Tissue Plasminogen Activator

Thrombolytic Therapy for AcuteIschemic Stroke in 4 Hospital Groups in Japan

Syoichiro Kono, MD,* Kentaro Deguchi, MD,* Nobutoshi Morimoto, MD,*

Tomoko Kurata, MD,* Shoko Deguchi, MD,* Tohru Yamashita, MD, PhD,*

Yoshio Ikeda, MD, PhD,* Tohru Matsuura, MD,* Hisashi Narai, MD, PhD,†

Nobuhiko Omori, MD, PhD,† Yasuhiro Manabe, MD, PhD,† Taijyun Yunoki, MD,‡

Yoshiki Takao, MD, PhD,‡ Sanami Kawata, MD, PhD,x Kenichi Kashihara, MD, PhD,xand Koji Abe, MD, PhD*

From the *Departmen

cine, Dentistry and Pha

Okayama, Japan; †Depa

Hospital Medical Center

logy, Kurashiki Heisei H

of Neurology, Okayama

Received March 16, 20

July 21, 2011.

Supported in part by

21390267 from theMinist

190

In October 2005 in Japan, the recombinant tissue plasminogen activator (tPA) alte-

plase was approved for patients with acute ischemic stroke within 3 hours of onset

at a dose of 0.6 mg/kg. The present study was undertaken to assess the safety and

efficacy of alteplase in Japan. Between October 2005 and December 2009, a total of

114 consecutive patients admitted to 4 hospitals received intravenous tPA within

3 hours of stroke onset. Clinical backgrounds and outcomes were investigated.

The patients were divided into 2 chronological groups: an early group, comprising

45 patients treated between October 2005 and December 2007, and a later group,

comprising 69 patients treated between January 2008 and December 2009. The

mean time from arrival at the hospital to the initiation of treatment was significantly

reduced in the later group, from 82.6 minutes to 70.9 minutes. Intracerebral hemor-

rhage (ICH) occurred in 26 patients (22.8%); compared with patients without ICH,

these patients had a significantly higher prevalence of cardiogenic embolism

(88.5% vs 58.0%); greater warfarin use (26.8% vs 6.8%); higher mean National Insti-

tutes of Health Stroke Scale (NIHSS) scores on admission (16 vs 10), at 3 days after

admission (14 vs 5), and at 7 days after admission (13.5 vs 3); and a lower Diffusion-

Weighted Imaging–Alberta Stroke Program Early CT Score (7.8 vs 9.1). Patients who

received edaravone had a higher prevalence of cardiogenic embolism (70.9% vs

36.4%), a higher recanalization rate (77.7% vs 36.4%), and lower NIHSS scores on

admission and at 3 and 7 days after admission compared with those who did not

receive edaravone. Our data suggest that administration of intravenous alteplase

0.6 mg/kg within 3 hours of stroke onset is safe and effective, that the NIHSS and

Diffusion-Weighted Imaging–Alberta Stroke Program Early CT Score are useful

predictors of ICH after tPA administration, and that warfarin-treated patients are

more likely to develop symptomatic ICH despite an International Normalized

Ratio ,1.7. Key Words: tPA—acute ischemic stroke—intracerebral hemorrhage—

DWI-ASPECTS—cardiogenic embolism—warfarin—edaravone.

� 2013 by National Stroke Association

t of Neurology, Graduate School of Medi-

rmaceutical Sciences, Okayama University,

rtment of Neurology, Okayama National

, Okayama, Japan; ‡Department of Neuro-

ospital, Okayama, Japan; and xDepartment

Kyokuto Hospital, Okayama, Japan.

11; revision received June 22, 2011; accepted

Grant-in-Aid for Scientific Research (B)

ry of Education, Science, Culture and Sports

of Japan, Grants-in-Aid from the Research Committee of CNS

Degenerative Diseases, and grants from theMinistry of Health, Labor,

and Welfare of Japan.

Address correspondence to Koji Abe, MD, PhD, Department of

Neurology, Graduate School of Medicine, Dentistry and Pharmaceu-

tical Sciences, Okayama University, 2-5-1 Shikatacho Kitaku,

Okayama 700-8558, Japan. E-mail: skono@cc.okayama-u.ac.jp.

1052-3057/$ - see front matter

� 2013 by National Stroke Association

doi:10.1016/j.jstrokecerebrovasdis.2011.07.016

Journal of Stroke and Cerebrovascular Diseases, Vol. 22, No. 3 (April), 2013: pp 190-196

TISSUE PLASMINOGEN ACTIVATOR THERAPY FOR STROKE IN OKAYAMA 191

Stroke is a serious disease leading to death and severe

disability. Although significant progress has been made

in the prevention and treatment of stroke, challenges re-

main. Because ischemic brain damage depends on both

the period after onset and the degree of ischemia, decreas-

ing the time between stroke onset to the start of treatment

is crucial to improving both factors simultaneously.

In October 2005, the recombinant tissue plasminogen

activator (tPA) alteplase was approved for patients with

acute ischemic stroke in Japan at a dose of 0.6 mg/kg.1

The free radical scavenger edaravone, approved in Japan

in 2001, has been shown to provide effective neuroprotec-

tive treatment in cerebral infarction. Thus, most patients

with ischemic stroke in Japan who received tPA treatment

also receive edaravone. Because intracerebral hemor-

rhage (ICH) is a frequent complication of tPA therapy,1

the use of tPA requires careful evaluation of both the risks

and potential benefits. We recently reported that edara-

vone prevents dissociation of the neurovascular unit,

reducing hemorrhagic transformation.2

During the more than 5 years since altepase was ap-

proved for use in Japan, several clinical parameters may

have changed, such as prognosis, frequency of hemor-

rhagic complications, time to treatment, and others. We

retrospectively analyzed such clinical parameters and

compared them in 2 groups of patients treated in our

4 hospitals: an early group, comprising 45 patients treated

between October 2005 and December 2007, and a later

group, comprising 69 patients treated between January

2008 and December 2009. We report the current status

of intravenous (IV) tPA therapy for acute ischemic stroke

in our 4 hospitals.

Materials and Methods

Patients

This study retrospectively evaluated 114 consecutive

patients (73 men and 41 women) aged 38-92 years

(mean age, 71.8 6 11.6 years) with acute cerebral infarc-

tion who were admitted to Okayama University Hospital,

Okayama National Hospital Medical Center, Kurashiki

Heisei Hospital, and Okayama Kyokuto Hospital be-

tween October 2005 and December 2009 (3 years and

2 months). Between 3 and 15 neurologists were present

during working hours in each hospital, and 1 neurologist

was reserved on call during weekends and after hours. In

cases of suspected tPA use, the stroke teammembers were

called and recruited before the arrival of the suspected

stroke patient. One hospital had a stroke care unit in

which stroke nurses served around the clock. In the other

hospitals, patients were treated in high care units or inten-

sive care units. All hospitals had a speech pathologist,

physiotherapist, occupational therapist, dietician, rehabili-

tation specialist, and medical social worker working as

a well-coordinated team. All patients received IV tPA

within 3 hours of onset.

Risk factors for acute ischemic stroke, including hyper-

tension, diabetes mellitus, hyperlipidemia, coronary

artery disease, a history of smoking for more than

2 months, and previous stroke, were evaluated. Other

clinical factors and outcomes were investigated as well,

including oral antithrombotic use at stroke onset due to

ischemic stroke, systolic and diastolic blood pressure be-

fore treatment, interval between arrival at the hospital

and the start of treatment, and use of edaravone.

Clinical Diagnosis and Evaluation

On admission, all patients were evaluated by a neurolo-

gist using the National Institutes of Health Stroke Scale

(NIHSS). Magnetic resonance imaging with diffusion-

weighted imaging and T2-weighted or fluid-attenuated

inversion recovery was performed to identify fresh infarc-

tions, and magnetic resonance angiography was done to

evaluate cerebral artery occlusions. Laboratory blood test-

ing, 12-lead electrocardiography, and plain chest X-rays

were performed in all patients. All patients were evalu-

ated with the NIHSS on admission, and 66 patients also

were evaluated at 3 days and 7 days after admission.

Infarct size was evaluated using the Diffusion-Weighted

Imaging–Alberta Stroke Program Early CT Score

(DWI-ASPECTS).3,4 Recanalization was determined by

magnetic resonance angiography at 24–72 hours after

treatment. Symptomatic ICH was defined using the Safe

Implementation of Thrombolysis in Stroke Monitoring

Study (SITS-MOST) criteria ($4-point increase in NIHSS

score from baseline or death within 36 hours and local

or remote parenchymal hemorrhage type 2 with a dense

hematoma of.30% of the lesion volumewith a significant

space-occupying effect).5

Statistical Analysis

The Student t test, Welch’s t test, and the c2 test were

used for statistical analyses. Multiple logistic regression

analysis was used to compare characteristics in the pa-

tients with and without ICH. A P value ,.05 was consid-

ered statistically significant.

Results

Of the 2936 patients presenting with stroke in our 4 hos-

pitals during the study period, 554 (18.9%) arrived at

hospital within 3 hours of onset, and 114 (3.9%) received

IV tPA. These 114 patients’ baseline clinical characteris-

tics, risk factors for stroke, type of cerebral infarction,

responsible occluded vessel, and clinical outcomes are

summarized in Table 1. Among these 114 patients,

77 (67.5%) had a cardiogenic embolism, 38 (33.3%) had

atherothrombosis, and 2 (1.8%) had a lacunar infarction.

The responsible occluded vessel was the middle cerebral

artery in 68 patients (59.6%) and the internal carotid

artery in 17 patients (14.9%). The mean NIHSS score on

Table 1. Clinical characteristics and outcomes of patient

Total (n5114)

n %

Male 73 64.0

Mean age 71.8

Medical history

Hypertension 50 43.9

Diabetes mellitus 22 19.3

Hyperlipidemia 21 18.4

Ischemic heart disease 16 14.0

Smoker 29 25.4

Stroke 24 21.1

Type of cerebral infarction

Cardiogenic embolism 77 67.5

Atherothrom bosis 38 33.3

Lacunar 2 1.8

Others 1 0.9

Responsible occluded vessel

Middle cerebral artery 68 59.6

Internal carotid artery 17 14.9

Anterior cerebral artery 2 1.8

Posterior cerebral artery 4 3.5

Basilar artery 6 5.3

Vertebral artery 4 3.5

Others 13 11.4

Systolic blood pressure 152.6

Diastolic blood pressure 79.2

Median NIHSS on admission (IQR) 10.5 (5-18.75)

Mean DW I-ASPECTS 8.7

Recanalization 84 73.7

Edaravone 103 90.4

Oral antiplatelet drug 33 28.9

Aspirin 16 14.0

Cilostazol 1 0.9

Clopidogrel 2 1.8

Ticlopidine 3 2.6

Aspirin + Clopidogrel 3 2.6

Aspirin + Ticlopidine 2 1.8

Aspirin + Cilostazol 3 2.6

Aspirin + Warfarin 3 2.6

Warfarin 13 11.4

Mean time from stroke onset to

arriving at hospital (min)

55.1

Mean time from arriving at hospital

to start of treatment (min)

75.5

Mean time from stroke onset to

start of treatment (min)

130.6

Bleeding complications 38 33.3

Asymptomatic intracerebral

hemorrhage

23 20.2

Symptomatic intracerebral

hemorrhage

3 2.6

Bleeding outside brain 12 10.5

Abbreviation: IQR, interquartile range.

Figure 1. When the total 114 patients were chronologically divided into 2

groups, 45 patients were treated from October 2005 to December 2007 (the

early group), and 69 patients from January 2008 to December 2009 (the later

group). Compared to the early group, the mean time from arriving at hospital

to start of treatment was significantly reduced in later group from 82.6 to

70.9 min (p , 0.05). As a result, our hospital groups became able to extend

to accept patients with mean time from stroke onset to arriving from 50.2 to

58.2 min in later group (p 5 n.s.).

S. KONO ET AL.192

admission was 12.8 6 8.7. Thirty-three patients (28.9%)

had received an oral antiplatelet drug, and 13 (11.4%)

had received warfarin. Bleeding complications were

seen in 38 patients (33.3%), and ICH in 26 patients

(22.8%).

The 114 patients were chronologically divided into

2 groups, an early group comprising 45 patients treated

between October 2005 and December 2007, and a later

group, comprising 69 patients treated between January

2008 and December 2009. For the entire study group,

the mean interval from stroke onset to arrival at the hos-

pital was 55.1 minutes, and the mean interval from hospi-

tal arrival to the initiation of treatment was 75.5 minutes

(Fig 1). Compared with the early group, the later group

had a significantly shorted mean interval from hospital

arrival to treatment initiation (70.9 minutes vs 82.6 min-

utes; P , .05). As a result, our hospitals were able to

accept patients with a mean time from stroke onset to

arrival of 50.2–58.2 minutes in the later group (Fig 1).

To analyze bleeding complications in detail, we further

divided the 114 patients into 2 groups, those with ICH and

those without ICH (Table 2). Thirty-eight patients had

bleeding complications (33.3%), 23 patients (20.2%) had

asymptomatic ICH, and 3 patients (2.6%) had symptom-

atic ICH (Table 1). Compared with the non-ICH group,

the ICH group had a significantly greater proportion of

cardiogenic embolism (88.5% vs 58.0%: P, .01), more fre-

quent recanalization (88.5% vs 69.3%), more frequent use

of warfarin (26.8% vs 6.8%; P , .01); higher mean NIHSS

Table 2. Comparison of intracerebral hemorrhage and non-hemorrhage groups

Intracerebral hemorrhage (n526) Non-hemorrhage (n588)

P valuen % n %

Male 16 61.5 57 64.8 0.7627

Mean age 74.5 71.1 0.1775

Medical history

Hypertension 8 30.8 42 47.7 0.1258

Diabetes mellitus 6 23.1 16 18.2 0.5784

Hyperlipidemia 4 15.4 17 19.3 0.6494

Ischemic heart disease 3 11.5 13 14.8 0.4794

Smoker 7 26.9 22 25.0 0.8432

Stroke 5 19.2 19 21.6 0.7954

Type of cerebral infarction

Cardiogenic embolism 23 88.5 51 58.0 0.0042**

Atherothrom bosis 3 11.5 34 38.6 0.0095**

Lacunar 0 0.0 2 2.3 0.9406

Others 0 0.0 1 1.1 0.7719

Responsible occluded vessel

Middle cerebral artery 15 57.7 53 60.2 0.8169

Internal carotid artery 7 26.9 10 11.4 0.0552

Anterior cerebral artery 0 0.0 2 2.3 0.9406

Posterior cerebral artery 2 7.7 2 2.3 0.2230

Basilar artery 1 3.8 5 5.7 0.5851

Vertebral artery 0 0.0 4 4.5 0.6170

Others 1 3.8 12 13.6 0.1509

Oral antiplatelet drug 4 15.4 28 31.8 0.1013

Aspirin 2 7.7 13 14.8 0.2827

Cilostazol 0 0.0 1 1.1 0.7719

Clopidogrel 0 0.0 2 2.3 0.9406

Ticlopidine 1 3.8 2 2.3 0.5436

Aspirin + Clopidogrel 0 0.0 3 3.4 0.4564

Aspirin + Ticlopidine 1 3.8 1 1.1 0.4057

Aspirin + Cilostazol 0 0.0 3 3.4 0.4564

Aspirin + Warfarin 0 0.0 3 3.4 0.4564

Warfarin 7 26.9 6 6.8 0.0098**

Mean time to treatment (min) 133.7 130.3 0.6459

Median NIHSS on admission (IQR) 16 (10.25-21.75) 10 (5-17.25) 0.0178*

Median NIHSS at 3days after onset (IQR) 14 (12.25-20) (n510) 5 (2-9.25) (n556) 0.0012**

Median NIHSS at 7days after onset (IQR) 13.5 (11.25-20.25) (n510) 3 (0.5-7) (n555) 0.0002**

Systolic blood pressure 150.9 149.9 0.8590

Diastolic blood pressure 81.2 78.5 0.4753

Edaravone 25 96.2 78 88.6 0.2321

Recanalization 23 88.5 61 69.3 0.0515

Mean DW I-ASPECTS 7.8 9.1 0.0077**

Abbreviation: IQR, interquartile range.

*P , 0.05.

**P , 0.01.

TISSUE PLASMINOGEN ACTIVATOR THERAPY FOR STROKE IN OKAYAMA 193

scores on admission (16 vs 10; P , .01), at 3 days after ad-

mission (14 vs 5; P , .01), and at 7 days after admission

(13.5 vs 3; P , .01), and lower mean DWI-ASPECTS (7.8

vs 9.1; P , .01) (Table 2). Serious symptom and large in-

farction range were more likely to raise the risk of ICH.

In the chronological group comparison, the rate of

bleeding complications did not differ in the early and

later groups (Table 3). In multiple logistic regression,

cardiogenic embolism (odds ratio [OR], 6.97; 95% confi-

dence interval [CI], 1.25-38.73; P 5 .027), baseline use of

warfarin (OR, 4.06; 95% CI, 1.08-15.26; P 5 .038), lower

DWI-ASPECTS (OR, 0.73; 95% CI, 0.58-0.92; P 5 .007),

and posterior cerebral artery infarction (OR, 24.64; 95%

CI, 2.79-222.53; P5 .004) were found to be significant pre-

dictors of ICH (Table 4).

A total of 103 of the 114 patients (90.4%) received edara-

voneasaneuroprotective therapy (Table5).Becauseof renal

dysfunction, 11 patients (9.6%) did not receive edaravone.

Table 3. Comparison between the former group and the latter group about bleeding complication

All years

(n5114)

The former group

(n545)

The latter group

(n569)

P Valuen % n % n %

Bleeding complications 38 33.3 16 35.6 22 31.9 0.6844

Asymptomatic intracerebral hemorrhage 23 20.2 9 20.0 14 20.3 0.9699

Symptomatic intracerebral hemorrhage 3 2.6 2 4.4 1 1.4 0.3431

Bleeding outside brain 12 10.5 5 11.1 7 10.1 0.5514

S. KONO ET AL.194

Compared with the non-edaravone group, the edaravone

group had a higher prevalence of cardiogenic embolism

(70.9% vs 36.4%; P , .05), a higher recanalization rate

(77.7% vs 36.4%; P , .01) and a lower mean NIHSS score

on admission and at 3 and 7 days after admission (Table 5).

Discussion

The baseline factors known to affect outcome in pa-

tients with acute ischemic stroke, including age, medical

history, type of cerebral infarction, and severity of

stroke, were comparable to those in reported in previous

studies.5-10 None of the baseline characteristics appeared

to favorably affect outcomes in the present study. The

number of patients receiving IV tPA has increased every

year in our hospitals. Compared with the early group,

the later group had a significantly shorter mean interval

from hospital arrival to initiation of treatment

(70.9 minutes vs 82.6 minutes; Fig 1). As a result, in the

later period, our hospitals were able to accept patients

with a mean time from stroke onset to arrival of 50.2-

58.2 minutes. According to the Japanese Fire and Disaster

Management Agency, the significant increase of the num-

ber of calls for ambulances in Japan causes delays in hos-

pital arrival. Traffic congestion is another factor in these

delays. In particular, the number of calls not requiring

urgent medical attention has been increasing. The average

interval from the emergency call to hospital arrival has

increased by 9 minutes over the past decade. Improved

patient knowledge of stroke risks and recognition of signs

and symptoms will help reduce the time from symptom

onset to hospital arrival.

Table 4. Logistic regression analyses testing the factors

predicting the likelihood of intracerebral haemorrhage

Odds

ratio 95% CI

Pvalue

Cardiogenic embolism 6.97 1.25-38.73 0.027

Warfarin 4.06 1.08-15.26 0.038

DWI-ASPECTS 0.73 0.58-0.92 0.007

Posterior cerebral artery 24.64 2.73-222.53 0.004

The frequency of symptomatic ICH in our study was

2.6% (Table 1), lower than that reported in the National In-

stitute of Neurological Disorders and Stroke study (6.4%)6

and Canadian Alteplase for Stroke Effectiveness Study

(CASES) (4.6%),7 and slightly higher than that reported

in SITS-MOST (1.7%)5 and The Japan post-Marketing Al-

teplase Registration Study (J-MARS) (1.3%).8 The differ-

ences in the definition of symptomatic ICH among

published studies hinder a direct comparison of the re-

sults of those studies. In the present study, the prevalence

of symptomatic ICH of 2.6% (Table 1) is comparable to

that reported in J-MARS8 when the European Coopera-

tive Acute Stroke Study9 and SITS-MOST5 criteria are ap-

plied. In the National Institute of Neurological Disorders

and Stroke study6 and CASES,7 symptomatic ICHwas de-

fined as any hemorrhage plus any neurologic deteriora-

tion (NIHSS score $1). In J-MARS, the prevalence of

symptomatic ICH was lower in centers with larger enroll-

ment ($20 cases) than in centers with smaller enrollment

(#19 cases) between October 2005 and October 2007.8 In

the present study, in 3 of our 4 hospitals, the number of

enrolled patients was ,20 in the first 2 years (Table 3),

resulting in a higher symptomatic rate of ICH than that

reported in SITS-MOST5 and J-MARS.8

Higher rates of symptomatic and nonsymptomatic ICH

were associated with a higher NIHSS score and infarction

range, defined by DWI-ASPECTS (Table 2), comparable to

those reported in other studies.10 The severer symptom

was (NIHSS was higher) and the larger range of ischemic

lesion was (ASPECTS was lower), the more frequency of

ICH raised (Table 2). This demonstrates that DWI-

ASPECTS and NIHSS are useful predictors of ICH after

tPA administration (Table 2).

In this study, baseline use of warfarin increased the risk

of symptomatic ICH despite an International Normalized

Ratio (INR) of ,1.7 (Table 2). Several mechanisms may

contribute to this effect. The fibrinolytic effects of tPA

may be enhanced by the anticoagulant effects of warfarin,

even at a subtherapeutic INR of ,1.7. Higher recanaliza-

tion rates with this combination (Table 2) may increase

the risk of reperfusion hemorrhage in the infarcted tissue.

Warfarin use is a marker for patients with cardiogenic

embolism, in whom hemorrhagic transformation is more

common and infarct volume is greater and stroke severity

Table 5. Comparison of edaravone and non-edaravone groups

Edaravone

(n5103)

Non-edaravone

(n511)

P valuen % n %

Male 66 64.1 7 63.6 0.6085

Mean age 71.8 71.9 0.9749

Type of cerebral infarction

Cardiogenic embolism 73 70.9 4 36.4 0.0267*

Atherothrombosis 31 30.1 7 63.6 0.0313*

Lacunar 2 1.9 0 0.0 0.8156

Others 1 1.0 0 0.0 0.9035

Mean time to treatment (min) 131.8 119.1 0.2243

Median NIHSS on admission (IQR) 10 (5-18.5) 16 (11.5-19.5) 0.0713

Median NIHSS at 3days after onset (IQR) 4 (1.5-11) (n555) 13 (9-18.5) (n511) 0.0082**

Median NIHSS at 7days after onset (IQR) 2 (0-7) (n554) 10 (6.5-15.5) (n511) 0.0207*

Median NIHSS reduction at 3days after onset (IQR) 3 (1-5.5) (n555) 1 (0.5-3) (n511) 0.4157

Median NIHSS reduction at 7days after onset (IQR) 4 (2.25-6.75) (n554) 2 (0-7) (n511) 0.9993

Recanalization 80 77.7 4 36.4 0.0071**

Bleeding complications 35 34.0 3 27.3 0.4681

Intracerebral hemorrhage 25 24.3 1 9.1 0.2780

Asymptomatic intracerebral hemorrhage 22 21.4 1 9.1 0.3033

Symptomatic intracerebral hemorrhage 3 2.9 0 0.0 0.7355

Abbreviation: IQR, interquartile range.

*P , 0.05.

**P , 0.01.

TISSUE PLASMINOGEN ACTIVATOR THERAPY FOR STROKE IN OKAYAMA 195

is high. Although the recanalization rate was higher in the

edaravone group compared with the non-edaravone

group, the rate of hemorrhagic transformation did not

differ between these 2 groups, suggesting the possibility

that edaravone can prevent hemorrhagic transformation.

Numerous in vitro and in vivo studies have shown that

the free radical scavenger edaravone inhibits brain

edema,11,12 tissue injury,13 delayed neuronal death,14 and

vascular endothelial cell injury after ischemia.15 We have

recently reported detachment of the neurovascular unit

by tPA and strong protection by edaravone against such

neurovascular unit destruction (ie, neurovascular protec-

tion).2 Combination therapy with tPA and edaravone

might have the potential to reduce hemorrhagic transfor-

mation in patients with acute ischemic stroke (Table 2).

In conclusion, our findings reported here demonstrate

that IV alteplase 0.6 mg/kg administered within 3 hours

of stroke onset should be safe and effective, with outcomes

and incidence of ICH (Table 1) compatiblewith previously

published data.5-10 DWI-ASPECTS and NIHSS were use-

ful predictors of ICH after tPA administration (Table 2).

Despite an INR of ,1.7, patients receiving warfarin treat-

ment aremore likely to experience symptomatic ICH after

IV tPA therapy (Table 2). Edaravonemight have the poten-

tial to reduce tPA-induced ICH (Table 5). Because this was

not a randomized controlled study, and the number of

patients without edaravone was small, renal dysfunction

(an exclusion criterion for edaravone use) might have

affected the efficacy of edaravone.

Acknowledgment: We thank the investigators and staff

at the centers involved in this study.

References

1. Yamaguchi T, Mori E, Minematsu K, et al. Alteplase at0.6 mg/kg for acute ischemic stroke within 3 hours ofonset: Japan Alteplase Clinical Trial (J-ACT). Stroke2006;37:1810-1815.

2. Yamashita T, Kamiya T, Deguchi K, et al. Dissociation andprotection of the neurovascular unit after thrombolysisand reperfusion in ischemic rat brain. J Cereb BloodFlow Metab 2009;29:715-725.

3. Barber PA, Hill MD, Eliasziw M, et al. Imaging of thebrain in acute ischaemic stroke: Comparison of computedtomography and magnetic resonance diffusion-weightedimaging. J Neurol Neurosurg Psychiatry 2005;76:1528-1533.

4. Morita N, Harada M, Uno M. Evaluation of initialdiffusion-weighted image findings in acute stroke pa-tients using a semiquantitative score. Magn Reson MedSci 2009;8:47-53.

5. Wahlgren N, Ahmed N, Davalos A, et al. Thrombolysiswith alteplase for acute ischaemic stroke in the Safe Imple-mentation of Thrombolysis in Stroke Monitoring Study(SITS-MOST): an observational study. Lancet 2007;369:275-282.

S. KONO ET AL.196

6. National Institute of Neurological Disorders and StrokertPA Stroke Study Group. Tissue plasminogen activatorforacute ischemic stroke.NEngl JMed1995;333:1581-1587.

7. Hill MD, Buchan AM. Thrombolysis for acute ischemicstroke. results of the Canadian Alteplase for Stroke Effec-tiveness Study. CMAJ 2005;172:1307-1312.

8. Nakagawara J, Minematsu K, Okada Y, et al. Thrombo-lysis with 0.6 mg/kg intravenous alteplase for acuteischemic stroke in routine clinical practice: The Japanpost-Marketing Alteplase Registration Study (J-MARS).Stroke 2010;41:1984-1989.

9. LarrueV, vonKummer RR,Muller A, et al. Risk factors forsevere hemorrhagic transformation in ischemic stroke pa-tients treatedwith recombinant tissueplasminogenactiva-tor: A secondary analysis of the European-AustralasianAcute Stroke Study (ECASS II). Stroke 2001;32:438-441.

10. Toyoda T, Koga M, Naganuma M, et al. Routine use ofintravenous low-dose recombinant tissue plasminogenactivator in Japanese patients: General outcomes and

prognostic factors from the SAMURAI register. Stroke2009;40:3591-3595.

11. Abe K, Yuki S, Kogure K. Strong attenuation of ischemicand postischemic brain edema in rats by a novel freeradical scavenger. Stroke 1988;19:480-485.

12. Nishi H, Watanabe T, Sakurai H, et al. Effect of MCI-186 on brain edema in rats. Stroke 1989;20:1236-1240.

13. Watanabe T, Yuki S, Egawa M, et al. Protective effects ofMCI-186 on cerebral ischemia: Possible involvementof free-radical scavenging and antioxidant actions.J Pharmacol Exp Ther 1994;268:1597-1604.

14. Yamamoto T, Yuki S, Watanabe T, et al. Delayed neuronaldeath prevented by inhibition of increased hydroxyl rad-ical formation in a transient cerebral ischemia. Brain Res1997;762:240-242.

15. Watanabe T, Morita I, Nishi H, et al. Preventive effect ofMCI-186 on 15-HPETE–induced vascular endothelialcell injury in vitro. Prostaglandins Leukot Essent FattyAcids 1988;33:81-87.