An association between smoking habits and blood pressure in normotensive Japanese men: a 5-year follow-up study

Abstract

Background: There are few studies on the effects of smoking on blood pressure (BP) that consider confounding factors such as age, obesity, lifestyle and blood chemistry. As such, we conducted a 5-year follow-up study to clarify the effects of smoking habits on blood pressure in normotensive Japanese men. Methods: The subjects were 2107 normotensive male steelworkers aged 40–54 years. They were classified using three indices: smoking habits in 1990, changes in smoking habits, and changes in amount of smoking. The associations between these indices and changes in blood pressure were evaluated using analysis of covariance (ANCOVA). As analyses, BMI, salt intake, physical activity, drinking habit, and results of blood chemistry in 1990 and their changes, age and blood pressure in 1990 were used as covariates. Results: The cumulative incidence of hypertension in smokers was lower than in non- and ex-smokers. The adjusted mean of change in blood pressure of smokers was significantly lower than in non- and ex-smokers. No significant difference between non-smokers and ex-smokers, or among groups of smokers who varied their amount of smoking was observed. Conclusion: Chronic smoking reduces changes in blood pressure and 5-year cumulative incidence of hypertension. However, no significant dose-dependent effect of smoking on changes in blood pressure was observed.

Introduction

Smoking has been shown to raise blood pressure (BP) by vasoconstriction and accelerated heart rate as an acute effect (Aronow et al., 1971, Cryer et al., 1976, Benowitz et al., 1982, Benowitz et al., 1984, De Cesaris et al., 1992), while the chronic effect of addictive smoking has not been elucidated in epidemiological studies. Different groups have reported that either no association existed between smoking habits and BP (Kagamimori et al., 1980, Arkwright et al., 1982, Criqui et al., 1982, Lang et al., 1983, Simons et al., 1984), that BP of smokers was lower than that of non-smokers (Higgins and Kjelsberg, 1967, Seltzer, 1974, Goldbourt and Medalie, 1977, Agner, 1983, Savdie et al., 1984, Green et al., 1986, Imamura et al., 1996), or that smoking raised BP (Elliott and Simpson, 1980, Dyer et al., 1982). In addition, a dose–effect relationship between daily smoking amount and BP was significant in some studies (Goldbourt and Medalie, 1977, Savdie et al., 1984, Imamura et al., 1996), while others failed to confirm this finding (Gofin et al., 1982, Green et al., 1986, Istvan et al., 1999). Thus, the effect of smoking habits on BP and, especially its precise nature, is still unclear.

One possible reason for such conflicting results is that there are many factors that impinge upon BP. Therefore, controlling for these various confounding factors is necessary to evaluate the effects of smoking habits on BP precisely. Therefore, we conducted a cross-sectional study that considered confounding factors (Okubo et al., 2002). The results of this study revealed that both systolic blood pressure (SBP) and diastolic blood pressure (DBP) of smokers were significantly lower than those of non-smokers and ex-smokers. In addition, no significant dose–effect association among smokers was observed.

Thus, we saw a need for a longitudinal evaluation of the long-term relationship of smoking habits with BP considering confounding factors. In addition, while the effect of cessation of smoking on BP has been reported (Higgins and Kjelsberg, 1967, Goldbourt and Medalie, 1977, Jacobs and Gottenborg, 1981, Green et al., 1986), there is no report investigating changes in BP in beginning smokers. Therefore, we conducted a 5-year follow-up study considering confounding factors of age, obesity, drinking habits, habitual exercise, salt intake, and blood chemistry, in normotensive middle-aged male workers to clarify the effects of smoking habits and their changes on BP.