Abstract
Background:
The proportion of patients with atrial fibrillation (AF) treated with anticoagulation varies from country to country. In Japan, little is known about regional differences in frequency of warfarin use or prognosis among patients with non-valvular AF (NVAF).
Methods and Results:
In J-RHYTHM Registry, the number of patients recruited from each of 10 geographic regions of Japan was based on region population density. A total of 7,406 NVAF patients were followed up prospectively for 2 years. At baseline, significant differences in various clinical characteristics including age, sex, type of AF, comorbidity, and CHADS2
score, were detected among the regions. The highest mean CHADS2
score was recorded in Shikoku. Frequency of warfarin use differed between the regions (P<0.001), with lower frequencies observed in Hokkaido and Shikoku. Baseline prothrombin time international normalized ratio differed slightly but significantly between the regions (P<0.05). On univariate analysis, frequency of thromboembolic events differed among the regions (P<0.001), with the highest rate seen in Shikoku. An inverse correlation was detected between frequency of thromboembolic and of major hemorrhagic events (P=0.062). On multivariate analysis, region emerged as an independent risk for thromboembolism.
Conclusions:
Thromboembolic risk, frequency of warfarin use, and intensity and quality of warfarin treatment differed significantly between geographic regions of Japan. Region was found to be an independent predictor of thromboembolic events. (Circ J 2016; 80: 1548–1555)
Atrial fibrillation (AF) is a common type of arrhythmia, and it is considered to be a risk factor for cardiogenic embolism. Although warfarin-based anticoagulation is effective at reducing the frequency of this complication, the proportion of AF patients treated with anticoagulation varies from country to country.1–5
In addition, regional differences in the proportions of AF patients treated with warfarin to prevent thromboembolism were detected in the USA.6
Furthermore, the intensity and quality of warfarin-based anticoagulation therapy also vary with geographic region around the world.4,7–9
In the warfarin arms of global clinical studies on direct oral anticoagulant (DOAC) use, the incidence rates of stroke and systemic embolic events were higher in Asian countries than in non-Asian countries.10–12
Little is known, however, about the regional differences in frequency of warfarin use or its outcomes among AF patients in Japan.
Editorial p 1520
The J-RHYTHM Registry is a prospective, observational study that has enrolled more than 7,000 patients with non-valvular AF (NVAF).13,14
One of the primary objectives of the J-RHYTHM Registry is to explore variations in the international normalized ratio (INR) of prothrombin time and the quality of anticoagulation between the 10 geographic regions of Japan, and to determine how these variations affect outcome.13
In the present study, the regional differences in the state of warfarin-based anticoagulation therapy and its outcomes in NVAF patients were examined.
Methods
Study Design
The study design of the J-RHYTHM Registry (UMIN Clinical Trials Registry, UMIN 000001569) has been reported previously.13,14
Briefly, the study protocol conformed to the Declaration of Helsinki and was approved by each of the participating institutions. Anti-thrombotic drugs and dosages were selected at the discretion of the treating physicians. Given that the patient registration started before the clinical use of DOAC for thromboembolism prevention was approved in Japan, warfarin was the only oral anticoagulant used in the present study. Anticoagulation intensity was determined based on INR of the patients receiving warfarin at baseline, and anticoagulation quality was assessed based on time in the therapeutic range (TTR) using the method developed by Rosendaal et al.15
Target INR was set at 1.6–2.6 for elderly patients aged ≥70 years and at 2.0–3.0 for patients aged <70 years according to the Japanese guidelines.16
Subjects
In order to avoid an unrepresentative regional distribution, the number of patients recruited from each of the 10 Japanese geographic regions was determined based on population density.13
The subjects consisted of a consecutive series of outpatients with AF of any type, who gave written informed consent. Patients were excluded from the present analysis if they had mitral stenosis, had undergone mechanical valve replacement, or were lost to follow-up during the 2-year follow-up period.
Outcomes
The endpoints were thromboembolic events and major hemorrhagic events. Thromboembolic events included symptomatic ischemic stroke, transient ischemic attack (TIA), and systemic embolism. Major hemorrhagic events included intracranial hemorrhaging and other hemorrhagic events that required hospitalization. The diagnostic criteria for each event have been described previously.13,14
Statistical Analysis
The data are given as percentage or mean±SD. Means were compared using ANOVA, and frequency was compared using chi-squared test. The correlation between the incidence rates of thromboembolic and of major hemorrhagic events was determined on linear regression analysis. Cox proportional hazards modeling was used to calculate hazard ratios (HR) and 95% CI for each risk factor. On multivariate analysis, region, age, sex, components of the CHADS2
score,17
type of AF, coronary artery disease, and the medications used at the baseline were included as explanatory variables. The Kaplan-Meier method was used to obtain estimates of the frequency of thromboembolic events in each region, and comparisons between these values were performed with the log-rank test. P<0.05 was considered significant. All statistical analysis was performed with JMP version 11 (SAS Institute, Cary, NC, USA).
Results
A total of 7,937 patients with AF were enrolled into the study between January 2009 and July 2009. Of these, 421 had mitral stenosis or had undergone mechanical valve replacement, and therefore, the remaining 7,516 patients constituted the NVAF group.18
Among the NVAF group, 110 patients (1.5%) were lost to follow-up, and the remaining 7,406 patients were used as the study group for the present analysis.
Clinical Characteristics
Baseline patient clinical characteristics according to geographic region are listed in
Table 1. Significant differences in various baseline characteristics, including age, sex, type of AF, comorbidities, CHADS2
score, blood pressure, and heart rate, were detected between the 10 regions. The highest mean CHADS2
score was seen in Shikoku, and the lowest mean CHADS2
score was recorded in Southern Kanto. The frequencies of warfarin, antiplatelet, antihypertensive drug, lipid-lowering drug, and hypoglycemic drug use differed significantly between the 10 regions (Table 2). Warfarin dosage also differed significantly between the 10 regions, with the lowest dosages seen in Hokuetsu and Shikoku. Baseline INR and TTR differed slightly, but significantly, between the 10 regions. Bucolome, a drug that interacts with warfarin to enhance its anticoagulation intensity,19
was preferentially used in Hokuetsu, which might explain why the lowest warfarin dosage was observed in this region.
Table 1.
Baseline Characteristics
| |
Hokkaido |
Tohoku |
Northern
Kanto |
Southern
Kanto |
Hokuetsu |
Chubu |
Kansai |
Chugoku |
Shikoku |
Kyushu |
P-value† |
| n |
396 |
655 |
678 |
1,582 |
304 |
1,114 |
948 |
612 |
384 |
733 |
|
| Age (years) |
71.3±9.4 |
69.5±10.0 |
69.9±9.3 |
69.1±9.8 |
70.0±9.7 |
69.6±9.9 |
69.8±9.9 |
71.0±10.6 |
71.4±9.9 |
69.7±10.5 |
<0.001 |
| Male |
288 (72.7) |
433 (66.1) |
501 (73.9) |
1,151 (72.8) |
215 (70.7) |
794 (71.3) |
683 (72.0) |
396 (64.7) |
269 (70.1) |
511 (69.7) |
0.002 |
| Type of AF |
| Paroxysmal |
123 (31.1) |
266 (40.6) |
225 (33.2) |
558 (35.3) |
82 (27.0) |
464 (41.7) |
373 (39.3) |
261 (42.6) |
147 (38.3) |
336 (45.8) |
|
| Persistent |
38 (9.6) |
66 (10.1) |
143 (21.1) |
253 (16.0) |
89 (29.3) |
93 (8.3) |
148 (15.6) |
102 (16.7) |
50 (13.0) |
99 (13.5) |
<0.001 |
| Permanent |
235 (59.3) |
323 (49.3) |
310 (45.7) |
771 (48.7) |
133 (43.8) |
557 (50.0) |
427 (45.0) |
249 (40.7) |
187 (48.7) |
298 (47.1) |
|
| Comorbidities |
| CAD |
58 (14.6) |
77 (11.8) |
59 (8.7) |
143 (9.0) |
18 (5.9) |
148 (13.3) |
89 (9.4) |
50 (8.2) |
54 (14.1) |
85 (11.6) |
<0.001 |
| Cardiomyopathy |
47 (11.9) |
61 (9.3) |
45 (6.6) |
116 (7.3) |
31 (10.2) |
99 (8.9) |
76 (8.0) |
65 (10.6) |
28 (7.3) |
66 (9.0) |
0.034 |
| HCM |
19 (4.8) |
20 (3.1) |
14 (2.1) |
48 (3.0) |
14 (4.6) |
45 (4.0) |
29 (3.1) |
25 (4.1) |
18 (4.7) |
32 (4.4) |
0.136 |
| DCM |
28 (7.1) |
41 (6.3) |
31 (4.6) |
68 (4.3) |
17 (5.6) |
54 (4.8) |
47 (5.0) |
40 (6.5) |
10 (2.6) |
34 (4.6) |
0.070 |
| CHD |
1 (0.3) |
13 (2.0) |
5 (0.7) |
16 (1.0) |
8 (2.6) |
10 (0.9) |
18 (1.9) |
8 (1.3) |
5 (1.3) |
12 (1.6) |
0.042 |
| COPD |
11 (2.8) |
13 (2.0) |
9 (1.3) |
25 (1.6) |
8 (2.6) |
21 (1.9) |
10 (1.1) |
11 (1.8) |
15 (3.9) |
8 (1.1) |
0.019 |
| Hyperthyroidism |
2 (0.5) |
13 (2.0) |
10 (1.5) |
25 (1.6) |
3 (1.0) |
23 (2.1) |
28 (3.0) |
7 (1.1) |
8 (2.1) |
12 (1.6) |
0.080 |
| Risk factors for stroke |
| Heart failure |
97 (24.8) |
186 (28.4) |
197 (29.1) |
341 (21.6) |
94 (30.9) |
349 (31.3) |
273 (28.8) |
185 (30.2) |
121 (31.5) |
212 (28.9) |
<0.001 |
| Hypertension |
216 (54.5) |
417 (63.7) |
421 (62.1) |
929 (58.7) |
157 (51.6) |
661 (59.3) |
561 (59.2) |
374 (61.1) |
282 (73.4) |
459 (62.6) |
<0.001 |
| Age (≥75 years) |
159 (40.2) |
226 (34.5) |
206 (32.4) |
487 (30.8) |
108 (35.5) |
383 (34.4) |
288 (30.4) |
256 (41.8) |
159 (41.4) |
279 (38.1) |
<0.001 |
| Diabetes mellitus |
81 (20.5) |
96 (14.7) |
112 (16.5) |
281 (17.8) |
44 (14.5) |
222 (19.8) |
189 (19.9) |
93 (15.2) |
99 (25.8) |
142 (19.4) |
<0.001 |
| Stroke/TIA |
62 (15.7) |
89 (13.6) |
97 (14.3) |
185 (11.7) |
38 (12.5) |
167 (15.0) |
115 (12.1) |
73 (11.9) |
80 (20.8) |
116 (15.8) |
<0.001 |
| CHADS2 score |
| 0 |
71 (17.9) |
85 (13.0) |
93 (13.7) |
309 (19.5) |
52 (17.1) |
170 (15.3) |
153 (16.1) |
83 (13.6) |
32 (8.3) |
109 (14.9) |
|
| 1 |
127 (32.1) |
239 (36.5) |
236 (34.8) |
586 (37.0) |
112 (36.8) |
347 (31.1) |
342 (36.1) |
195 (31.9) |
97 (25.3) |
231 (31.5) |
|
| 2 |
100 (25.3) |
196 (29.9) |
205 (30.2) |
380 (24.0) |
83 (27.3) |
330 (29.6) |
247 (26.1) |
198 (32.4) |
121 (31.5) |
196 (26.7) |
|
| 3 |
61 (15.4) |
84 (12.8) |
94 (13.9) |
195 (12.3) |
32 (10.5) |
195 (14.8) |
137 (14.5) |
94 (15.4) |
77 (20.1) |
120 (16.4) |
<0.001 |
| 4 |
21 (5.3) |
37 (5.6) |
34 (5.0) |
86 (5.4) |
20 (6.6) |
69 (6.2) |
53 (5.6) |
30 (4.9) |
38 (9.9) |
48 (6.5) |
|
| 5 |
13 (3.3) |
12 (1.8) |
16 (2.4) |
23 (1.5) |
5 (1.6) |
27 (2.4) |
14 (1.5) |
11 (1.8) |
15 (3.9) |
25 (3.4) |
|
| 6 |
3 (0.8) |
2 (0.3) |
0 (0.0) |
3 (0.2) |
0 (0.0) |
6 (0.5) |
2 (0.2) |
1 (0.2) |
4 (1.0) |
4 (0.5) |
|
| Mean |
1.7±1.3 |
1.7±1.2 |
1.7±1.2 |
1.5±1.2 |
1.6±1.2 |
1.8±1.3 |
1.6±1.2 |
1.7±1.2 |
2.1±1.3 |
1.8±1.3 |
<0.001 |
| Systolic BP (mmHg) |
124.9±13.3 |
126.8±14.1 |
127.1±15.7 |
124.5±15.5 |
124.0±16.3 |
125.8±16.9 |
126.9±17.7 |
127.7±16.6 |
129.2±10.6 |
124.5±16.6 |
<0.001 |
| Diastolic BP (mmHg) |
75.3±10.6 |
73.2±10.1 |
74.7±10.9 |
73.4±10.7 |
74.8±48.5 |
72.1±11.2 |
76.2±26.3 |
74.2±11.6 |
73.1±11.9 |
71.1±11.7 |
<0.001 |
| Heart rate (beats/min) |
72.7±12.1 |
71.8±11.7 |
73.6±13.6 |
73.4±13.9 |
71.3±11.6 |
71.6±13.8 |
72.2±13.1 |
73.1±13.0 |
72.3±13.3 |
71.6±12.9 |
0.001 |
Data given as n (%) or mean±SD. †Comparisons between the 10 regions. AF, atrial fibrillation; BP, blood pressure; CAD, coronary artery disease; CHADS2, congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, and history of stroke or TIA; CHD, congenital heart disease; COPD, chronic obstructive pulmonary disease; DCM, dilated cardiomyopathy; HCM, hypertrophic cardiomyopathy; TIA, transient ischemic attack.
Table 2.
Anti-Thrombotic Therapy and Other Medications at Enrollment
| |
Hokkaido |
Tohoku |
Northern
Kanto |
Southern
Kanto |
Hokuetsu |
Chubu |
Kansai |
Chugoku |
Shikoku |
Kyushu |
P-value† |
| Warfarin |
314 (79.3) |
617 (94.2) |
581 (85.7) |
1,356 (85.7) |
287 (94.4) |
927 (83.2) |
826 (87.1) |
529 (86.4) |
311 (81.0) |
656 (89.5) |
<0.001 |
| Dosage (mg/day) |
2.6±1.1 |
3.1±1.2 |
2.8±1.1 |
2.9±1.2 |
2.5±1.3 |
2.9±1.2 |
2.9±1.2 |
2.9±1.2 |
2.5±1.0 |
3.0±1.2 |
<0.001 |
| INR |
| <1.6 |
97 (30.9) |
142 (23.0) |
147 (25.3) |
362 (26.7) |
69 (24.0) |
272 (29.3) |
204 (24.7) |
140 (26.5) |
81 (26.0) |
156 (23.8) |
|
| 1.6–1.99 |
103 (32.8) |
250 (40.5) |
208 (35.8) |
508 (37.5) |
107 (37.3) |
337 (36.4) |
292 (35.4) |
201 (38.0) |
111 (35.7) |
231 (35.2) |
|
| 2.0–2.59 |
80 (25.6) |
187 (30.3) |
174 (29.9) |
376 (27.7) |
94 (32.8) |
248 (26.8) |
239 (28.9) |
154 (29.1) |
91 (29.3) |
211 (32.2) |
0.026 |
| 2.6–2.99 |
21 (6.7) |
31 (5.0) |
34 (5.9) |
72 (5.3) |
10 (3.5) |
52 (5.6) |
61 (7.4) |
21 (4.0) |
24 (7.7) |
37 (5.6) |
|
| ≥3.0 |
13 (4.1) |
7 (1.1) |
18 (3.1) |
38 (2.8) |
7 (2.4) |
18 (1.9) |
30 (3.6) |
13 (2.5) |
4 (1.3) |
21 (3.2) |
|
| Mean |
1.91±0.54 |
1.89±0.44 |
1.92±0.48 |
1.89±0.48 |
1.93±0.49 |
1.87±0.51 |
1.95±0.55 |
1.89±0.45 |
1.91±0.46 |
1.94±0.50 |
0.030 |
| TTR (%)‡ |
58.3±29.7 |
60.7±28.9 |
57.7±28.9 |
58.5±29.2 |
61.7±28.5 |
57.2±29.4 |
59.2±29.1 |
60.9±29.8 |
59.5±28.7 |
62.2±29.6 |
0.035 |
| [n for TTR] |
[306] |
[587] |
[546] |
[1,292] |
[275] |
[878] |
[779] |
[504] |
[293] |
[604] |
|
| Antiplatelets |
207 (52.3) |
155 (23.7) |
185 (27.3) |
406 (25.7) |
54 (17.8) |
347 (31.1) |
199 (21.0) |
101 (16.5) |
127 (33.1) |
156 (21.3) |
<0.001 |
| Aspirin |
178 (44.9) |
138 (21.1) |
151 (22.3) |
354 (22.4) |
43 (14.1) |
311 (27.9) |
185 (19.5) |
83 (13.6) |
104 (27.1) |
128 (17.5) |
<0.001 |
| Others |
49 (12.4) |
33 (5.0) |
47 (6.9) |
85 (5.4) |
13 (4.3) |
70 (6.3) |
22 (2.3) |
33 (5.4) |
42 (10.9) |
44 (6.0) |
<0.001 |
| Warfarin+antiplatelets |
162 (40.9) |
126 (19.2) |
133 (19.6) |
272 (17.2) |
46 (15.1) |
228 (20.5) |
142 (15.0) |
72 (11.8) |
87 (22.7) |
126 (17.2) |
<0.001 |
| Bucolome |
13 (4.1) |
7 (1.1) |
29 (5.0) |
85 (6.3) |
81 (28.2) |
11 (1.2) |
18 (2.2) |
4 (0.8) |
22 (7.1) |
1 (0.2) |
<0.001 |
| Anti-hypertensive drugs |
241 (60.9) |
483 (73.7) |
493 (72.7) |
1,106 (69,9) |
218 (71.7) |
835 (75.0) |
679 (71.6) |
439 (71.7) |
290 (75.5) |
570 (77.8) |
<0.001 |
| ARB/ACEI |
158 (39.9) |
360 (55.0) |
358 (52.8) |
817 (51.6) |
157 (51.6) |
617 (55.4) |
519 (54.0) |
342 (55.9) |
209 (54.4) |
397 (54.2) |
<0.001 |
| Others |
167 (42.2) |
309 (47.2) |
316 (46.6) |
735 (46.5) |
146 (48.0) |
595 (53.4) |
386 (40.7) |
278 (45.4) |
216 (56.3) |
397 (54.2) |
<0.001 |
| Lipid-lowering drugs |
114 (28.8) |
157 (34.0) |
192 (28.3) |
422 (26.7) |
51 (16.8) |
352 (31.6) |
252 (26.6) |
132 (21.6) |
113 (29.4) |
225 (30.7) |
<0.001 |
| Statins |
104 (26.3) |
141 (21.5) |
177 (26.1) |
370 (23.4) |
45 (14.8) |
317 (28.5) |
223 (23.5) |
111 (18.1) |
102 (26.6) |
205 (28.0) |
<0.001 |
| Others |
14 (3.5) |
20 (3.1) |
22 (3.2) |
62 (3.9) |
6 (2.0) |
42 (3.9) |
33 (3.5) |
21 (3.4) |
13 (3.4) |
22 (3.0) |
0.880 |
| Hypoglycemic drugs |
55 (13.9) |
56 (8.5) |
73 (10.8) |
158 (10.0) |
21 (6.9) |
133 (11.9) |
129 (13.6) |
65 (10.6) |
47 (12.2) |
75 (10.2) |
0.006 |
Data given as n (%) or mean±SD. †Comparison between the 10 regions. ‡Target INR was 2.0–3.0 (<70 years old) or 1.6–2.6 (≥70 years old). ARB, angiotensin II receptor blockers; ACEI, angiotensin-converting enzyme inhibitors; INR, international normalized ratio of prothrombin time; TTR, time in therapeutic range.
The 2-year incidence rates of thromboembolic and major hemorrhagic events are summarized in
Table 3. The frequency of thromboembolic events differed significantly among the 10 regions; it was highest in Shikoku and lowest in Hokuetsu. On Kaplan-Meier analysis, the frequency of thromboembolic events differed significantly among the regions (Figure 1). The frequency of major hemorrhagic events, however, did not differ among the regions. The combined frequency of thromboembolic and hemorrhagic events did not differ between the 10 regions (Table 3). A marginally significant inverse correlation was detected between the frequencies of thromboembolic and major hemorrhagic events (P=0.062,
Figure 2).
Table 3.
Two-Year Incidence Rates
| |
Hokkaido |
Tohoku |
Northern
Kanto |
Southern
Kanto |
Hokuetsu |
Chubu |
Kansai |
Chugoku |
Shikoku |
Kyushu |
P-value† |
| Thromboembolism |
10 (2.5) |
5 (0.8) |
20 (2.9) |
24 (1.5) |
1 (0.3) |
24 (2.2) |
7 (0.7) |
11 (1.8) |
14 (3.6) |
10 (1.4) |
<0.001 |
| Major hemorrhage |
3 (0.8) |
10 (1.5) |
9 (1.3) |
27 (1.7) |
8 (2.6) |
23 (2.1) |
21 (2.2) |
16 (2.6) |
5 (1.3) |
18 (2.5) |
0.352 |
| Thromboembolism+major hemorrhage |
13 (3.3) |
15 (2.3) |
29 (4.3) |
51 (3.2) |
9 (3.0) |
47 (4.2) |
28 (3.0) |
27 (4.4) |
19 (4.9) |
28 (3.8) |
0.281 |
Data given as n (%/2 years). †Comparisons between the 10 regions.
The adjusted HR for thromboembolic events obtained on multivariate Cox proportional hazards modeling are shown in
Tables 4,5. Region was found to be an independent risk factor for thromboembolism. When Shikoku was used as the reference, the Tohoku, Hokuetsu, and Kansai regions were found to be independent predictors of lower thromboembolic event rates (Table 5). As expected, age and prior ischemic stroke or TIA emerged as independent risk factors for thromboembolism (Table 4).20
Warfarin use was associated with a lower risk of thromboembolism. In addition, permanent AF was found to be associated with a higher risk of thromboembolic events.
Table 4.
Multivariate Indicators of Thromboembolic Events
| Explanatory variables |
HR (95% CI) |
P-value |
| Region |
Table 5 |
0.003† |
| Age (per year) |
1.05 (1.03–1.07) |
<0.001 |
| Sex (female) |
0.82 (0.54–1.21) |
0.320 |
| Persistent AF‡ |
1.14 (0.57–2.14) |
0.702 |
| Permanent AF‡ |
1.98 (1.30–3.10) |
0.001 |
| Heart failure |
0.85 (0.56–1.27) |
0.437 |
| Hypertension |
0.90 (0.59–1.39) |
0.640 |
| Diabetes mellitus |
1.17 (0.75–1.78) |
0.486 |
| Stroke/TIA |
1.64 (1.06–2.48) |
0.028 |
| CAD |
0.99 (0.53–1.72) |
0.971 |
| Warfarin use |
0.42 (0.27–0.69) |
<0.001 |
| Antiplatelet use |
0.95 (0.61–1.44) |
0.803 |
| Bucolome |
0.89 (0.22–2.51) |
0.892 |
| Anti-hypertensive drugs |
1.17 (0.78–1.78) |
0.448 |
| Lipid lowering drugs |
0.88 (0.56–1.35) |
0.565 |
†Likelihood ratio test; ‡vs. paroxysmal AF. AF, atrial fibrillation; CAD, coronary artery disease; CI, confidence interval; HR, hazard ratio; TIA, transient ischemic attack.
Table 5.
Regional Differences in Incidence Rate of Thromboembolic Events
| Region |
HR |
95% CI |
P-value |
| Hokkaido |
0.70 |
0.30–1.58 |
0.387 |
| Tohoku |
0.28 |
0.09–0.75 |
0.010 |
| Northern Kanto |
1.06 |
0.54–2.15 |
0.867 |
| Southern Kanto |
0.52 |
0.27–1.04 |
0.064 |
| Hokuetsu |
0.12 |
0.01–0.61 |
0.007 |
| Chubu |
0.72 |
0.37–1.43 |
0.338 |
| Kansai |
0.26 |
0.09–0.64 |
0.003 |
| Chugoku |
0.61 |
0.27–1.35 |
0.221 |
| Shikoku |
1.00 (reference) |
|
|
| Kyushu |
0.48 |
0.21–1.09 |
0.078 |
Abbreviations as in Table 4.
Discussion
Major Findings
The major findings of the present study are as follows. First, regional differences were detected in many of the baseline clinical characteristics and in the baseline usage frequencies of various drugs between the 10 geographic regions of Japan. Second, on univariate analysis regional differences were detected in the frequency of thromboembolic events, but not of major hemorrhagic events. The frequency of thromboembolic events was highest in Shikoku. On multivariate analysis, region emerged as an independent predictor of thromboembolic events.
Regional Differences in Comorbidities and Medication Usage
The prevalence rates of certain diseases might vary between the different regions of Japan. On analysis of a stroke patient database run by the Ministry of Health, Labor, and Welfare, Japan, in 2002 the age-adjusted frequency of hospital or clinic visits due to stroke was highest in Ehime prefecture for men and in Kochi prefecture for women.21
Both prefectures are located on Shikoku. In the present study, the frequency of a history of cerebral infarction or TIA was highest in Shikoku, as expected from the Government database.21
In addition, the patients in Shikoku had higher mean age, CHADS2
score, and systolic blood pressure as well as higher frequency of coronary artery disease, hypertrophic cardiomyopathy, heart failure, hypertension, and diabetes.
In a previous study on warfarin use in AF patients in the USA, the frequency of warfarin use in AF patients was lowest in the southern part of the USA.6
In the present study, the regional differences in the frequency of warfarin use were numerically small, but statistically significant (Table 2). The frequency of warfarin use was lowest in Hokkaido, followed by Shikoku. Baseline INR and TTR also differed significantly between the regions. Furthermore, the baseline usage frequency of medications such as angiotensin II receptor blockers, angiotensin-converting enzyme inhibitors, and statins differed significantly between the regions.
Regional Differences in Event Rates
Taken together, the clinical characteristics of the AF patients in Shikoku might have contributed to the fact that the highest incidence rate of thromboembolic events was seen in this region (Table 3). Regional differences in the incidence rate of thromboembolic events, however, were still detected after adjusting for clinical characteristics (Tables 4,
5). Therefore, the AF patients in Shikoku might have been more prone to falling into a prothrombotic state, or some unmeasured confounding factors might have contributed to the regional differences in the frequency of thromboembolic events. It is worth noting that an inverse correlation was detected between the incidence rates of thromboembolic events and of major hemorrhagic events, although it was only marginally significant (P=0.062;
Figure 2).
Mean baseline INR and TTR were around 1.90 and 60%, respectively. In some regions, for example, Hokuetsu, Kansai, and Kyushu, anticoagulation intensity and quality were sufficiently high to prevent thromboembolic events, although they would have increased the risk of major hemorrhagic events (Table 3). In other regions, however, for example, Shikoku, Hokkaido, and Northern Kanto, anticoagulation intensity and quality were not high enough to prevent thromboembolic events, but this resulted in a lower rate of major hemorrhagic events. Factors other than anticoagulation intensity and quality might also have been involved in this trade-off between thromboembolic and major hemorrhagic events. One possible factor was a regional difference in international sensitivity index (ISI) of the thromboplastin reagent in the J-RHYTHM Registry. Median ISI was 1.09 for the whole institutions, but it was 1.64 and 1.58 for institutions in Hokuetsu and Shikoku, respectively.22
Higher ISI in Hokuetsu and Shikoku could have induced large variations in INR, and thereby affected the event rates.
Study Limitations
This study had several limitations, although the objective had been predetermined before the first patients were enrolled in the J-RHYTHM Registry.13
First, the number of patients recruited from each of the 10 geographic regions of Japan was based on population density.13
Accordingly, the number of patients enrolled was <400 in 3 regions; that is, Hokkaido, Hokuetsu, and Shikoku, and these small numbers of patients might have affected the present results. In addition, the enrolled patients might not have been representative of the NVAF patients in each region. The number of institutions that participated in this Registry was relatively small in some regions, that is, 7 in Hokkaido, 8 in Hokuetsu and 9 in Shikoku. Therefore, a deviation of one or 2 institutions might have had a large impact on the data of the region. Inter-institutional differences might have contributed to the regional difference in addition to patient factors. Second, >80% of the patients received warfarin; therefore, it is possible that many of the patients included in this study were warfarin experienced. This selection bias would have affected the event rate data because warfarin-experienced patients have lower thromboembolic event rates than warfarin-naïve patients.23,24
We did not have any data on the proportions of warfarin-naïve and -experienced patients in the J-RHYTHM Registry. Third, warfarin was the only anticoagulant used. Had the study included patients treated with DOAC instead of warfarin, the results could have been different.25
Only data on baseline medication were used for analysis; therefore, the effects of changes in the anticoagulants or other medications on outcomes were not taken into consideration. Finally, permanent NVAF emerged as an independent predictor of thromboembolic events, as reported previously.20
In the previous subanalysis of the J-RHYTHM Registry, however, permanent NVAF was not associated with thromboembolic events on multivariate analysis, when a different set of explanatory variables was selected.26
Therefore, the finding that permanent NAVF is an independent risk factor for thromboembolic events should be interpreted cautiously.
Conclusions
Thromboembolic risk level, frequency of warfarin use, and intensity and quality of warfarin treatment in NVAF patients differed between the 10 geographic regions of Japan. Region was found to be an independent predictor of thromboembolic events, even after adjusting for baseline clinical characteristics. Unmeasured confounding factors, however, might have been involved in the regional differences in the frequency of thromboembolic events. Further studies are needed to elucidate the mechanisms underlying the regional differences observed in the present study.
Acknowledgments
The present study was supported by a grant from the Japan Heart Foundation, Tokyo, Japan. The participating physicians are listed in References 13 and 14.
Disclosures
H.I. received remuneration from Daiichi-Sankyo, Bayer Healthcare, and Bristol-Meyers Squibb; H.A. received research funding from Boehringer Ingelheim and remuneration from Bayer Healthcare, Boehringer Ingelheim, and Daiichi-Sankyo; K.O. received research funding from Boehringer Ingelheim and Daiichi-Sankyo and remuneration from Boehringer Ingelheim, Bayer Healthcare, Daiichi-Sankyo, and Pfizer; T.Y. received research funding from Daiichi-Sankyo, Bayer Healthcare, Tanabe-Mitsubishi, and Bristol-Meyers Squibb and remuneration from Boehringer Ingelheim, Daiichi-Sankyo, Pfizer, Bayer Healthcare, Bristol-Myers Squibb, and Eisai; H.O. received remuneration from Daiichi-Sankyo and Bayer Healthcare; I. Kubota received research funding from Daiichi-Sankyo and remuneration from Daiichi-Sankyo; K.M. received remuneration from Boehringer Ingelheim; M.C. received remuneration from Boehringer Ingelheim; Y. Koretsune received remuneration from Daiichi-Sankyo, Bayer Healthcare, Bristol-Meyers Squibb. Boehringer Ingelheim, and Pfizer; Y.O. received remuneration from Bristol-Meyers Squibb. Boehringer Ingelheim, and Pfizer; and S.B. received remuneration from Bayer Healthcare, Daiichi-Sankyo, and Boehringer Ingelheim.
References
- 1.
Le Heuzey JY, Breithardt G, Camm J, Crijins H, Dorian P, Kowey PR, et al. The RecordAF study: Design, baseline data, and profile of patients according to chosen treatment strategy for atrial fibrillation. Am J Cardiol 2010; 105: 687–693.
- 2.
Amerena J, Chen SA, Sriratanasathavorn C, Chao JG, Huang D, Omar R, et al. Insights into management of atrial fibrillation in Asia Pacific gained from baseline data from Registry on Cardiac rhythm disORDers (RecordAF-Asia Pacific [AP]) registry. Am J Cardiol 2012; 109: 378–382.
- 3.
Kakkar AK, Mueller I, Bassand JP, Fitzmaurice DA, Goldhaber SZ, Goto S, et al. Risk profiles and antithrombotic treatment of patients newly diagnosed with atrial fibrillation at risk of stroke: Perspectives from the international, observational, prospective GARFIELD Registry. PLOS One 2013; 8: e63479, doi:10.1371/journal.pone.0063479.
- 4.
Shimada YJ, Yamashita T, Koretsune Y, Kimura T, Abe K, Sasaki S, et al. Effects of regional differences in Asia on efficacy and safety of edoxaban compared with warfarin: Insights from the ENGAGE AF-TIMI 48 trial. Circ J 2015; 79: 2560–2567.
- 5.
Huisman MV, Rothman KJ, Paquette M, Teutsch C, Diener HC, Dubner SJ, et al. Antithrombotic treatment patterns in patients with newly diagnosed nonvalvular atrial fibrillation: The GLORIA-AF Registry, phase II. Am J Med 2015; 128: 1306–1313.
- 6.
Stafford RS, Singer DE. National patterns of warfarin use in atrial fibrillation. Arch Intern Med 1996; 156: 2537–2541.
- 7.
Ogawa S, Aonuma K, Tse HF, Huang D, Huang JL, Kalman J, et al. The APHRS’s 2013 statement on antithrombotic therapy of patients with nonvalvular atrial fibrillation. J Arrhythmia 2013; 29: 190–200.
- 8.
Singer DE, Hellkamp AS, Piccini JP, Mahaffey KW, Lokhnygina Y, Pan G, et al. Impact of global geographic region on time in therapeutic range on warfarin anticoagulation therapy: Data from the ROCKET AF clinical trial. J Am Heart Assoc 2013; 2: e000067, doi:10.1161/JAHA.112.000067.
- 9.
Lip GYH, Laroche C, Boriani G, Dan GA, Santini M, Kalarus Z, et al. Regional differences in presentation and treatment of patients with atrial fibrillation in Europe: A report from the EURObservational Research Programme Atrial Fibrillation (EORP-AF) Pilot General Registry. Europace 2015; 17: 194–206.
- 10.
Hori M, Connolly SJ, Zhu J, Liu LS, Lau CP, Pais P, et al. Dabigatran versus warfarin: Effects on ischemic and hemorrhagic strokes and bleeding in Asians and non-Asians with atrial fibrillation. Stroke 2013; 44: 1891–1896.
- 11.
Wong KSL, Hu DY, Oomman A, Tan RS, Patel MR, Singer DE, et al. Rivaroxaban for stroke prevention in east Asian patients from the ROCKET AF trial. Stroke 2014; 45: 1739–1747.
- 12.
Goto S, Zhu J, Liu L, Oh BH, Wojdyla DM, Aylward P, et al. Efficacy and safety of apixaban compared with warfarin for stroke prevention in patients with atrial fibrillation from East Asia: A subanalysis of the Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation (ARISTOTLE) trial. Am Heart J 2014; 168: 303–309.
- 13.
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.
- 14.
Atarashi H, Inoue H, Oukmura 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.
- 15.
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.
- 16.
JCS Joint Working Group. Guidelines for pharmacotherapy of atrial fibrillation (JCS 2013): Digest version. Circ J 2014; 78: 1997–2021.
- 17.
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.
- 18.
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.
- 19.
Obata H, Watanabe H, Ito M, Hirono S, Hanawa H, Kodama M, et al. Effects of combination therapy with warfarin and bucolome for anticoagulation in patients with atrial fibrillation. Circ J 2011; 75: 201–203.
- 20.
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.
- 21.
The Ministry of Health, Labor and Welfare, Japan. [Statistics of stroke.] http://www.mhlw.go.jp/bunya/shakaihosho/iryouseido01/pdf/kannrenjigyou-m.pdf (accessed February 17, 2016) (in Japanese).
- 22.
Kodani E, Okumura K, Inoue H, Yamashita T, Atarashi H, Origasa H. Present status and problems of international sensitivity index (ISI) on the management of international normalized ratio of prothrombin time (INR): Subanalysis of the J-RHYTHM Registry. Jpn J Electrocardiol 2011; 31: 225–233 (Abstract in English).
- 23.
Garcia DA, Wallentin L, Lopes RD, Thomas L, Alexander JH, Hylek EM, et al. Apixaban versus warfarin in patients with atrial fibrillation according to prior warfarin use: Results from the apixaban for reduction in stroke and other thromboembolic events in atrial fibrillation trial. Am Heart J 2013; 166: 549–558.
- 24.
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.
- 25.
Kodani E, Atarashi H, Inoue H, Okumura K, Yamashita T, Origasa H, et al. Beneficial effect of non-vitamin K antagonist oral anticoagulants in patients with non-valvular atrial fibrillation: Results of the J-RHYTHM Registry 2. Circ J 2016; 80: 843–851.
- 26.
Inoue H, Atarashi H, Okumura K, Yamashita T, Kumagai N, Origasa H; for the J-RHYTHM Registry Investigators. Thromboembolic events in paroxysmal versus permanent non-valvular atrial fibrillation: A sub-analysis of the J-RHYTHM Registry. Circ J 2014; 78: 2388–2393.
