Secular Trends in Ischemic Stroke Characteristics in a Rapidly Developed Country Results From the Korean Stroke Registry Study (Secular Trends in Korean Stroke)

Abstract

A dynamic change in industry, lifestyle, and healthcare structure brings a corresponding change in disease patterns. Limited data exist with respect to secular trends in stroke epidemiology in Korea, a rapidly developed country.
We analyzed individual patient data registered the Korean Stroke Registry, a nationwide hospital-based stroke database, between January 2002 and November 2010. Mortality data were obtained from a national death certificate system. Linear or logistic regression analyses were performed to assess secular trends. A total of 46 098 patients were included in this study. Mean ± SD age was 66.1 ± 12.3 years, and 57.6% of the patients were men. Over the 9-year period, patient ages steadily increased by 0.24 year annually (P<0.001). Risk factor proportions of hypertension, diabetes, smoking, and prior stroke declined slightly (P<0.05 for all). However, dyslipidemia frequency showed a complex pattern of an initial decline and then an increase. For relative proportions of subtypes, cardioembolism increased, small vessel occlusion decreased, and large artery atherosclerosis remained stable. Still, intracranial stenosis overwhelms extracranial stenosis, but extracranial stenosis is on the rise. Arrival within 3 hours increased from 20% to 29%, and reperfusion therapy increased from 5.3% to 7.0%. Age-adjusted all-cause mortality did not decrease at 30 days but decreased at 1 year over time.
During the first decade of 21st century, stroke characteristics in Korea changed, likely because of increased lifespan, westernized lifestyle, and improved public awareness. Stroke experts need to cope with these distinguishing trends to establish a better strategy for prevention and acute therapy.

Full text

Secular Trends in Ischemic Stroke Characteristics in a
Rapidly Developed Country
Results From the Korean Stroke Registry Study (Secular Trends in
Korean Stroke)
Keun-Hwa Jung, MD, PhD; Seung-Hoon Lee, MD, PhD; Beom Joon Kim, MD;
Kyung-Ho Yu, MD, PhD; Keun-Sik Hong, MD, PhD; Byung-Chul Lee, MD, PhD;
Jae-Kyu Roh, MD, PhD; Korean Stroke Registry Study Group
Background—A dynamic change in industry, lifestyle, and healthcare structure brings a corresponding change in disease
patterns. Limited data exist with respect to secular trends in stroke epidemiology in Korea, a rapidly developed country.
Methods and Results—We analyzed individual patient data registered the Korean Stroke Registry, a nationwide
hospital-based stroke database, between January 2002 and November 2010. Mortality data were obtained from a
national death certificate system. Linear or logistic regression analyses were performed to assess secular trends. A total
of 46 098 patients were included in this study. Mean⫾SD age was 66.1⫾12.3 years, and 57.6% of the patients were men.
Over the 9-year period, patient ages steadily increased by 0.24 year annually (P⬍0.001). Risk factor proportions of
hypertension, diabetes, smoking, and prior stroke declined slightly (P⬍0.05 for all). However, dyslipidemia frequency
showed a complex pattern of an initial decline and then an increase. For relative proportions of subtypes,
cardioembolism increased, small vessel occlusion decreased, and large artery atherosclerosis remained stable. Still,
intracranial stenosis overwhelms extracranial stenosis, but extracranial stenosis is on the rise. Arrival within 3 hours
increased from 20% to 29%, and reperfusion therapy increased from 5.3% to 7.0%. Age-adjusted all-cause mortality did
not decrease at 30 days but decreased at 1 year over time.
Conclusions—During the first decade of 21st century, stroke characteristics in Korea changed, likely because of increased
lifespan, westernized lifestyle, and improved public awareness. Stroke experts need to cope with these distinguishing trends
to establish a better strategy for prevention and acute therapy. (Circ Cardiovasc Qual Outcomes. 2012;5:327-334.)
Key Words: ischemia 䡲stroke 䡲trends 䡲Korea 䡲registries
Stroke is a major health burden worldwide.
1–5
Expansion
of the aged population and alarmingly increasing vascu-
lar risk factor prevalence in the child and middle-aged
populations undoubtedly forecast an exponential increase of
stroke burden in the near future.
3,4
Rates of cardiovascular
risk factors and disease phenotypes vary not only by ethnic
group, but also across the socioeconomic state of a nation.
5
The Korean economy has continued a stable and strong
growth during the past 3 or 4 decades, and rapid industrial
and societal changes have imposed a great impact on disease
patterns.
6
It is widely supposed, but has not been systemati-
cally explored, that multiple domains of risk factors, stroke
subtypes, and public awareness of and response to acute
stroke have changed in Korea. Experiences from Korea can
be informative in establishing stroke prevention strategies
and acute care systems in other rapidly developing countries.
Hospital-based stroke registries have well-recognized lim-
itations in generalizability and ability to estimate incidence
and prevalence of populations. However, in contrast to
population-based epidemiological studies, they generally en-
sure more-accurate and comprehensive data with regard to
demographics, risk factors, stroke etiology, admission pat-
tern, provided care, and clinical outcomes. Accordingly,
stroke registries are well suited to defining the secular trends
of these multiple domains. The Korean Stroke Registry
(KSR) is a multicenter, prospective, hospital-based stroke
registry. Launched in 2002, the KSR has collected nationwide
data to gain insight into stroke diagnosis and care.
7
In the
Received August 25, 2011; accepted February 21, 2012.
From the Department of Neurology, Seoul National University Hospital, Seoul, South Korea (K.-H.J., S.-H.L., B.J.K., J.-K.R.); Department of
Neurology, Hallym University Sacred Heart Hospital, Anyang, South Korea (K.-H.Y., B.-C.L.); and Department of Neurology, Ilsan Paik Hospital, Inje
University, Goyang, South Korea (K.-S.H.).
The online-only Data Supplement is available at http://circoutcomes.ahajournals.org/lookup/suppl/doi:10.1161/CIRCOUTCOMES.111.963736/-/DC1.
Correspondence to Jae-Kyu Roh, MD, PhD, Department of Neurology, Seoul National University Hospital, 101, Daehangno, Jongno-gu, Seoul
110 –744, South Korea, e-mail rohjk@snu.ac.kr or Byung-Chul Lee, MD, PhD, Department of Neurology, Hallym University, College of Medicine, 896,
Pyungchon, Anyang City, 431-070, South Korea, e-mail ssbrain@hallym.ac.kr.
© 2012 American Heart Association, Inc.
Circ Cardiovasc Qual Outcomes is available at http://circoutcomes.ahajournals.org DOI: 10.1161/CIRCOUTCOMES.111.963736
327
Downloaded from http://ahajournals.org by on August 7, 2018

present study, we aimed to evaluate recent national trends in
clinical characteristics, delay in seeking medical care, reper-
fusion therapy, and mortality in patients with acute ischemic
stroke in Korea.
WHAT IS KNOWN
●Rapid industrial and societal changes can lead to
dynamic changes in risk factors, stroke phenotypes,
and treatment.
●Korea has achieved a remarkably high level of
economic growth in a short period of time.
WHAT THE STUDY ADDS
●The ages of patients with stroke are steadily increas-
ing over time.
●Cardioembolic stroke continues to increase.
●Extracranial artery stenosis is on the rise.
●Arriving at the hospital within 3 hours and reperfu-
sion therapy are increasing.
●Age-adjusted all-cause mortality at 1 year is
decreasing.
Methods
Registry Characteristics
In 2002, the Korean Stroke Society launched a task force comprising
stroke specialists in neurology from 26 major university hospitals
throughout the country as well as recruited hospitals. To date, the 31
educational general hospitals in Korea have phased in the registry
system (online-only Data Supplement Table I). The participating
hospitals have registered patients with ischemic stroke or transient
ischemic attack (TIA) within 7 days after symptom onset. Stroke was
diagnosed through clinical symptoms and relevant findings on CT
and MRI scans and magnetic resonance angiography. TIA was
defined as a sudden focal neurological deficit that completely
resolves within 24 hours. The KSR provides a Web-based database
(www.strokedb.or.kr) in which each participating hospital can reg-
ister patient data in real time. An attending physician initially enters
the data into the registry, and all records are regularly monitored
within a 1- to 2-week interval by dedicated faculty members with
stroke expertise at each hospital for completeness and consistency.
The steering committee of KSR has monitored data consistency
regularly within a 3-month interval and has encouraged consistent
registration. Nonetheless, unexpected circumstances, such as change
of individual hospital policies and turnover of dedicated staff at each
hospital could cause inconsistent registration during a certain period.
The guidelines and procedures of this registry were approved by the
Institutional Review Board of the Hallym University Sacred Heart
Hospital as a delegate of participating hospitals.
Data Acquisition
We acquired all registered data from 31 centers between January
2002 and November 2010. Exclusion criteria were missing demo-
graphic information; incomplete (⬎10%) documentation for all
categories; duplication; protocol violations, such as admission after 7
days from onset; and age ⬍16 years. For each patient, we collected
data on demographics, risk factors, Trial of RRG 10172 in Acute
Stroke Treatment (TOAST) stroke subtypes, stroke onset, prehospi-
tal delays, initial stroke severity as measured by the National
Institute of Health Stroke Scale (NIHSS), radiological findings of
intracranial and extracranial stenosis relevant for the index stroke,
and in-hospital management. We also obtained mortality data for
each patient from the Korean National Vital Statistics system.
Definition of Variables
Vascular risk factors included hypertension, diabetes mellitus, dys-
lipidemia, potential source of cardioembolism (CE), smoking, and
history of previous stroke. Hypertension was defined as a systolic
blood pressure ⬎140 mm Hg, a diastolic blood pressure
⬎90 mm Hg, and current use of antihypertensive agents. Diabetes
mellitus was diagnosed through relevant clinical or drug history or
biochemical evidence of at least 2 measurements of fasting blood
glucose readings of ⬎126 mg/dL. Dyslipidemia was defined as
current use of lipid-lowering agents or at least 2 abnormal serum
lipid measurements (total cholesterol ⬎240 mg/dL or low-density
lipoprotein cholesterol ⬎160 mg/dL). The presence of a potential
source of CE was defined as having 1 of the following high-risk
cardiac conditions: mechanical prosthetic valve, mitral stenosis with
atrial fibrillation, atrial fibrillation, left atrial or atrial appendage
thrombus, sick sinus syndrome, recent myocardial infarction (⬍4
weeks), left ventricular thrombus, dilated cardiomyopathy, akinetic
left ventricular segment, atrial myxoma, and infective endocarditis.
8
Smoking was defined as a smoking history of ⬎2 pack-years or
current smoker.
The stroke mechanisms were determined by TOAST criteria as
follows
8
: large artery atherosclerosis (LAA), small vessel occlusion
(SVO), CE, stroke of other etiology, and stroke of undetermined
etiology. TIA was not classified for a certain etiology in the TOAST
criteria. The KSR provides an algorithm protocol for TOAST
classification to enhance standardization. All records on stroke
subtypes were monitored by faculty members at each hospital for
internal consistency and were refined by a regular discussion by the
steering committee for external consistency. In LAA subtype, the
location of stenosis relevant to the index stroke was classified into
intracranial arterial stenosis of anterior circulation (A-ICAS), ex-
tracranial arterial stenosis of anterior circulation (A-ECAS), and
posterior circulation stenosis. The A-ICAS was defined as ⬎50%
luminal narrowing or occlusion of proximal portions of the middle
cerebral artery, anterior cerebral artery, and intracranial portion of
the internal carotid artery.
9
The A-ECAS was defined as ⬎50%
luminal narrowing or occlusion of extracranial portions of the
internal carotid artery and common carotid artery.
10
The posterior
circulation stenosis was defined as ⬎50% luminal narrowing or
occlusion of the vertebral artery, basilar artery, and posterior cerebral
artery.
Statistical Analysis
Baseline characteristics were summarized by mean⫾SD or number
and percentage for each calendar year of stroke diagnosis. Annual
trends in age, time interval after onset, and NIHSS scores were
estimated by linear regression analyses. Age trend was analyzed
separately for men and women. For the binary categorical variables,
log-linear Poisson models were used to estimate rates of increase by
calendar year. Calendar year was treated as a continuous variable in
the analysis. The mortality change was adjusted for age using
10-year age intervals. Results are presented as relative risks (RRs)
with 95% CIs. All Pvalues are 2 sided, and a 5% level of statistical
significance was used. Statistical analyses were performed with SAS
STAT version 9.1 software.
Results
Baseline Characteristics of the Patients
A total of 53 542 patients were registered in the KSR during
the study period. After a review of the records, we excluded
7444 patients because of missing demographic information
and incomplete (⬎10%) documentation for all categories
(n⫽2922); duplicate registrations (n⫽2061) from early refer-
ral to another hospital, recurrences, or registration errors;
protocol violations (n⫽2461), such as admission after 7 days
of onset; and age ⬍16 years. The numbers of excluded cases
according to each hospital and year are shown in online-only
Data Supplement Table I. There were no significant changes
328 Circ Cardiovasc Qual Outcomes May 2012
Downloaded from http://ahajournals.org by on August 7, 2018

in the exclusion number among the years or the hospitals, but
clinical characteristics of the excluded cases differed from
those of the study population with respect to age, risk factors,
NIHSS score, and prehospital delay (online-only Data Sup-
plement Table II).
Finally, 46 098 patients with ischemic stroke and TIA were
included in the current study. The mean age was 66.1⫾12.3
years, and 26 566 (57.6%) were men. Women were older than
men (69.2⫾11.9 versus 63.9⫾12.1 years, P⬍0.001). There
were 42 981 (93.2%) ischemic strokes and 3117 (6.8%) TIAs.
Hypertension was the most common vascular risk factor (29
266 patients [63.5%]) followed by smoking (33.0%), diabetes
mellitus (30.1%), prior stroke (19.7%), potential sources of
CE (19.4%), and dyslipidemia (19.3%). Detailed data for the
baseline characteristics of the study population are shown in
online-only Data Supplement Table II.
Secular Trends
Demographic and Risk Factors
Table 1 and Figure 1 show the secular trends by age, sex, and
vascular risk factors expressed as the regression coefficient
(
␤
) and RR from linear and log-linear Poisson regression
analysis, respectively. Ages at stroke onset have increased
over time (
␤
⫽0.237 per calendar-year; 95% CI, 0.186 –
0.288; P⬍0.001). When stratified by sex, the increasing trend
was more prominent in women than in men (Figure 1). The
proportion by sex was not significantly changed (P⫽0.193).
For risk factors, frequencies of hypertension, diabetes melli-
tus, smoking, and prior stroke have decreased statistically
(P⬍0.001), but the changes were not substantial. A complex
pattern was noted for dyslipidemia. Its frequency decreased
slowly from 2002 to 2007 and then increased again after
2008. Frequency of potential source of CE remained stable
during the study period.
Stroke Subtypes
Of 46 098 patients, 36 191 with ischemic stroke were
subjected to the stroke subtype analysis after excluding 3117
with TIA (6.8%) and 6790 with missing values based on
TOAST classification (14.7%). LAA stroke was the most
frequent stroke subtype (36.1%) followed by SVO (25.4%),
CE (17.1%), and other (stroke of other etiology, 1.8%; stroke
of undetermined etiology, 19.6%). In Poisson regression
models, the relative proportion of SVO decreased, with an
RR of 0.96 per year (95% CI, 0.95– 0.97) between 2002 and
2010, whereas CE increased, with an RR of 1.06 per year
(95% CI, 1.05–1.08) during the same period (P⬍0.001)
(Table 2). This trend in the stroke subtype is also shown in
Figure 2A. The relative proportion of LAA did not show any
directional trend across time periods over the 9 years of the
study. Although stroke of undetermined etiology proportions
have decreased (RR, 0.99; P⫽0.021), extensive investigation
still failed to define stroke mechanisms in one sixth of the
patients. Stroke of other etiology proportions have increased
(RR, 1.08; P⬍0.001), but its contributions were minute at ⬍3%.
Distribution of Stenoocclusive Lesions
Of 13 061 patients with LAA subtype, angiographic data
were available for 11 123 (85.2%). Stenoocclusive lesions
relevant for the index stroke were classified into A-ICAS,
A-ECAS, A-ICAS⫹A-ECAS (patients with carotid territory
stroke having tandem lesions), and posterior circulation
stenosis. Between 2000 and 2010, relative proportions of
A-ECAS (RR, 1.07; P⬍0.001) and posterior circulation
stenosis (RR, 1.04; P⬍0.001) have increased over time,
Table 1. Regression Models of Demographic and Risk Factors, Thrombolysis, and Mortality by Calendar Year of Stroke Diagnosis in
Korea, 2002 to 2010
2002
(n⫽798)
2003
(n⫽2738)
2004
(n⫽4100)
2005
(n⫽4791)
2006
(n⫽5501)
2007
(n⫽5521)
2008
(n⫽8262)
2009
(n⫽8294)
2010
(n⫽6093)
␤
or
RR 95% CI PValue
Age, y 64.5⫾11.6 64.8⫾11.8 65.4⫾12.1 65.9⫾12.1 66.0⫾12.0 66.0⫾12.4 66.4⫾12.4 66.6⫾12.5 66.7⫾12.6 0.237 0.186– 0.288 ⬍0.001
Male sex 439 (55.0) 1557 (56.9) 2390 (58.3) 2716 (56.7) 3152 (57.3) 3170 (57.4) 4768 (57.7) 4785 (57.7) 3589 (58.9) 1.00 1.00–1.01 0.193
TIA 65 (8.1) 173 (6.3) 265 (6.5) 344 (7.2) 312 (5.7) 347 (6.3) 551 (6.7) 614 (7.4) 446 (7.3) 1.02 1.00–1.03 0.040
HT 527 (66.0) 1805 (65.9) 2729 (66.6) 3046 (63.6) 3486 (63.4) 3356 (60.8) 5201 (63.0) 5340 (64.4) 3776 (62.0) 0.99 0.99–1.00 0.014
DM 250 (31.3) 835 (30.5) 1285 (31.3) 1478 (30.8) 1643 (29.9) 1661 (30.1) 2426 (29.4) 2522 (30.4) 1763 (28.9) 0.99 0.98–1.00 0.042
Dyslipidemia 161 (20.2) 551 (20.1) 938 (22.9) 943 (19.7) 1161 (21.1) 1130 (20.5) 1226 (14.8) 1418 (17.1) 1359 (22.3) 0.98 0.97–0.99 ⬍0.0001
PSCE 147 (18.4) 494 (18.0) 890 (21.7) 921 (19.2) 1032 (18.8) 999 (18.1) 1590 (19.2) 1746 (21.1) 1146 (18.8) 1.00 0.99–1.01 0.601
Smoking 254 (31.8) 980 (35.8) 1496 (36.5) 1651 (34.5) 1840 (33.4) 1790 (32.4) 2531 (30.6) 2737 (33.0) 1917 (31.5) 0.98 0.98–0.99 ⬍0.0001
Previous stroke 190 (23.8) 595 (21.7) 921 (22.5) 985 (20.6) 1032 (18.8) 975 (17.7) 1648 (19.9) 1596 (19.2) 1150 (18.9) 0.98 0.97–0.99 ⬍0.0001
Visit (⬍3 h) 161 (20.2) 624 (22.8) 921 (22.5) 1061 (22.1) 1315 (23.9) 1509 (27.3) 2250 (27.2) 2216 (26.7) 1743 (28.6) 1.04 1.03–1.05 ⬍0.001
Any thrombolysis 42 (5.3) 115 (4.2) 156 (3.8) 252 (5.3) 373 (6.8) 448 (8.1) 630 (7.6) 681 (8.2) 429 (7.0) 1.09 1.07–1.11 ⬍0.001
rt-PA 35 (4.4) 102 (3.7) 131 (3.2) 194 (4.0) 293 (5.3) 350 (6.3) 533 (6.5) 577 (7.0) 365 (6.0) 1.10 1.08–1.12 ⬍0.001
30-d mortality 32 (4.0) 110 (4.0) 147 (3.6) 170 (3.6) 195 (3.5) 202 (3.7) 344 (4.2) 335 (4.0) 236 (3.9) … … …
UA ………………………1.01 0.99–1.04 0.211
AA ………………………1.00 0.98–1.02 0.966
1-y mortality 89 (11.2) 369 (13.5) 499 (12.2) 559 (11.7) 639 (11.6) 668 (12.1) 1031 (12.5) 1040 (12.5) 633 (10.4) … … …
UA ………………………0.99 0.98–1.00 0.109
AA ………………………0.97 0.96–0.99 ⬍0.001
Data are presented as mean⫾SD or n (%), unless otherwise indicated. Pvalue and age were analyzed by linear regression and other categorical variables by
log-linear Poisson regression. RR indicates relative risk; TIA, transient ischemic attack; HT, hypertension; DM, diabetes mellitus; PSCE, potential source of
cardioembolism; rt-PA, recombinant tissue-type plasminogen activator; UA, unadjusted; AA, age adjusted.
Jung et al Secular Trends in Korean Stroke 329
Downloaded from http://ahajournals.org by on August 7, 2018

whereas the A-ICAS proportion has declined (RR, 0.95;
P⬍0.001) (Table 2). This trend in the LAA subtype is also
shown in Figure 2B.
Interval of Onset to Arrival and Thrombolysis Rate
For all patients, the average time interval between symptom
onset and admission was 28.4⫾36.0 hours. As shown in
online-only Data Supplement Figure I, the time interval has
significantly decreased from 31.7⫾37.6 to 26.8⫾33.9 hours
between 2002 and 2010 (P⬍0.001 for trend). The proportion
of patients admitted within 3 hours of symptom onset has
increased over time (RR, 1.04; P⬍0.001) (Table 1, Figure
3A) from 20.2% in 2002 to 28.6% in 2010. In concert with
the decline in prehospital delay, intravenous recombinant
tissue-type plasminogen activator use increased from 4.4% in
2002 to 6.0% in 2010, and any reperfusion therapy increased
from 5.3% in 2002 to 7.0% in 2010 (Table 1, Figure 3B).
Stroke Severity on Admission and Mortality
The average NIHSS score was 4 (interquartile range, 2–9),
indicating that the majority of enrolled patients had a mild to
moderate stroke severity. For patients arriving within 24
hours from onset, the admission NIHSS score markedly
declined between 2002 and 2005 (P⬍0.001) (online-only
Data Supplement Figure II), but it did not further decrease
after 2005. Overall, all-cause mortality rates were 4% at 30
days and 12% at 1 year (Table 1). Age-adjusted analyses did
not show any significant decline in 30-day mortality
(P⫽0.966) but demonstrated a significant decline in 1-year
mortality (P⬍0.001) (Figure 3C).
Discussion
Changing trends in stroke epidemiology in Korea should be
understood in the context of the country’s extremely rapid
economic growth and globalization. Korea has achieved a
remarkably high level of economic growth in a short period
of time. Its economy is already the 10th largest in the world,
and the rapid pace of its economic development over the past
3 decades is the highest according to the Organization for
Economic Cooperation and Development.
6
In this period of
transition, rapid industrial and societal changes could lead to
dynamic changes in stroke phenotypes. The current nation-
wide stroke registry study indicates that even during a 9-year
period within the first decade of 21st century, there were notable
changes in age, risk factors, stroke subtypes, the public seeking
out acute medical care, and reperfusion therapy. Understanding
these dynamic trends in patient- and care-related domains could
Figure 1. Secular trends for demographic and risk factors among Korean patients with ischemic stroke (2002–2010). The graphs repre-
sent age, sex, TIA, and cerebrovascular risk profiles by time period. Estimates were from linear regression (age) or Poisson regression
analysis (other variables), with the 2002 group as the reference. DM indicates diabetes mellitus; HT, hypertension; PSCE, potential
source of cardioembolism; RR, relative risk; TIA, transient ischemic attack.
330 Circ Cardiovasc Qual Outcomes May 2012
Downloaded from http://ahajournals.org by on August 7, 2018

help to guide health providers and policymakers in developing
guidelines for stroke prevention and care in Korea as well as in
other rapidly developing countries.
The secular changes in demographic and risk factors are
likely to impose changes in stroke subtypes, treatment, and
outcomes.
11
Mean age of patients with stroke increased by
⬎2 years from 64.5 to 66.7 years during the 9-year period,
and the increase was greater in women than in men. Popula-
tion aging in Korea is projected to be the fastest in the
Organization for Economic Cooperation and Development
area.
6
Aging of the present stroke population might be
attributable to an increase in life expectancy of the general
Korean population, which increased by ⬎4 years from 76.5
years (men, 72.8 years; women, 80.0 years) in 2001 to 80.6
Figure 2. Secular trends for stroke subtypes and distribution of stenoocclusive lesions on angiography in the LAA subtype among
Korean patients with ischemic stroke (2002–2010). A, Subtypes of ischemic stroke by time period. B, Distribution of stenoocclusive
lesions by time period. Data were from 11 123 subjects who had the LAA subtype. Lesion distribution was classified into A-ICAS,
A-ECAS, A-ICAS⫹A-ECAS, and P-CAS. Estimates were from Poisson regression models, with the 2002 group as the reference.
A-ECAS indicates extracranial arterial stenosis of anterior circulation; A-ICAS, intracranial arterial stenosis of anterior circulation; CE,
cardioembolism; LAA, large artery atherosclerosis; P-CAS, posterior circulation stenosis; RR, relative risk; SOE, stroke of other deter-
mined etiology; SUE, stroke of undetermined etiology; SVO, small vessel occlusion.
Table 2. Regression Models of Stroke Subtypes and Distribution of Stenoocclusive Lesions by Calendar Year of Stroke Diagnosis in
Korea, 2002 to 2010
2002
(n⫽700)
2003
(n⫽2510)
2004
(n⫽3774)
2005
(n⫽4229)
2006
(n⫽4290)
2007
(n⫽4277)
2008
(n⫽6082)
2009
(n⫽6406)
2010
(n⫽3923) RR 95% CI PValue
LAA 262 (37.4) 950 (37.8) 1299 (34.4) 1494 (35.3) 1462 (34.1) 1572 (36.8) 2204 (36.2) 2363 (36.9) 1455 (37.1) 1.01 1.00 –1.01 0.174
SVO 224 (32.0) 706 (28.1) 1124 (29.8) 1174 (27.8) 1142 (26.6) 1052 (24.6) 1436 (23.6) 1433 (22.4) 915 (23.3) 0.96 0.95–0.97 ⬍0.0001
CE 103 (14.7) 331 (13.2) 589 (15.6) 583 (13.8) 634 (14.8) 744 (17.4) 1157 (19.0) 1252 (19.5) 803 (20.5) 1.06 1.05–1.08 ⬍0.0001
SOE 9 (1.3) 22 (0.9) 79 (2.1) 73 (1.7) 62 (1.4) 68 (1.6) 96 (1.6) 140 (2.2) 101 (2.6) 1.08 1.04–1.11 ⬍0.0001
SUE 102 (14.6) 501 (20.0) 683 (18.1) 905 (21.4) 990 (23.1) 841 (19.7) 1189 (19.5) 1218 (19.0) 649 (16.5) 0.99 0.98–1.00 0.021
LAA Subgroup
2002
(n⫽193)
2003
(n⫽756)
2004
(n⫽1125)
2005
(n⫽1329)
2006
(n⫽1208)
2007
(n⫽1345)
2008
(n⫽1909)
2009
(n⫽2106)
2010
(n⫽1152) RR 95% CI PValue
A-ICAS 96 (39.7) 404 (53.4) 570 (50.7) 634 (47.7) 565 (46.8) 633 (47.1) 746 (39.1) 824 (39.1) 452 (39.2) 0.95 0.94–0.97 ⬍0.0001
A-ECAS 12 (6.2) 54 (7.1) 113 (10.0) 152 (11.4) 142 (11.8) 184 (13.7) 268 (14.0) 270 (12.8) 153 (13.3) 1.07 1.04–1.09 ⬍0.0001
A-ICAS⫹A-ECAS 13 (6.7) 55 (7.3) 83 (7.4) 103 (7.8) 111 (9.2) 90 (6.7) 136 (7.1) 170 (8.1) 74 (6.4) 0.99 0.96–1.02 0.650
P-CAS 72 (37.3) 243 (32.1) 359 (31.9) 440 (33.1) 390 (32.3) 438 (32.6) 759 (39.8) 842 (40.0) 473 (41.1) 1.04 1.03–1.06 ⬍0.0001
Data are presented as n (%), unless otherwise indicated. Pvalue was analyzed by log-linear Poisson regression. RR indicates relative risk; LAA, large artery
atherosclerosis; SVO, small vessel occlusion; CE, cardioembolism; SOE, stroke of other etiology; SUE, stroke of undetermined etiology; A-ICAS, intracranial arterial
stenosis of anterior circulation; A-ECAS, extracranial arterial stenosis of anterior circulation; P-CAS, posterior circulation stenosis.
Jung et al Secular Trends in Korean Stroke 331
Downloaded from http://ahajournals.org by on August 7, 2018

years in 2009 (men, 77.0 years; women, 83.8 years).
12
The
steep increase of age at stroke onset as well as the general
population’s life expectancy indicates a continuing increase
of elderly patients with stroke who are generally at higher risk
for poor prognosis and less responsive to intervention. Ac-
cordingly, societal burden of the elderly stroke population
would be a significant concern in coming decades in Korea.
Meanwhile, risk factor proportions of hypertension, diabe-
tes, smoking, and prior stroke have declined slightly. A recent
systematic review indicates a slowly increasing trend in the
prevalence in hypertension, diabetes mellitus, and dyslipid-
emia in the community.
13
However, the prevalence in the risk
factors can vary depending on survey area, time, and age, sex,
and disease distribution of the study population. The ideal
stroke registry would be population based. Unfortunately, to
date, there has been no systematic investigation for the
prevalence of risk factors with respect to Korean ischemic
stroke. In practice, it is very difficult to establish such an ideal
stroke registry because of methodological and financial lim-
itations. The current study is the first detailed analysis to our
knowledge that is based on the KSR, which ensures more-
accurate and comprehensive nationwide data.
As widely recognized, hypertension was the most predom-
inant risk factor in the population, but its prevalence in the
stroke population decreased over time. In elderly patients, the
occurrence of stroke attributable to hypertension is lower than
in younger patients.
14
The increase of elderly patients with
stroke might be one factor that explains the declining trend of
hypertension frequency in the stroke population. On the other
hand, hypertension is more closely associated with SVO than
with CE.
15
Thus, the relative decrease of SVO and increase of
CE would be another factor associated with the declining
trend of hypertension frequency. However, taking its high
prevalence and proven treatment efficacy in the elderly into
account, the finding does not negate the priority of hyperten-
sion control for stroke prevention.
Smoking is also a leading risk factor in all stroke sub-
types.
16
It should be noted that one third of the patients with
stroke were current smokers or had a history of smoking of
⬎2 years, despite its decreasing trend. Compared with the
global stroke population enrolled in REACH (Reduction of
Atherothrombosis for Continued Health) Registry,
17
the pres-
ent patients with stroke had a lower rate of current or previous
smoking. This low rate is likely attributable to a very low rate
of smoking in women. The National Health and Nutrition
Survey in 2005 reported that smoking prevalence was 53.3%
in men and 5.8% in women aged ⬎15 years in Korea.
18
This
gender difference in smoking prevalence in Korea is much
higher than in the US population (22.3% in men and 17.4% in
women).
19
When restricted to the male stroke population, the
rate of current or previous smoking was 52.6% and compa-
rable to or higher than those of other countries. Because
smoking is a major preventable risk factor, educational efforts
to prevent or stop smoking should be further encouraged.
In the study, dyslipidemia was found in less than one fourth
of patients with stroke, which is lower than the 30% to 50%
observed in other studies.
17,20,21
However, it should be noted
that we used a high threshold of total cholesterol (⬎240
mg/dL) and low-density lipoprotein cholesterol (⬎160 mg/
dL) to define dyslipidemia. When we started the KSR
registry, definitions of hypercholesterolemia were not unified
worldwide, and our threshold was the insurance reimburse-
ment criteria for treating hypercholesterolemia. Therefore,
with contemporary criteria applied, the dyslipidemia preva-
lence would be greater. Hypercholesterolemia within the
REACH Registry was present in 67% of the population in
Eastern Europe, whereas it is 45% in Asia. For carotid plaque,
the prevalence rates ranged from 45.1% in Western Europe to
20.5% in Asia.
17
The present observation of a recent increase
in dyslipidemia is consonant with the findings of an earlier
study that demonstrated an increasing trend in dyslipidemia
prevalence in Asians.
22
This changing trend indicates an
Figure 3. Secular trends for visit within a 3-hour window, thrombolytic therapy, and mortality among Korean patients with ischemic
stroke (2002–2010). A, Patients who sought evaluation at the hospital within 3 hours after the onset of symptoms by time period. B,
Patients who underwent any thrombolytic and rt-PA therapy by time period. C, Changes in 30-day and 1-year mortality after incident
stroke by time period. Estimates were from Poisson regression models, with the 2002 group as the reference. rt-PA indicates recombi-
nant tissue-type plasminogen activator.
332 Circ Cardiovasc Qual Outcomes May 2012
Downloaded from http://ahajournals.org by on August 7, 2018

adverse transition of vascular risk factors to a more athero-
genic profile.
Although the relative proportion of SVO decreased, the
proportion of CE increased from 14.7% to 20.5%, which is
less than, but still approaching, the 25.6% of CE finding in
the German Stroke Data Bank.
23
Increased physician aware-
ness of the importance of CE source detection leading to a
more-extensive work-up for this condition would partly
explain the increase of CE. Actually, the performance fre-
quency of diagnostic modalities, including transthoracic
echocardiography, transesophageal echocardiography, and
Holter monitoring, tended to increase across time periods
over the 9 years studied (online-only Data Supplement Figure
III, online-only Data Supplement Table III), but the changes
were not substantial. A prior study reported that even after
adjusting for age, atrial fibrillation incidence increased during
the past 2 decades.
24
If the incidence continuously increases,
the projected number of adults with atrial fibrillation would
increase by 3-fold from 2000 to 2050.
25
Although atrial
fibrillation was rare (0.1%– 0.5%) in middle-aged patients,
the prevalence was 3.5% to 9% in elderly patients.
6,7
The
increasing trend of CE in Korea is likely to be caused by the
increased number of elderly patients with atrial fibrilla-
tion.
25,26
The geographical variations of atrial fibrillation
within the REACH Registry have been reported to range from
7.7% in Asia to 16% in North America.
17
However, our
registry, in which the potential source of CE was classified
only into a high-risk or medium-risk source, did not allow us
to delineate the change of exact atrial fibrillation prevalence.
CE stroke usually is more disabling and fatal and leads to a
higher risk for recurrent stroke than other stroke subtypes.
11
In Korea, CE stroke is expected to continue increasing and
should be aggressively prevented with optimal antithrombotic
therapies.
The frequencies of distribution of intracranial and extracra-
nial artery stenosis vary among different races and ethnici-
ties.
27,28
The ratio of intracranial stenosis to extracranial
stenosis was reported be 4:1 in Asian patients with stroke.
29,30
In Korean patients with stroke, the A-ICAS is still nearly 3
times the A-ECAS, but the ratio markedly decreased from
6.4-fold to 2.9-fold, and the extracranial stenosis is on the
rise. Intracranial and extracranial artery stenoses have differ-
ent risk factors and different pathophysiologies.
29,30
In an
autopsy study of patients with ischemic stroke, unstable
plaque was a common cause of extracranial carotid occlusion,
but in intracranial occlusion, fibrocalcific narrowing was the
main cause of occlusion rather than unstable plaque.
29
Westernization of diet, increasing cholesterol level, and
obesity in the general population may contribute to the
increase of extracranial carotid disease in Korea. Of note,
significant relevant A-ICAS⫹A-ECAS was found in 7% of
the LAA subtype. Given that tandem lesions are common in
Asian patients with stroke and carry a higher risk for further
vascular events or poor outcomes,
31,32
this patient subgroup
would be the subject of future drug or intervention trials.
Prehospital delay is the most important obstacle in
thrombolytic therapy. During the study period, hospital ar-
rival within a 3-hour window improved by a relative 43%
increase from 20% to 29%, and the thrombolysis rate im-
proved by 1.3-fold from 5.3% to 7.0%. Improved acute care
systems and educational programs in the general population
and high-risk patients might contribute to these favoring
trends. Actually, in 2007, the number of hospital beds;
hospital length of stay; and number of physicians, nursing
personnel, and consulting physicians in Korea were beyond
those of the Organization for Economic Cooperation and
Development average and comparable to a high-income
country.
33
However, patients arriving within the 3-hour win-
dow are still fewer than those of other Asian and western
countries.
34–36
Given that patient or bystander awareness of
stroke symptoms have been associated with shorter prehos-
pital delays,
35,37
expert endeavors to improve public aware-
ness should be continued. Early mortality at 30 days was
unchanged over time, whereas long-term mortality at 1 year
modestly, but significantly decreased. The absence of de-
tailed information regarding the cause of death did not allow
us to fully explore the reason for the discrepant trends
between early and long-term mortality rates. The secular
improvement of the overall healthcare system could decrease
a long-term mortality rate, but the increase in comorbidities
and CE stroke according to advancing age could negate the
beneficial effects of other factors. Further studies are required
to disclose the precise reasons for mortality trends.
This study has several limitations. The findings were
derived from a hospital-based registry, and most participating
centers were neurology training hospitals. Therefore, the
extrapolation of the findings to the general Korean stroke
population might be limited. This study did not include all
consecutive patients. Of all registered patients, ⬇5% were
excluded from this analysis because of missing information.
In addition, data from patients with stroke who arrived at the
emergency department but were not hospitalize were not
captured. Nonetheless, this nationwide hospital-based regis-
try has many advantages in that it can clarify diagnoses,
classify subtypes, assess risk factors related to etiology, and
observe treatment and outcomes.
Acknowledgments
We are grateful to the Medical Research Collaborating Center at
Seoul National University Hospital for the contribution of statistical
methods.
Sources of Funding
The KSR is supported by the Korean Stroke Association (2002–present).
Disclosures
None.
References
1. Rosamond W, Flegal K, Furie K, Go A, Greenlund K, Haase N, Hailpern
SM,Ho M, Howard V, Kissela B, Kittner S, Lloyd-Jones D, McDermott
M, Meigs J, Moy C, Nichol G, O’Donnell C, Roger V, Sorlie P, Stein-
berger J, Thom T, Wilson M, Hong Y; American Heart Association
Statistics Committee and Stroke Statistics Subcommittee. Heart disease
and stroke statistics—2008 update. Circulation. 2008;117:e25– e146.
2. Mathers CD, Lopez AD, Murray CJL. The burden of disease and mor-
tality by condition: data, methods, and results for 2001. In Lopez AD,
Mathers CD, Ezzati M, Jamison DT, Murray CJL, eds. Global Burden of
Disease and Risk Factors. Washington, DC: World Bank; 2006:45–240.
3. Sarti C, Rastenyte D, Cepaitis Z, Tuomilehto J. International trends in
mortality from stroke, 1968 –1994. Stroke. 2000;31:1588 –1601.
Jung et al Secular Trends in Korean Stroke 333
Downloaded from http://ahajournals.org by on August 7, 2018

4. Hong KS, Kim J, Cho YJ, Seo SY, Hwang SI, Kim SC, Kim JE, Kim A,
Cho JY, Park HK, Bae HJ, Yang MH, Jang MS, Han MK, Lee J, Kang
DW, Park JM, Koo J, Yu KH, Oh MS, Lee BC. Burden of ischemic stroke
in Korea: analysis of disability-adjusted life years lost. J Clin Neurol.
2011;7:77– 84.
5. Bhatt DL, Steg PG, Ohman EM, Hirsch AT, Ikeda Y, Mas JL, Goto S,
Liau CS, Richard AJ, Ro¨ther J, Wilson PW; REACH Registry Investi-
gators. International prevalence, recognition, and treatment of cardiovas-
cular risk factors in outpatients with atherothrombosis. JAMA. 2006;295:
180 –189.
6. Organization for Economic Cooperation and Development. OECD
Economic Surveys: Korea 2010. OECD Web site. http://www.oecd.org/
dataoecd/14/34/45432048.pdf. Accessed April 2011.
7. Lee BC, Roh JK; Korean Stroke Registry. International experience in
stroke registries: Korean Stroke Registry. Am J Prev Med. 2006;31:
S243–S245.
8. Adams HP Jr, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL,
Marsh EE III. Classification of subtype of acute ischemic stroke. Defi-
nitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172
in Acute Stroke Treatment. Stroke. 1993;24:35– 41.
9. Samuels OB, Joseph GJ, Lynn MJ, Smith HA, Chimowitz MI. A stan-
dardized method for measuring intracranial arterial stenosis. AJNR Am J
Neuroradiol. 2000;21:643– 646.
10. North American Symptomatic Carotid Endarterectomy Trial Collabo-
rators. Beneficial effect of carotid endarterectomy in symptomatic
patients with high-grade stenosis. N Engl J Med. 1991;325:445– 453.
11. Kolominsky-Rabas PL, Weber M, Gefeller O, Neundoerfer B, Heu-
schmann PU. Epidemiology of ischemic stroke subtypes according to
TOAST criteria: incidence, recurrence, and long-term survival in ische-
mic stroke subtypes: a population-based study. Stroke. 2001;32:
2735–2740.
12. Statistics Korea. 2009 Life Tables for Korea. Statistics Korea Web site.
http://kostat.go.kr/portal/english/news/1/1/index.board?bmode⫽read
&bSeq⫽&aSeq⫽244586&pageNo⫽1&rowNum⫽10&navCount⫽10
&currPg⫽&sTarget⫽title&sTxt⫽. Accessed April 2011.
13. Park TH, Kim MK, Lee KB, Park JM, Lee SJ, Jung KH, Cho YJ, Lee JS,
Lee JY, Bae HJ. Prevalence of risk factors for ischemic stroke in Korean:
a systematic review. J Korean Neurol Assoc. 2009;27:19 –27.
14. Curb JD, Abbott RD, MacLean CJ, Rodriguez BL, Burchfiel CM, Sharp
DS, Ross GW, Yano K. Age-related changes in stroke risk in men with
hypertension and normal blood pressure. Stroke. 1996;27:819 – 824.
15. Inzitari D, Giordano GP, Ancona AL, Pracucci G, Mascalchi M,
Amaducci L. Leukoaraiosis, intracerebral hemorrhage, and arterial hyper-
tension. Stroke. 1990;21:1419 –1423.
16. Donnan GA, McNeil JJ, Adena MA, Doyle AE, O’Malley HM, Neill GC.
Smoking as a risk factor for cerebral ischaemia. Lancet. 1989;2:643–647.
17. Ro¨ther J, Alberts MJ, Touze´ E, Mas JL, Hill MD, Michel P, Bhatt DL,
Aichner FT, Goto S, Matsumoto M, Ohman EM, Okada Y, Uchiyama S,
D’Agostino R, Hirsch AT, Wilson PW, Steg PG; REACH Registry
Investigators. Risk factor profile and management of cerebrovascular
patients in the REACH Registry. Cerebrovasc Dis. 2008;25:366 –374.
18. Report on 2005 National Health and Nutrition Survey. Gyeonggi-do,
Korea: Ministry of Health and Social Welfare; 2006.
19. Goldstein LB, Bushnell CD, Adams RJ, Appel LJ, Braun LT, Chaturvedi
S, Creager MA, Culebras A, Eckel RH, Hart RG, Hinchey JA, Howard
VJ, Jauch EC, Levine SR, Meschia JF, Moore WS, Nixon JV, Pearson
TA; American Heart Association Stroke Council; Council on Cardio-
vascular Nursing; Council on Epidemiology and Prevention; Council for
High Blood Pressure Research; Council on Peripheral Vascular Disease;
Interdisciplinary Council on Quality of Care and Outcomes Research.
Guidelines for the primary prevention of stroke: a guideline for healthcare
professionals from the American Heart Association/American Stroke
Association. Stroke. 2011;42:517–584.
20. Deleu D, Hamad AA, Kamram S, El Siddig A, Al Hail H, Hamdy SM.
Ethnic variations in risk factor profile, pattern and recurrence of non-
cardioembolic ischemic stroke. Arch Med Res. 2006;37:655– 662.
21. Sacco RL, Boden-Albala B, Abel G, Lin IF, Elkind M, Hauser WA, Paik
MC, Shea S. Race-ethnic disparities in the impact of stroke risk factors:
the Northern Manhattan Stroke Study. Stroke. 2001;32:1725–1731.
22. Gunarathne A, Patel JV, Potluri R, Gill PS, Hughes EA, Lip GY. Secular
trends in the cardiovascular risk profile and mortality of stroke
admissions in an inner city, multiethnic population in the United
Kingdom (1997–2005). J Hum Hypertens. 2008;22:18 –23.
23. Grau AJ, Weimar C, Buggle F, Heinrich A, Goertler M, Neumaier S,
Glahn J, Brandt T, Hacke W, Diener HC. Risk factors, outcome, and
treatment in subtypes of ischemic stroke: the German Stroke Data Bank.
Stroke. 2001;32:2559 –2566.
24. Miyasaka Y, Barnes ME, Gersh BJ, Cha SS, Bailey KR, Abhayaratna
WP, Seward JB, Tsang TS. Secular trends in incidence of atrial fibril-
lation in Olmsted County, Minnesota, 1980 to 2000, and implications on
the projections for future prevalence. Circulation. 2006;114:119 –125.
25. Nakayama T, Yokoyama T, Yoshiike N, Zaman MM, Date C, Tanaka H,
Detels R. Population attributable fraction of stroke incidence in
middle-aged and elderly people: contributions of hypertension, smoking
and atrial fibrillation. Neuroepidemiology. 2000;19:217–226.
26. Go AS, Hylek EM, Phillips KA, Chang Y, Henault LE, Selby JV, Singer
DE. Prevalence of diagnosed atrial fibrillation in adults: national impli-
cations for rhythm management and stroke prevention: the AnTicoagu-
lation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA.
2001;285:2370 –2375.
27. Li H, Wong KS. Racial distribution of intracranial and extracranial
atherosclerosis. J Clin Neurosci. 2003;10:30 –34.
28. De Silva DA, Woon FP, Lee MP, Chen CP, Chang HM, Wong MC. South
Asian patients with ischemic stroke: intracranial large arteries are the
predominant site of disease. Stroke. 2007;38:2592–2594.
29. Lammie GA, Sandercock PA, Dennis MS. Recently occluded intracranial
and extracranial carotid arteries. Relevance of the unstable atherosclerotic
plaque. Stroke. 1999;30:1319 –1325.
30. Suwanwela NC, Chutinetr A. Risk factors for atherosclerosis of cervico-
cerebral arteries: intracranial versus extracranial. Neuroepidemiology.
2003;22:37– 40.
31. Man BL, Fu YP, Chan YY, Lam W, Hui CF, Leung WH, Wong KS. Use
of magnetic resonance angiography to predict long-term outcomes of
ischemic stroke patients with concurrent stenoses in Hong Kong. Cere-
brovasc Dis. 2009;28:112–118.
32. Wong KS, Li H. Long-term mortality and recurrent stroke risk among
Chinese stroke patients with predominant intracranial atherosclerosis.
Stroke. 2003;34:2361–2366.
33. Organization for Economic Cooperation and Development. OECD Health
Data 2009: Statistics and Indicators for 30 Countries. Paris, France:
Organization for Economic Cooperation and Development; 2009.
34. Kimura K, Kazui S, Minematsu K, Yamaguchi T; Japan Multicenter
Stroke Investigator’s Collaboration. Analysis of 16,922 patients with
acute ischemic stroke and transient ischemic attack in Japan. A
hospital-based prospective registration study. Cerebrovasc Dis. 2004;18:
47–56.
35. Agyeman O, Nedeltchev K, Arnold M, Fischer U, Remonda L, Isenegger
J, Schroth G, Mattle HP. Time to admission in acute ischemic stroke and
transient ischemic attack. Stroke. 2006;37:963–966.
36. Derex L, Adeleine P, Nighoghossian N, Honnorat J, Trouillas P. Factors
influencing early admission in a French stroke unit. Stroke. 2002;33:
153–159.
37. Kim YS, Park SS, Bae HJ, Cho AH, Cho YJ, Han MK, Heo JH, Kang K,
Kim DE, Kim HY, Kim GM, Kwon SU, Kwon HM, Lee BC, Lee KB,
Lee SH, Lee SH, Lee YS, Nam HS, Oh MS, Park JM, Rha JH, Yu KH,
Yoon BW. Stroke awareness decreases prehospital delay after acute
ischemic stroke in Korea. BMC Neurol. 2011;11:2.
334 Circ Cardiovasc Qual Outcomes May 2012
Downloaded from http://ahajournals.org by on August 7, 2018