Circ J 2004; 68: 417 – 421

Attitudes of Japanese Cardiologists Toward Anticoagulation

for Nonvalvular Atrial Fibrillation and Reasons
for Its Underuse

Hiroshi Inoue, MD; Takashi Nozawa, MD; Ken Okumura, MD*;

Atsushi Iwasa, MD*; Jong-Dae Lee, MD**; Akihiko Shimizu, MD†;
Motonobu Hayano, MD††; Katsusuke Yano, MD††

Background Although warfarin reduces embolic events in patients with nonvalvular atrial fibrillation
(NVAF), it is used less frequently in Japan and so the aim of the present study was to determine the attitudes of
Japanese cardiologists toward antithrombotic therapy for NVAF patients.
Methods and Results Subjects were NVAF patients enrolled in a prospective study in 1999. Clinical charac-
teristics, type of NVAF and antithrombotic therapy, risk factors for embolism, and contraindications to warfarin
were analyzed. Risk factors included advanced age (>75 years), hypertension, diabetes mellitus, congestive heart
failure, and prior embolic events. Contraindications to warfarin included bleeding tendency, malignant tumors
and others. Among 509 patients (66.6±10.3 years old), 359 had at least one risk factor for embolism and of these
359 patients, 200 (55.7%) received warfarin (ie, modest adherence to the guideline for antithrombotic therapy).
There were 159 patients who had at least one risk factor but did not receive warfarin; 70.4% of these received
antiplatelet drugs. Contraindications were found in only 22.6% and paroxysmal nature of NVAF seemed a possi-
ble reason for non-use of warfarin in 47.2% of 159 patients.
Conclusions In Japan warfarin is not used extensively for treatment of NVAF patients having risk factors and
the reasons for not using antithrombotic therapy seemed inappropriate in most of patients. (Circ J 2004; 68:
417 – 421)
Key Words: Antiplatelet; Embolism; Nonvalvular atrial fibrillation; Risk factors; Warfarin

R
andomized clinical trials clearly show that adjusted
dosage of warfarin decreases embolic episodes by
two-thirds in patients with nonvalvular atrial fib-
rillation (NVAF), and that aspirin only does so to a modest
extent.1 Together with progressive accumulation of data
supporting the efficacy of warfarin, its use has increased
gradually,2–7 but was still less than 50% even among pa-
tients with NVAF who were complicated with risk factors
for embolism.3,5,8 Previous clinical studies indicated that
50–60% of patients with NVAF did not have contraindica-
tions to warfarin use,4 and a recent guideline recommended
appropriate antithrombotic therapy and intensities of anti-
coagulation with warfarin according to the age and other
clinical characteristics of patients with NVAF.9
There are clear ethnic differences in the perceptions of
atrial fibrillation (AF) and anticoagulation therapy among
patients living in England,10 and possible ethnic differences
in the incidence of stroke.11 Therefore, there could be dif-
(Received December 26, 2003; revised manuscript received February
17, 2004; accepted March 3, 2004)
Second Department of Internal Medicine, Toyama Medical & Phar-
maceutical University, Toyama, *Second Department of Internal
Medicine, Hirosaki University, Hirosaki, **First Department of In-
ternal Medicine, Fukui Medical University, Fukui, †Faculty of Health
Science, Yamaguchi University School of Medicine, Ube, and ††Third
Department of Internal Medicine, Nagasaki University, Nagasaki,
Japan
Mailing address: Hiroshi Inoue, MD, The Second department of In-
ternal Medicine, Toyama Medical & Pharmaceutical University, 2630
Sugitani, Toyama 930-0194, Japan. E-mail: hiroshi@ms.toyama-mpu.
ac.jp
ferences between Japan and the USA in cardiologists’ atti-
tude toward antithrombotic therapy in patients with NVAF.
In the early 90 s Japanese cardiologists used warfarin in less
than 20% of patients with AF,12,13 so the aim of the present
study was to determine current attitudes toward warfarin
use, the proportion of adherence to the guideline for anti-
thrombotic therapy and reasons for non-use of warfarin in
patients with NVAF.
Methods
Study Subjects
We started a prospective, cooperative study from 1999
to determine the clinical efficacy of measuring hemostatic
markers to identify patients with NVAF who would be at
increased risk for embolic complication.14 The design of the
study was approved by the ethics committee of each uni-
versity hospital. Patients with NVAF treated in 1999 at 5
tertiary, university hospitals were subjects of the present
study. A total of 509 patients (66.6±10.3 years old, 330 men
and 179 women) with NVAF were enrolled in the study
after giving informed consent.14 Baseline clinical character-
istics, including diabetes mellitus, hypertension, New York
Heart Association functional status, prior cerebral infarc-
tion (CI) or transient ischemic attack (TIA), status of
antithrombotic therapy, and type of AF, were determined.
The selection of antithrombotic drugs was left to the treat-
ing cardiologist’s decision. Intensity of anticoagulation
with warfarin expressed as the international normalized
ratio (INR) of prothrombin time was determined at the time

Circulation Journal Vol.68, May 2004

418 INOUE H et al.

Table 1 Clinical Characteristics

All patients Warfarin Antiplatelets* No antithrombotic drugs

(n=509) (n=263) (n=163) (n=83)
Age (years) 66.6±10.3 67.0±9.0 67.5±11.0 63.4±12.0
Male 64.8 67.7 58.9 67.4
Paroxysmal AF 39.3 27.4 44.2 67.5
Risk factors
>75 years old 20.4 19.0 24.5 16.9
Hypertension 42.8 45.2 41.1 38.6
CHF 22.2 28.5 17.8 10.8
DM 15.3 16.7 13.5 13.3
Prior embolism 19.6 27.0 13.5 8.4
≥ one risk factor 70.5 76.0 68.7 56.7

Figs are mean ± standard deviation or %.

AF, atrial fibrillation; CHF, congestive heart failure, ie, New York Heart Association functional class II or more; DM, diabetes
mellitus.
*Antiplatelet therapy included aspirin and ticlopidine.

Table 2 Risk Factors and Distribution of Antithrombotic Therapy Among the Patients With NVAF

Distribution of antithrombotic therapy (%)

Recommendation of
n Warfarin
Antiplatelets None the guideline (9)
(INR)
Risk factors*
Present 337 57.9 29.4 12.7 Warfarin
(1.87±0.82)
Absent
Age
<65 years 88 37.5 35.2 27.3 Aspirin
(2.00±0.79)
65–75 years 62 48.4 32.3 19.3 Aspirin > Warfarin
(1.66±0.53)
>75 years 22 22.7 59.1 18.2 Warfarin > Aspirin
(1.90±0.80)
INR, international normalized ratio expressed as mean ± SD.
*Hypertension, diabetes mellitus, prior embolism, congestive heart failure.

of enrollment. received antiplatelet drugs such as aspirin and ticlopidine,

Patients were considered to have risk factors for embol-
ism if they had any of the following: previous CI including
TIA or systemic embolic event, hypertension, congestive
heart failure, diabetes mellitus or age >75 years.9 Patients
were considered to have a contraindication to warfarin ther-
apy if any of the following conditions were present: prior
history of bleeding, bleeding tendency, active gastric or
duodenal ulcer, malignant tumors, arteriovenous malfor-
mation, dementia, poor compliance, impaired hepatic func-
tion, or recurrent falls.15 The decision regarding impairment
of hepatic function was left to the cardiologist’s judgment.
Treating cardiologists were asked to give specific reason or
reasons for not giving warfarin to the patients. The prefer-
ence of the treating cardiologists in the selection of anti-
thrombotic therapy was not included in this analysis.
Data are presented as mean value ± standard deviation, or
prevalence. Nonparametric data were compared with chi-
square test. A p-value less than 0.05 was considered statis-
tically significant.
Results
The clinical characteristics and antithrombotic therapy
of the 509 patients with NVAF are summarized in Table 1.
Of the 509 patients, 263 (51.7%) received anticoagulation
with warfarin. Of the remaining 246 patients, two-thirds
and one-third did not receive either anticoagulant or anti-
platelet treatment. Paroxysmal AF was more frequently
seen in the patient group not receiving any antithrombotic
treatment (p<0.001). In contrast, prior embolism and con-
gestive heart failure were found more in the patient group
receiving warfarin (p<0.001). Of the 263 patients receiving
warfarin, 76.0% had at least one risk factor for embolism.
Surprisingly, 56.7% of patients who did not receive any
antithrombotic therapy had at least one risk factor.
Table 2 summarizes the distribution of antithrombotic
therapy and intensities of anticoagulation in relation to risk
factors and age. Those without 4 specific conditions in-
creasing embolic risk were divided into 3 groups according
to their age.9 Nearly 60% of patients with risk factors other
than advanced age were treated according to the guideline;9
that is, they received warfarin. Among patients who did not
have any of the 4 specific risk factors, the treatment of a
considerable number did not adhere to the guideline. For
instance, of those who were 75 years old or younger, ap-
proximately 40–50% received warfarin compared with
only one-fifth of those older than 75 years. In total, 55% of
patients having one or more risk factors for embolism,
including advanced age, received warfarin.
Mean intensities of anticoagulation with warfarin ranged
from INR 1.66 to 2.00, which were slightly lower than those
recommended by the guideline,9 especially for those with

Circulation Journal Vol.68, May 2004

Warfarin Use and AF
Table 3 Reasons for Non-Use of Warfarin
All patients
(n=246)
Contraindications
Bleeding tendency
Active ulcer
Malignant tumor
AV malformation
Dementia
Poor compliance
Hepatic dysfunction
Recurrent falls
Any of the above contraindications
Paroxysmal AF†
No apparent reason
Figs are %. AV, arteriovenous; AF, atrial fibrillation.
*Risk factors for embolism included advanced age (>75 years), hypertension, diabetes mellitus, congestive heart failure, and prior
embolic events.
†Although these patients did not have any contraindications to warfarin, warfarin was not given possibly because of the
paroxysmal nature of the AF.
any of the 4 specific risk factors (INR: 1.87±0.80, Table 2).
Table 3 summarizes the reasons for not using warfarin.
There were 159 patients who had risk factors including
advanced age (>75 years), but were not given warfarin, and
contraindications to warfarin use were found in only 22.6%
of these patients. Paroxysmal AF was found in 47.2%, and
this could be the possible reason for not giving warfarin.
The remaining patients (30.2%) in whom anticoagulation
was indicated and who had a chronic form of AF, but did
not have contraindications, did not receive warfarin.
Discussion
Major Findings
First, only 55% of NVAF patients having risk factors for
embolism, including advanced age, received anticoagula-
tion with warfarin; that is, warfarin use was still insufficient
among patients with NVAF who were treated by cardiolo-
gists at tertiary university hospitals in Japan. The intensities
of warfarin therapy in the present study were slightly lower
than those recommended by the guideline.9 Lower anticoag-
ulation intensities could be based on data from prospective,
secondary prevention trials performed in Japan.16
Second, in many NVAF patients without the 4 specific
conditions that could increase embolic risk, the antithrom-
botic therapy did not follow the guideline.9 Warfarin rather
than antiplatelet drugs was given more frequently to pa-
tients aged 75 years old or younger. Only one fifth of pa-
tients aged more than 75 years were given warfarin.
Third, contraindications were found in only 22.6% of
NVAF patients who had at least one risk factor for embol-
ism, including advanced age, but were not given warfarin.
Nearly half of these NVAF patients were probably not
given warfarin because of the paroxysmal nature of their
AF. The appropriateness of this decision is discussed later.
Frequency of Warfarin Use
With progressive accumulation of results of randomized
clinical trials of effects of antithrombotic therapy in pa-
tients with NVAF,1 warfarin use has increased gradually in
the USA,2–7 although a recent analysis revealed that its use
in that country remains at approximately 50% of NVAF
patients,4 which is also true for warfarin use in Japan. Previ-
ously warfarin use was 8–17% of patients with AF,12,13 but
Circulation Journal Vol.68, May 2004
419
Patients with risk factors for embolism*
(n=159)
3.3 4.4
5.3 6.9
5.7 5.7
0.8 1.3
2.4 3.1
12.2 14.5
6.5 4.4
2.4 2.5
18.3 22.6
52.0 47.2
29.7 30.2
had increased to 51.7% of NVAF patients in the present
study.
The present study indicated that adherence to the guide-
line of antithrombotic therapy in patients with NVAF was
still insufficient. Only 55% of NVAF patients who had risk
factors including advanced age were anticoagulated with
warfarin. In contrast, Bradley et al showed that warfarin
was prescribed to 67% of 750 patients with AF in whom
warfarin was indicated and those authors attributed the
higher rate of warfarin use, at least in part, to the teaching
hospital environment of their study.7 The present study was
also carried out at university hospitals, but warfarin use
was a little short of that of Bradley et al.7
Advanced age itself is considered a risk factor for
embolism9 and in the present study there were 104 patients
with NVAF aged >75 years, of whom 48% received
warfarin. However, when analysis was confined to patients
having only the risk factor of advanced age (>75 years),
warfarin was given to only 23% of them. Warfarin use was
less frequent for older patients in previous studies,2–4,8,17
possibly because of higher risk of major bleeding, recurrent
falls and other complications in the elderly patients.15,17
However, warfarin can be used safely and efficiently
among older patients (≥90 years old) who did not have risk
factors for intracranial bleeding.18
In contrast, warfarin was given more frequently than anti-
platelet drugs to NVAF patients without any risk factors for
embolism, including advanced age; 38% of 88 NVAF
patients younger than 65 years old and without risk factors
were given warfarin, a finding of possible overuse of
warfarin and could be partly related to the conviction of
some cardiologists that the presence of NVAF itself implies
an increased risk of embolism. Lower risk of bleeding with
anticoagulation in younger patients would promote warfarin
use in these patients without risk factors for embolism.
Intensities of Anticoagulation
The intensity of anticoagulation with warfarin deserves
some comments. Based on clinical trials, an INR of 2–3 is
recommended for prevention of embolic events in patients
with NVAF,19,20 but for Japanese patients, an INR of 1.6–2.6
is considered appropriate, based on prospective, secondary
prevention trials.16 Therefore, in the present study the inten-
sity of anticoagulation among patients with risk factors
420
followed the Japanese guideline well.16
Reasons for Non-Use of Warfarin
Previous studies indicate that approximately 50–60% of
NVAF patients do not have contraindications to anticoagu-
lation with warfarin,4,5 and the risk of bleeding seems the
leading cause of hesitation in using warfarin.4,17 Our results
were quite different from those previous studies. In the
present study, patients’ poor compliance was the most fre-
quent contraindication (14.5%) to warfarin use, compared
with only 9% of patients not receiving warfarin in the study
of Bradley et al.7
Nearly half of the present patients who had risk factors
for embolism but no contraindications to warfarin had
paroxysmal AF and the paroxysmal nature of the AF might
be the reason why the cardiologist did not prescribe
warfarin therapy. Bradley et al showed that warfarin was
not given to 45% of 494 patients because of paroxysmal or
lone AF at a veterans affairs medical center.7 However, the
decision to not give warfarin to patients because of the
paroxysmal nature of their AF could be inappropriate. Pro-
spective trials have indicated that the incidence of embolic
events did not differ between patients with paroxysmal and
chronic AF,21,22 and this is also true for Japanese patients
with AF.23
Surprisingly, there were 30% of NVAF patients who had
risk factors for embolism but did not have any apparent
reasons for the non-use of warfarin including paroxysmal
AF. This frequency of non-use of warfarin was, however,
similar to that in the study by Bradley et al.7
Study Limitations
First, in the analyses of contraindications, patients’
refusal and their life expectancy were not included as
reasons for non-use of warfarin. Cardiologists’ hesitation in
using warfarin because of the frequent, complicated moni-
toring of anticoagulation intensity was not included in the
present study either. Second, we could not be certain that
patients with paroxysmal AF and risk factors for embolism
had not been given anticoagulant therapy because of the
paroxysmal nature of their AF. The incidence of CI might
be lower in Japanese subjects than in Caucasian sub-
jects,11,24 which could lead, at least in part, to the lower rate
of warfarin use, especially in patients with paroxysmal AF,
in the present study. However, a previous epidemiologic
study found that the incidence of embolism among
Japanese patients with NAVF did not differ from that
among Caucasian patients with NVAF.25 Third, the present
data were collected from tertiary university hospitals, and
might not represent nationwide trends of warfarin use in
Japan. Finally, the mean age of the patients in the present
study was 66.6 years, which was lower than in previous
studies4,5,15 and could affect the proportion of warfarin use.
Clinical Implications
Although limited for the reasons given, the present study
clearly indicates warfarin is still insufficiently used as anti-
thrombotic therapy for patients with NVAF in Japan. Many
NVAF patients having risk factors for embolism are not
given warfarin for inappropriate reasons; the paroxysmal
nature of AF appears to be the leading cause of non-use of
warfarin. Although transesophageal echocardiographic
findings might be useful for risk stratification of embolism
and more sophisticated selection of antithrombotic drugs in
NVAF patients,26,27 we believe that the attitude of Japanese
INOUE H et al.
cardiologists toward anticoagulation in patients with NVAF
needs to be changed.
Acknowledgments
This study was supported by a grant-in-aid for Scientific Research from
the Japanese Ministry of Education, Science and Culture, Japan.
The following investigators participated in this study. Toyama Medical
& Pharmaceutical University: Tadakazu Hirai, MD, Akira Fujiki, MD,
Koichi Mizumaki, MD, Hidetsugu Asanoi, MD; Hirosaki University:
Shingo Sasaki, MD, Takumi Higuma, MD; Fukui Medical University:
Hiromasa Shimizu, MD, Hiroyasu Uzui, MD; Yamaguchi University:
Toshihiko Yamagata, MD, Masunori Matsuzaki, MD; and Nagasaki Uni-
versity: Genji Toda, MD, Shojiro Isomoto, MD, Norihiro Komiya, MD.
References
1. 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.
2. Stafford RS, Singer DE. National patterns of warfarin use in atrial
fibrillation. Arch Intern Med 1996; 156: 2537 – 2541.
3. Stafford RS, Singer DE. Recent national patterns of warfarin use in
atrial fibrillation. Circulation 1998; 97: 1231 – 1233.
4. Smith NL, Psaty BM, Furberg CD, White R, Lima JA, Newman AB,
et al. Temporal trends in the use of anticoagulants among older adults
with atrial fibrillation. Arch Intern Med 1999; 159: 1574 – 1578.
5. Albers GW, Yim JM, Belew KM, Bittar N, Hattemer CR, Phillips
BG, et al. Status of antithrombotic therapy for patients with atrial
fibrillation in university hospitals. Arch Intern Med 1996; 156:
2311 – 2316.
6. Munschauer FE, Priore RL, Hens M, Castilone A. Thromboembolism
prophylaxis in chronic atrial fibrillation: Practice patterns in commu-
nity and tertiary-care hospitals. Stroke 1997; 28: 72 – 76.
7. Bradley BC, Perdue KS, Tisdel KA, Gilligan DM. Frequency of anti-
coagulation for atrial fibrillation and reasons for its non-use at a
veterans affairs medical center. Am J Cardiol 2000; 85: 568 – 572.
8. Brass LM, Krumholz HM, Scinto JM, Radford M. Warfarin use
among patients with atrial fibrillation. Stroke 1997; 28: 2382 – 2389.
9. Cairns JA. Atrial fibrillation: Antithrombotic therapy. In: Yusuf S,
Cairns JA, Camm AJ, Fallen EL, Gersh BJ, editors. Evidence-based
cardiology. London: BMJ Books; 1998; 544 – 552.
10. Lip GYH, Kamath S, Jafri M, Mohammed A, Bareford D. Ethnic
differences in patient perceptions of atrial fibrillation and anticoagu-
lation therapy: The West Birmingham Atrial Fibrillation Project.
Stroke 2002; 33: 238 – 244.
11. Rodriguez BL, D’Agostino R, Abbott RD, Kagan A, Burchfiel CM,
Yano K, et al. Risk of hospitalized stroke in men enrolled in the
Honolulu heart program and the Framingham study: A comparison
of incidence and risk factor effects. Stroke 2002; 33: 230 – 237.
12. Research Group for Antiarrhythmic Drug Therapy. Survey of atrial
fibrillation and thromboembolism in the elderly: A multicenter coop-
erative study. J Cardiol 1999; 33: 27 – 35 (English abstract).
13. Tomita F, Kohya T, Sakurai M, Kaji T, Yokoshiki H, Sato M, et al.
for the Hokkaido Atrial Fibrillation Study Group. Prevalence and
clinical characteristics of patients with atrial fibrillation: Analysis of
20,000 cases in Japan. Jpn Circ J 2000; 64: 653 – 658.
14. Nozawa T, Inoue H, Iwasa A, Okumura K, Lee JD, Shimizu A, et al.
Effects of anticoagulation intensity on hemostatic markers in patients
with non-valvular atrial fibrillation. Circ J 2004; 68: 29 – 34.
15. McCormick D, Gurwitz JH, Goldberg RJ, Becker R, Tate JP, Elwell
A, et al. Prevalence and quality of warfarin use for patients with atrial
fibrillation in the long-term care setting. Arch Intern Med 2001; 161:
2458 – 2463.
16. Yasaka M, Minematsu K, Yamaguchi T. Optimal intensity of inter-
national normalized ratio in warfarin therapy for secondary preven-
tion of stroke in patients with non-valvular atrial fibrillation. Intern
Med 2001; 40: 1183 – 1188.
17. McCrory DC, Matchar DB, Samsa G, Sanders LL, Pritchett ELC.
Physician attitudes about anticoagulation for nonvalvular atrial fibril-
lation in the elderly. Arch Intern Med 1995; 155: 277 – 281.
18. Joffe HV, Goldhaber SZ. Effectiveness and safety of long-term anti-
coagulation of patients ≥90 years of age with atrial fibrillation. Am J
Cardiol 2002; 90: 1397 – 1398.
19. Hylek EM, Skates SJ, Sheehan MA, Singer DE. An analysis of the
lowest effective intensity of prophylactic anticoagulation for patients
with nonrheumatic atrial fibrillation. N Engl J Med 1996; 335: 540 –
546.
Circulation Journal Vol.68, May 2004
Warfarin Use and AF
20. Hylek EM, Singer DE. Risk factors for intracranial hemorrhage in
outpatients taking warfarin. Ann Intern Med 1994; 120: 897 – 902.
21. The Stroke Prevention in Atrial Fibrillation Investigators. Predictors
of thromboembolism in atrial fibrillation: I. Clinical features of pa-
tients at risk. Ann Intern Med 1992; 116: 1 – 5.
22. The Boston Area Anticoagulation Trial for Atrial Fibrillation Inves-
tigators. The effect of low-dose warfarin on the risk of stroke in
patients with nonrheumatic atrial fibrillation. N Engl J Med 1990;
323: 1505 – 1511.
23. Inoue H, Atarashi H for the Research Group for Antiarrhythmic Drug
Therapy. Risk factors for thromboembolism in patients with parox-
ysmal atrial fibrillation. Am J Cardiol 2000; 86: 852 – 855.
24. Tohgi H, Tajima T, Konno T, Towada S, Kamata A, Yamazaki M.
The risk of cerebral infarction in nonvalvular atrial fibrillation:
Circulation Journal Vol.68, May 2004
421
Effects of age, hypertension and antihypertensive treatment. Eur
Neurol 1991; 31: 126 – 130.
25. Tanaka H, Hayashi M, Date C, Imai K, Asada M, Shoji H, et al.
Epidemiologic studies of stroke in Shibata, a Japanese provincial
city: Preliminary report on risk factors for cerebral infarction. Stroke
1985; 16: 773 – 780.
26. Shinokawa N, Hirai T, Takashima S, Kameyama T, Nakagawa K,
Asanoi H, et al. A transesophageal echocardiographic study on risk
factors for stroke in elderly patients with atrial fibrillation: A com-
parison with young patients. Chest 2001; 121: 840 – 846.
27. Abe Y, Asakura T, Sakamoto N, Ishikawa S, Muroi S, Saitoh F, et al.
Embolic attack in patients with atrial fibrillation and atrial thrombus
depends on the character of the thrombus. Circ J 2003; 67: 203 – 208.