Exercising to the point of inducing a sweat at least four times per week may have a protective effect against stroke, according to the results of a large, population-based, longitudinal study of people in the "Stroke Belt" of the Southern United States.
The study, conducted by Michelle N. McDonnell, Ph.D., of the University of South Australia, Adelaide, and her colleagues, confirms previous reports about the effect of exercise on stroke risk, but it is the first to find a dose-response effect of exercise on stroke risk only in men and not in women. Men who exercised to the point of sweating at least four times per week had a significantly lower risk of stroke than did men who exercised one to three times per week. However, women who exercised at either level of frequency did not have a significantly lower risk of stroke, compared with women who did not exercise (Stroke 2013 July 18 [doi:10.1161/STROKEAHA.113.001538]).
The Reasons for Geographic and Racial Differences in Stroke (REGARDS) study assessed 27,348 participants aged 45 years or older who self-reported their frequency of moderate to vigorous physical activity within three categories: none, one to three times per week, and four or more times per week. Intensity of physical activity was determined by whether the exercise induced sweat in participants.
The researchers obtained demographic information and medical history via telephone, and conducted in-person follow-up interviews to take participants’ physical measurements. They conducted follow-up interviews with participants every 6 months to assess for potential stroke risk for a mean of 5.7 years, collecting data on demographics, socioeconomic factors, stroke risk factors (including mass index, smoking status, alcohol use, diabetes, and hypertension), and physical health limitations.
A total of 918 confirmed stroke/transient ischemic attack cases occurred over the follow-up period, with a significant association between activity and stroke. The most physically active participants had a lower body mass index and prevalence of diabetes and were male, white, and highly educated with high incomes. Participants who reported no physical activity had a 20% greater risk for stroke than did participants who exercised at least four times per week (hazard ratio = 1.20; 95% confidence interval, 1.01-1.42), after adjustment for age, sex, race, and age-race interaction. This remained the same after further adjustments were made for region, urban/rural residence, and socioeconomic status.
The hazard ratio was 1.16 for individuals who did not exercise, compared with those who did so one to three times per week, but it was not statistically significant and was further reduced when additional adjustments were made for region, urban/rural residence, and socioeconomic status.
Exercising at least four times per week was similarly protective against both ischemic and hemorrhagic stroke.
A statistical model in the study that added stroke risk factors attenuated the effects of exercising at least four times per week to the point where it was no longer significant. Individuals who regularly participated in physical activity were more likely to be moderate or heavy consumers of alcohol, leading the investigators to suggest that exercise’s benefit on reducing stroke risk "is likely to be explained for the most part by the positive effect that physical activity has on body mass index, hypertension, and diabetes mellitus, although physical activity also has the potential to reduce stroke risk because of other biological actions (e.g., improving endothelial function and reducing platelet activity)," the investigators wrote.
When Dr. McDonnell and her associates stratified the participants according to sex, they found that men who exercised one to three times per week had a significantly higher risk of stroke than did men who exercised at least four times per week, with hazard ratios ranging from 1.26 to 1.30. Surprisingly, men who did not exercise had no significantly increased risk of stroke. For exercise frequencies of one to three times per week and four or more times per week, women had no change in the risk of stroke, compared with women who did not exercise.
The authors identified a few limitations to this study. First, the "sweat question," though considered a valid assessment of activity, may not account for light to moderate intensity activities such as walking or cycling. Second, the researchers’ measure of physical activity does not measure duration or type of activity. Third, physical activities were categorized at the beginning of the study, with stroke events occurring years later. Lastly, some stroke cases could have been missed if the participants did not seek medical care.