Vasomotor symptoms and insulin resistance in the study of women's health across the nation

摘要

ABSTRACT Classic symptoms of the menopausal transition are vasomotor symptoms (VMSs), such as hot flashes and night sweats. The association of VMSs with cardiovascular disease (CVD) and other health outcomes is unclear. Although some evidence exists for a relationship between VMSs and risk of CVD, findings have been mixed or observed only in certain populations. Relatively few studies have examined the possible association between VMSs and elevated body mass index (BMI) or blood glucose, insulin resistance, and other metabolic factors. The aim of this longitudinal, community-based cohort study was to determine whether VMSs (hot flashesight sweats) were associated with insulin resistance and blood glucose. Associations were examined in mixed models over 8 years* controlling for BMI, demographics, cardiovascular risk factors, medications, and the reproductive hormones estrogen (E2) and follicle-stimulating hormone (FSH). Data were obtained from the Study of Women's Health Across the Nation, which followed midlife women through the menopausal transition. Primary models included 3075 women, aged 42 to 52 years at entry who met criteria for inclusion. At each SWAN (Study of Women's Health Across the Nations) visit, VMSs were assessed using a questionnaire on the same day as the blood draw. Each participant was asked how often she experienced hot flashes or night sweats during the past 2 weeks: none, 1 to 5 days, 6 or more days, and the past 2 weeks. Data for hot flashes and night sweats were analyzed separately. Physical measures were blood pressure, height, and weight. Fasting blood was used to measure serum glucose and insulin levels as well as E2 and FSH. The Homeostasis Model Assessment (HOMA) index was used as an estimate of insulin resistance. Hot flashes were associated with a higher HOMA index (no hot flashes vs hot flashes: 1-5 days: % difference, 2.37; 95% confidence interval, 0.36-4.43, P = 0.02; and >6 days: % difference, 5.91; 95% confidence interval, 3.17-8.72, P < 0.0001) in multivariable models after adjustment for BMI. Associations persisted after additional adjustment for E2 and FSH. Findings were similar for night sweats. Although associations between hot flashes and glucose were also significant, the magnitude of the association was small. These findings show that VMSs are associated with insulin resistance as measured by the HOMA index and with glucose to a lesser extent. Understanding relations between VMSs and HOMA may contribute to understanding potential mechanisms linking hot flashes to risk of CVD.