PURPOSE: To examine the risk factors for subarachnoid hemorrhage, for few prospective data have been available on risk factors of subarachnoid hemorrhage among Japanese. METHODS: A 9.4-year cohort study was conducted on 12,372 men and women, aged 40-69 years, free of history of stroke in six communities in Japan. RESULTS: During the follow-up, seventy-one incidents of subarachnoid hemorrhage occurred. After adjusting for age, serum total cholesterol, body mass index, history of diabetes mellitus, blood pressure category and drinking category, current smokers had a significantly increased risk of subarachnoid hemorrhage compared with 'never-smokers': multivariate relative risk (RR) [95% confidence interval (95% CI)] = 3.1(1.4-6.9) for women, 1.5(0.4-5.3) for men and 2.7(1.3-5.5)for men and women. Hypertensives had a significantly increased risk of subarachnoid hemorrhage compared with normotensives: multivariate RR(95% CI) = 3.1(1.6-6.0) for women, 4.3(1.5-12.0) for men and 3.4(2.0-5.9) for men and women. For men, heavy drinkers (> = 69 g per day ethanol), had a significantly increased risk of subarachnoid hemorrhage compared with 'never-drinkers': multivariate RR(95% CI) = 4.3(1.1-16.8). For women, the relationship between heavy drinking and risk was not examined due to the small number of heavy drinkers (n = 13). Multivariate RR(95% CI) for men who smoked currently and drank heavily compared with men who did not smoke currently nor drink heavily was 6.0(1.8-20.1). Multivariate RR(95% CI) for hypertensive men who drank heavily compared with normotensive or borderline hypertensive men who did not drink heavily was 13.0(3.9-43.9). Multivariate RR(95% CI) for hypertensive men and women who smoked currently compared with normotensive or borderline hypertensive men and women who did not smoke currently was 6.1(1.3-28.7) for men and 6.3(1.4-28.0) for women. Furthermore, the combination of smoking, hypertension and heavy drinking raised the risk of subarachnoid hemorrhage substantially: multivariate RR(95% CI) = 17.5(3.4-90.2). CONCLUSION: Smoking and hypertension appeared to increase the risk of subarachnoid hemorrhage. For men, although the association of smoking with the risk of subarachnoid hemorrhage was weak compared with women, the combination of smoking, hypertension and heavy drinking increased the risk substantially. For women, smoking appeared to be an independent risk of subarachnoid hemorrhage, and the combination of smoking and hypertension increased the risk in additive fashion.
PURPOSE: To examine the risk factors for subarachnoid hemorrhage, for few prospective data have been available on risk factors of subarachnoid hemorrhage among Japanese. METHODS: A 9.4-year cohort study was conducted on 12,372 men and women, aged 40-69 years, free of history of stroke in six communities in Japan. RESULTS: During the follow-up, seventy-one incidents of subarachnoid hemorrhage occurred. After adjusting for age, serum total cholesterol, body mass index, history of diabetes mellitus, blood pressure category and drinking category, current smokers had a significantly increased risk of subarachnoid hemorrhage compared with 'never-smokers': multivariate relative risk (RR) [95% confidence interval (95% CI)] = 3.1(1.4-6.9) for women, 1.5(0.4-5.3) for men and 2.7(1.3-5.5)for men and women. Hypertensives had a significantly increased risk of subarachnoid hemorrhage compared with normotensives: multivariate RR(95% CI) = 3.1(1.6-6.0) for women, 4.3(1.5-12.0) for men and 3.4(2.0-5.9) for men and women. For men, heavy drinkers (> = 69 g per day ethanol), had a significantly increased risk of subarachnoid hemorrhage compared with 'never-drinkers': multivariate RR(95% CI) = 4.3(1.1-16.8). For women, the relationship between heavy drinking and risk was not examined due to the small number of heavy drinkers (n = 13). Multivariate RR(95% CI) for men who smoked currently and drank heavily compared with men who did not smoke currently nor drink heavily was 6.0(1.8-20.1). Multivariate RR(95% CI) for hypertensivemen who drank heavily compared with normotensive or borderline hypertensivemen who did not drink heavily was 13.0(3.9-43.9). Multivariate RR(95% CI) for hypertensivemen and women who smoked currently compared with normotensive or borderline hypertensivemen and women who did not smoke currently was 6.1(1.3-28.7) for men and 6.3(1.4-28.0) for women. Furthermore, the combination of smoking, hypertension and heavy drinking raised the risk of subarachnoid hemorrhage substantially: multivariate RR(95% CI) = 17.5(3.4-90.2). CONCLUSION: Smoking and hypertension appeared to increase the risk of subarachnoid hemorrhage. For men, although the association of smoking with the risk of subarachnoid hemorrhage was weak compared with women, the combination of smoking, hypertension and heavy drinking increased the risk substantially. For women, smoking appeared to be an independent risk of subarachnoid hemorrhage, and the combination of smoking and hypertension increased the risk in additive fashion.