Invited Commentary: Women’s Reproductive Life Spans and Subsequent Inflammatory Profiles—How to Best Measure Reproductive Life Span and the Need for Baseline Assessments

Abstract

Inflammatory processes are known to drive the pathogenesis of several chronic diseases, including cardiovascular disease and Alzheimer disease, as well as all-cause mortality. Emerging research indicates that women who have a longer reproductive life span—roughly determined as the period from menarche to menopause or lifetime number of ovulatory cycles after accounting for anovulatory time spans—are at lower risk for these inflammation-related chronic diseases. The paradox is that ovulation is known to induce acute inflammation. Given the limited research assessing the relationship between reproductive life span and later inflammatory profiles, Huang et al. (Am J Epidemiol. 2020;189(7):660–670) set out to investigate this relationship within 2 of the most robust longitudinal cohort studies of women, the Nurses’ Health studies. They found that after adjustment for other inflammation-related factors, including adiposity, exercise, and diet, lifetime ovulatory years was associated with lower C-reactive protein levels in both premenopausal and postmenopausal women. Huang et al. call attention to several challenges in research on women’s reproductive life spans, including how to appropriately capture lifetime ovulatory cycles and the need for repeated measurements of inflammatory biomarkers across the life course if we wish to understand pathogenic processes linking reproductive factors to subsequent chronic disease.

Abbreviations

CRP

C-reactive protein

LOC

lifetime ovulatory cycles

LOY

lifetime ovulatory years

Inflammatory processes are known to drive the pathogenesis of several chronic diseases, including cardiovascular disease (1) and Alzheimer disease (2, 3), as well as all-cause mortality (4). Cardiovascular disease is the leading cause of death among women in the United States, with 1 in 3 women dying from heart disease or stroke (5). Women are also disproportionately affected by Alzheimer disease and related dementia, with lifetime risk being 1 in 5 for women compared with 1 in 10 for men (6). Studies show that women’s longer life span only partially explains this risk difference. Variation in exposures throughout the life course also contributes, including variations in a woman’s menstrual cycle function, ability to achieve and maintain a healthy pregnancy, and menopausal transition experience. A woman’s reproductive history is thought to be an important modifier of inflammatory activity (7), but few studies assessing associations between women’s reproductive life histories and inflammatory biomarkers have been conducted.

In the current issue of the Journal, Huang et al. (8) evaluate the association between estimated lifetime ovulatory years (LOY) and levels of several inflammatory biomarkers, including C-reactive protein (CRP), interleukin 6, and soluble necrosis factor α receptor 2, among women from 2 large prospective cohort studies, the Nurses’ Health studies. The authors found that after adjustment for other inflammation-related factors, including adiposity, exercise, and diet, LOY was associated with lower CRP levels in both premenopausal and postmenopausal women (8). They thus conclude that although many reproductive events induce acute inflammation (e.g., ovulation), higher LOY is associated with lower chronic inflammation across women’s life spans. The paper by Huang et al. brings to our attention the important role that LOY may play in chronic disease risk for women, as well as some methodological challenges regarding 1) how to best capture all of the reproductive history elements that make up LOY and 2) how to determine causation when relying on single versus repeated measurements of inflammatory biomarkers.

REPRODUCTIVE HISTORY AND LOY

Defining female reproductive life span is challenging. The emerging use of administrative medical databases in population-based research, including women’s reproductive health, is attractive because of their low cost, large size, and representative nature (9, 10). However, specific to capturing female reproductive life span, they are limited, and thus we rely on large, prospective cohort studies such as the Nurses’ Health studies. Huang et al. should be commended for their detailed assessment of female reproductive life span as captured via LOY. Strengths of their assessment include biennial self-reports of parity (pregnancies lasting longer than 6 months), oral contraceptive use (duration and brand, for calculation of cumulative estrogen and progestin dose), menopausal status, and surgical removal of the uterus/ovaries, as well as a baseline report of age at menarche (8). However, there are several other factors in addition to parity, oral contraceptive use, and age at menarche/menopause that should be captured in future studies in order to truly capture female reproductive life span, including all conceptions, pregnancy losses (and time of loss), lactation, and menstrual cycle length.

Calculation of lifetime ovulatory cycles (LOC), as opposed to LOY, is a more robust measurement of the exposure that Huang et al. are basing their hypothesis on (i.e., “reproductive factors that disrupt hormonal profiles and reduce the number of ovulatory cycles” (8, p. 000)). For example, Tung et al. (11), in their assessment of the association between anovulation factors and pre- and postmenopausal ovarian cancer risk, calculated LOC in the following manner: subtracting the total duration (years) of any anovulatory periods due to pregnancy, lactation, oral contraceptive use, and amenorrhea from the total duration (years) of menstruation (age at menopause (or age at interview for premenopausal women) minus age at menarche) and then multiplying by the estimated number of menstrual cycles per year, based upon a woman’s cycle length (365/cycle length). Calculation of LOC, taking into account the duration of all pregnancies, not just those that lasted longer than 6 months, as well as periods of breastfeeding, has been undertaken in several other cancer studies for over 2 decades (11–13), and this measure should be equally applied to inflammation-related chronic disease research.

Considering first- and second-trimester pregnancy losses in addition to time spent breastfeeding is important not only for better capturing women’s total estrogen exposure, which is thought to influence risks of cardiometabolic disease, Alzheimer disease, and reproductive cancer (11–15), but also because of the potential long-term immunological benefits of early pregnancy (7) and the insulin sensitivity benefits of breastfeeding (16, 17). Specifically, the rather crude process of subtracting 1 year of LOY for every pregnancy that lasted more than 6 months fails to account for the dramatic increase in regulatory T-cells during the first trimester of pregnancy that prevents excessive inflammation (7). Furthermore, beyond a woman’s ability to maintain a pregnancy, we must begin collecting information on women’s ability to conceive. The most common causes of female infertility, including polycystic ovary syndrome, endometriosis, fibroids, and diminished ovarian reserve, have all been linked to chronic morbidity and adverse events later in life (18). Women with high LOY due to choosing not to conceive versus not being able to conceive are different in terms of their long-term chronic disease risk.

NEED FOR REPEATED MEASUREMENTS IN PROSPECTIVE COHORT STUDIES OF BIOMARKERS

Given their finding that LOY is linked with lower subsequent CRP levels, Huang et al. conclude that reproductive history may be an important source of inflammation in women’s later years, comparable to diet and lifestyle factors (8). The questions that remain to be answered include, How do women’s early reproductive inflammatory profiles affect their later inflammatory profiles, and how do women’s menstrual cycle function and reproductive events mediate this relationship? Inflammatory processes are implicated in the pathogenesis of women’s reproductive disorders, including endometriosis (19), polycystic ovary syndrome (20), and pregnancy loss (21). Specific to CRP, findings from the Effects of Aspirin in Gestation and Reproduction (EAGeR) Study demonstrated that lean women with inflammation measured by high-sensitivity CRP had lower fecundity, with aspirin therapy restoring fecundity (21). The power of better understanding the mediating effects of reproductive disorders and therapies used to treat these disorders is not only to “clarify the potentially pathogenic role of inflammation linking reproductive factors to chronic disease risk” as Huang et al. point out (8, p. 000) but additionally to understand how lifestyle modifications and therapeutic treatments can mitigate chronic disease risk. This understanding is particularly key given that modifiable factors beyond oral contraceptive use and pregnancy are needed for potential alteration of women’s reproductive life spans.