Smoking and the Risk of Hemorrhagic Stroke in Men

Tobias Kurth, MD, MSc; Carlos S. Kase, MD; Klaus Berger, MD, MPH, MSc;
Elke S. Schaeffner, MD, MSc; Julie E. Buring, ScD; J. Michael Gaziano, MD, MPH

Background and PurposeSmoking is an established risk factor for ischemic stroke and subarachnoid hemorrhage
(SAH), but the impact of smoking on intracerebral hemorrhage (ICH) is less clear.
MethodsProspective cohort study among 22 022 US male physicians participating in the Physicians Health Study.
Incidence of stroke was measured by self-report and confirmed by medical record review. We used Cox proportional-
hazards models to evaluate the association of smoking with risk of total hemorrhagic stroke, ICH, and SAH. We
categorized smoking into 4 groups: never, past, or current smokers of 20 or of 20 cigarettes per day.
ResultsDuring 17.8 years of follow-up, 108 ICHs and 31 SAHs occurred. Never smokers and past smokers had equal
rates of ICH and SAH. Current smokers of 20 cigarettes per day had multivariable-adjusted relative risks of 1.65 (95%
CI, 0.61 to 4.50) for total hemorrhagic stroke, 1.60 (95% CI, 0.50 to 5.07) for ICH, and 1.75 (95% CI, 0.24 to 13.09)
for SAH when compared with never smokers. Current smokers of 20 cigarettes had relative risks of 2.36 (95% CI,
Downloaded from http://stroke.ahajournals.org/ by guest on October 5, 2017

1.38 to 4.02) for total hemorrhagic stroke, 2.06 (95% CI, 1.08 to 3.96) for ICH, and 3.22 (95% CI, 1.26 to 8.18) for SAH
when compared with never smokers.
ConclusionsThis prospective study suggests an increased risk of total hemorrhagic stroke, ICH, and SAH in current
cigarette smokers with a graded increase in risk that depended on how many cigarettes were smoked. The effect of
smoking on ICH is of about the same magnitude as the effect of smoking on ischemic stroke. Our results add to the
multiple health benefits that can be accrued by abstaining from cigarette smoking. (Stroke. 2003;34:1151-1155.)
Key Words: cigarette smoking cohort study intracerebral hemorrhage
risk factors subarachnoid hemorrhage

emorrhagic stroke accounts for 20% of all stroke among 22 000 US male physicians 40 to 84 years of age at
H cases, and the incidence of intracerebral hemorrhage
(ICH) is at least twice that of SAH.1 Despite attempts to
entry over a time period of 17 years of follow-up.

improve its prognosis by medical or neurosurgical treatments, Methods

hemorrhagic stroke has high long-term disability and a
mortality of 40 to 50%.13 Hence, identification and pre-
vention of modifiable risk factors such as smoking remain of
great importance.
Cigarette smoking is a well-established risk factor for
ischemic stroke4 6 and subarachnoid hemorrhage (SAH).711
However, the association between smoking and ICH remains
controversial.8,1220 One study found a positive association,20
another found an inverse association,8 and most of the
remaining studies showed no significant association between
cigarette smoking and ICH.12,1519
The Physicians Health Study (PHS)21 provided the oppor-
tunity to assess prospectively the association between smok-
ing and the incidence of hemorrhagic stroke and its subtypes
Study subjects were all participants of the PHS, a completed
randomized trial of aspirin and beta carotene in the primary preven-
tion of cardiovascular disease and cancer.21,22 Briefly, the entire
study population consisted of 22 067 apparently healthy male phy-
sicians with no history of cardiovascular disease, cancer (except
nonmelanoma skin cancer), or other major illnesses who began
participation in 1982. After the scheduled end of the PHS in 1995,
participants were followed up until March 2002, which is the date of
inclusion of data used for our analyses. Morbidity and mortality data
were available for 99% of the study participants.
Baseline information was self-reported and collected by a mailed
questionnaire that asked about many demographic, medical history,
and lifestyle variables, including information about usual alcohol
consumption, cigarette smoking, and exercise. Every 6 months for
the first year and annually thereafter, participants were sent
follow-up questionnaires asking about study compliance, medical
history (including stroke), and health behaviors during the study
period.

Received September 16, 2002; final revision received October 28, 2002; accepted November 13, 2002.
From Division of Preventive Medicine, Department of Medicine, Brigham and Womens Hospital (T.K., K.B., E.S.S., J.E.B., J.M.G), and Department
of Ambulatory Care and Prevention (J.E.B.), Harvard Medical School, Boston Mass; Department of Epidemiology, Harvard School of Public Health,
Boston, Mass (T.K., J.E.B.); Department of Neurology, Boston University Medical Center, Boston, Mass (C.S.K.); Institute of Epidemiology and Social
Medicine, University of Muenster, Muenster, Germany (K.B.); Department of Nephrology, University of Freiburg, Freiburg, Germany (E.S.S.); and
Massachusetts Veterans Epidemiology Research and Information Center, Boston VA Healthcare System, Boston Mass (J.M.G.).
This study was presented in part at the 54th Annual Meeting of the American Academy of Neurology in Denver, Colo, April 16, 2002.
Correspondence to Tobias Kurth, MD, MSc, Brigham and Womens Hospital, Division of Preventive Medicine, 900 Commonwealth Ave E, Boston,
MA 02215-1204. E-mail tkurth@rics.bwh.harvard.edu
2003 American Heart Association, Inc.
Stroke is available at http://www.strokeaha.org DOI: 10.1161/01.STR.0000065200.93070.32

1151
 1152 Stroke May 2003
Evaluation of Smoking Status
In the baseline and 24-, 60-, and 144-month questionnaires, partic-
ipants were asked about their smoking habits, including information
about the amount of cigarettes smoked. With this information, we
categorized smoking into never, past, or current. Current smoking
was further categorized into 20 and 20 cigarettes per day. This
categorization was used for each of the 4 assessments.
The 40 participants for whom information on smoking status was
missing were not included in the analyses. Baseline information on
the amount of cigarettes smoked was missing for 27 of the current
smokers. However, because this information became available at
follow-up, they were included in the analyses.
Evaluation of Stroke Cases
All incident nonfatal and fatal stroke cases that occurred during the
study period were considered for this analysis. Participants who
reported stroke on a questionnaire were asked permission to review
their medical records. The End Points Committee confirmed a
diagnosis of stroke only after review of the medical records and
results of laboratory tests. A stroke was defined as a focal neurolog-
ical deficit of sudden onset and vascular mechanism lasting 24
Downloaded from http://stroke.ahajournals.org/ by guest on October 5, 2017
hours. Stroke was classified according to the criteria established by
the National Survey of Stroke23 into ischemic, hemorrhagic, and
unknown subtype. Hemorrhagic stroke was further classified into
ICH or SAH. Stroke classification was performed on the basis of
medical records, reports of brain imaging, and the judgment of the
neurologist on the End Points Committee. Hemorrhagic strokes that
directly followed an intervention or procedure (n2) or occurred
under anticoagulant treatment (n15) were not considered primary
hemorrhagic stroke cases and were excluded from analysis. Cases of
fatal stroke were documented by evidence of a cerebrovascular
mechanism obtained from all available sources, including hospital
records and death certificates. The interrater agreement in the
classification of major stroke types and hemorrhagic subtypes was
excellent over the entire study period.24,25
At follow-up, 5 participants reported stroke events before random-
ization. These participants, as well as the 40 for whom information
on smoking status was not available at baseline, were excluded,
leaving a total sample of 22 022 subjects for this analysis.
Statistical Analysis
We compared the characteristics of participants with respect to their
smoking status using direct standardization to adjust categorical
variables for age in 5-year age groups and the general linear models
procedure (SAS version 8.2, SAS Institute) to adjust continuous
baseline measurements for age. We used Coxs proportional-hazards
model26 to evaluate the association of smoking status and incident
stroke. Separate models were developed for hemorrhagic stroke,
ICH, and SAH. We additionally evaluated the association between
smoking and ischemic stroke. Smoking status was treated as a
time-varying variable in the regression models. This approach
incorporated changes in smoking habits over time.
On the basis of known pathophysiological mechanisms and
published findings, potential confounders were identified for inclu-
sion in the multivariable models. A distinction was made between
variables considered potential confounders and those considered
intermediate stepsie, variables suspected to be in the biological
pathway between smoking and hemorrhagic stroke (body mass
index, hypertension, and diabetes). However, because we wanted to
measure the contribution of smoking to hemorrhagic stroke sepa-
rately from the intermediate variables, analyses were performed
separately with and without controlling for these variables.
We calculated age- and multivariable-adjusted hazard ratios as a
measure for the relative risk (RR) for total hemorrhagic strokes, ICH,
SAH, and ischemic stroke. The first multivariable regression model
controlled for age (continuous), alcohol consumption (1 drink per
week, 2 to 6 drinks per week, 1 drink per day), exercise (1 or 1
times per week), parental history of myocardial infarction before 60
years of age, and randomized treatment assignment. The second
multivariable regression model controlled for all factors in model 1
plus body mass index (continuous), hypertension (defined as self-
reported systolic blood pressure of 140 mm Hg, diastolic blood
pressure 90 mm Hg, or current treatment of hypertension regard-
less of blood pressure), history of diabetes, and history of elevated
cholesterol (240 mg/dL).
Because only 31 total SAH cases occurred and thus the number of
variables for inclusion in the models was restricted,27 we adjusted
only for the most important confounders. In the first multivariable
model, we adjusted only for age, exercise, and randomized treatment
assignment; in the second model, we also adjusted for hypertension.
We evaluated effect measure modification of smoking status by
hypertension (yes, no), alcohol consumption (1 drink per week, 2
to 6 drinks per week, 1 drink per day), and age (60 or 60
years). We tested the proportionality assumption of the models and
found no violation.
Results
During a mean of 17.8 years of follow-up (385 419 person-
years), a total of 1069 strokes occurred, of which 139 were
hemorrhagic (108 ICH, 31 SAH), 913 were ischemic, and 17
were undefined. At baseline, 10 918 participants (49.6%)
self-reported to have never smoked, 8668 (39.4%) reported
past smoking, 849 (3.9%) reported current smoking of 20
cigarettes, and 1560 (7.1%) reported current smoking of 20
cigarettes. The age-adjusted baseline characteristics of partic-
ipants according to their smoking status are shown in Table 1.
The age-adjusted and multivariable-adjusted RRs of total
hemorrhagic stroke, ICH, and SAH according to time-varying
smoking status are summarized in Table 2. Compared with
never smokers, past smokers had no increase in the risk of
total hemorrhagic stroke, ICH, or SAH. The multivariable-
adjusted risk of developing a hemorrhagic stroke was 65%
higher among current smokers of 20 cigarettes per day and
136% higher for current smokers of 20 cigarettes per day
compared with never smokers. Smoking 20 cigarettes was
associated with a 1.6-fold increase in the risk of ICH and a
1.8-fold increase in the risk of SAH compared with never
smokers. Current smokers of 20 cigarettes had a 2.1-fold
increased risk of developing an ICH and a 3.2-foldincreased
risk of developing an SAH compared with never smokers.
We additionally compared all current smokers to never and
past smokers. Current smoking yielded age-adjusted RRs for
total hemorrhagic stroke, ICH, and SAH of 2.33 (95% CI,
1.43 to 3.80), 1.95 (95% CI, 1.06 to 3.58), and 3.61 (95% CI,
1.54 to 8.50), respectively. The multivariable adjusted (model
1) RRs were 2.41 (95% CI, 1.47 to 3.95), 1.98 (95% CI, 1.07
to 3.65), and 3.53 (95% CI, 1.50 to 8.32), respectively.
The multivariable-adjusted RRs (model 1) of ischemic
stroke for past smokers, current smokers of 20 cigarettes
per day, and current smokers 20 cigarettes per day were
0.99 (95% CI, 0.86 to 1.14), 1.56 (95% CI, 1.03 to 2.37), and
2.25 (95% CI, 1.80 to 2.81), respectively, compared with
never smokers. Current smoking was associated with a 2-fold
increase in the risk of ischemic stroke (RR, 2.11; 95% CI,
1.72 to 2.60) compared with never and past smokers.
Additional adjustment for potential biological mediators
(model 2) did not substantially change the effect estimate of
smoking status with any of the stroke subtypes. No significant
effect measure modification was observed between smoking
and age, hypertension, or alcohol consumption with respect to
total hemorrhagic stroke, ICH, or SAH.
 Kurth et al Smoking and the Risk of Hemorrhagic Stroke 1153

TABLE 1. Age-Adjusted* Baseline Characteristics of Study Participants According to Smoking Status

Never Past Current Smokers Current Smokers
Smokers Smokers (20 cigarettes/d) (20 cigarettes/d)
Variable (n10 918) (n8668) (n849) (n1560) P
Age (SD), y 52.2 (9.5) 54.5 (9.5) 52.7 (9.3) 53.2 (8.7) 0.001
Blood pressure (SE), mm Hg
Systolic 125.5 (0.11) 126.4 (0.13) 126.5 (0.41) 128.5 (0.30) 0.001
Diastolic 78.6 (0.08) 78.9 (0.09) 78.6 (0.27) 79.7 (0.20) 0.001
Hypertension, % 22.3 25.0 23.0 26.6 0.001
Body mass index, %
25 kg/m2 58.7 55.0 56.6 50.8 0.001
2529.9 kg/m2 36.2 39.2 38.5 41.7
30 kg/m2 5.1 5.8 5.0 7.5
Exercise 1 time/wk, % 73.3 73.2 70.1 60.9 0.001
Alcohol consumption, %
1/wk 48.3 31.1 30.9 34.3 0.001
Downloaded from http://stroke.ahajournals.org/ by guest on October 5, 2017

26/wk 33.0 38.1 37.1 28.0

1/d 17.8 30.0 30.9 37.2
Diabetes, % 2.3 2.4 3.0 2.9 0.074
History of high cholesterol, % 11.6 12.4 11.1 11.1 0.99
Family history of myocardial 9.0 9.3 10.4 10.8 0.01
infarction, %
Random aspirin assignment, % 49.7 50.3 51.0 48.8 0.99
Continuous values are means, and all other values are frequencies unless stated otherwise. n21 995.
*Adjusted for age (40 44, 45 49, 50 54, 5559, 60 64, 65 69, and 70 years).
P value from general linear models for continuous variables and Mantel-Haenszel 2 test using row mean score differences for
categorical variables.
Hypertension was defined as self-reported systolic blood pressure 140 mm Hg or diastolic blood pressure 90 mm Hg or current
treatment of hypertension (regardless of blood pressure).
History of elevated total cholesterol level 240 mg/dL.
Family history of myocardial infarction 60 years of age.

Discussion compared with never smokers. This observation may be ex-

We found an increased overall risk of total hemorrhagic stroke,
ICH, SAH, and ischemic stroke in current smokers that in-
creased with the amount of cigarettes smoked per day. Those
who smoked 20 cigarettes per day had a multivariable-
adjusted 2-fold increase in the risk of total hemorrhagic stroke
(RR, 2.36; 95% CI, 1.38 to 4.02) and ICH (RR, 2.06; 95% CI,
1.08 to 3.96) and 3-fold increase in SAH (RR, 3.22; 95% CI,
1.26 to 8.18) compared with never smokers. The magnitude of
the effect of smoking on ICH is about the same as the effect of
smoking on ischemic stroke (RR, 2.25; 95% CI, 1.72 to 2.60).
Additional adjustment for potential biological mediators (body
mass index, hypertension, diabetes, and cholesterol) did not
change the observed association between smoking and any of
the stroke subtypes.
Studies published in the 1980s and 1990s established smoking
as a strong risk factor for ischemic stroke4 6,11,28 and SAH.711 In
contrast, the association between smoking and ICH is less clear.
A case-control study by Gill et al20 showed an 80% increased
risk of ICH among male smokers compared with nonsmokers
and a 30% increased risk for female smokers. This result,
however, did not reach statistical significance, and the 95% CIs
were wide. In contrast, a study from Korea by Park et al8 showed
a significant inverse association between cigarette smoking and
ICH. Current or past smokers had a 64% risk reduction of ICH
plained in part by the fact that the current and past smokers (and
not past and never smokers) were collapsed into 1 group, and
past smokers may have had a healthier lifestyle after they quit
smoking than the average never smoker. Juvela13 reported that
smoking is less prevalent in ICH patients compared with SAH
patients. Because this was a case series, no effect estimate of the
risk of ICH among smokers could be calculated. Other studies,
most of them case-control studies, did not find a significantly
increased risk of ICH among smokers.12,1519
A recent population-based case-control study by Woo et al29
suggested an association between smoking and specific loca-
tions of ICH. In their univariate analysis, they showed that
current smoking was associated with lobar ICH (but not with
nonlobar ICH) compared with never smokers. In their multivari-
able models, however, this effect of smoking on lobar ICH was
not significant. Because we did not have data on the location of
ICH in our cohort, we were unable to evaluate this potential
association.
There are several proposed mechanisms by which smoking
could cause stroke. One mechanism that links smoking to
ischemic stroke is structural artery wall damage and carotid
atherosclerosis,30 leading to thrombosis or embolic phenome-
na.31 The mechanism of thrombogenesis is a short-term effect of
smoking, which includes increased fibrinogen levels and platelet
 1154 Stroke May 2003

TABLE 2. RR of Total Hemorrhagic Stroke and Hemorrhagic Stroke Subtypes According to

Smoking Status
Current Smokers
Never Past
Smokers Smokers 20 Cigarettes/d 20 Cigarettes/d
Variable (n10 918) (n8668) (n849) (n1560)
RR (95%CI) RR (95%CI) RR (95%CI)
Total hemorrhagic stroke
Cases (n139), n 68 50 6 15
Age adjusted* 1.00 0.80 (0.571.14) 1.64 (0.604.45) 2.34 (1.383.96)
Model 1 1.00 0.76 (0.531.09) 1.65 (0.614.50) 2.36 (1.384.02)
Model 2 1.00 0.72 (0.491.06) 1.87 (0.685.11) 2.52 (1.454.39)
ICH
Cases (n108), n 54 40 4 10
Age adjusted* 1.00 0.81 (0.551.20) 1.60 (0.515.09) 2.04 (1.073.89)
Model 1 1.00 0.80 (0.541.20) 1.60 (0.505.07) 2.06 (1.083.96)
Model 2 1.00 0.75 (0.491.15) 1.80 (0.565.74) 2.08 (1.054.13)
Downloaded from http://stroke.ahajournals.org/ by guest on October 5, 2017

SAH
Cases (n31), n 14 10 2 5
Age adjusted* 1.00 0.79 (0.371.68) 1.78 (0.2413.31) 3.31 (1.318.38)
Model 1 1.00 0.79 (0.371.68) 1.75 (0.2413.09) 3.22 (1.268.18)
Model 2 1.00 0.83 (0.381.79) 1.81 (0.2413.57) 3.40 (1.328.71)
*Adjusted for age as continuous term.
Model 1 was adjusted for age (continuous), exercise (1 vs 1 times/wk), parental history of myocardial infarction
before 60 years of age, alcohol consumption (1 drink/wk, 2 6 drinks/wk, 1 drink/d), and randomized treatment
assignment.
Model 2 was adjusted for variables in model 1 plus body mass index (continuous), history of hypertension (self-reported
systolic blood pressure 140 mm Hg, or diastolic blood pressure 90 mm Hg or current treatment of hypertension
regardless of blood pressure), history of diabetes, and history of elevated cholesterol 240 mg/dL.
Model 1 for SAH was adjusted for age, exercise, and randomized treatment assignment (categorized as before).
Model 2 for SAH was adjusted for all variables as in model 1 for SAH plus history of hypertension (defined as before).

aggregability,32,33 elevated hematocrit values, and reduced cere-
bral blood flow as a result of arterial vasoconstriction.34 The role
of these factors in promoting stroke in smokers is supported by
the observed reduction in risk of ischemic stroke after smoking
cessation.4,5,28,34,35 Structural damage to the arterial wall may
also play a role in the development of ICH and SAH. For SAH,
there is strong evidence for an association of smoking with
aneurysm formation,36 growth,36 and rupture.37 Juvela et al36,37
showed that current smoking and recent heavy alcohol use were
strong risk factors for aneurysmal SAH in men and women. In
addition, other factors related to the biology of cerebral aneu-
rysms have been linked to smoking: The presence of multiple
cerebral aneurysms in subjects with38 and without39 SAH was
highly correlated with smoking status and female sex, and the
risk of formation of new aneurysms and growth of known
aneurysms was strongly associated with smoking and female sex
for growth of 1 mm and only with smoking for growth of
3 mm over a mean follow-up period of 19 years. These data
provide further support for the notion that smoking relates to
SAH and aneurysm biology via a direct effect on the cerebral
vasculature. For ICH, the observed increased risk in smokers has
less biological support than for SAH, and the possible mecha-
nisms for the association are at this time speculative. Among
potential mediators of the association between current smoking
and ICH, the known increase in blood pressure40 and evidence
for arterial wall damage30 in smokers may play a role in causing
the rupture of small intraparenchymal arteries that cause ICH.
The strength of this study is that, given the large number of
incident stroke cases, we had sufficient power to prospectively
evaluate the association of smoking and hemorrhagic stroke sub-
types. All stroke cases were confirmed by detailed medical record
review, and interrater agreement in the classification of stroke was
excellent over the entire study period.24,25 Furthermore, 90% of
the hemorrhagic stroke cases had confirmatory imaging data, and
the remaining cases were diagnosed from positive evidence from
lumbar puncture and/or autopsy reports. None of the hemorrhagic
cases were only self-reported with no further diagnostic evidence of
ICH or SAH. Information about smoking status and the amount of
cigarettes smoked was updated during the follow-up period; thus,
changes in smoking status or in the amount of cigarettes smoked
could be incorporated into the analysis.
There are several limitations to this study. One stems from the
use of data that relate only to mostly (92.1%) white male physicians.
Thus, extrapolation of the results to women or minority populations
is only speculative. In addition, the low prevalence of current
smokers in the PHS may have resulted in an underestimate of the
true effect of smoking on stroke. Also, participants in the PHS who
suffered a fatal smoking-related outcome (eg, myocardial infarc-
tion) add to the underestimation of the effect of smoking on stroke.
The information on all covariates, like all data in the PHS, was
collected by self-administered questionnaires. Although physicians
 Kurth et al
are likely to provide health-related information appropriately, some
of the information may not be accurate. In several validation studies
of other cohorts, however, self-reports of smoking41 and other
cardiovascular risk factors were reliable.42,43 Because we had very
few cases of SAH, the effect estimate of smoking on SAH is
imprecise despite the fact that it was statistically significant. We had
40 subjects with missing information on smoking status at baseline,
5 of whom had a subsequent stroke; because all 5 stroke events
were nonhemorrhagic, the missing data had no impact on the
conclusions we reached on the association between smoking and
hemorrhagic stroke.
In summary, this prospective study with 17 years of follow-up
and updated smoking information establishes smoking as a risk
factor for ICH. The effect of smoking on ICH is of about the same
magnitude as smoking on ischemic stroke but less powerful than
smoking on SAH. Our results add to the multiple health benefits
that can be accrued by abstaining from cigarette smoking.
Downloaded from http://stroke.ahajournals.org/ by guest on October 5, 2017
Acknowledgments
This study was supported by grants CA 34944 and CA 40360 from
the National Cancer Institute, Bethesda, Md, and grants HL-26490
and HL-34595 form the National Heart, Lung, and Blood Institute,
Bethesda. We are indebted to the participants in the PHS for their
outstanding commitment and cooperation and to the entire PHS staff
for their expert and unfailing assistance.
References
1. Broderick JP, Brott T, Tomsick T, Miller R, Huster G. Intracerebral
hemorrhage more than twice as common as subarachnoid hemorrhage.
J Neurosurg. 1993;78:188 11.
2. Thom TJ. Stroke mortality trends: an international perspective. Ann Epi-
demiol. 1993;3:509518.
3. Fogelholm R, Nuutila M, Vuorela AL. Primary intracerebral hemorrhage in the
Jyvskyl region, central Finland, 198589: incidence, case fatality rate, and
functional outcome. J Neurol Neurosurg Psychiatry. 1992;55:546552.
4. Abbott RD, Yin Y, Reed DM, Yano K. Risk of stroke in male cigarette
smokers. N Engl J Med. 1986;315:717720.
5. Colditz GA, Bonita R, Stampfer MJ, Willett WC, Rosner B, Speizer FE,
Hennekens CH. Cigarette smoking and risk of stroke in middle-aged women.
N Engl J Med. 1988;318:937941.
6. Robbins AS, Manson JE, Lee IM, Satterfield S, Hennekens CH. Cigarette
smoking and stroke in a cohort of U.S. male physicians. Ann Intern Med.
1994;120:458 462.
7. Teunissen LL, Rinkel GJ, Algra A, van Gijn J. Risk factors for subarachnoid
hemorrhage: a systematic review. Stroke. 1996;27:544 549.
8. Park JK, Kim HJ, Chang SJ, Koh SB, Koh SY. Risk factors for hemorrhagic
stroke in Wonju, Korea. Yonsei Med J. 1998;39:229235.
9. Knekt P, Reunanen A, Aho K, Heliovaara M, Rissanen A, Aromaa A,
Impivaara O. Risk factors for subarachnoid hemorrhage in a longitudinal
population study. J Clin Epidemiol. 1991;44:933939.
10. Longstreth WT Jr, Koepsell TD, Yerby MS, van Belle G. Risk factors for
subarachnoid hemorrhage. Stroke. 1985;16:377385.
11. Bonita R. Cigarette smoking, hypertension and the risk of subarachnoid hemor-
rhage: a population-based case-control study. Stroke. 1986;17:831835.
12. Leppala JM, Virtamo J, Fogelholm R, Albanes D, Heinonen OP. Different
risk factors for different stroke subtypes: association of blood pressure,
cholesterol, and antioxidants. Stroke. 1999;30:25352540.
13. Juvela S. Prevalence of risk factors in spontaneous intracerebral hemorrhage and
aneurysmal subarachnoid hemorrhage. Arch Neurol. 1996;53:734740.
14. Zodpey SP, Tiwari RR, Kulkarni HR. Risk factors for hemorrhagic stroke: a
case-control study. Public Health. 2000;114:177182.
15. Thrift AG, McNeil JJ, Donnan GA. The risk of intracerebral haemorrhage
with smoking: the Melbourne Risk Factor Study Group. Cerebrovasc Dis.
1999;9:3439.
16. Juvela S, Hillbom M, Palomaki H. Risk factors for spontaneous intracerebral
hemorrhage. Stroke. 1995;26:15581564.
17. Jorgensen HS, Nakayama H, Raaschou HO, Olsen TS. Intracerebral hemor-
rhage versus infarction: stroke severity, risk factors, and prognosis. Ann
Neurol. 1995;38:4550.
Smoking and the Risk of Hemorrhagic Stroke 1155
18. Suh I, Jee SH, Kim HC, Nam CM, Kim IS, Appel LJ. Low serum cholesterol
and haemorrhagic stroke in men: Korea Medical Insurance Corporation
Study. Lancet. 2001;357:922925.
19. Fogelholm R, Murros K. Cigarette smoking and risk of primary intracerebral
haemorrhage: a population-based case-control study. Acta Neurol Scand.
1993;87:367370.
20. Gill JS, Shipley MJ, Tsementzis SA, Hornby R, Gill SK, Hitchcock ER,
Beevers DG. Cigarette smoking: a risk factor for hemorrhagic and nonhem-
orrhagic stroke. Arch Intern Med. 1989;149:20532057.
21. Steering Committee of the Physicians Health Study Research Group. Final
report on the aspirin component of the ongoing Physicians Health Study:
Steering Committee of the Physicians Health Study Research Group. N Engl
J Med. 1989;321:129135.
22. Hennekens CH, Buring JE, Manson JE, Stampfer MJ, Rosner B, Cook NR,
Belanger C, LaMotte F, Gaziano JM, Ridker PM, et al. Lack of effect of
long-term supplementation with beta carotene on the incidence of malignant
neoplasms and cardiovascular disease. N Engl J Med. 1996;334:11451149.
23. Walker AE, Robins M, Weinfeld FD. The National Survey of Stroke: clinical
findings. Stroke. 1981;12(suppl):I-13I-44.
24. Berger K, Kase CS, Buring JE. Interobserver agreement in the classification of
stroke in the Physicians Health Study. Stroke. 1996;27:238242.
25. Kurth T, Gaziano JM, Berger K, Kase CS, Rexrode KM, Cook NR, Buring
JE, Manson JE. Body mass index and the risk of stroke in men. Arch Intern
Med. 2002;162:2557-2562.
26. Cox DR. Regression models and life-tables. J Roy Stat Soc Ser B. 1972;34:
187220.
27. Harrell FE Jr, Lee KL, Mark DB. Multivariable prognostic models: issues in
developing models, evaluating assumptions and adequacy, and measuring
and reducing errors. Stat Med. 1996;15:361387.
28. Shinton R, Beevers G. Meta-analysis of relation between cigarette smoking
and stroke. BMJ. 1989;298:789794.
29. Woo D, Sauerbeck LR, Kissela BM, Khoury JC, Szaflarski JP, Gebel J,
Shukla R, Pancioli AM, Jauch EC, Menon AG, Deka R, Carrozzella JA,
Moomaw CJ, Fontaine RN, Broderick JP. Genetic and environmental risk
factors for intracerebral hemorrhage: preliminary results of a
population-based study. Stroke. 2002;33:11901196.
30. Howard G, Wagenknecht LE, Burke GL, Diez-Roux A, Evans GW,
McGovern P, Nieto FJ, Tell GS. Cigarette smoking and progression of
atherosclerosis: the Atherosclerosis Risk in Communities (ARIC) Study.
JAMA. 1998;279:119124.
31. Surgeon General of the United States. The Surgeon Generals 1990 Report on the
Health Benefits of Smoking Cessation: Executive Summary. MMWR Recomm
Rep. Washington, DC: Office of the Surgeon General; 1990;39:i-xv, 112.
32. Wannamethee G, Perry IJ, Shaper AG. Haematocrit, hypertension and risk of
stroke. J Intern Med. 1994;235:163168.
33. Kannel WB, Wolf PA, Castelli WP, DAgostino RB. Fibrinogen and risk of
cardiovascular disease: the Framingham Study. JAMA. 1987;258:11831186.
34. Wannamethee SG, Shaper AG, Whincup PH, Walker M. Smoking cessation and
the risk of stroke in middle-aged men. JAMA. 1995;274:155160.
35. Kawachi I, Colditz GA, Stampfer MJ, Willett WC, Manson JE, Rosner B,
Speizer FE, Hennekens CH. Smoking cessation and decreased risk of stroke
in women. JAMA. 1993;269:232236.
36. Juvela S, Poussa K, Porras M. Factors affecting formation and growth of intra-
cranial aneurysms: a long-term follow-up study. Stroke. 2001;32:485491.
37. Juvela S, Hillbom M, Numminen H, Koskinen P. Cigarette smoking and
alcohol consumption as risk factors for aneurysmal subarachnoid hemor-
rhage. Stroke. 1993;24:639646.
38. Juvela S. Risk factors for multiple intracranial aneurysms. Stroke. 2000;31:
392397.
39. Qureshi AI, Suarez JI, Parekh PD, Sung G, Geocadin R, Bhardwaj A,
Tamargo RJ, Ulatowski JA. Risk factors for multiple intracranial aneurysms.
Neurosurgery. 1998;43:2226.
40. Omvik P. How smoking affects blood pressure. Blood Pressure. 1996;5:7177.
41. Patrick DL, Cheadle A, Thompson DC, Diehr P, Koepsell T, Kinne S. The
validity of self-reported smoking: a review and meta-analysis. Am J Public
Health. 1994;84:10861093.
42. Rimm EB, Stampfer MJ, Colditz GA, Chute CG, Litin LB, Willett WC.
Validity of self-reported waist and hip circumferences in men and women.
Epidemiology. 1990;1:466473.
43. Colditz GA, Martin P, Stampfer MJ, Willett WC, Sampson L, Rosner B,
Hennekens CH, Speizer FE. Validation of questionnaire information on risk
factors and disease outcomes in a prospective cohort study of women. Am J
Epidemiol. 1986;123:894900.
 Smoking and the Risk of Hemorrhagic Stroke in Men
Tobias Kurth, Carlos S. Kase, Klaus Berger, Elke S. Schaeffner, Julie E. Buring and J. Michael
Gaziano
Downloaded from http://stroke.ahajournals.org/ by guest on October 5, 2017

Stroke. 2003;34:1151-1155; originally published online March 27, 2003;

doi: 10.1161/01.STR.0000065200.93070.32
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright 2003 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/34/5/1151

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/