662
Transient Cerebral Ischemic Attacks in the
Young and Middle Aged
A Population Study
STURLA
E. JOHNSON, M.D.,* AND HA YARD SKRE, M.D.t
SUMMARY During an investigation of coronary risk factors, a population 20 to 54 years old in Troms0,
Northern Norway was screened for transient ischemic cerebral attacks. Three simple screening questions
were used. Sixteen thousand six hundred and twenty-one subjects participated in the study. Among the
responders, a sample of 501 were evaluated neurologicaliy and 10 men and 16 women identified as TIA
cases. Mean age for men was 41.3 years, for women 33 years. Five women (mean age 24.4) had the events
during pregnancy, pointing to pregnancy as a period of risk. Five-year incidence was found to be 2.5. per
1000. Clinical expressions and ratio carotid to vertebral-basilar TLA hardly differed from that found hi
materials of older patients. All 26 remained stroke-free during a mean observation period of 55 months.
Known risk factors like hypertension, carotid stenosis and cardiac disease were found in only a few. Five
women had low blood pressure. It is suggested that TLA in younger age groups may constitute a separate
entity where, among other, haemodynamk factors and pregnancy play a role.
Stroke Vol 17, No 4, 1986
Downloaded from http://stroke.ahajournals.org/ by guest on June 15, 2017
TRANSIENT ISCHEMIC ATTACKS, the most important warning sign of brain infarction in old age, is
occasionally seen even among young and middle aged
subjects. The cause and pathogenesis of this potentially hazardous condition remain unknown, but hypertension and atherosclerosis are thought to play a major
role. Other suspected causes include mitral valve prolapse, lipid abnormalities, increased coagulability and
polycythemia.1-2 Earlier investigations of TIA frequency deal mainly with subjects over 60 years. Estimates of incidence stem from data gained from records
and diagnostic registers.3"7
There are no known studies of TIA from screenings
of defined populations under 50-55 years. Knowledge
is sparse concerning the frequency, clinical course and
prognosis of the condition in these younger age strata.
Such knowledge may give clues to pathogenesis and
ultimately aid prevention. A recent survey of coronary
risk factors in a population in Norway* gave us the
opportunity to study TIA in males 20 to 54 years and
females 20 to 50 years old.
Material and Methods
The study took place in the municipality of Troms0
in Northern Norway 1979-80 and was linked to the
mass chest x-ray examination carried out that year by
the National Mass Radiography Service. The followup was completed by July 1981. Eligible were all men
born 1925 to 1959 and women bom 1930 to 1959,
registered in the municipality, in all 21,329 persons.
With the invitation went a 2-page questionnaire asking for information mostly on symptoms related to
heart disease, but including a question on stroke. This
questionnaire was delivered when the participants reFrom the Department of Neurology, University Hospital, N-9012
Troms0, Norway,* and Department of Neurology, Akershus Central
Hospital, N-1474 Nordbyhagen, Norway.
Sturla E. Johnson, M.D.* is a Consultant. Havard Skre, M.D.t is
Professor and Chairman.
Address correspondence to: Dr. Sturla E. Johnson, Consultant, Department of Neurology, University Hospital, N-9012 Troms0, Norway.
Received July 9, 1985; accepted December 3, 1985.
ported for the first screening. In addition to chest xray, this consisted of weight, height and blood pressure
measurement and collection of venous non-fasting
blood samples for assessment of total cholesterol,
HDL cholesterol, triglycerides and glucose levels. The
participants were now given a second questionnaire
with 45 questions on past and present health, previous
illnesses, dietary habits, cigarette smoking, alcohol
consumption, as well as social conditions. They were
asked to fill in the form at home and return it by mail.
A separate section had questions aimed at detecting
transient ischemic attacks in the brain. These were:
"Have you ever had: 1. Sudden paralysis or numbness
in one side of body or face, in hand or foot? 2. Sudden
loss of ability to speak? 3. Sudden loss of eyesight,
completely or partially, or sudden double vision?" A
yes/no reply was required in each instance. A random
sample of 50% of those giving an affirmative answer
were chosen for the follow-up study. They received a
letter in which we explained the object of the study and
invited them to come for an interview and medical
examination by either one of the authors.
The guide-lines set down by the Study Group of
Heyman et al9 were used for the identification of cases,
TIA being defined as episodes of temporary focal cerebral dysfunction of vascular origin. Symptoms commonly last 2 to 30 minutes and clear rapidly leaving no
residue. There may be only one attack or multiple
attacks at different intervals. TIA are classified according to which vascular territory is involved, the carotid
artery or the vertebral-basilar artery system, each giving rise to different symptoms. In practice this differentiation can be difficult.
Each person was thoroughly questioned about the
incident(s). Symptoms were recorded as accurately as
possible, as were general health and other circumstances around the event. Special note was made of
illness at the time and any use of pharmaceutical drugs
and alcohol was recorded. Whenever the history suggested focal cerebral ischemia, a full clinical neurological examination was carried out. This included a
general medical examination focusing on signs of car-
A POPULATION STUDY/yo/i/iTon and Skre
Downloaded from http://stroke.ahajournals.org/ by guest on June 15, 2017
diovascular and haeraatological disorders in particular.
Blood pressure was measured with the patient recumbent for 2 minutes. All subjects met fasting. A venous
blood sample was drawn from the cubital vein. A total
of 275 samples, including all stroke and TLA cases,
have been analysed for lipoproteins and lipoprotein
markers. The results will appear elsewhere.
Identification of TLA in the present study was made
after a close scrutiny of data collected during the interview. Whenever available, hospital records were
checked. Our hospital is the only one in this area and
records existed for 50% of those interviewed.
Vertigo alone as symptom was considered too unspecific to merit inclusion, as were syncope and drop
attacks. Headache before or after the event or a history
of recent migraine attacks also meant exclusion although headache can be a feature of TLA. There was
one exception: A 35 year old women with classical
migraine over 9 years whose attacks were stable and
uniform and occurred 2-3 times a year. She was included on the basis of an episode with complete halfsided sensory loss of 10 minutes duration and without
headache. Identification was in some cases difficult
and required a second telephone interview in order to
get more details. When in doubt, two or more identical
episodes would mean inclusion. One episode was accepted when there could be little doubt about the ischemic nature like in transient hemiparesis or monocular
blindness.
Results
Of the 21,329 persons summoned for the mass x-ray
examination, 16,621 responded, delivering their answers to the 2 page primary questionnaire. Twenty-six
replied affirmatively to the question referring to
stroke. All of these were invited for a follow-up with
the intention of verifying the diagnosis. Twenty-one
were seen and examined. An overview of numbers is
given in table 1. The results appear in table 2. Five of
the 12 patients with thromboembolic stroke reported
episodes of TLA in addition. Since these cases are not
randomly selected, they were not included in the subsequent TLA group. Fourteen thousand six hundred
and sixty-seven of the 16,621 persons participating in
the mass x-ray examination returned the second questionnaire. One thousand, one hundred sixty-seven had
given an affirmative answer to one or more of the TLA
questions. Half of these, 588 persons, all born on even
dates, were invited for further medical investigations,
TABLE 1 The Tromsj Study. Sixteen Thousand Six Hundred and
Twenty-one Men and Women 20 to 54 Years Old
Attended first screening
16621
Self-reported stroke
26
Postal questionnaires returned
14667
Affirmative answers to TIA questions
1167
Selected for follow-up
588
Attended foUow-up (85%)
501
TIA-cases found
26
663
TABLE 2 Verification of 26 Self-reported Strokes in 17 Men
(32-54 years) and 9 Women (28-50 years). Results of Follow-up
and Review of Records
Thromboembolic infarction
12
Transient ischemic attack
1
Subarachnoid haemorrhage
7
Traumatic brain haemorrhage
1
Non-cerebrovascular conditions
4
Unknown
1
Total
26
and 501 (85%) responded. Four hundred ninety-four
subjects were seen personally by the authors while 7
could not attend for different reasons, but were interviewed by telephone. Of the 87 non-responders, one
had died of coronary heart attack before the follow-up,
15 had moved and 11 declined.
As expected a large majority of those seen had no
history suggesting transient cerebral ischemia. Table 3
summarizes the diagnoses of this false positive group.
The group includes 3 persons with previously unrecognized minor strokes. In the same group were a number
of patients with specific non-cerebrovascular neurological conditions which had been diagnosed during
earlier admissions to the hospital.
Twenty-six cases of TLA, 10 men and 16 women,
were identified. Six of these already had the diagnosis
from earlier admissions or visits to the outpatient clinic. For 25 subjects the events took place between 1969
and 1980. The exception was a woman whose event
occurred during pregnancy 30 years earlier. Age at the
time of first TLA averaged 41.3 years for men (range
29-54) and 33 years for women (18-48). Five young
women had TLA during the second half of pregnancy.
If we leave out these who pathogenetically may belong
to a different category, mean age rises to 36.8 years
and equals that of men born after 1930 (37.2 years).
Incidence of TLA over age is given in figure 1.
Only men and women born after 1st of January 1930
with first events 1975 through 1979 were considered
for the estimation of incidence, in allfivemen and nine
TABLE 3 Classification of "False Positives" in Response to TIA
Questions
Acute or intermittent numbness/weakness due to
cervical spondylosis, shoulder conditions, nerve
entrapment, myosis
31.5
Sensory/motor symptoms due to lumbar disc lesions
or ortopedic/rheumatic conditions
7.0
Migraine
8.9
Facial palsy (Bell's)
7.6
Psychoneurosis, hyperventilation syndrome
11.4
Eye disorders e.g. squint, refraction errors
9.5
Dizziness, near-syncope w/visual disturbances
5.5
Specific neurological disorders, e.g. stroke,
epilepsy, traumatic and neoplastic brain lesions,
MS, poliomyelitis, meningitis, polyneuropathy
and myotonic dystrophy
7.0
Unrelated or non-classifiable disorders
11.6
Total
100.0
n = 473.
STROKE
664
VOL
17, No 4, JULY-AUGUST
>3
3.4
1986
JL,>3.7
2.5
>2.4
1.5
M
Age
<25
30
35
40
45
50
54
FIGURE 2. TIA episodes over age. Figures on top of columns
denote mean number of attacks. In 50—54 year group men only.
Age
<25
30
35
40
45
50
54
Downloaded from http://stroke.ahajournals.org/ by guest on June 15, 2017
FIGURE 1. Age at the time of first TIA. 10 men, 16 women
(shaded). Figures on top of columns indicate total numbers.
Age 50-54 years covers men only.
women. A 5-year overall incidence (first event) of TIA
for men and women between 20 and 50 years was
estimated to be about 2.5 per 1000.
In the TIA group, the majority were in good health
at the time of the event. Four were being treated for
hypertension and two of these had coronary heart attacks within five years of the cerebral event. In two of
the others cardiac murmurs were heard on auscultation
and one of these had rheumatic fever as a child. Six
were markedly overweight, one of whom had undergone jejuno-ileal bypass surgery. The rest were generally healthy (table 4). None of the women were using
contraceptive pills. Five young women (mean age
24.6) had TIA during pregnancy, four of them during
the last trimester. One woman experienced similar TIA
during 2 successive pregnancies. The examination revealed a bruit over the corresponding carotid artery. In
two of the others were found possible bruits. None
were under the influence of alcohol when the TIA
occurred. Mostly, multiple attacks occurred within a
few days, but 4 subjects reported recurrent episodes
over more than four years. Figure 2 shows mean attack
number over age. There were no overall sex differences. Ten subjects experienced only one attack
whereas 4 subjects had more than five.
TABLE 4 Follow-up Study of 26 TIA-cases20-54 Years: Some
Clinical Findings
Systolic
BP, (mm Hg) mean
Systolic
Systolic
BP ^ 145. n =
BP 5 110. n =
131.3 (160-105)
6
5
Diastolic
Diastolic
BP, mean
BP S 110. n =
85.0 (110-70)
3
Heart disease
Overweight
Carotid bruit,
definite
possible
impalpable pulse
2
6
1
2
2
All the common manifestations of TLA were seen,
like monocular blindness, hemiparesis, hemianesthesia and aphasia (table 5). Seventeen subjects experienced sensory symptoms either alone or in combination with other symptoms. In 20 subjects symptoms
were thought to come from ischemia in the carotid
artery territory. The vertebral-basilar territory was implicated in theremaining6. Although the events might
have taken place several years earlier, all remembered
them well and could give detailed accounts. A similar
experience of good patient memory is reported from
the Evans County Study.12 In 17 out of 23 symptoms
had caused sufficient alarm to make them consult a
doctor. Eight were soon afterwards referred to our
department where 6 underwent extra- and intracranial
vessel angiography. Another 2 have since the study
been admitted for angiography because of recurrent
TIA. In all 8 angiographies were performed. Bilateral
carotid aneurysms were demonstrated in one person
with fibromuscular dysplasia. Another had extensive
atheromatous lesions and stenosis in both internal carotids. Endarterectomy was carried out as an emergency. No abnormality was seen in the other 6.
At the time of neurological evaluation mean observation time was 55 months (range 12-132) without
occurrence of stroke in 25 subjects. One person, a
woman, has remained stroke-free for 30 years.
Over the last ten years an average of 60 patients
TABLE 5 Symptoms Due to TIA in 26 Subjects 20-54 Years
8
Numbness in half of face + arm, half of body
2
Numbness in half of face + arm with reduced vision
1
Numbness in half of face + arm with loss of speech
2
Parestesia and numbness in half of face + arm
2
Hemiparesis
3
Paresis and numbness in face and arm
3
Paresis and parestesia in tongue, arm and leg
1
Paresis and numbness in one arm, with dysarthry
Amaurosis fugax (1st event). Hemianopia and parestesia
1
in one arm (2nd event)
2
Bilaterally reduced vision or blindness
1
Hemianopia
26
Total
A POPULATION STUDY/Johnson and Skre
annually have been admitted to the neurological, medical and surgical departments under the diagnosis TLA
(ICD no.435.99). The hospital's databased diagnostic
register has been checked to see if any TLA-cases admitted who were eligible for the follow-up study, had
in fact been missed. None were found.
Downloaded from http://stroke.ahajournals.org/ by guest on June 15, 2017
Discussion
Two methods are available if one wishes to study the
occurrence of TLA in a population. Most studies so far
have made use of data from medical records or, indirectly, from diagnostic registers. This means that only
persons who seek medical attention for their symptoms
are included. One would expect such estimates to be
too low since a number of patients presumably never
report their incidents. TLA symptoms, as the present
study shows, are often sensory, very short-lived, appear only once and may not cause much alarm. One
can safely assume that the younger the patient, the less
regard he pays such symptoms. A population screening with the use of self-administered questionnaire is
equally open to error. The form may have been incompletely filled in, misunderstood or lost or events may
have been forgotten. It is impossible to say to what
extent this affects results. From our experience the
general public is motivated for health screenings once
the object is made clear and people take care to answer
the questions correctly. A great many said they welcomed the chance to get a medical check. This is also
reflected in the good tum-out.
As for the 1954 subjects (11.8%) who did not respond to the second questionnaire, these do not differ
in age from the rest. We suspect that a lower incidence
of TLA would have been found in this group had it been
followed up. Persons who experience symptoms are
probably more alert to the need for a medical check
than others and therefore likely to be found among the
responders.
The questions used in the Troms0 Study may seem
few and unspecific if compared with those used and
recommended by other investigators.'110 In a Danish
study in some ways similar to the present, 5 simple
questions were used." The nature of the Troms0 Study
ruled out the use of additional questions. In our view
too many and detailed questions can also cause errors
and lead to higher non-response rate. It was decided
instead to use few questions and to follow up with a
personal evaluation of a large number of the responders. This was done by qualified neurologists with access to hospital records. We therefore believe a high
degree of case ascertainment was achieved.
As expected the Troms0 Study showed transient
ischemic attacks to be rare in persons under 50-54.
Earlier population studies differ from the present in
method and age selection and directly comparable estimates of incidence are not available. The Danish Study
linked to an investigation of heart disease in a Copenhagen suburb gives an estimate of 4/1000/year for the
age 50-59 years." The Evans County survey covers
subjects over 40 years and a sample of 15 to 39 year
olds. Prevalence of TLA at the time of follow-up was
665
8/1000 under 54 years. There was a racial difference,
whites having a higher rate than blacks.12 Japanese
men in Hawaii over 45 years had an incidence of
0.7/1000 in the 45-49 year group.13 A recent Swedish
survey based on data from hospital admissions, suggests an incidence as low as 4/100000/year under the
age 55.6 The authors assume that nearly all TLA cases
are referred for specialist evaluation, a view which is
clearly open for debate. In comparison, The Rochester
Study,3 also based on medical records, gives an incidence of 16/100000/year in the 45-54 year group. Allowing for the uncertainties connected with population
screenings like the present and the small number of
TLA cases found, our estimate of a 5-year incidence of
2.5/1000 seems reasonable. It must, however, be taken only as a pointer. The real incidence is likely to be
higher for reasons stated earlier. Also, the figure does
not include persons who suffered stroke or died after
TLA. An interesting detail is the comparison between
male/female ratios in the Swedish and Troms0 materials. There seems to be a trend towards a female dominance in the younger age strata and the reverse in the
older age groups in both materials. Since our cases are
so few, this must be taken with all due reservations.
A surprising number (17 out of 26) experienced
sensory deficits, either alone or in combination with
other symptoms. Karp12reported8 with sensory symptoms out of 28 but 18 of these were over 55 years old.
This could imply that sensory TLA make up a larger
proportion of cases in the young than in those of higher
age. The number of cases with paresis of the limbs
corresponds well in the two materials (9 in the Troms0
Study against 8 in the other). The attacks were mostly
short, lasting less than 30 minutes in 20 subjects. In 11
the duration was less than 5 min.
The ratio of carotid to vertebral/basilar territory TLA
was 4:. This differs from the 2:1 ratio found in a
middle and old age patient material in Rochester,14 and
in a Dutch material of similar age" but is equal to that
found in a retirement community in Seal Beach.4
In contrast to the attention given stroke, not much is
found in the literature concerning transient ischemic
attacks during pregnancy. Sporadic cases have been
described but more often regarded as manifestations of
migraine than TLA.1316 None of our cases suffered
from migraine during or after the pregnancy. In our
experience TLA during the latter half of pregnancy is
not uncommon. We have since the study encountered
several additional cases, referred from the Department
of Obstetrics. A report is about to be published.l7 This
very interesting phenomenon should, in our view, be
further investigated with regard to pathogenesis and
prognostic implications. All the five women in the
study had uneventful deliveries.
There is a puzzling trend towards a higher TLA incidence in the age groups around 40 years, corresponding roughly with the highest mean recurrence rate.
There is also a high incidence and TLA recurrence rate
in age group 25-29, corresponding to the peak of TLA
incidence in pregnancies. Since the number of cases is
so small, such trends should not be overemphasized,
STROKE
666
Downloaded from http://stroke.ahajournals.org/ by guest on June 15, 2017
but they suggest two possibilities: Causes leading to
TIA episodes in the actual age groups exert their maximum influence around the 40 years mark, or during the
childbearing periods in the female. Further, these episodes are evidently of a more benign type than those
occurring in older age, where approximately one third
of TIA cases may experience a stroke later, half of
them within the following month.18 Also other observations suggest that our "young" TIA cases constitute
special etiological groups, since risk factors like hypertension, overweight, carotid stenosis, cardiac diseases, etc., were largely non-existent and hardly different from the population average.
This is in line with findings in another Norwegian
study of cerebrovascular accidents, which comprised
non-hypertensive males younger than 55 years of age
admitted to hospital. TLA incidence peaked around 40
years, while stroke incidence prevailed later. In this
study the antigen Ag(x), which is of a lipoprotein nature (Beta fraction), was abundant in TLA cases (59%
positive, as against 35% in the population.20 Lack of
this factor is regarded a risk factor in cardiac infarction, its role in TIA is obscure. Another possible risk
factor in TLA may be low blood pressure in affected
women. Of the 16 females in our series with TIA, 5
had systolic blood pressures =Sl 10 mm Hg, while such
lowfigureswere not found in 14 age-matched controls
(women with Bell's palsy from the same study).
In a Dutch study from 1981,19 comprising 225 women (28% below 55 years), significant correlation was
observed between low BP and TLA risk in all age
groups.
On the basis of our ownfindingsset in a context with
other observations, it is reasonable to believe that early
TIA mayrepresenta rather benign condition, and perhaps constitute a separate entity(ies), where, among
other, haemodynamic factors and pregnancy play a
role.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
References
1. Marshall J: The cause and prognosis of strokes in people under SO
years. J Neurol Sci S3: 473-488, 1982
2. Mettinger KL, Sdderstrom CE: Pathogenetic profile of TIA before
55. J Neurol Sci 36: 341-348, 1978
3. Whisnant JP: A population study of stroke and TIA: Rochester,
20.
VOL 17, No 4, JULY-AUGUST 1986
Minnesota. In: Stroke. Proceedings of the 9th Pfizer Intern. Symp.
Gillingham, Mawdsley & Williams (eds.). Churchill Livingstone,
Edinburgh pp 21-39, 1976
Friedman GD, Wilson WS, Mosier JM, Colandrea MA, Nihaman
MZ: Transient ischemic attacks in a community. JAMA 210:
1428-1434, 1969
Terent A: A prospective epidemiological survey of cerebrovascular
disease in a Swedish community. Uppsala J Med Sci 84: 235-246,
1979
Mettinger KL, Sdderstrom CE, Allander E: Epidemiology of acute
cerebrovascular disease before the age of 55 in the Stockholm
County 1973-77: I. Stroke 15: 795-801, 1984
Zupping R, Roose M: Epidemiology of cerebrovascular disease in
Tartu, Estonia, USSR, 1970 through 1973. Stroke 7: 187-190,
1976
Thelle DS, Amesen E, F0rde OH: The Tromsd Heart Study: Does
coffee raise serum cholesterol? N Engl J Med 308: 1454-1457,
1983
Heyman A (Chairman) and The Study Group on TIA criterion and
detection: Transient focal cerebral ischemia: Epidemiological and
clinical aspects. Stroke 5: 277-287, 1974
Wilkinson WE, Heyman A, Burch JG, Ostfeld A, Labarthe DR,
Leviton A, O'Fallon WM, Whisnant J: Use of self-administered
questionnaire for detection of transient cerebral ischemic attacks: I.
Ann Neurol 6: 40-46, 1979
Boysen G, Jensen G, Schnohr P: Frequency of focal cerebral transient ischemic attacks during a 12 month period. Stroke 10:
533-534, 1979
Karp HR, Heyman A, Heyden S, Bartel AG, Tyroler HA, Hames
CG: Transient cerebral ischemia. Prevalence and prognosis in a
biracial rural community. JAMA 235: 125-128, 1973
Rhoads GG, Popper JS, Kagan A, Yano K: Incidence of transient
cerebral ischemic attack in Hawaii Japanese men. Stroke 11:
21-26, 1980
Cartlidge NEF, Whisnant JP, Elveback LR: Carotid and vertebralbasilar transient cerebral ischemic attacks. Mayo Clin Proc 52:
117-120, 1977
Hart RG, Miller VT: Cerebral infarction in young adults: A practical approach. Stroke 14: 110-114, 1983
Fisher CM: Cerebral ischemia — less familiar types. Clin Neurosurg 18: 267-336, 1971
Johnson SE, Monstad P, Mellgren SI, Verelst M: Transient ischemic attacks related to pregnancy and puerperium. 1985 (Submitted
for publication)
Hier DB, Caplan LR: Stroke — who is at risk? Sandorama 29-32,
HI, 1981
Herman B, de Waard F, Collette HJA: Unexpected trends in the
analysis of a questionnaire and interview procedure to detect transient cerebral ischemic attack in a female population. Clin Neurol
Neurosurg 83-4: 225-237, 1981
Skre H, Berg K, Nyland H, Kobro G: Some lipoprotein markers
and other variables in males aged 20-55 years with cerebro-vascular accident (CVA). Proceedings of the Norwegian Neurological
Society, September 20, 1984. Acta Neurol Scand, 1985 (in press)
Transient cerebral ischemic attacks in the young and middle aged. A population study.
S E Johnson and H Skre
Stroke. 1986;17:662-666
doi: 10.1161/01.STR.17.4.662
Downloaded from http://stroke.ahajournals.org/ by guest on June 15, 2017
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1986 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/17/4/662
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click Request
Permissions in the middle column of the Web page under Services. Further information about this process is
available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/