Strengthening the Case for the Influence of Sleep on Cardiovascular Health in Women

In this issue of the Journal of Women's Health, Sands-Lincoln and colleagues explore the relationship between sleep duration, insomnia, and incident coronary heart disease (CHD) and cardiovascular disease (CVD) in post-menopausal women. ¹ The study found that participants with high insomnia scores had a modest yet significant increased risk for the incidence of CHD and CVD in fully adjusted models. Mechanisms previously proposed to explain the relationship between disturbed sleep and CHD/CVD include increased sympathetic nervous system activity and disturbed immune function that increases inflammation. ² The association between insomnia and CHD found in the Sands et al. study was not as strong as a previous study with middle-aged Chinese participants. ³ The exclusive inclusion of post-menopausal women could have played some role in the differences. Menopause has been found to be associated with increased risk of CVD. ⁴ Estrogen deficiency during menopause is thought to induce endothelial and vascular dysfunction through reduced compliance of the large arteries. ⁵ Sleep complaints also increase after menopause, and estrogen replacement therapy has been shown to improve objective sleep quality. ⁶ An interesting line for future inquiry would be the role of estrogen in the relationship between sleep disruption and CHD/CVD in women.

This is the first study to explore and identify an interaction between sleep duration and insomnia. Women who reported both long sleep duration and insomnia had significantly higher risks of suffering from incident CHD/CVD. This was a rare occurrence with only 0.6% of the women reporting sleeping ≥10 hours, but the finding is very interesting and raises a number of questions. We would expect that difficulty falling and staying asleep would decrease the amount of sleep, but these women had very long sleep durations. Did the women incorrectly report time in bed as opposed to actual sleep duration? Or if they actually did get that much sleep, was the insomnia due to their already having more than enough sleep? Did the women suffer from a sleep disorder with daytime sleepiness, causing them to desire to sleep for long periods and to feel frustrated when they could not?

The authors also found both short and long sleep duration to be significantly associated with CHD and CVD after controlling for age and race, but the associations fell out of statistical significance after controlling for multiple other covariates. It is likely that some of the covariates included in the multivariate models were actually partial mediators of the relationship. Short sleep duration has been shown to be associated with obesity, diabetes, hypertension, and hyperlipidemia, ⁷ so these variables could have mediated the relationship between short sleep duration and CHD/CVD. A portion of the attenuation in the hazards ratios (HR) was therefore likely to have been due to the inclusion in the multivariate models of these variables. Short sleep duration has also been theorized to mediate the association between socioeconomic status and poor health, ⁸ so another portion of the attenuation could be attributable to the inclusion of variables such as income and education that are associated with diabetes, body weight, and hypertension. The older age of the study participants could also help explain the lack of statistically significant associations for sleep duration. Age has been shown to be an effect modifier of the relationship between sleep duration and obesity and between sleep duration and hypertension. Sleep duration is more strongly associated with hypertension incidence and obesity in middle-aged subjects than in elderly subjects. ⁷ The participants' reported sleep durations post-menopause are also unlikely to accurately reflect their sleep durations over the course of their lifetimes when they could have developed chronic medical conditions that increased their risk for CHD/CVD. If chronic medical conditions were present at study baseline then the resultant increases in pro- and/or anti-inflammatory cytokines could have functioned to either shorten or lengthen their sleep durations.

The findings from the present study highlight the importance of sleep in women's health. Future research on the relationship between sleep and CHD would ideally have repeated measures of sleep parameters, confounders, and mediators over the life-course to appropriately gauge their influence. If insomnia and abnormal sleep duration increase the risk for CHD, then interventions that improve sleep quality and normalize sleep duration could serve as primary preventative measures.