Impact of Alcohol Consumption and Cigarette Smoking on Stroke Among the Elderly in Taiwan
Background and Purpose We investigated the influence of alcohol consumption and cigarette smoking on all types of stroke and cerebral infarction, in particular among a representative sample of elderly residents in Taiwan.
Methods This study was a component of a nationwide survey of health and living status of residents aged 65 years or older in Taiwan in which subjects received detailed physical, neurological, and laboratory examinations. Inquiries were made about medical history, and information on the amount and duration of drinking and smoking was obtained. Diagnoses of stroke were made according to the results of brain computed tomography at the onset of disease or were based on criteria established by the World Health Organization.
Results Of the 2600 subjects, there were 155 elderly persons with stroke (prevalence, 6%). Excessive drinking of more than 367.6 g/wk of alcohol was associated with a high prevalence of cerebral infarction. Consumption of ≤367.5 g/wk of alcohol did not have an influence on stroke prevalence. The relationship between duration of alcohol drinking and stroke was equivocal. More than 30 pack-years of cigarette smoking was a significant risk factor for all types of stroke and cerebral infarction in particular. Using multiple logistic regression to control for possible confounders, it was found that smoking was an independent risk factors for all stroke and was of borderline significance for cerebral infarction. Although excessive drinking was a significant risk factor for cerebral infarction in univariate analysis, this effect was lost after adjustment for other confounders.
Conclusions Cigarette smoking was a more important risk factor for stroke and cerebral infarction than excessive drinking of alcohol.
Alcohol has long been recognized to be associated with a wide range of neurological complications.1 For cerebrovascular diseases, alcohol consumption has been reported to be a possible risk factor for thromboembolic stroke,2 3 a less well-documented risk factor for stroke,4 5 not an independent risk factor for cerebral infarction in middle-aged and elderly patients,6 and a protective factor for stroke for light or moderate consumers.7 8 9 10 11 12 13 Cigarette smoking is now generally accepted as a major risk factor for both thrombotic and hemorrhagic stroke,14 15 16 17 18 although a few reports showed a negative correlation or demonstrated insignificant increased risk for stroke.19 Because there have been few observations on the relationship between alcohol consumption or cigarette smoking and stroke in Chinese populations and because stroke is the most prevalent cause of death among elderly persons in Taiwan,20 we investigated the relationship between these habits and the occurrence of stroke using data from a nationwide survey.
Subjects and Methods
A survey of persons older than 65 years was conducted from April 1989 through June 1991 to evaluate their physical and living status. Samples were drawn from four areas in Taiwan: Taipei, Taichung, Kaohsiung, and Hualien, representing the northern, central, southern, and eastern parts of the island. A random sample of two subjects was chosen from each “Li” (the smallest administration unit) in these cities: one man and one woman aged 65 years or older whose names were the first to appear in the selected census registration books.
Subjects were invited to come to one hospital in each city to receive extensive physical, neurological, and laboratory examinations. A questionnaire on past medical history, current living status, alcohol consumption, and cigarette smoking was also administered in that session. If subjects had difficulty with transportation to the hospital, physicians were sent to their homes to interview and examine them.
Stroke was classified as cerebral hemorrhage or infarction according to the results of brain computed tomography (CT) at the onset of disease. Strokes in subjects without brain CTs were categorized as unclassified if they fulfilled the World Health Organization criteria for stroke.21 Subjects with stroke were analyzed as a whole group (including cerebral hemorrhage, cerebral infarction, and unclassified stroke) and as a subgroup of those with cerebral infarction.
Among subjects without stroke, hypertension was defined as the use of antihypertensive drugs or a systolic blood pressure of >160 mm Hg and/or diastolic blood pressure of >95 mm Hg measured at the survey. Diabetes mellitus was defined as the use of oral hypoglycemic agents or insulin or fasting blood glucose of >140 mg/dL measured at the survey. Hypercholesterolemia was defined as fasting serum cholesterol levels of >220 mg/dL measured at the survey. For subjects who had already suffered from stroke at the time of the survey, results of physical examinations at the first stroke; history of hypertension, diabetes mellitus, and hypercholesterolemia; and habits of drinking and smoking before the onset of the first stroke were obtained from the questionnaire. Spouses or close relatives and friends were interviewed for those with difficulty in communication. Previous medical records were reviewed in a few cases.
Subjects were classified as nondrinkers if they had never drunk alcohol and as ex-drinkers if they had stopped drinking for more than 6 months. Shaohsing wine, which contains 17.5 g alcohol per 100 mL, is the most popular in Taiwan; Taiwan beer (4.5 g per 100 mL) and other spirits (40 g per 100 mL) were also widely consumed. Effects of various beverages were not analyzed separately because it has been reported that the influences of beer, wine, or liquor on stroke do not differ.8 Amount of alcohol consumption was classified as “light” if the subject consumed <122.5 g/wk of alcohol, as “moderate” if 122.6 to 245 g/wk, as “heavy” if 245.1 to 367.5 g/wk, and as “excessive” if >367.6 g/wk. Subjects were classified as nonsmokers if they had never smoked and as ex-smokers if they had ceased smoking for more than 6 months. Current smokers were classified as “light smokers” if they smoked ≤19 cigarettes per day and as “heavy smokers” if they smoked >20 cigarettes per day. Lifetime cigarette exposure was assessed by pack-years.
Prevalence odds ratio (POR) and 95% confidence intervals (CI) were used to estimate the strength of association between risk factors and stroke.22 Stratified analyses and multiple logistic regression were used to adjust for the effects of potential confounders.23 Regression coefficients were transformed to POR for comparison with the results from univariate analyses.
Some subjects who accepted the invitation for examinations did not come to the hospitals. Accrual of subjects stopped after data on 2600 elderly persons had been examined (1322 men and 1278 women). One hundred fifty-five subjects (91 men and 64 women) had one or more episodes of stroke before the survey; 21 (13.5%) were cerebral hemorrhage, 115 (74.2%) were cerebral infarction, and 19 (12.3%) were unclassified (Table 1⇓). Seven subjects had history of transient ischemic attacks and were excluded from subsequent analyses.
Mean±SD age of the subjects was 71.7±4.5 years for men and 71.1±4.5 years for women. Overall prevalence of stroke was 6.0% (95% CI, 5.1 to 7.0%); it was 6.9% among men and 5.0% among women. Prevalence of all stroke was higher in men than in women, with a POR of 1.37 (95% CI, 0.99 to 1.92, adjusted for age). Prevalence of cerebral infarction was also higher among men, with a POR of 1.45 (95% CI, 0.99 to 2.13, adjusted for age). No significant difference in prevalence of stroke between each 5-year age group was found (data not shown).
In the whole group, excessive drinkers who drank >367.6 g/wk of alcohol were associated with an increased prevalence of cerebral infarction (POR, 2.86) and an almost significant increase in prevalence of all stroke (POR, 2.18) (Table 2⇓). Drinking of <367.5 g/wk of alcohol did not increase the prevalence for stroke. Ex-drinkers were associated with a significantly increased prevalence of all stroke (POR, 2.55) and a borderline increase in prevalence of cerebral infarction (POR, 1.94). For men, significant increases were found in excessive drinkers for prevalence of cerebral infarction (POR, 2.51) and in ex-drinkers for prevalence of all stroke (POR, 2.19). The relationship between duration of alcohol drinking and prevalence of stroke was inconclusive (data not shown).
Current smokers had a significant increase in prevalence for cerebral infarction (POR, 1.53) in the whole group (Table 3⇓). Moreover, heavy smoking (≥20 cigarettes per day) was associated with a significantly higher prevalence for all stroke (POR, 1.65) and cerebral infarction (POR, 1.96) in the whole group. Among men, heavy smoking was a significant risk factor for cerebral infarction (POR, 1.84). Those with lifetime exposure of more than 30 pack-years had a higher prevalence of all stroke (POR, 1.71) and cerebral infarction (POR, 1.95), but these effects were not definite among men (Table 4⇓).
A stratified analysis was performed to evaluate the potential confounding effect of smoking on the relationship between alcohol consumption and stroke (Table 5⇓). Prevalence of stroke or cerebral infarction was not associated with alcohol consumption among heavy smokers. Only in the group of ex-drinkers was there an increased prevalence of stroke among nonsmokers.
The individual relations of hypertension, diabetes mellitus, and hypercholesterolemia to stroke were also evaluated separately (Table 6⇓). Hypertension was a significant risk factor for all stroke and cerebral infarction (POR, 3.57 and 3.46, respectively). Diabetes mellitus was associated with increased prevalence for all stroke and cerebral infarction (POR, 1.43 and 1.48, respectively). Hypercholesterolemia was not found to be a risk factor for either all stroke or cerebral infarction.
Multiple logistic regression was used to control for the effects of potential confounders, including sex, age, hypertension, diabetes mellitus, smoking, and drinking (Table 7⇓). Age was not a significant risk factor when evaluated in 5-year age groups, but it became a significant risk factor for all stroke and cerebral infarction (POR, 1.05 and 1.04, respectively) when coded as a continuous variable in the multivariate model. POR associated with 5-year increments of age was 1.28 (1.055) for all strokes and 1.22 (1.045) for cerebral infarction. After we controlled for potential confounders, hypertension and diabetes remained as significant risk factors for all stroke and for cerebral infarction. Smoking was also associated with increased prevalence for all stroke (POR, 1.71), and its association with cerebral infarction was of borderline significance (POR, 1.72). Effects of alcohol consumption became insignificant after controlling for the effects of other confounders.
The role of alcohol on the occurrence of stroke has been the subject of much controversy. It has been reported that consumption of >300 g/wk of alcohol is associated with an increased risk of stroke,7 drinking of >252 g/wk of alcohol increases the risk for all stroke,9 and heavy consumption of alcohol (>400 g/wk) predisposes to both hemorrhagic and nonhemorrhagic stroke.10 However, other studies found that there were no significant associations between any level of alcohol use even when drinkers consumed a level of >300 g/wk5 or that drinking >290 g/wk of alcohol had little adverse effect.12 Findings from the present study suggest that excessive drinking of alcohol of >367.6 g/wk might be a risk factor for all stroke and cerebral infarction, although this finding could as well be explained by the confounding effects of smoking.
For light to moderate drinking of alcohol, most studies have found a protective effect for stroke.7 8 9 10 11 12 13 The exception was a study that reported all levels of alcohol consumption to be associated with a modest increased risk for stroke.3 In our observations, those who drank <367.5 g/wk of alcohol seemed to have lower risk for all stroke and cerebral infarction. Due to the limited sample size and hence the wide CI of the effect estimates, we were not able to confirm this protective role for moderate drinking.
Ex-drinkers had a significant increase in risk for all stroke and cerebral infarction in this study. A possible explanation is that some of the ex-drinkers were at high risk for stroke and had abstained from drinking because of medical advice. Because we did not ascertain the reason for abstinence, we cannot test the validity of this hypothesis. It has been reported in one study that the risk of hemorrhagic stroke in ex-drinkers was found to be equal to that of those who drank little or did not drink at all, but comparable data for ischemic stroke is not available.9
Alcohol certainly exerts an influence on the pathogenesis of stroke through complex effects. It has been demonstrated that a moderate amount of alcohol increases prostacyclin levels, augments the fibrinolytic system, depresses concentration of low-density lipoprotein, increases high-density lipoprotein, impairs platelet function, elevates cerebral blood flow, impairs aggregation and red blood cell deformity, and increases hematocrit levels.1 7 11 We did not evaluate the effects of changes in these intermediate factors on stroke.
Previous studies have shown that smokers have an increased risk of stroke.14 16 18 The risk for thromboembolic stroke was increased by two to three times and the risk for hemorrhagic stroke by four to six times,15 the risk of cerebral infarction was increased by more than three times in 16 years of follow-up,24 or there was a 4.2-fold increase in risk of cerebral infarction.25 There is also evidence of a dose-response relationship between the number of cigarettes smoked and the risk of stroke.14 16 18 26 The OR for those smoking 1 to 20 cigarettes per day has been reported to be 3.3 (95% CI, 2.0 to 5.5), and for the heavier smokers who smoked >20 cigarettes a day, it was 5.6 (95% CI, 3.2 to 9.9).14 In a study of stroke among women, OR was reported to be 2.2 (95% CI, 1.5 to 3.3) for those smoking 1 to 14 cigarettes per day, whereas the OR was 3.7 (95% CI, 2.7 to 5.1) for those who smoked >25 cigarettes per day.16 Our study showed that smoking <19 cigarettes a day did not have a remarkable effect on stroke, but smoking >20 cigarettes in a day was associated with a significant risk for all stroke and cerebral infarction.
For ex-smokers, it has been reported that the risk of stroke is significantly decreased after they cease smoking15 ; the longer the patient abstains from smoking, the more likely carotid artery atherosclerosis will be of a less severe degree.17 Among ex-smokers, risks of cardiovascular mortality were similar to risks in those who never smoked, regardless of the number of years they had smoked.27 We did not find an increased risk for stroke among ex-smokers in this study.
Pack-year is a useful parameter to represent the combination of the amount and duration of smoking. Pack-year was reported to be positively correlated with the extent of extracranial carotid artery atherosclerosis in a study using duplex ultrasonography for the assessment of sclerotic changes.28 Those men who had more than 40 pack-years of exposure were reported to have significantly higher cardiovascular mortality rate.27 We have found that smoking of >30 pack-years was associated with a significantly high risk of all stroke and cerebral infarction.
Although excessive drinking and heavy smoking were recognized as significant risk factors for stroke when assessed in univariate analyses in this study, the effects of alcohol consumption became equivocal when adjusted for the effects of smoking and other confounders. Similar results also have been observed in other studies.6 17 The influence of cigarette smoking on the occurrence of stroke is obviously greater than that of alcohol consumption.
The internal validity of this study is limited by its cross-sectional nature. For those with stroke, every effort was made to solicit information before the onset of the first episode of stroke. Validity of such recalled information may not be comparable with concurrent information obtained from healthy subjects. Moreover, the validity of self-reported smoking and alcohol consumption may be questionable, and the recall bias between healthy subjects and those with stroke may also contribute to the positive findings.
This study was supported by a grant (DOH 80-25) from the Department of Health, Executive Yuan, Republic of China.
- Received August 16, 1994.
- Revision received February 14, 1995.
- Accepted February 14, 1995.
- Copyright © 1995 by American Heart Association
Gorelick PB. The status of alcohol as a risk factor for stroke. Stroke. 1989;20:1608-1610.
Balow J, Alter M, Resch JA. Cerebral thromboembolism: a clinical appraisal of 100 cases. Neurology. 1966;16:559-564.
Ben-Shlomo Y, Markowe H, Shipley M, Marmot MG. Stroke risk from alcohol consumption using different control groups. Stroke. 1992;23:1093-1098.
Dykin ML, Wolf PA, Barnett HJM, Bergan JJ, Hass WK, Kannel WB, Kuller L, Kurtzke JF, Sundt TM. Risk factors in stroke: a statement for physicians by the Subcommittee on Risk Factors and Stroke of the Stroke Council. Stroke. 1984;15:1105-1111.
Gorelick PB, Redin MB, Langenberg P, Hier DB, Costigan J, Gomez I, Spontak S. Is acute alcohol ingestion a risk factor for ischemic stroke? Results of a controlled study in middle-aged and elderly stroke patients at three urban centers. Stroke. 1987;18:359-364.
Gill JS, Shipley MJ, Hornby RH, Gill SK, Beevers DG. A community case-control study of alcohol consumption in stroke. Int J Epidemiol. 1988;17:542-547.
Camargo CA. Moderate alcohol consumption and stroke: the epidemiologic evidence. Stroke. 1989;20:1611-1626.
Rodgers H, Aitken D, French JM, Curless RH, Bates D, James OFW. Alcohol and stroke: a case-control study of drinking habits past and present. Stroke. 1993;24:1473-1477.
Shinton R, Sagar G, Beevers G. The relation of alcohol consumption to cardiovascular risk factors and stroke: the west Birmingham stroke project. J Neurol Neurosurg Psychiatry. 1993;56:458-462.
Bogousslavsky J, Melle GV, Despland PA, Regli F. Alcohol consumption and carotid atherosclerosis in the Lausanne stroke registry. Stroke. 1990;21:715-720.
Bonita R, Scragg R, Stewart A. Jackson R, Beaglehole R. Cigarette smoking and risk of premature stroke in men and women. Br Med J. 1986;293:6-8.
Whisnant JP, Homer D, Ingall TJ, Baker HL, O’Fallon WM, Wiebers DO. Duration of cigarette smoking is the strongest predictor of severe extracranial carotid artery atherosclerosis. Stroke. 1990;21:707-714.
Tuomilehto J, Bonita R, Stewart A, Nissinen A, Salonen JT. Hypertension, cigarette smoking and the decline in stroke incidence in eastern Finland. Stroke. 1991;22:7-11.
Brust JCM. Stroke and substance abuse. In: Barnett HJM, Mohr JP, Stein BM, Yatsu FM, eds. Stroke: Pathophysiology, Diagnosis, and Management. 2nd ed. London, England: Churchill Livingstone; 1992:881-883.
National Health Administration. Executive Yuan, Republic of China. Health Statistics: I, Vital Statistics, 1974-1988. Taipei, Taiwan, National Health Administration, 1975-1989.
Report of the WHO Task Force on Stroke and Other Cerebrovascular Disorders. Special report from the World Health Organization. Stroke—1989: recommendations on stroke prevention, diagnosis, and therapy. Stroke. 1989;20:1407-1431.
Rothman KJ. Modern Epidemiology. Boston, Mass: Little, Brown & Co; 1986:32-34, 217-219.
Hosmer DW, Lemeshow S. Applied Logistic Regression. New York, NY: John Wiley & Sons; 1989:25-75.
Kannel WB. Current status of the epidemiology of brain infarction associated with occlusive arterial disease. Stroke. 1971;2:295-318.
Salonen JT, Puska P, Tuomilehto J, Homan K. Relation of blood pressure, serum lipids, and smoking to the risk of cerebral stroke: a longitudinal study in eastern Finland. Stroke. 1982;13:327-333.
Dempsey RJ, Moore RW. Amount of smoking independently predicts carotid artery atherosclerosis severity. Stroke. 1992;23:693-696.
Crouse JL, Tool JF, McKinney WN, Dignan MB, Howard G, Kahl FR, McMahan MR, Harpold GH. Risk factors for extracranial carotid artery atherosclerosis. Stroke. 1987;18:990-996.