Background and purpose: Some previous Japanese cohort studies failed to show an association between smoking and stroke risk. Because such an association has been noted in other populations, this issue should be re-examined in a recent representative Japanese cohort with a higher total cholesterol level.
Methods: A total of 9638 men and women aged 30 years and older without a history of cardiovascular disease (CVD) at baseline in 1980 were followed-up for 14 years.
Results: We observed 203 stroke deaths (107 cerebral infarctions, 45 cerebral hemorrhages, and 51 others), 191 heart disease deaths, and 413 CVD deaths. The average serum total cholesterol level was approximately 4.91 mmol/L. Cox proportional hazard ratios were calculated adjusting for age, systolic blood pressure, and other conventional risk factors. The hazard ratios for men who smoked 1 to 20 cigarettes/day for all strokes, cerebral infarction, and cerebral hemorrhage were 1.60 (95% CI, 0.91 to 2.79), 2.97 (CI, 1.27 to 6.98), and 0.42 (CI, 0.16 to 1.09), respectively, and for those who smoked > or =21 cigarettes/day, they were 2.17 (CI, 1.09 to 4.30), 3.26 (CI, 1.11 to 9.56), and 0.68 (CI, 0.20 to 2.33), respectively. For women who smoked > or =21 cigarettes/day, the hazard ratio for all strokes was 3.91 (CI, 1.18 to 12.90). For CVD, all heart disease, and ischemic heart disease, the hazard risks of smoking were significant (1.49 to 4.25) for men but not significant for women.
Conclusions: Smoking in a cohort with moderate serum total cholesterol level was a potent risk factor for stroke, especially cerebral infarction, for both men and women, and for CVD and ischemic heart disease for men.