. The study design and subjects were taken over from the J-RHYTHM Registry.⁵^,⁶ Briefly, a consecutive series of outpatients with AF of any type was enrolled, regardless of the antithrombotic drugs being used at the time of enrollment. Antithrombotic drugs and dosages were selected at the discretion of the treating cardiologists. Of 7,937 patients enrolled from 158 institutions in the J-RHYTHM Registry, 909 patients in 27 institutions did not give consent to extend the follow-up period; a duplicated case was also withdrawn from the J-RHYTHM Registry. The remaining 7,027 patients were therefore enrolled in the J-RHYTHM Registry 2 (Figure 1).

Figure 1.

Study patient enrollment and exclusion. *Valvular AF includes mitral stenosis and/or mechanical prosthetic valve. AF, atrial fibrillation; OAC, oral anticoagulant; NOAC, non-vitamin K antagonist oral anticoagulant.

Follow-up and Endpoint Definitions

The patients were followed for a total of 5 years, including the 2 years of the J-RHYTHM Registry, or until an endpoint, whichever occurred first. The thromboembolic endpoints included symptomatic ischemic stroke, TIA, and systemic embolism. Major hemorrhage, including intracranial, gastrointestinal, and other hemorrhages requiring hospitalization, was selected as the safety endpoint. All-cause and cardiovascular deaths were also determined. Information on event type and antithrombotic therapy, including anticoagulant and antiplatelet drugs, were obtained for any event that occurred during the follow-up period. The diagnostic criteria for each event have been described elsewhere.⁵^,⁶

Patient Groupings

To clarify the event rates in patients receiving NOACs, the patients were divided into 3 groups according to the final status of anticoagulation therapy at the time of the event or at the end of follow-up: patients taking warfarin (Warfarin group), any NOAC (NOACs group), and no anticoagulant (No-OAC group). Event rates were compared among these 3 groups. In addition, event rates in subgroups based on the final status of antiplatelet therapy in each group were determined. Event rates according to the baseline PT-INR value (<1.6, 1.6–1.99, 2.0–2.59, 2.6–2.99, and ≥3.0) in patients who continued warfarin throughout the follow-period were also evaluated as reported previously.⁷

Statistical Analysis

Data are presented as mean±standard deviation (SD). The statistical significance of differences in the mean values was analyzed using Student’s t-test or analysis of variance (ANOVA), as appropriate. Frequencies of parameters or events were compared using chi-square or Fisher’s exact test, as appropriate. Kaplan-Meier curves for time to events were compared with log-rank tests. A Cox proportional hazard model was used to investigate the influence of warfarin therapy. Odds ratios (ORs) for each event in the Warfarin and NOACs groups were calculated by multivariate logistic regression analysis adjusted for the components of the CHA₂DS₂-VASc score (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, history of ischemic stroke or TIA, vascular disease [coronary artery disease], age 65–74 years, and female sex)²³ and antiplatelet use, using the No-OAC group as a reference. Two-sided P-values <0.05 were considered statistically significant. All statistical analyses were performed with the IBM SPSS Statistics for Windows, version 23.0 (IBM Corp, Armonk, NY, USA).

Results

Of the 7,027 patients with AF who had been enrolled in this extended study, 364 were excluded for valvular AF, including mitral stenosis and/or mechanical prosthetic valves.²⁴ Of the remaining 6,663 patients with NVAF, 47 (0.7%) were lost to follow-up. Therefore, a total of 6,616 patients with NVAF were included in the analyses (Figure 1).

Baseline Characteristics and Antithrombotic Therapy

The baseline characteristics of the patients and their antithrombotic therapy at baseline in the 3 groups are shown in Table 1. Mean age, CHADS₂ score, and prevalence of permanent AF, coronary artery disease, cardiomyopathy, and heart failure were significantly different among the 3 groups, with the lowest CHADS₂ score (1.4±1.1) in the NOACs group (Table 1). The frequency of baseline warfarin therapy was, of course, highest (95.5%) in the Warfarin group. By contrast, antiplatelet use was most frequent at baseline in the No-OAC group, and 332 patients (44.1%) in this group discontinued warfarin because of any reason. In the NOACs group, 776 patients (84.1%) switched from warfarin, while 147(15.9%) newly started anticoagulation therapy with NOAC during the extended follow-up period (Table 1). Consequently, 3,786 patients who were taking warfarin at baseline continued warfarin and 421 patients did not receive any anticoagulation therapy throughout the follow-up period. The characteristics of these 4,207 patients are shown in Table S1.

Table 1. Baseline Characteristics of the AF Patients and Antithrombotic Therapy at the Start of the J-RHYTHM Registry

	Overall	Final status of OAC therapy in the J-RHYTHM Registry 2			P value*
	Overall	No-OAC	Warfarin	NOACs	P value*
No. of patients	6,616	753	3,964	923
Age, years	69.7±9.9	68.3±12.0	70.1±9.4	67.1±8.3	<0.001
Sex, male	4,700 (71.0)	539 (71.6)	2,824 (71.2)	668 (72.4)	0.789
Type of atrial fibrillation
Paroxysmal	2,569 (38.8)	445 (59.1)	1,360 (34.3)	384 (41.6)	<0.001
Persistent	974 (14.7)	90 (12.0)	385 (9.7)	139 (15.1)
Permanent	3,073 (46.4)	218 (29.0)	2,019 (50.9)	400 (43.3)
Comorbidities
Coronary artery disease	697 (10.5)	65 (8.6)	434 (10.9)	72 (7.8)	0.006
Cardiomyopathy	562 (8.5)	39 (5.2)	389 (9.8)	65 (7.0)	<0.001
HCM	234 (3.5)	18 (2.4)	166 (4.2)	24 (2.6)	0.009
DCM	328 (5.0)	21 (2.8)	223 (5.6)	41 (4.4)	0.003
Congenital heart disease	92 (1.4)	6 (1.8)	61 (1.5)	13 (1.4)	0.288
COPD	122 (1.8)	14 (1.9)	71 (1.8)	11 (1.2)	0.420
Hyperthyroidism	118 (1.8)	13 (1.7)	76 (1.9)	14 (1.5)	0.698
Risk factors for stroke
Heart failure	1,800 (27.2)	160 (21.2)	1,194 (30.1)	179 (19.4)	<0.001
Hypertension	3,979 (60.1)	437 (58.0)	2,422 (61.1)	544 (58.9)	0.184
Age (≥75 years)	2,251 (34.0)	247 (32.8)	1,399 (35.3)	174 (18.9)	<0.001
Diabetes mellitus	1,202 (18.2)	118 (15.7)	741 (18.7)	167 (18.1)	0.143
Stroke/TIA	911 (13.8)	88 (11.7)	582 (14.7)	121 (13.1)	0.065
CHADS₂ score
0	1,048 (15.8)	179 (23.8)	532 (13.4)	194 (21.0)	<0.001
1	2,275 (34.4)	260 (34.5)	1,320 (33.3)	367 (39.8)
2	1,820 (27.5)	163 (21.6)	1,175 (29.6)	215 (23.3)
3	929 (14.0)	82 (10.9)	604 (15.2)	95 (10.3)
4	388 (5.9)	46 (6.1)	237 (6.0)	38 (4.1)
5	136 (2.1)	21 (2.8)	86 (2.2)	12 (1.3)
6	20 (0.3)	2 (0.3)	10 (0.3)	2 (0.2)
Mean	1.7±1.2	1.5±1.3	1.7±1.2	1.4±1.1	<0.001
Antithrombotic therapy
Warfarin	5,737 (86.7)	332 (44.1)	3,786 (95.5)	776 (84.1)	<0.001
Antiplatelet	1,710 (25.8)	310 (41.2)	917 (23.1)	252 (27.3)	<0.001
Aspirin	1,499 (22.7)	281 (37.3)	797 (20.1)	223 (24.2)	<0.001
Other	366 (5.5)	53 (7.0)	209 (5.3)	50 (5.4)	0.149
Warfarin+antiplatelet	1,193 (18.0)	73 (9.7)	822 (20.7)	154 (16.7)	<0.001

Data are no. of patients (%) or mean±SD. *Comparison among 3 groups of No-OAC, Warfarin, and NOACs. CHADS₂, congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, and history of stroke or TIA; COPD, chronic obstructive pulmonary disease; DCM, dilated cardiomyopathy; HCM, hypertrophic cardiomyopathy; NOAC, non-vitamin K antagonist oral anticoagulant; OAC, oral anticoagulant; TIA, transient ischemic attack.

Changes in Antithrombotic Therapy During Follow-up

During the 3-year extended follow-up period, NOAC use increased year by year, in parallel with decreasing warfarin use. In the last follow-up year (2014), the frequency of NOAC use was 20% among 4,404 patients. Rivaroxaban was administered most frequently, followed by dabigatran and apixaban (Table 2). No patient was receiving edoxaban at the end of the follow-up period in 2014. Finally, as shown in Table 1 and Table 2, 3,964 (59.9%) of 6,616 patients were taking warfarin (Warfarin group), while 923 (14.0%) received any NOAC at the time of an event or at the end of the follow-up period (NOACs group). In contrast, 753 (11.4%) were not on any anticoagulation therapy (No-OAC group). Unfortunately, information regarding anticoagulation therapy at the time of an event or at the end of follow-up was lacking in 976 patients (Table 2); these patients were therefore excluded from the subsequent analysis.

Table 2. Changes in Antithrombotic Therapy During Extended Follow-up Period

	Baseline	Extended follow-up period			P value*	Final**
	2009	2012	2013	2014	P value*	Final**
No. of patients	6,616	5,442	4,850	4,404		6,616
No-OAC	879 (13.3)	599 (11.0)	487 (10.0)	416 (9.4)	<0.001	753 (11.4)
Any OAC	5,737 (86.7)	4,435 (81.5)	4,038 (83.3)	3,698 (84.0)	<0.001	4,887 (73.9)
Warfarin	5,737 (86.7)	4,103 (75.4)	3,413 (70.4)	2,801 (63.6)	<0.001	3,964 (59.9)
NOAC	0 (0.0)	332 (6.1)	625 (12.9)	897 (20.4)	<0.001	923 (14.0)
Dabigatran	0 (0.0)	295 (5.4)	351 (7.2)	311 (7.1)	<0.001	325 (4.9)
Rivaroxaban	0 (0.0)	20 (0.4)	261 (5.4)	394 (8.9)		403 (6.1)
Apixaban	0 (0.0)	0 (0.0)	6 (0.1)	183 (4.2)		184 (2.8)
Edoxaban	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)		0 (0.0)
Unknown	0 (0.0)	17 (0.3)	7 (0.1)	9 (0.2)		11 (0.2)
Unknown OAC	0 (0.0)	408 (7.5)	325 (6.7)	290 (6.6)	<0.001	976 (14.8)
Antiplatelet	1,710 (25.8)	1,092 (20.1)	892 (18.4)	757 (17.2)	<0.001	1,194 (18.0)
Antiplatelet only	517 (7.8)	263 (4.8)	196 (4.0)	145 (3.3)	<0.001	301 (4.5)
OAC+antiplatelet	1,193 (18.0)	829 (15.2)	696 (14.4)	612 (13.9)	<0.001	893 (13.5)

Data are no. of patients (%). *P for trend among years of 2009, 2012, 2013, and 2014. **At the time of an event or at the end of follow-up. Dabigatran, rivaroxaban, and apixaban have been available since March 14, 2011; April 18, 2012; and February 26, 2013; respectively. Edoxaban can be used in Japan for the prevention of ischemic stroke and systemic embolism in patients with nonvalvular atrial fibrillation from September 26, 2014. Abbreviations as in Table 1.

Event Rates and Final Anticoagulation Therapy

During the 5-year follow-up period (median 1,935 days, 28,933 person·years), thromboembolic events occurred in 271 patients (4.1%), major hemorrhage in 303 (4.6%), and all-cause death in 430 (6.5%), including cardiovascular death in 115 (1.7%) (Table 3). The overall incidence rate of each event was 0.9, 1.0, 1.5, and 0.4/100 person·years, respectively.

Table 3. Incidence of Events

	Overall	Final status of OAC			P value*	NOACs			P value**
	Overall	No-OAC	Warfarin	NOACs	P value*	Dabigatran	Rivaroxaban	Apixaban	P value**
No. of patients	6,616	753	3,964	923		325	403	184
Thromboembolism	271 (4.1)	45 (6.0)	196 (4.9)	19 (2.1)	<0.001	6 (1.8)	10 (2.5)	2 (1.1)	0.519
Cerebral infarction	229	39	163	16		5	9	1
TIA	17	1	15	1		0	1	0
Systemic embolism	25	5	18	2		1	0	1
Major hemorrhage	303 (4.6)	36 (4.8)	233 (5.9)	22 (2.4)	<0.001	7 (2.0)	11 (2.7)	4 (2.2)	0.857
Intracranial	103	10	83	3		2	1	0
Gastrointestinal	95	11	70	10		4	4	2
Other	105	15	80	9		1	6	2
Thromboembolism+major hemorrhage	574 (8.7)	81 (10.8)	429 (10.8)	41 (4.4)	<0.001	13 (4.0)	21 (5.2)	6 (3.3)	0.516
All-cause death	430 (6.5)	105 (13.9)	230 (5.8)	13 (1.4)	<0.001	7 (2.2)	2 (0.5)	1 (0.5)	0.074
Cardiovascular death	115 (1.7)	27 (3.6)	73 (1.8)	3 (0.3)	<0.001	1 (0.3)	1 (0.2)	1 (0.5)	0.842

Data are no. of patients (%). *Comparison among 3 groups of No-OAC, Warfarin, and NOACs, except for 976 patients lacking data of OAC at the time of an event or at the end of follow-up. **Comparison among 3 NOAC groups, except for 11 patients lacking information on NOAC. Abbreviations as in Table 1.

Event rates in the 3 groups according to the final anticoagulation therapy are shown in Table 3. There were significant differences in all events among the groups (P<0.001 for each, Table 3). Notably, all event rates in the NOACs group were lower than those in the No-OAC and Warfarin groups. There were no significant differences in the event rates among the 3 NOAC groups (Table 3). When patients were divided into 2 subgroups according to the final status of antiplatelet therapy, antiplatelet use did not influence any event rate in the NOACs group (Figure 2). Unexpectedly, antiplatelet use was associated with an increase in thromboembolic events in the No-OAC and Warfarin groups, and also in all-cause and cardiovascular mortality rates in the Warfarin group (Figure 2). Multivariate logistic regression analysis revealed that the Warfarin group was at lower risk for all-cause and cardiovascular death as compared with the No-OAC group after adjusting for multiple confounding factors including the components of the CHA₂DS₂-VASc score and antiplatelet use. By contrast, the NOACs group was at lower risk for all events after adjusting for multiple confounders (Table 4).

Figure 2.

Five-year incidence of thromboembolism (A), major hemorrhage (B), all-cause death (C), and cardiovascular death (D) based on the final status of antiplatelet therapy. See text for details. P-values: comparison between patients with and without antiplatelet drugs at the end of the follow-up in each group. OAC, oral anticoagulant; NOAC, non-vitamin K antagonist oral anticoagulant.

Table 4. Adjusted ORs for Events

	Final status of OAC
	No-OAC (reference)	Warfarin		NOACs
	No-OAC (reference)	OR (95% CI)	P value	OR (95% CI)	P value
Thromboembolism	1.00	0.90 (0.64–1.28)	0.553	0.42 (0.24–0.74)	0.003
Major hemorrhage	1.00	1.20 (0.83–1.73)	0.343	0.53 (0.31–0.93)	0.027
Thromboembolism+major hemorrhage	1.00	1.03 (0.79–1.34)	0.816	0.45 (0.30–0.68)	<0.001
All-cause death	1.00	0.30 (0.23–0.39)	<0.001	0.10 (0.06–0.18)	<0.001
Cardiovascular death	1.00	0.44 (0.27–0.70)	0.001	0.12 (0.04–0.40)	0.001

ORs are adjusted for the components of the CHA₂DS₂-VASc score (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, history of stroke or transient ischemic attack, vascular disease [coronary artery disease], age 65–74 years, and female sex) and antiplatelet use. CI, confidence interval; OR, odds ratio. Other abbreviations as in Table 1.

Event Rates and Long-Term Warfarin Therapy

The 5-year incidence rates of thromboembolism and all-cause death were significantly lower in the 3,786 patients who continued warfarin therapy throughout the study compared with the 421 patients who did not receive any anticoagulation therapy (5.0% vs. 7.8%, P=0.021 and 5.8% vs. 10.2%, P<0.001, respectively). The incidence of major hemorrhage was higher in patients on warfarin therapy compared with those in the No-OAC group (5.9% vs. 2.6%, P=0.007) (Table S2). The Kaplan-Meier curves for these patients are shown in Figure 3. Event-free rates of thromboembolism and all-cause death were significantly higher in patients on continuous warfarin therapy than in those without any OAC (P=0.002 and P<0.001, respectively, by log-tank test), whereas the event-free rate of major hemorrhage was lower in patients on continuous warfarin therapy than in those without any OAC (P=0.024 by log-tank test) (Figure 3).

Figure 3.

Kaplan-Meier curves for thromboembolism (A), major hemorrhage (B), all-cause death (C), and cardiovascular death (D) in 3,786 patients on continuous warfarin therapy and in 421 patients without any OAC treatment throughout the follow-up period. Insets show the magnified event-free curves with a 10-fold scale. See text for details. P-values: comparisons between patients with warfarin and without any OAC by log-rank test. OAC, oral anticoagulant.

Rates of thromboembolism decreased (P=0.006 for trend), but those of major hemorrhage increased (P<0.001 for trend) along with an increase in PT-INR values at baseline in the 3,786 patients on continuous warfarin therapy compared with those in patents without any OAC (Figure S1). After adjusting for confounding factors of the components of the CHA₂DS₂-VASc score and antiplatelet use, hazard ratios (HRs) of each PT-INR subgroup for thromboembolism and major hemorrhage showed significant trends. In addition, the HRs for thromboembolism were significantly lower in all PT-INR subgroups, but those for major hemorrhage were significantly higher in the PT-INR ≥2.6 subgroups compared with the reference group not receiving any OAC therapy throughout the study (Figure 4).

Figure 4.

Hazard ratios for thromboembolism (Left) and major hemorrhage (Right) in patients without OACs and in patients with different anticoagulation levels at baseline. Hazard ratios are adjusted for the components of CHA₂DS₂-VASc score and antiplatelet drug use, using 421 patients in the No-OAC group as a reference. See text for details. P-values: comparison among 6 groups (No-OAC and 5 subgroups of PT-INR) for trend. PT-INR, international normalized ratio of prothrombin time; OAC oral anticoagulant.

Discussion

The major findings of the present study were as follows. First, the overall incidence of each event was quite low for the 5-year follow-up period in Japanese patients with NVAF. Second, all event rates in the NOACs group were extremely low compared with those in the No-OAC and Warfarin groups. In addition, NOAC use was an independent factor for reduction of all events. Third, the beneficial effects of warfarin lasted for 5 years when event rates were analyzed in patients who did not change anticoagulation therapy throughout the follow-up period (continuous Warfarin vs. continuous No-OAC). Long-term warfarin therapy was effective at preventing thromboembolism, but was associated with an increased incidence of major hemorrhage. Fourth, the incidence of thromboembolism decreased, but that of major hemorrhage increased along with an increase in baseline PT-INR values compared with those for the No-OAC patients.

Effectiveness of NOACs

In the present study, all event rates in the NOACs group were extremely low compared with those in the Warfarin group. After their approval for clinical use, several real-world reports on NOACs have been published. In a recent report from Hong Kong,²⁵ the incidence of stroke and intracranial hemorrhage was 2.24 and 0.32/100 person·years, respectively, in 8,754 Chinese patients administered dabigatran. Another study from the United States, using Medicare data from 134,414 patients on dabigatran,²⁶ reported the incidence of ischemic stroke and major hemorrhage as 1.13 and 4.27/100 person·years, respectively. In the Xarelto^® for Prevention of Stroke in Patients with Atrial Fibrillation (XANTUS) study,²⁷ the rates of these events in 6,784 patients, mostly European, on rivaroxaban therapy were 0.7 and 2.1/100 person·years, respectively. In other recent reports on rivaroxaban from the Dresden NOAC registry of 1,776 patients²⁸ and real-world outcomes in the USA of 27,467 patients,²⁹ the incidence of major hemorrhage was 3.1 and 2.86/100 person·years, respectively. In contrast, in the present study the overall rate of thromboembolism and major hemorrhage was 0.9 and 1.0/100 person·years, respectively, and those in the NOACs group were further lower. In addition, the incidence of major hemorrhage was rather higher in the No-OAC group than in the NOACs group (Table 3). Antiplatelet drug use was not the reason, because antiplatelet drugs did not increase the incidence of major hemorrhage in the No-OAC group in contrast to that of thromboembolism (Figures 2A,B).³⁰ It was difficult to rule out the possibility that the differences in potential bleeding risk, observation period, and unmeasured confounding factors between the No-OAC and NOACs groups could have influenced the higher incidence of major hemorrhage in the No-OAC group, despite the OR for major hemorrhage being significantly lower in the NOACs group after adjusting for known confounding factors (Table 4). By contrast, the incidence of thromboembolism was higher in patients with than in those without taking antiplatelet drugs in the No-OAC and Warfarin groups (Figure 2A). All-cause and cardiovascular mortality rates were also higher in patients taking antiplatelet drugs in the Warfarin group (Figures 2C,D). We do not have a plausible explanation for these unexpected findings. A possible explanation is that antiplatelet drugs were generally administered to patients who had any comorbidities such as a history of stroke/TIA, coronary artery disease, and/or other atherosclerotic diseases, thereby resulting in increased event rates in patients receiving antiplatelet drugs. Although antiplatelet drug use and other unknown confounding factors might have affected the incidence of events as mentioned before, the ORs for thromboembolism and all-cause and cardiovascular death in the NOACs group were significantly lower even after adjusting for multiple confounding factors, including antiplatelet use (Table 4). These results indicate that NOACs may be effective at preventing events of all types. Furthermore, these findings strongly support the Japanese guidelines for the management of AF³¹ that recommend use of NOACs when both warfarin and NOACs are indicated.

Effectiveness of Long-Term Warfarin Therapy

Previously, the beneficial effects of warfarin therapy for preventing thromboembolism were observed for 2 years in the J-RHYTHM Registry.⁷ However, in the present study, there was no significant difference in the rate of thromboembolism between the No-OAC and Warfarin groups at the end of the follow-up period (Table 3). One possible explanation for this observation was that warfarin was switched to NOAC, especially in low-risk patients in good clinical condition, since the risk for thromboembolism in the NOACs group was lower than that in the Warfarin group (Table 1). The present study extended the follow-up period to investigate the long-term effects of warfarin therapy. Anticoagulation therapies were changed in more than 1,000 patients during the extended follow-up period (Tables 1,2). It therefore became more difficult to clearly determine the long-term effects of warfarin based on the baseline status of warfarin therapy. Therefore, we compared event rates and time-to-event curves in the limited number of patients who did not change anticoagulation therapy from baseline to the end of follow-up (ie, 3,786 patients with continuous warfarin and 421 without any anticoagulation therapy throughout the study period). As shown in Figure 3 and Table S2, the beneficial effects of warfarin to prevent thromboembolism lasted for 5 years. However, long-term use of warfarin was associated with an increasing incidence of major hemorrhage. Although these results were consistent with our previous report,⁷ they differed from the 1-year outcomes of the Fushimi AF Registry, in which anticoagulation therapy did not affect the incidence of either thromboembolism or major hemorrhage.³² This discrepancy could be explained by differences between study populations rather than follow-up period. In contrast to the J-RHYTHM Registry, the Fushimi AF Registry included a large number of patients from general hospitals in an urban community.³²^,³³

The results of the present study revealed that the baseline PT-INR level still influenced the incidence of thromboembolism and major hemorrhage (Figure S1), even after adjusting for multiple confounding factors (Figure 4). Importantly, the adjusted HRs for thromboembolism were significantly lower in all PT-INR subgroups, while those for major hemorrhage were significantly higher for PT-INR values ≥2.6 compared with the reference group not receiving any anticoagulation therapy. These results support our previous reports on target INR values in the J-RHYTHM Registry.⁷^,⁸^,¹¹^,¹²

Study Limitations

First, this registry study was performed in a single country and subjects were enrolled from 131 selected institutions in Japan. The clinical backgrounds of the patients in this registry may not be extrapolated to a general Japanese patient population with NVAF³²^,³³ because most of the participating physicians and institutions specialized in cardiology. Second, although the original J-RHYTHM Registry started with 7,937 patients, the number of patients participating in the extended follow-up decreased to 6,616 patients with NVAF. Additionally, 976 patients were lacking data regarding anticoagulation status at the end of the study; therefore only 5,640 patients were available for on-treatment analysis at the end of follow-up. Third, the quality of warfarin therapy was evaluated using only baseline PT-INR values. Because PT-INR values were not recorded during each visit during the extended follow-up period, neither time in the therapeutic range (TTR)³⁴ nor the intensity of anticoagulation during the extended follow-up period was determined in the present study. The TTR according to the Japanese guidelines²⁴ was 59.3±29.2% in the first 2 years of the J-RHYTHM Registry.⁷^,¹¹^,¹² Finally, although we found that event rates in the NOACs group were quite low, the selection of antithrombotic drugs was not randomized, because of the observational nature of the study. In addition, the number of patients in the NOACs group was relatively small (n=923) and the follow-up period after starting NOACs was shorter than that for patients who continued to take warfarin. Further observations with larger real-world populations and longer follow-up periods are necessary to clarify the efficacy and safety of NOACs among Japanese patients with NVAF.

Conclusions

In Japanese NVAF patients, long-term use of warfarin may be beneficial for preventing thromboembolism at the expense of major hemorrhage. In contrast, NOAC use could be beneficial for reducing event rates of all types.

Disclosures

The J-RHYTHM registry and the J-RHYTHM registry 2 were supported by grants from the Japan Heart Foundation (12080025 and 12110015). This research is partially supported by the Practical Research Project for Life-Style related Diseases including Cardiovascular Diseases and Diabetes Mellitus from the Japan Agency for Medical Research and Development (AMED, 15656344). The following co-authors have potential conflicts of interest: H.A. has received research funding from Boehringer Ingelheim, and remuneration from Bayer Healthcare, Boehringer Ingelheim, and Daiichi-Sankyo; H.I. has received research funding from Boehringer Ingelheim and Daiichi-Sankyo, and remuneration from Daiichi-Sankyo, Bayer Healthcare, and Boehringer Ingelheim; K.O. has received research funding from Boehringer Ingelheim and Daiichi-Sankyo, and remuneration from Boehringer Ingelheim, Bayer Healthcare, Daiichi-Sankyo, and Pfizer; T.Y. has received research funding from Daiichi-Sankyo, Bayer, Bristol-Meyers Squibb and remuneration from Nippon Boehringer, Bristol-Myers Squibb, Pfizer, Bayer, Daiichi-Sankyo, Tanabe-Mitsubishi, Ono Pharmaceutical, Eisai and Toa-Eiyo; H.O. has received lecture fees from Daiichi-Sankyo.

Acknowledgments

The present study was presented in part at the 80^th Annual Scientific Meeting of the Japanese Circulation Society (Sendai, Japan, March 18, 2016). Executive Committee of the J-RHYTHM Registry 2 comprised Okumura K, Inoue H, Atarashi H, and Yamashita T. Local Executive Committee was as follows: Sakurai M, Kawamura Y (Hokkaido); Okumura K, Kubota I (Tohoku); Kaneko Y, Matsumoto K (North Kanto); Ogawa S, Atarashi H, Yamashita T (South Kanto); Inoue H, Chinushi M (Hokuetsu); Kodama I, Watanabe E (Chubu); Koretsune Y, Okuyama Y (Kansai); Shimizu A, Adachi M (Chugoku); Bando S, Fukatani M (Shikoku); Saikawa T, Chishaki A (Kyushu); Statistical advisor: Origasa H. Other investigators are listed in references 6 and 35.

Supplementary Files

Supplementary File 1

Table S1. Characteristics of patients based on anticoagulation therapy throughout the follow-up period

Table S2. Five-year incidence of events based on anticoagulation therapy throughout the follow-up period

Figure S1. Five-year incidence rates of thromboembolism (Left) and major hemorrhage (Right) in patients without any OAC and in patients with different anticoagulation levels at baseline.

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-16-0066

References

1. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: The Framingham Study. Stroke 1991; 22: 983–988.
2. Feinberg WM, Blackshear JL, Laupacis A, Kronmal R, Hart RG. Prevalence, age distribution, and gender of patients with atrial fibrillation: Analysis and implications. Arch Intern Med 1995; 155: 469–473.
3. 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.
4. Hart RG, Pearce LA, Aguilar MI. Meta-analysis: Antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med 2007; 146: 857–867.
5. 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.
6. 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.
7. 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.
8. Kodani E, Atarashi H, Inoue H, Okumura K, Yamashita T. Target intensity of anticoagulation with warfarin in Japanese patients with valvular atrial fibrillation: Subanalysis of the J-RHYTHM Registry. Circ J 2015; 79: 325–330.
9. Fuster V, Ryden 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 (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation): Developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Circulation 2006; 114: e257–e354, doi:10.1161/CIRCULATIONAHA.106.177292.
10. 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.
11. Kodani E, Atarashi H, Inoue H, Okumura K, Yamashita T, Origasa H; J-RHYTHM Registry Investigators. Secondary prevention of stroke with warfarin in patients with non-valvular atrial fibrillation: Subanalysis of the J-RHYTHM Registry. J Stroke Cerebrovasc Dis 2015 December 22, doi:10.1016/j.jstrokecerebrovasdis.2015.11.020.
12. Kodani E, Atarashi H, Inoue H, Okumura K, Yamashita T, Origasa H. Use of warfarin in elderly patients with non-valvular atrial fibrillation: Subanalysis of the J-RHYTHM Registry. Circ J 2015; 79: 2345–2352.
13. 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: 1139–1151.
14. Connolly SJ, Ezekowitz MD, Yusuf S, Reilly PA, Wallentin L. Newly identified events in the RE-LY trial. N Engl J Med 2010; 363: 1875–1876.
15. Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011; 365: 883–891.
16. Hori M, Matsumoto M, Tanahashi N, Momomura S, Uchiyama S, Goto S, et al. Rivaroxaban vs. warfarin in Japanese patients with atrial fibrillation: The J-ROCKET AF study. Circ J 2012; 76: 2104–2111.
17. Granger CB, Alexander JH, McMurray JJ, Lopes RD, Hylek EM, Hanna M, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2011; 365: 981–992.
18. Giugliano RP, Ruff CT, Braunwald E, Murphy SA, Wiviott SD, Halperin JL, et al. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2013; 369: 2093–2104.
19. Barnes GD, Lucas E, Alexander GC, Goldberger ZD. National trends in ambulatory oral anticoagulant use. Am J Med 2015; 128: 1300–1305.e2, doi:10.1016/j.amjmed.2015.05.044.
20. Olesen JB, Sorensen R, Hansen ML, Lamberts M, Weeke P, Mikkelsen AP, et al. Non-vitamin K antagonist oral anticoagulation agents in anticoagulant naive atrial fibrillation patients: Danish nationwide descriptive data 2011–2013. Europace 2015; 17: 187–193.
21. Ogawa S, Ikeda T, Kitazono T, Nakagawara J, Minematsu K, Miyamoto S, et al. Present profiles of novel anticoagulant use in Japanese patients with atrial fibrillation: Insights from the rivaroxaban postmarketing surveillance registry. J Stroke Cerebrovasc Dis 2014; 23: 2520–2526.
22. Miyamoto K, Aiba T, Nakajima I, Yamada Y, Okamura H, Noda T, et al. Efficacy and safety of novel anticoagulant dabigatran in clinical practice for Japanese patients with non-valvular atrial fibrillation. J Arrhythmia 2014; 30: 58–64.
23. 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.
24. JCS Joint Working Group. Guidelines for pharmacotherapy of atrial fibrillation (JCS 2008): Digest version. Circ J 2010; 74: 2479–2500.
25. 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.
26. Graham DJ, Reichman ME, Wernecke M, Zhang R, Southworth MR, Levenson M, et al. Cardiovascular, bleeding, and mortality risks in elderly Medicare patients treated with dabigatran or warfarin for nonvalvular atrial fibrillation. Circulation 2015; 131: 157–164.
27. Camm AJ, Amarenco P, Haas S, Hess S, Kirchhof P, Kuhls S, et al. XANTUS: A real-world, prospective, observational study of patients treated with rivaroxaban for stroke prevention in atrial fibrillation. Eur Heart J 2015 September 1, doi:10.1093/eurheartj/ehv466.
28. Beyer-Westendorf J, Forster K, Pannach S, Ebertz F, Gelbricht V, Thieme C, et al. Rates, management, and outcome of rivaroxaban bleeding in daily care: Results from the Dresden NOAC registry. Blood 2014; 124: 955–962.
29. Tamayo S, Frank Peacock W, Patel M, Sicignano N, Hopf KP, Fields LE, et al. Characterizing major bleeding in patients with nonvalvular atrial fibrillation: A pharmacovigilance study of 27 467 patients taking rivaroxaban. Clin Cardiol 2015; 38: 63–68.
30. Ogawa H, Akao M, Suzuki S, Yamashita T, Okumura K, Atarashi H, et al. Antiplatelet therapy in Japanese patients with atrial fibrillation without oral anticoagulants: Pooled analysis of Shinken Database, J-RHYTHM registry and Fushimi AF registry. Int J Cardiol 2015; 190: 344–346.
31. JCS Joint Working Group. Guidelines for pharmacotherapy of atrial fibrillation (JCS 2013): Digest version. Circ J 2014; 78: 1997–2021.
32. Akao M, Chun YH, Esato M, Abe M, Tsuji H, Wada H, et al. Inappropriate use of oral anticoagulants for patients with atrial fibrillation. Circ J 2014; 78: 2166–2172.
33. Akao M, Chun YH, Wada H, Esato M, Hashimoto T, Abe M, et al. Current status of clinical background of patients with atrial fibrillation in a community-based survey: The Fushimi AF Registry. J Cardiol 2013; 61: 260–266.
34. Rosendaal FR, Cannegieter SC, van der Meer FJ, Briet E. A method to determine the optimal intensity of oral anticoagulant therapy. Thromb Haemost 1993; 69: 236–239.
35. 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.

Corresponding author

Register with J-STAGE for free!