It has been reported that high prevalence of metabolic syndrome and increased risk of diabetes in postmenopausal women. This may be hormone imbalance caused by ovarian failure in menopause periods. Although many studies have reported the effect of hormone replacement therapy (HRT) on chronic diseases, the risk and benefits of HRT are still controversial. Most of studies are focused on older white women, and studies target for younger menopausal Asian women are insufficient. In this study, we investigated the effect of HRT on metabolic syndrome components according to diabetes status in Korean women.

This study was conducted in the Health Examinees (HEXA) study, a multicenter-based cohort using baseline data from 2004 to 2013. A total of 47,307 nondiabetic and 4003 diabetic postmenopausal women aged 40–69 years were included after excluding participants without information of menopause status, HRT and diabetes. Data was collected using interviewer-administered self-reported questionnaires. HRT was categorized into never, past and current users. Diabetes was defined as fasting glucose level ≥ 126 or reported medication use (hyperglycemic drug or insulin) by considering diagnosis of diabetes at recruitment. The mean differences of metabolic syndrome components[fasting glucose (FG), total cholesterol, high-density lipoprotein (HDL), triglyceride (TG), systolic and diastolic blood pressure (SBP and DBP), body mass index (BMI), waist circumstance (WC), waist-hip ratio (WHR) and visceral fat] were assessed by using a general linear model (GLM) and Tukey's multiple comparisons tests by adjusting for age, education, smoking, drinking, physical activity, hypertension, dyslipidemia and family history of diabetes or hypertension. The associations of HRT on metabolic syndrome components were performed stratified by diabetes status and the duration and the period of initiation of HRT after the menopause were considered among current users.

HRT use was associated with lower FG, total cholesterol, TG, SBP, DBP, BMI, WC, WHR and visceral fat among nondiabetic women, which were more stronger among current HRT users; mean differences for FG (past vs. never: 0.43 [95% CI: 0.16, 0.69] and current vs. never: 2.98 [95% CI: 2.57, 3.39]), total cholesterol (past vs. never: −0.08 [95% CI: −0.96, 0.81] and current vs. never: 11.39 [95% CI: 10.02, 12.76]), TG (past vs. never: 2.13 [95% CI: 0.16, 4.10] and current vs. never: 12.30 [9.26, 15.33]), WC (past vs. never: 0.77 [95% CI: 0.56, 0.97] and current vs. never: 1.26 [95% CI: 0.94, 1.57]), and visceral fat (past vs. never: 0.06 [95% CI: 0.04, 0.09] and current vs. never: 0.11 [95% CI: 0.07, 0.16]). The associations of HRT use on FG, total cholesterol, TG, SBP and WC were still significant among diabetic women. Considering the duration and the period of initiation of HRT after the menopause among current users, we could not confirm the differential associations between the groups, though.

Healthy postmenopausal women could be expected to reduce of metabolic syndrome components by using HRT. It is still necessary to further study to evaluate the risks and benefits of HRT applied for younger healthy menopausal women.