Hanna Kische1, Ralf Ewert1, Ingo Fietze1, Stefan Gross1, Henri Wallaschofski1, Henry Völzke1, Marcus Dörr1, Matthias Nauck1, Anne Obst1, Beate Stubbe1, Thomas Penzel1, Robin Haring1. 1. Institute of Clinical Chemistry and Laboratory Medicine (H.K., H.W., M.N., R.H.), Departments of Internal Medicine B (R.E., A.O., B.S.) and Cardiology (S.G., M.D.), and Institute for Community Medicine (H.V.), University Medicine Greifswald, D-17475 Greifswald, Germany; Interdisciplinary Sleep Center (I.F., T.P.), Charité, University Medicine, 10117 Berlin, Germany; German Centre for Cardiovascular Research (S.G., H.V., M.D., M.N.) and German Centre for Diabetes Research (H.V.), partner site Greifswald, 13092 Greifswald, Germany; and Faculty of Applied Public Health (R.H.), European University of Applied Sciences, 18051 Rostock, Germany.
Abstract
CONTEXT AND OBJECTIVES: Associations between sex hormones and sleep habits originate mainly from small and selected patient-based samples. We examined data from a population-based sample with various sleep characteristics and the major part of sex hormones measured by mass spectrometry. DESIGN, SETTING, AND PARTICIPANTS: We used data from 204 men and 213 women of the cross-sectional Study of Health in Pomerania-TREND. MAIN OUTCOME AND MEASURES: Associations of total T (TT) and free T, androstenedione (ASD), estrone, estradiol (E2), dehydroepiandrosterone-sulphate, SHBG, and E2 to TT ratio with sleep measures (including total sleep time, sleep efficiency, wake after sleep onset, apnea-hypopnea index [AHI], Insomnia Severity Index, Epworth Sleepiness Scale, and Pittsburgh Sleep Quality Index) were assessed by sex-specific multivariable regression models. RESULTS: In men, age-adjusted associations of TT (odds ratio 0.62; 95% confidence interval (CI) 0.46-0.83), free T, and SHBG with AHI were rendered nonsignificant after multivariable adjustment. In multivariable analyses, ASD was associated with Epworth Sleepiness Scale (β-coefficient per SD increase in ASD: -0.71; 95% CI: -1.18 to -0.25). In women, multivariable analyses showed positive associations of dehydroepiandrosterone-sulphate with wake after sleep onset (β-coefficient: .16; 95% CI 0.03-0.28) and of E2 and E2 to TT ratio with Epworth Sleepiness Scale. Additionally, free T and SHBG were associated with AHI in multivariable models among premenopausal women. CONCLUSIONS: The present cross-sectional, population-based study observed sex-specific associations of androgens, E2, and SHBG with sleep apnea and daytime sleepiness. However, multivariable-adjusted analyses confirmed the impact of body composition and health-related lifestyle on the association between sex hormones and sleep.
CONTEXT AND OBJECTIVES: Associations between sex hormones and sleep habits originate mainly from small and selected patient-based samples. We examined data from a population-based sample with various sleep characteristics and the major part of sex hormones measured by mass spectrometry. DESIGN, SETTING, AND PARTICIPANTS: We used data from 204 men and 213 women of the cross-sectional Study of Health in Pomerania-TREND. MAIN OUTCOME AND MEASURES: Associations of total T (TT) and free T, androstenedione (ASD), estrone, estradiol (E2), dehydroepiandrosterone-sulphate, SHBG, and E2 to TT ratio with sleep measures (including total sleep time, sleep efficiency, wake after sleep onset, apnea-hypopnea index [AHI], Insomnia Severity Index, Epworth Sleepiness Scale, and Pittsburgh Sleep Quality Index) were assessed by sex-specific multivariable regression models. RESULTS: In men, age-adjusted associations of TT (odds ratio 0.62; 95% confidence interval (CI) 0.46-0.83), free T, and SHBG with AHI were rendered nonsignificant after multivariable adjustment. In multivariable analyses, ASD was associated with Epworth Sleepiness Scale (β-coefficient per SD increase in ASD: -0.71; 95% CI: -1.18 to -0.25). In women, multivariable analyses showed positive associations of dehydroepiandrosterone-sulphate with wake after sleep onset (β-coefficient: .16; 95% CI 0.03-0.28) and of E2 and E2 to TT ratio with Epworth Sleepiness Scale. Additionally, free T and SHBG were associated with AHI in multivariable models among premenopausal women. CONCLUSIONS: The present cross-sectional, population-based study observed sex-specific associations of androgens, E2, and SHBG with sleep apnea and daytime sleepiness. However, multivariable-adjusted analyses confirmed the impact of body composition and health-related lifestyle on the association between sex hormones and sleep.