What influences the age of menopause?

Despite significant changes in the age of menarche over recent decades, no such change has been seen in the age of menopause. Although there may be minor differences across different ethnicities, the age of menopause has remained remarkably static over time. Complex processes determine the rate of loss of ovarian follicles over our reproductive lives1. We understand that genetic, medical and environmental factors may influence the age of menopause but that understanding is rudimentary at best. A better appreciation of the processes that regulate ovarian reserve is vital to our understanding of premature ovarian insufficiency (POI) and menopause generally. Intervention may then be possible to prevent oocyte atresia and early loss of fertility.

A woman’s pool of follicles reaches its peak prior to birth and minimal decline is seen leading up to puberty. FSH, LH, estrogens, androgens, activins, inhibins and anti-Müllerian hormone (AMH) all have roles in follicle growth, development and atresia. FSH is a survival signal that prevents atresia for the developing follicles but, for every one oocyte incorporated into a follicle, nine others are lost.

Estimates of the heritability of the age of menopause vary from 30 to 85%2. X chromosome abnormalities and mutations in genes involved in follicle formation and hormone action have been associated with the onset of POI. Turner syndrome and Fragile X syndrome are good examples. Carriers of the BRCA genes reach menopause on average 3 years before other women3. Other genes involved in DNA repair and cell death may have a small but additive role in determining age at menopause. In this regard, genome-wide association studies have shown a potential role for genes coding for cytokine or inflammasome production4. We know that a range of autoimmune conditions is more common among women with POI. The presence of ovarian autoantibodies has been noted in many women with a premature menopause but whether this is a normal association of ovarian aging or the cause of it remains unclear.

Many of the studies examining the effect of environmental factors on age of menopause have relied on retrospective self-reporting, with all the associated flaws that come with that method of data collection. However, we have been able to glean some useful information about factors over which we have some control. For instance, smoking accounts for 5% of the risk of early menopause and this is calculated to be similar to the summed effect of the top 17 genetic variants1. AMH is lower and declines more steeply in smokers, bringing forward menopause by approximately 1 year5.

In a study of over 2000 premenopausal women, alcohol, socioeconomic status, exercise and body composition appeared to play no role in determining age of menopause. Very low (< 2.5 kg) or high (> 4 kg) birth weight has been associated with an earlier than average menopause but breast-feeding or weight at age 2 years has no impact6. Use of the contraceptive pill and longer cycles have been linked with a later age at menopause but the effect of parity remains controversial.

In a cross-sectional study of Latin-American women, type 2 diabetes was associated with a three-fold higher incidence of menopause before 45 years compared to non-diabetic women7. With the growing incidence of type 2 diabetes, this effect on ovarian function and the related changes in cardiovascular health are a concern. Use of chemotherapeutic agents, particularly alkylating agents, for treatment of cancer or autoimmune conditions is clearly associated with early ovarian demise. Although this outcome may be discussed with and understood by women undergoing treatment for cancer, many women with conditions such as systemic lupus erythematosus may not be warned of the impact of treatment on ovarian reserve.

Risk-reducing surgery for women with BRCA mutations has typically involved bilateral oophorectomy and resulted in premature menopause. A recent move towards early bilateral salpingectomy as an alternative option is not well studied and its effect on ovarian aging unclear. Retrospective and cross-sectional studies of women presenting for in vitro fertilization have demonstrated lower AMH and higher FSH levels in women who had undergone bilateral salpingectomy compared to untreated women8^,9. In light of the evidence that even a simple hysterectomy and other forms of minor ovarian surgery, such as drilling, bring forward the age of menopause, it is possible that a relatively 'benign' procedure such as tubal surgery may have a deleterious effect in a group of women already predisposed to an early menopause10^,11. Balancing the risks of cancer against the long-term health risks and changes in quality of life due to an early menopause can be a challenge.

Polycystic ovarian syndrome (PCOS) is a common condition that affects women of all ethnicities. Data would suggest that it is associated with a delay in menopause of approximately 2 years, making it one of the few conditions that extends a woman’s reproductive life12. An increased pool of follicles, slower decreases in AMH over time, higher LH levels and genetic factors have all been described in women with PCOS. The genetic variants seen in these women and which are linked to later age at menopause are also associated with the characteristic ultrasound and gonadotropin changes seen in PCOS.

The science of determining age at menopause is at an embryonic stage. However, knowledge of the role of risk factors, particularly environmental and iatrogenic ones, does give us the opportunity to intervene. A later menopause may, for many women, be associated with a significant decrease in all-cause mortality but an early menopause predicts considerable long-term health risks. Awareness of the risks enables early intervention for adverse lifestyle factors and enables balanced decision-making with medical interventions.