Original Paper
Cerebrovasc Dis 2007;23:117–120
DOI: 10.1159/000097047
Recurrence after Ischemic Stroke in
Chinese Patients: Impact of Uncontrolled
Modifiable Risk Factors
Gelin Xu Xinfeng Liu Wentao Wu Renliang Zhang Qin Yin
Department of Neurology, Jinling Hospital, Nanjing University School of Medicine, Nanjing, PR China
Key Words
Stroke recurrence
Hypertension
Stroke risk determinants

Atrial fibrillation
Smoking
Abstract
Background: Data concerning stroke occurrence and recur-
rence in China are extremely rare. This study was designed
to analyze determinants of stroke recurrence in a cohort of
Chinese patients. Methods: Subjects were patients with
ischemic stroke registered in the Nanjing Stroke Registry
Program. Modifiable risk factors for stroke were identified
and stratified into 3 levels: without, controlled and uncon-
trolled. Cox proportional hazard model was used to detect
influence factors for stroke recurrence. Results: First-year re-
currence rate was 11.2% in the registered patients. Hyper-
tension, atrial fibrillation (AF) and smoking were associated
with increased risk of recurrence. Controlling hypertension
and AF each halved recurrent risk (p ! 0.001). Ceasing smok-
ing for more than 1 year reduced hazard ratio of recurrence
from 1.71 to 1.39 (p ! 0.05). Controlling blood sugar level in
diabetics did not significantly change recurrent risk (hazard
ratio, 1.69 vs. 1.64, p 1 0.05). Conclusions: The recurrence
rate is higher in Chinese patients with ischemic stroke com-
pared with the one reported in western populations. Failure
to control some modifiable risk factors in Chinese patients
may be responsible for this discrepancy.
Copyright © 2007 S. Karger AG, Basel
© 2007 S. Karger AG, Basel
1015–9770/07/0233–0117$23.50/0
Fax +41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/ced
Received: March 24, 2006
Accepted: July 20, 2006
Published online: November 15, 2006
Stroke recurrence and its predictors have rarely been
investigated in China. Stroke prevention and manage-
ment guidelines designed for Chinese were tailored to a
considerable degree on the basis of epidemiological, etio-
logical and clinical profiles mirrored from western popu-
lations. But previous studies of stroke incidence and re-
currence discovered significant variances across coun-
tries of different socioeconomic backgrounds and among
populations of different ethnic origins [1, 2]. These vari-
ances emphasize the need for population-specific preven-
tive strategies on account of influence factors for stroke
occurrence and recurrence. Based on the data bank of the
first stroke registry program in mainland China, we in-
vestigated in this study the 1st-year recurrence and risk
determinants in a cohort of Chinese stroke patients.
Subjects and Methods
Subjects were patients registered in the Nanjing Stroke Regis-
try Program (NSRP). The detailed protocol for the NSRP has been
published previously [3]. All patients registered in the NSRP were
evaluated for eligibility for enrollment. Inclusion criteria includ-
ed having first-ever ischemic stroke, being evaluated by a neu-
rologist within 7 days of stroke onset, having at least one CT or
MRI scan during hospitalization, being aged 18 years or older.
Because status of risk factor controlling in patients who died of
index stroke was mostly unknown, they were excluded from data
analysis. This study was approved by the Ethical Review Board of
Jinling Hospital.
Xinfeng Liu
Department of Neurology, Jinling Hospital
Nanjing University School of Medicine, 305 East Zhongshan Road
Nanjing, 210002, Jiangsu Province (PR China)
Tel. +86 25 8086 0454, Fax +86 25 8480 1861, E-Mail xfliu2@gmail.com
 Baseline assessment and diagnosis were made, optimally
within 7 days after stroke onset. Only ischemic stroke was in-
cluded in the study, which was further classified, with reference
to TOAST criteria, as lacunar, atherothrombolic, cardioembolic,
and stroke of other determined or undetermined causes [4]. Hy-
pertension, diabetes mellitus (DM), and hyperlipidemia were di-
agnosed according to established criteria. The included patients
were followed up at 2 or 3 monthly intervals after index stroke.
Modifiable risk factors for stroke were reevaluated at each follow-
up. Previously diagnosed hypertension was regarded as controlled
when blood pressure was lower than 140/90 mm Hg. DM was re-
garded as controlled when fasting serum glucose level was lower
than 126 mg/dl. Hyperlipidemia was regarded as controlled when
blood cholesterol level was lower than 240 mg/dl and triglyceride
level lower than 300 mg/dl. Atrial fibrillation (AF) was regarded
as under control when the international normalized ratio (INR)
was kept in the range of 2.0–3.0. Patients were considered as being
under appropriate antiplatelet therapy when they received aspirin
at a daily dose between 50 and 325 mg, or clopidogrel at 75 mg, or
ticlopidine at 500 mg after the index stroke.
Statistical analysis was performed utilizing SPSS
10.0 (Statis-
tical Product and Service Solutions Inc, Chicago, Ill., USA). Means
of parameter data were compared using one-way ANOVA. Unad-
justed statistical analyses were performed using
2 analysis for
dichotomous results. Cox proportional hazards model was used
to evaluate the influences on stroke recurrence imposed by poten-
tial risk factors. The cumulative risk of stroke recurrence by levels
of modifiable factors was estimated using Kaplan-Meier analysis.
The 1st-year recurrence risk was compared with log rank test.
Results
A total of 1,432 patients with first-ever ischemic stroke
had been registered in the NSRP and 834 patients who
met the inclusion criteria were included in data analysis.
There were 160 with a second stroke during the 1-year
follow-up, of whom 12 (7.5%) were hemorrhagic. The 1st-
year general recurrence rate was 11.2% (160/1,432). Of the
834 enrolled patients, 556 (66.7%) were male and 278
(33.3%) were female. Their ages ranged from 19 to 97
(68.5 8 12.4) years; 734 (88.1%) patients had an MRI
scan and 812 (97.4%) had a CT scan. Based on the index
stroke, 181 (21.7%) patients were classified as lacunar, 226
(27.1%) as atherothrombolic, 192 (23.0%) as cardioem-
bolic, and 235 (28.2%) as stroke of undetermined or oth-
er determined causes. Hundred and twenty-one (14.5%)
patients had a history of transient ischemic attacks (TIA).
Prevalence of hypertension, DM, hyperlipidemia and AF
was 48.7, 29.1, 19.8 and 15.3%, of which 17.4, 10.6, 5.6 and
6.1%, respectively, were uncontrolled. Nearly half (45.7%)
of the patients were not on antiplatelets after the index
stroke. Three hundred and sixty-nine (44.2%) patients
had a history of smoking, of whom 211 (25.3%) were cur-
118 Cerebrovasc Dis 2007;23:117–120
rent smokers and 158 (18.9%) were former smokers. More
than one third (36.7%) of the patients were regular alco-
hol drinkers.
Age, gender, educational level and regional residency
were found to have no remarkable influence on stroke re-
currence (p 1 0.05). Subtypes of ischemic stroke were as-
sociated with different levels of recurrent risk (p ! 0.001).
The impact of high blood pressure on stroke recurrence
was very significant when hypertension was stratified
into 3 levels (p ! 0.001). DM had a borderline significant
impact on stroke recurrence (p = 0.054). Patients with a
history of TIA and AF were more prone to have a recur-
rence than patients without (p ! 0.05). Smoking was
scarcely a significant influence factor for recurrence (p =
0.046). Alcohol intake did not influence recurrent risk
significantly (p = 0.266). Antiplatelet treatment was nega-
tively associated with stroke recurrence (p ! 0.05).
After univariate Cox proportional hazard evaluation,
6 factors, subtype of stroke, hypertension, AF, history of
TIA, smoking and antiplatelet treatment, entered multi-
variate analysis. Because DM had a borderline significant
influence on stroke recurrence (p = 0.054) and the impact
of modifiable risk factors was an emphasis of this study,
we also included DM in multivariate analysis. The results
are shown in table 1. Influence of hypertension and AF
on stroke recurrence reached a very significant level (p !
0.001). Patients with uncontrolled hypertension had a
higher risk for recurrence (hazard ratio, HR = 3.15, 95%
CI = 2.15–4.62). Controlling blood pressure in hyperten-
sives halved the risk for recurrence (HR, 1.54 vs. 3.15).
Patients with uncontrolled AF, not surprisingly, bear a
high risk for recurrence (HR = 4.70, 95% CI = 2.96–7.46).
Maintaining INR between 2.0 and 3.0 also halved the risk
for recurrence in AF patients (HR, 2.38 vs. 4.70). Recur-
rent risk in patients with controlled and uncontrolled
DM was very similar (HR, 1.64 vs. 1.69). Multivariate
analysis indicated that the impact of smoking on recur-
rence was significant (p ! 0.05). The risk is slightly high-
er in current smokers than in former smokers (HR, 1.71
vs. 1.39). History of TIA mildly increased the risk of
stroke recurrence (HR = 1.59, 95% CI = 1.07–2.38, p !
0.05). Antiplatelets slightly but significantly decreased
recurrence risk (HR for without antiplatelet = 1.41, 95%
CI = 1.02–1.94, p ! 0.05). The 1st-year cumulative risks
for recurrence by different levels of 4 modifiable risk
factors were evaluated by Kaplan-Meier analysis (fig. 1).
Log rank test identified significantly increased recurrent
risks among patients with uncontrolled hypertension
and AF (p ! 0.001). The risk for recurrence was also high-
er in current smokers than in non- or former smokers
Xu/Liu/Wu/Zhang/Yin
Table 1. Multivariate analysis of factors affecting 1st-year stroke
recurrence

Variable HR (95% CI) p

Subtype of stroke
Lacunar 1.00 0.001
Atherothrombolic 3.24 (1.78–5.92)
Cardioembolic 2.55 (1.35–4.82)
Undetermined 3.18 (1.74–5.81)
Hypertension
Without 1.00 0.000
Controlled 1.54 (1.05–2.24)
Uncontrolled 3.15 (2.15–4.62)
DM
Without 1.00 0.011
Controlled 1.64 (1.10–2.45)
Uncontrolled 1.69 (1.06–2.68)
AF
Without 1.00 0.000
Controlled 2.38 (1.39–4.08)
Uncontrolled 4.70 (2.96–7.46)
History of TIA
Without 1.00 0.023
With 1.59 (1.07–2.38)
Smoking
Nonsmoker 1.00 0.015
Former smoker 1.39 (0.92–2.11)
Current smoker 1.71 (1.18–2.47)
Antiplatelet treatment
With 1.00 0.036
Without 1.41 (1.02–1.94)

(p ! 0.05). No significant differences in recurrent risk

were detected between patients with different levels of
DM (p 1 0.05).

Discussion

The present study detected a 1st-year recurrence rate

of 11.2% in a cohort of Chinese patients with ischemic
stroke. This rate is higher compared with that reported
in some western populations [5]. Prevalence and failure
to treat some modifiable risk factors in Chinese patients
were postulated to be responsible for this increased risk.
After evaluating the impact of several modifiable risk
factors on stroke recurrence, we confirmed that uncon-
trolled hypertension and AF as well as current smoking Fig. 1. One-year recurrence rate at different levels of 4 risk factors
are responsible, at least partly, for the higher risk of stroke estimated by Kaplan-Meier analysis. Log rank test identified very
significant differences in recurrent risk between patients with
recurrence in Chinese patients. different levels of hypertension (HT) and AF (p ! 0.001). Recur-
Results of previous studies concerning the impact of rent risk also differed significantly among non-, former and cur-
hypertension on stroke recurrence were inconsistent [5– rent smokers (p ! 0.05).

Stroke Recurrence in Chinese Cerebrovasc Dis 2007;23:117–120 119

8]. Many of these studies utilized history of hypertension
or blood pressure at admission as a variant for risk evalu-
ation. Few of them emphasized the impact of long-term
uncontrolled hypertension on stroke recurrence. Preva-
lence of uncontrolled modifiable risk factors in this co-
hort of Chinese patients provided us a chance, which is
becoming less and less feasible in western populations
thanks to the increasing awareness and improving health-
care [9], to investigate their impact on stroke recur-
rence.
AF is associated with a high rate of ischemic stroke.
Oral anticoagulation can markedly decrease the risk for
stroke occurrence and recurrence in patients with AF
[10]. In this study, patients with uncontrolled AF had a
HR of 4.70 for stroke recurrence. By maintaining INR
between 2.0 and 3.0, the risk for recurrence almost halved.
These results are consistent with other studies [7, 10].
Reports concerning the impact of smoking on stroke
recurrence are equivocal. The Oxfordshire Community
Stroke Project detected an increased risk for recurrence
among smokers with first-ever stroke (HR 1.66, 95%
CI = 1.10–2.51) [11]. A predominant number of other
studies, however, did not associate smoking with stroke
recurrence [5–7, 12]. Most of these studies employed his-
tory of smoking or current smoking as a variant in risk
evaluation. A group of patients with intermediate status,
those who had stopped smoking, were assigned either as
smokers or nonsmokers. This arrangement may weaken
the power of differences because previous studies indi-
cated that former smokers bear an intervenient risk for
stroke [13]. When patients were stratified into non-, for-
mer and current smokers, a slightly but significantly
higher recurrent risk was detected for current smokers.
In conclusion, recurrence rate is higher in Chinese
stroke patients than in White patients. Failure to control
some modifiable risk factors in Chinese patients may be
responsible for this discrepancy. Further studies are war-
ranted to disclose the causes behind these uncontrolled
risk factors.

References

1 Malmgren R, Warlow C, Bamford J, Sander-
cock P: Geographical and secular trends in
stroke incidence. Lancet 1987; 2:1196–1200.
2 Bonita R, Stewart A, Beaglehole R: Interna-
tional trends in stroke mortality: 1970–1985.
Stroke 1990;21:989–992.
3 Liu X, Xu G, Wu W, Zhang R, Yin Q, Zhu W:
Subtypes and one-year survival of first-ever
stroke in Chinese patients: The Nanjing
Stroke Registry. Cerebrovasc Dis 2006; 22:
130–136.
4 Adams HP Jr, Bendixen BH, Kappelle LJ,
Biller J, Love BB, Gordon DL, Marsh EE 3rd:
Classification of subtype of acute ischemic
stroke. Definitions for use in a multicenter
clinical trial. TOAST. Trial of Org 10172 in
Acute Stroke Treatment. Stroke 1993;24:35–
41.
5 Moroney JT, Bagiella E, Paik MC, Sacco RL,
Desmond DW: Risk factors for early recur-
rence after ischemic stroke: the role of stroke
syndrome and subtype. Stroke 1998; 29:
2118–2124.
6 Appelros P, Nydevik I, Viitanen M: Poor out-
come after first-ever stroke: predictors for
death, dependency, and recurrent stroke
within the first year. Stroke 2003; 34: 122–
126.
7 Hillen T, Coshall C, Tilling K, Rudd AG, Mc-
Govern R, Wolfe CD; South London Stroke
Register: Cause of stroke recurrence is mul-
tifactorial: patterns, risk factors, and out-
comes of stroke recurrence in the South Lon-
don Stroke Register. Stroke 2003; 34:
1457–1463.
8 Friday G, Alter M, Lai SM: Control of hyper-
tension and risk of stroke recurrence. Stroke
2002;33:2652–2657.
9 Wang TJ, Vasan RS: Epidemiology of uncon-
trolled hypertension in the United States.
Circulation 2005;112:1651–1662.
10 Secondary prevention in non-rheumatic
atrial fibrillation after transient ischaemic
attack or minor stroke: EAFT (European
Atrial Fibrillation Trial) Study Group. Lan-
cet 1993;342:1255–1262.
11 Burn J, Dennis M, Bamford J, Sandercock P,
Wade D, Warlow C: Long-term risk of recur-
rent stroke after a first-ever stroke. The Ox-
fordshire Community Stroke Project. Stroke
1994;25:333–337.
12 Leira EC, Chang KC, Davis PH, Clarke WR,
Woolson RF, Hansen MD, Adams HP Jr: Can
we predict early recurrence in acute stroke?
Cerebrovasc Dis 2004;18:139–144.
13 Abbott RD, Yin Y, Reed DM, Yano K: Risk of
stroke in male cigarette smokers. N Engl J
Med 1986;315:717–720.

120 Cerebrovasc Dis 2007;23:117–120 Xu/Liu/Wu/Zhang/Yin