This study was conducted using data from KNHANES 2013–2014. The KNHANES is a cross-sectional and nationally representative survey that was designed to evaluate the health and nutritional status in the Korean population of all ages. All participants provided written informed consent, and the institutional review board of the Korea Centers for Disease Control and Prevention approved the study protocol (2013-07CON-03-4C, 2013-12EXP-03-5C). The KNHANES is based on complex, multistage stratified and clustered samples of the Korean population. We selected 3,023 participants among 3,219 premenopausal women aged 20 to 55 years who had complete information on marital status, household income, education level, smoking status, alcohol consumption, physical activity, number of pregnancies, use of oral contraceptives, duration of lactation, and duration of estrogen exposure.

Participants answered a self-reported questionnaire on their current age, socioeconomic variables (marital status, household income, education level), lifestyle variables (smoking status, alcohol consumption, physical activity), and reproductive variables (number of pregnancies, use of oral contraceptives, duration of lactation). Early menarche was defined as age at the first menstrual period less than 12 years, and late menarche was defined as age at the first menstrual period greater than or equal to 16 years [4,16]. Participants were divided into 3 groups based on age at menarche: <12 years, 12–15 years, ≥16 years.

Marital status was divided into two groups. Those who had a spouse or were living with a partner were classified as married, and all others were classified as unmarried. Household income was classified by quartiles based on monthly average family equivalent income: high, higher-middle, lower-middle, and lower. Education level was divided into two groups: high school graduate or less, and college graduate or more.

Smoking status was divided into three groups. (1) Non-smokers were defined as individuals who had never smoked or had smoked less than 100 cigarettes in their lifetime. (2) Past smokers were defined as individuals who had smoked more than 100 cigarettes in the past, but do not currently smoke. (3) Current smokers were defined as individuals who had smoked more than 100 cigarettes and continued to smoke at the time of questionnaire completion. Those who drank more than 5 glasses of alcohol per session and more frequently than twice a week were classified into the high risk drinking group according to the KNHANES recommendations. The physically active group was defined by walking for more than 30 minutes per session and more than five days a week.

The number of pregnancies was divided into two groups, based on pregnancies more than or equal to 3 times, including abortion; use of oral contraceptives was divided into two groups, based on oral contraceptive use for more than 1 month [4]. Duration of lactation was divided into 4 groups: <1 month, 1–6 months, 6–12 months, and ≥12 months [17]. Duration of estrogen exposure was calculated from the current age minus the age at menarche.

Physical examinations and blood tests were performed by the survey team to assess MetS. The body mass index (BMI) was calculated by dividing the measured weight by the square of the height. Plasma glucose, triglycerides, and HDL-C were measured using a Hitachi Automatic Analyzer 7600-210 (Hitachi, Tokyo, Japan) after an 8–12-hour overnight fast. MetS was defined according to International Diabetes Federation and National Heart, Lung, and Blood Institute/American Heart Association guidelines as any three or more of the following [18]: (1) waist circumference for Korean women ≥85 cm [19]; (2) fasting triglycerides ≥150 mg/dL; (3) HDL-C <50 mg/dL; (4) systolic blood pressure ≥130 mm Hg or diastolic blood pressure ≥85 mm Hg or taking medicine for high blood pressure; and (5) fasting plasma glucose ≥100 mg/dL or taking medicine for high glucose.

Statistical analysis was performed using sampling weights according to KNHANES guidelines. Because the prevalence of MetS increases as age increases, participants were divided into two groups based on age at the time of the survey: 20–39 years and 40–55 years. In order to compare the characteristics of participants according to age at menarche, continuous variables were reported as mean and standard deviation using analysis of variance, and categorical variables were reported as percentage using the chi-square test. Post-hoc testing using the Bonferroni method was performed to examine differences between the components of MetS in three groups according to the age at menarche. To investigate the independent relationship between age at menarche and MetS, we used multivariable logistic regression analysis. We examined the adjusted odds ratios (ORs) and 95% confidence intervals (CIs), controlling for current age, socioeconomic status, lifestyle variables, reproductive status, and duration of estrogen exposure. SAS ver. 9.2 (SAS Institute Inc., Cary, NC, USA) was used to perform all statistical analyses. P-values less than 0.05 were considered statistically significant.

Table 1 shows general characteristics according to the age at menarche. The mean current age was 37.29 years, and the mean age at menarche was 13.56 years. Younger women reported earlier menarche (mean current age: 30.98 years with menarche at <12 years, 37.09 years with menarche at 12–15 years, and 43.96 years with menarche at ≥16 years). Women with early menarche were less likely to marry, had more education, and were physically active. They were also less likely to be pregnant 3 times or more and lactate for more than 12 months (Table 1).

The prevalence of MetS was 8.7% with menarche at <12 years, 8.3% with menarche at 12–15 years, and 15.3% with menarche at ≥16 years.

Table 1 shows that age at the time of survey affected the prevalence and components of MetS. The ages were significantly different among the three groups according to age at menarche. Therefore, we compared the prevalence and components of MetS after dividing the subjects into two groups based on age at the time of survey. Women with early and late menarche were more likely to have MetS after categorization into two groups based on age at the time of survey (6.0%, 3.7%, and 4.2% in those aged 20–39 years; 23.4%, 14.1%, and 19.2% in those aged 40–55 years).

Table 2 shows the distribution of each component of MetS according to the age at menarche after adjusting for current age of participants. In those aged 20–39 years, BMI and waist circumference were significantly higher in women with menarche at <12 years of age than in those with menarche at 12–15 years of age or older. HDL-C, fasting triglycerides, fasting plasma glucose, and blood pressures were not significantly different according to age at menarche. In those aged 40–55 years, systolic blood pressure was higher in women with menarche at ≥16 years than in those with menarche at 12–15 years or younger. There were no significant differences in the other components of MetS among the three groups (Table 2).

Table 3 shows the risk of MetS according to the age at menarche after adjusting for confounders. In those aged 20–39 years, compared with women with menarche at 12–15 years, the risk of MetS did not increase in those with early and late menarche after adjusting for current age (OR, 1.922; 95% CI, 0.951–3.885 with menarche at <12 years; OR, 1.053; 95% CI, 0.317–3.500 with menarche at ≥16 years). However, in model 2, which was adjusted for marital status, household income, education level, smoking status, high-risk drinking, and physical activity, in addition to current age, the risk of MetS was significantly increased in women with early menarche (OR, 2.035; 95% CI, 1.011–4.094). In model 3, which was further adjusted for number of pregnancies, history of oral contraceptive use, and duration of lactation, the risk of MetS did not increase in those with early and late menarche (OR, 1.767; 95% CI, 0.718–4.351 with menarche at <12 years of age; OR, 0.743; 95% CI, 0.171–3.219 with menarche at ≥16 years of age). In model 4, which was further adjusted for duration of estrogen exposure, the risk of MetS with menarche at <12 years of age tended to decrease compared to that in other models, but there was no statistically significant difference. In those aged 40–55 years, the results were the same as in those aged 20–39 years (Table 3).

Age at menarche did not show a significant correlation with MetS after analysis of age at menarche as a continuous variable (data not shown).