Introduction: Age at menopause and the postmenopausal state are frequently linked to cardiovascular disease (CVD) risk. Anti-Müllerian hormone (AMH) is produced by ovarian antral follicles and its concentration in peripheral blood is a quantitative estimation of the size of the antral follicle pool, thereby providing a measure of ovarian reserve before the end of the reproductive lifespan. We and others have shown that AMH is related to age at menopause. Literature on whether AMH is related to CVD risk is still scarce and heterogeneous.
Hypothesis: We hypothesized that a low ovarian reserve, measured by low levels of AMH, is associated with higher risk of CVD risk in premenopausal women.
Methods: We used data from 2,338 premenopausal women who participated in the second follow-up round of the population-based Doetinchem Cohort Study. In a cross-sectional analysis, CVD risk was compared between women in different AMH level-based categories. AMH was measured in stored serum samples using a Gen-II ELISA assay (Beckman-Coulter, Sinsheim, Germany) in a single laboratory. CVD risk was assessed in three ways: levels of single cardiovascular risk factors (systolic and diastolic blood pressure, total and HDL cholesterol, glucose), a metabolic risk factor score, and predicted 10- and 30-year risk using Framingham prediction rules. The association of AMH categories with CVD risk outcomes was studied using linear regression models, adjusted for age, age-squared, current estrogen use, current smoking, parity, menstrual cycle regularity, current pregnancy and BMI.
Results: AMH was unmeasurable in 456, and measurable, but below the limit of quantification (0.16 μg/L) in 186 women. In the remaining 1,696 women AMH level could be quantified and the median [IQR] AMH was 1.34 [0.24-5.14] μg/L. After adjustment, women with quantifiable AMH had 0.11 mmol/l (95% CI -0.23; 0.01) lower total cholesterol level than women with unmeasurable and unquantifiable AMH levels. Women with quantifiable AMH had on average 0.11 (95% CI 0.01-0.21) fewer metabolic risk factors compared to women with unmeasurable AMH, adjusted for confounders. Women with quantifiable AMH also had a 5% (95% CI -0.01; 11.0) lower hazard of 10- and 30-year CVD occurrence compared to women with unmeasurable AMH.
Conclusions: This study indicates that premenopausal women with very low ovarian reserve, expressed by unmeasurable serum AMH levels, may have an increased CVD risk. Longitudinal research with clinical endpoints and more sensitive AMH assays are necessary to confirm a relationship between AMH and CVD risk.