Early menopause and cardiovascular risk factors: a cross-sectional and longitudinal study

Abstract

Objective: The aim of the study is to evaluate the cross-sectional and longitudinal association of early natural menopause with changes in cardiovascular risk factors (CVRFs).

Methods: Postmenopausal women from the Swiss CoLaus study, reporting age at natural menopause (ANM) and having CVRFs measurements (blood lipids, blood pressure, glucose, homeostatic model assessment for insulin resistance [HOMA-IR], and inflammatory markers) at baseline (2003-2006) and first follow-up (2009-2012) were eligible for analysis. Age at natural menopause was analyzed as a continuous variable and in categories (ANM <45 and ≥45 y old). Linear regression analysis and linear mixed models were used to assess whether ANM is associated cross-sectionally and longitudinally with changes in CVRFs. Models were adjusted for demographic characteristics, lifestyle-related factors, time since menopause, medication, and clinical conditions.

Results: We analyzed 981 postmenopausal women. The cross-sectional analysis showed that women with ANM younger than 45 years had lower diastolic blood pressure (β = -3.76 mm Hg; 95% confidence interval [CI] = -5.86 to -1.65) compared with women whose ANM was 45 years or older. In the longitudinal analysis, ANM younger than 45 years was associated with changes in log insulin (β = 0.26; 95% CI = 0.08 to 0.45) and log homeostatic model assessment for insulin resistance levels (β = 0.28; 95% CI = 0.08 to 0.48). No associations were found between ANM and other CVRFs.

Conclusions: Early menopause may be associated with changes in glucose metabolism, while it may have little to no impact on other CVRFs. Larger longitudinal studies are needed to replicate our findings.

FIG 1

Longitudinal changes in insulin and HOMA-IR. (A) Changes in log insulin in women with early menopause compared with women with menopause at average or late menopause. (B) Changes in log HOMA-IR in women with early menopause compared with women with average or late menopause. Model 3: using linear mixed models' regression, adjusted for chronological age, drinking status (yes vs. no), smoking (never, current, former), physical activity (once, twice, three times per week and none), history of cardiovascular diseases (yes vs. no), diabetes (yes vs. no), age at menarche, use of antihypertensive medication, diabetes medication, hormone therapy and statins (yes vs. no), body mass index, and interaction term ANM and follow-up time in years. Early menopause <45 years old, average and late ≥45 years old (Ref). ANM, age at natural menopause; HOMA-IR, homeostatic model assessment for insulin resistance.