Early menopause, hysterectomy, and biological aging: Health and Retirement Study

Abstract

Objective:

This study examines the association between the timing of menopause and hysterectomy and biological aging, focusing on epigenetic and physiological aging markers.

Methods:

Data were analyzed from women aged 56 and over in the nationally representative Health and Retirement Study (HRS). Regressions of biological aging measured by accelerated epigenetic aging and biological age based on physiological dysregulation on menopause/hysterectomy history were conducted to examine associations of normal-aged and early menopause with and without hysterectomy with biological aging.

Results:

Hysterectomy, whether following normal-aged or early menopause, or in younger ages, was significantly associated with markers of accelerated biological aging. Women with early menopause or hysterectomy showed accelerated epigenetic aging. Early menopause was associated with accelerated physiological dysregulation only when combined with hysterectomy, suggesting that surgical menopause may be related to accelerated systemic aging processes. Epigenetic clocks were associated with early cellular and molecular aging changes linked to natural early menopause, while physiological dysregulation was associated with the cumulative systemic impacts related to hysterectomy.

Conclusions:

This study highlights associations between reproductive history and biological aging. These findings underscore the importance of considering both natural and surgical factors in menopause in evaluating aging-related health risks and suggest avenues for targeted interventions to mitigate health risks in women with these reproductive histories.

Menopause is a significant event in a woman’s life, marking the end of reproductive years and triggering a range of physiological changes and health changes such as the onset of cardiovascular disease,^1,2 osteoporosis,^3,4 type 2 diabetes,⁵ and cognitive decline.^6,7 While menopause has been examined as an indicator of reproductive health and physiological, cognitive and functional aging, recent research has begun to focus on its broader impact on accelerated biological aging at the cellular level.^8-10

Studies have linked menopause to accelerated epigenetic aging, with the timing of menopause playing a crucial role in determining its association with biological aging. While menopause typically occurs between the ages of 45 and 55, with an average onset at 52 in the United States,¹¹ some women experience early menopause. Studies show that early-onset menopause is related to faster epigenetic aging,^8,9 while later onset menopause is related to higher estimated granulocyte levels in DNA methylation (DNAm), which may suggest protective immune function and inflammation regulation.¹⁰ Hysterectomy also warrants attention in this context. Hysterectomy is a common procedure such that 41.8% of women aged 75 and older in the United States in 2021 had at some time received a hysterectomy.¹² This procedure often coincides with the menopause transition due to an increased prevalence of gynecologic conditions during the premenopausal and menopausal years,¹³ introducing additional health risks. In addition to potential complications from surgery, hysterectomy even when performed with ovarian conservation, has been associated with disruption of ovarian physiology, including earlier ovarian failure¹⁴ and hormonal changes (eg, estrogen and androgen levels) that may influence biological aging and systemic health.^15,16 This can, in turn, accelerate biological aging. These hormonal shifts have been linked to increased risks of cardiovascular,^17,18 metabolic,¹⁹ immune and inflammatory,^20,21 skeletal^22,23 health, all of which are critical factors in the biological aging of postmenopausal women. Hysterectomy, especially when performed early in life, is linked with an even faster decline in estrogen levels than natural menopause, potentially exacerbating these age-related changes.^8,24 Furthermore, hysterectomy may impact biological and epigenetic aging through its effects on the endocrine system and brain function, through disruption of the uterus-ovary-brain endocrine axis, as suggested by cognitive and hormonal changes observed in animal models, while the stress of surgery itself could also contribute to accelerated aging processes.²⁵ The uterus, even when not involved in reproduction, plays a critical systemic role beyond fertility^25,26 and underscores the complex biological implications of hysterectomy on overall health and aging trajectories. This suggests the importance of examining the combined effects of hysterectomy and menopausal timing on biological aging, yet this relationship remains largely unexplored.

Given the potential links between menopause, hysterectomy, and accelerated biological aging, it is essential to explore how these life events are related to cellular and physiological processes over time. Epigenetic aging, which involves changes in the chemical modifications of DNA, particularly through methylation, reflects cumulative cellular damage and can offer insights into the biological impacts of hormonal shifts associated with menopause and hysterectomy. These epigenetic modifications do not alter the genetic code itself but influence gene expression, making epigenetic aging a valuable tool for studying menopause’s role in heightened risks for age-related diseases and making it a potential target for interventions aimed at slowing down postmenopausal aging. Epigenetic clocks like PhenoAge, GrimAge, and DunedinPACE, which are based on health outcomes and mortality risk, may be highly sensitive to the molecular-level changes²⁷ that natural menopause and/or surgical menopause may trigger because these clocks capture early cellular aging, especially through methylation patterns linked to inflammation, oxidative stress, cardiovascular risk²⁸ and altered immune cell composition of blood,^29,30 and thus may reveal accelerated aging following menopause.

In addition to epigenetic aging, biological age provides information about another level of physiology and a direct assessment of multisystem physiological changes linked to health outcomes.³¹ While epigenetic measures reflect molecular changes, biological age, based on clinical and biochemical indicators including cardiovascular and metabolic health, lipid metabolism, immune and inflammatory markers, liver, kidney and respiratory function, hematological parameters, and more, captures downstream effects that are closely tied to age-related, multisystem physiological changes that are linked to health outcomes such as increased cardiovascular risk,^32,33 metabolic and inflammatory dysfunction,³⁴ hormone changes, and bone density loss, which are common postmenopausal outcomes. Together, these aging measures would provide complementary insights into the biological aging process in relation to menopause and hysterectomy.

Socioeconomic status and lifestyle factors are known to accelerate biological aging and menopause age. Research shows that those with lower levels of education^35-37 and race/ethnic minorities such as Black and Hispanic persons³⁸ are more likely to have faster biological aging, both epigenetic and physiological. These groups also tend to experience earlier menopause,^39-41 which could influence the relationship between menopause and biological aging. Lifestyle factors such as smoking are also associated with earlier menopause onset,^39,42-44 occurrence of hysterectomy,⁴⁵ and accelerated aging.^35,37,46,47 Thus, considering lifetime smoking history, education level, and race/ethnicity is important to disentangle the specific association of menopause and hysterectomy with biological aging, ensuring that observed associations are not confounded by these factors.

This study focuses on the relationship between the timing of menopausal age and hysterectomy with accelerated epigenetic aging and biological age. Specifically, we examine whether early menopause, hysterectomy, or the combination of both are related to accelerated aging and how early menopause with and without hysterectomy is associated with a higher risk of accelerated biological aging as measured by both epigenetic and physiological indicators. By examining the independent and combined associations of menopausal timing with hysterectomy, this study aims to provide insights that may inform interventions designed to mitigate the risks of accelerated aging in women who experienced early menopause and/or underwent hysterectomy.

METHODS

Data

The data come from the Health Retirement Study (HRS), a nationally representative longitudinal study of more than 22,000 Americans over the age of 50 conducted biannually. The survey covers economic, physical, and mental health, social stresses, marital and family status, financial support situations, labor market status, and retirement planning. The HRS which is collected by the Survey Research Center at the University of Michigan and supported by the National Institute on Aging of the National Institutes of Health (NIH) employs rigorous methodologies to ensure data validity, reproducibility, and minimal interviewer variation.⁴⁸ To ensure representativeness, the HRS utilizes a multistage area probability sample, oversampling minority populations such as African American and Hispanic households. Data collection is conducted primarily through face-to-face and telephone interviews, as well as self-administered questionnaires, all adhering to standardized protocols implemented by trained interviewers. To minimize interviewer variation, the HRS implements standardized training programs for interviewers and employs computer-assisted interviewing systems, ensuring uniform administration of survey instruments and reducing interviewer-related variability. Regular quality control measures, such as reinterviews and validation studies, are conducted to monitor and address any inconsistencies arising from observer variation.

The 2016 wave of HRS included venous blood collection by phlebotomists in a home visit about 2 months after a core interview. The biomarker data were collected from 9,193 community-dwelling respondents. Centrifuged blood from the field was sent cold to the Advanced Research and Diagnostics Laboratory at the University of Minnesota and assayed in this lab for aging-related biomarkers. Epigenetic aging measures were derived from the HRS 2016 venous blood innovation subsample (N=4,018), which, when weighted, is a representative subsample of community-dwelling self-respondents. In addition to biomarkers from venous blood sample (VBS) collection, measures from physical measures data and dried blood spot data were also used for the construction of the expanded biological age measure based on physiological and clinical chemistry measures. Physical measures and dried blood spots were collected at every other wave for a random half sample and data for the other half sample were collected at the alternate wave, thus we used data from 2014 and 2016 to make data for the full sample.

Menopause and hysterectomy questions, including the age at menopause and at hysterectomy, were first asked in 2008 to self-interviewees only, but not to proxy interviewees in HRS. After 2008, women who previously said they had a hysterectomy were not asked the questions about hysterectomy. The data on menopause and hysterectomy were gathered from all survey years (2008-2022), with earlier waves used to supplement missing data and to allow us to include new participants.

Out of 17,190 women with information on menopause, 2,349 had epigenetic data; 4,094 had the variables needed for the measure indicating physiological dysregulation; 1,736 had both epigenetic and the physiologically based biological age data; 1,660 had a VBS subsample weight. Those who were not postmenopausal by 2016 were excluded (n=6). Those who had missing information for the timing of menopause (n=168) and those whose reported menopause age was older than hysterectomy age (n=25) were treated as missing due to potential misreporting given that having a hysterectomy before natural menopause was coded as having menopause. With additional missing cases from covariates such as education (n=6) and smoking pack years (n=12), the final sample consisted of 1,443 postmenopausal women (Supplemental Fig. 1, Supplemental Digital Content 1, http://links.lww.com/MENO/B366). The sample in this analysis was slightly younger (68.3 y) than the initial sample with epigenetic data (69.1 y) or biological age data based on physiology (68.4 y).

Measures

Early menopause and hysterectomy

For those who did not provide the exact age of menopause, subsequent questions on the range of menopause age were asked to bracket the age: older than/about/younger than age 50, 45, and 55. We assigned mid-interval or close values for these answers: 48 for those who answered their menopause age was younger than 50; 47 for those whose menopause was between 45 and 50; 52 for those with menopause between 50 and 55; 52 for those reporting menopause about age 50; 56 for those about 55; and 47 for those about 45. If the respondents’ answers to menopausal age were different across waves, we took the age reported at the most recent wave. If a respondent answered that she had a hysterectomy in any wave up to 2016, we coded her as having a hysterectomy and used the age of the hysterectomy reported in that wave. Hysterectomy is defined as surgery to remove the uterus or womb regardless of the removal of ovaries. Because ovarian removal leads to a more abrupt decline in estrogen and other hormones; hysterectomy without ovarian status data should be interpreted with caution.

Early menopause was defined as menopause occurring at age 44 or younger. While the Office on Women’s Health defines early menopause as menopause occurring before age 45 and menopause occurring before age 40 as premature menopause, we use the term “early menopause” to include all cases of menopause occurring before age 45. Women whose hysterectomy occurred before age 45 were not asked the question about the age of menopause so based on the timing of the hysterectomy, they were defined as having early menopause (n=359). To examine the combined effect of the timing of menopause and hysterectomy, we developed 4 categories by combining menopause timing and hysterectomy status: normal-aged menopause and no hysterectomy; normal-aged menopause and hysterectomy; early menopause and no hysterectomy; and early menopause and hysterectomy This operationalization aimed to account for early hormonal disturbances and the potential health impacts associated with surgically induced early menopause. It aligns with epidemiological definitions that recognize hysterectomy as an indicator of premature or early menopause transition, while also distinguishing the potential differences in health outcomes between naturally occurring early menopause and menopause induced by a surgical procedure.

Composite factor score of accelerated epigenetic aging

Second-generation and third-generation epigenetic clocks—PhenoAge, GrimAge, and DunedinPACE—were used in this study. These clocks are estimated from DNA methylation (DNAm) patterns that correlate with various aging-related phenotypes, health risks, mortality, and changes in health status.⁴⁹ These clocks have little overlap in the Cytosine-phosphate-Guanine (CpG) sites included but are assumed to reflect an aging process related to health outcomes.⁵⁰ Accelerated epigenetic aging was calculated as the residuals from a regression of PhenoAge and GrimAge on chronological age, resulting in measures termed accelerated PhenoAge and accelerated GrimAge. DunedinPACE is a measure of the rate of the change in epigenetic age, indicating the amount of aging per year relative to a year per year of time or chronological age, so it is initially an indication of accelerated or decelerated aging. A confirmatory factor analysis (CFA) was conducted to derive a latent factor representing a composite measure of epigenetic aging, using accelerated DNAm GrimAge, DNAm PhenoAge, and DunedinPACE as indicators. CFA is well-suited to this analysis, as it allows for the estimation of a latent factor that captures the shared variance among multiple measures of epigenetic aging, aligning with the theoretical understanding that these indicators reflect a common underlying, unobserved aging process.⁵¹ The latent factor score, reflecting the common variance derived from the CFA model, was used in our analyses.

Accelerated “expanded biological age”

“Expanded biological age” indicating physiological dysregulation, or the breakdown of normal functioning of multiple physiological systems in the body that occurs with aging, was estimated based on an approach pioneered by Levine,⁵² and then expanded as more biomarkers of aging became available.³¹ This measure combines 22 biomarkers from original biological age,⁵² phenotypic age⁵³ and the Targeting Aging with Metformin (TAME) assays⁵⁴ that represent different aspects of physiological systems with changes linked to aging. Our measure of biological aging includes many markers related to organ aging and systemic biological aging. These biomarkers are integrated into a summary measure to assess how deviations from normal biomarker functioning ranges can collectively indicate greater physiological dysregulation. The systems and the biomarkers used in the expanded biological age are as follows: cardiovascular functioning [systolic blood pressure, N-terminal pro-B-type natriuretic peptide (NT-proBNP, logged)], metabolic functioning [total cholesterol, hemoglobin A1c (HbA1c)], immune load [cytomegalovirus (CMV), white blood cell count (WBC), lymphocyte percentage, CD4 to CD8 ratio], liver function [alkaline phosphatase], kidney function [blood urea nitrogen (BUN), cystatin C)], lung function [peak expiratory flow rate], inflammation [albumin, c-reactive protein (CRP, logged), interleukin-6 (IL6, logged), tumor necrosis factor receptor I (TNFRI), interleukin-10 (IL10), interleukin-1 receptor antagonist (IL1Ra), transforming growth factor beta (TGFB), insulin signaling [insulin-like growth factor 1 (IGF1)], and hematological measures [mean cell volume (MCV), red cell distribution width (RDW)].³¹ Each biomarker’s value for a person is compared with the expected value for someone of their chronological age. The algorithm then adjusts for how closely each biomarker aligns with age. By combining the information from all the biomarkers, the algorithm produces a biological age that should match the average chronological age across the sample. The residual from regressing this summary expanded biological age measure on chronological age is “accelerated expanded biological age” which indicates whether a person is aging physiologically more slowly or more quickly than the rest of the sample.³¹

Covariates

Sociodemographic factors including education (less than high school graduation, high school graduation, some college, and college graduation or higher), race/ethnicity (non-Hispanic White/other, which includes American Indian, Alaskan Native, Asian, Native Hawaiian Pacific Islander, and any other, non-Hispanic Black, and Hispanic), and lifetime smoking pack years were controlled in the analysis. Due to restricted data on specific race categories within the “other” group and its small sample size (n=42 of 1,443), we combined this category with non-Hispanic White.

Analysis

We first show the age-adjusted mean of the composite accelerated epigenetic age factor score and accelerated expanded biological age by the categories reflecting the timing of menopause and hysterectomy. Then, we ran ordinary least squares (OLS) regressions to examine how accelerated epigenetic age and expanded biological age differed for those with normal-aged menopause with hysterectomy, early menopause without hysterectomy and early menopause with hysterectomy, compared to the reference group (normal-aged menopause without hysterectomy), with age, race/ethnicity, education and smoking pack years controlled. The analyses were weighted using the VBS subsample weight for epigenetic analyses. All data preparation and analyses were conducted in SAS 9.4 and RStudio version 2024.09.0.

In our analysis of the HRS data, we have taken several measures to ensure that our findings are valid and reproducible. We carefully selected variables that accurately represent the constructs under investigation, such as age at menopause and hysterectomy status from multiple waves, ensuring they align with established definitions and prior research. We also controlled for potential confounding variables, including education, race/ethnicity, and smoking history, to isolate the associations of early menopause and hysterectomy with accelerated biological aging. For reproducibility, we utilized published expanded biological age³¹ and epigenetic age measures created with HRS and validated by various studies, and we documented thoroughly how we came up with the final analytic sample (Supplemental Fig. 1, Supplemental Digital Content 1, http://links.lww.com/MENO/B366).

RESULTS

Descriptive statistics for the sample are presented in Table 1. About 9.6% of the sample were non-Hispanic Black and another 9.7% were Hispanic. More than half of the sample had some college education or higher. About 30.5% of our sample experienced early menopause, that is, before age 45, and 32.7% had a hysterectomy. There are 4 groups defined by the timing of menopause and hysterectomy: 59.8% of the sample had normal-aged menopause, at age 45 or older, and did not have a hysterectomy; 9.7% experienced menopause at 45 or older and had a hysterectomy after menopause (postmenopausal hysterectomy with normal menopause age); 7.5% experienced early menopause and did not have hysterectomy; and 23.0% experienced early menopause followed by hysterectomy or hysterectomy at an age younger than 45.

TABLE 1.

Sample description (N=1,443)

	Mean (SD), range	%
Age in 2016	68.3 (8.7), 56-90
%Race/ethnicity
Non-Hispanic White/Other a		80.7
Non-Hispanic Black		9.6
Hispanic		9.7
%Education
High school graduation or less		13.1
High school graduation		31.5
Some college		27.4
College graduation or higher		28.0
Smoking pack years	9.7 (18.2), 0-140.1
%Menopause timing/hysterectomy
Normal-aged menopause/no hysterectomy		59.8
Normal-aged menopause/hysterectomy		9.7
Early menopause/no hysterectomy		7.5
Early menopause/hysterectomy		23.0
Epigenetic clocks factor score	0.0 (1.0), -3.4~4.6
Expanded biological age	68.2 (11.9), 44.3-121.0

^a Other categories include American Indian, Alaskan Native, Asian, Native Hawaiian Pacific Islander, and any other.

The rate of early menopause in our sample (30.5%) differs from that of other reports due to our broader definition of early menopause. One study conducted in Olmsted County, MN, examined a random sample of women aged 18-50 years between 1988 and 2007, with follow-up through 2021, and estimated that ∼10% experienced early menopause.⁵⁵ According to the Office on Women’s Health, about 6% (5% early menopause and 1% premature menopause) of women experience early/premature menopause before age 45.^56,57 In these reports, early menopause is defined as natural menopause or bilateral oophorectomy before age 45 while our definition includes any hysterectomy occurring before age 45, including cases with hysterectomy regardless of whether the ovaries were removed. Given that the reported prevalence of bilateral oophorectomy in the United States is about 10.5%⁵⁸ and the prevalence of hysterectomy in our sample is about 32.7%, we calculate that about 22.2% of women in our sample have had hysterectomy without oophorectomy. This broader definition of early menopause, which includes hysterectomy without ovary removal, would lead to an estimated prevalence of early menopause of around 28%, which is consistent with the observed 30.5% in our sample.

Age-adjusted mean accelerated epigenetic and biological age by timing of menopause and hysterectomy

Fig. 1 presents the age-adjusted mean scores for accelerated epigenetic age (epigenetic age factor score) and expanded biological age across groups defined by menopause timing and hysterectomy status. Women who experienced early menopause, both with and without hysterectomy, had accelerated epigenetic aging as indicated by a higher mean epigenetic factor score of 0.26 for early menopause without hysterectomy (P=0.0017) and 0.15 for early menopause with hysterectomy (P=0.0004), compared to those who experienced menopause at a normal age and no hysterectomy (−0.11). This suggests that early menopause, regardless of hysterectomy status, is associated with more advanced epigenetic aging. In terms of biological age, those with early menopause with hysterectomy [or those with hysterectomy in younger ages (before age 45)] had significantly higher accelerated biological age (1.34) compared to those with normal-aged menopause without hysterectomy (−0.83, P=0.0002). This indicates that early menopause and hysterectomy, in combination, may be related to accelerated biological age.

FIG. 1.

Age-adjusted mean epigenetic age factor score and accelerated expanded biological age by menopause timing and hysterectomy group. Hyst indicates hysterectomy; Meno, menopause; Normal-aged Meno/No Hyst, normal-aged menopause and no hysterectomy; Normal-aged Meno/Hyst, normal-aged menopause and hysterectomy after menopause; Early Meno/No Hyst, early menopause and no hysterectomy; Early Meno/Hyst, early menopause and hysterectomy after menopause or hysterectomy before age 45.

Regression of accelerated epigenetic and biological age on timing of menopause and hysterectomy

The multivariate analysis that examined the association between the timing of menopause/hysterectomy and biological aging was conducted with controls for other potentially confounding factors including age, education, race/ethnicity, and smoking (Fig. 2). The regression analysis showed that both hysterectomy and early menopause were related to accelerated biological aging, though their associations differed across epigenetic and physiologically based biological age measures. For epigenetic aging, normal-aged menopause followed by hysterectomy was significantly associated with an increase in the epigenetic clock factor score (b=0.20, P=0.0193), suggesting that a hysterectomy is related to accelerated epigenetic aging, even when it occurs after menopause at a normal age. Early menopause without hysterectomy showed an even stronger association (b=0.28, P=0.0030), highlighting the crucial association of menopause timing with accelerating epigenetic aging. In contrast, no statistically significant association was observed between early menopause with hysterectomy and epigenetic aging. This suggests that the combined effects of early menopause and hysterectomy may differ from their individual associations with epigenetic aging. For biological age, the associations differed slightly. Normal-aged menopause followed by hysterectomy was associated with a statistically significantly higher accelerated biological age (b=1.59, P=0.0269), suggesting a strong physiological association between aging and having a hysterectomy even when performed after menopause. Although early menopause without hysterectomy was not related to biological age acceleration, early menopause with hysterectomy had a significant association with accelerated aging (b=1.35, P=0.0099) relative to normal-aged menopause without hysterectomy. The findings indicate that hysterectomy was associated with accelerated biological age, both independently and in combination with early menopause. These findings on epigenetic age and biological age suggest that both the timing of menopause and hysterectomy are associated with biological aging, but their relationships differ depending on the measure and level at which aging is assessed. The findings emphasize that hysterectomy followed by normal-aged menopause has a strong association with both epigenetic aging and accelerated biological age, while early menopause followed by hysterectomy or early hysterectomy is significantly linked to biological age acceleration but not epigenetic aging, and early menopause without hysterectomy was related to epigenetic aging but not to biological age acceleration. While hysterectomy appears to be associated with accelerated biological age regardless of its timing relative to menopause, the results suggest complex interactions that warrant further investigation into how these factors are related to physiological aging.

FIG. 2.

The ordinary least squares regression coefficients on menopause timing and hysterectomy of epigenetic clock factor score and accelerated expanded biological age. Compared with normal-aged menopause/no hysterectomy. Age, race/ethnicity, education, and smoking (pack year) controlled. Hyst indicates hysterectomy; Meno, menopause; NS, not significant, significance level at P value.

DISCUSSION

This study examined the relationship between the timing of menopause, hysterectomy status, and accelerated aging, as measured by a summary factor score of three epigenetic clocks and expanded biological age, a measure reflecting physiological dysregulation. Our focus was on the association of reproductive histories of postmenopausal women with underlying biological aging, as assessed by molecular, cellular, and physiological dysregulation-based measures. The results emphasize the importance of considering both menopause timing and hysterectomy when evaluating aging trajectories in women. Our findings align with previous research that identifies menopause as significantly associated with accelerated biological aging^8-10 and support the idea that menopause timing and hysterectomy can be independently and interactively associated with biological aging.

Women who experienced normal-aged menopause without hysterectomy demonstrated lower levels of accelerated biological aging, as indicated by negative age-adjusted mean values in bivariate analysis in both epigenetic clock score and expanded biological age. This finding supports prior studies suggesting that later menopause and/or no hysterectomy are associated with a reduced risk of accelerated aging,⁸ potentially due to prolonged exposure to estrogen, which plays a beneficial role in maintaining cardiovascular, metabolic, and cognitive health.^3,5,10

Early menopause and hysterectomy, independently and/or in combination, showed unique associations with accelerated aging measures. However, an important consideration is that our study does not distinguish between hysterectomy with ovarian conservation and hysterectomy with oophorectomy. Because oophorectomy results in an abrupt decline in estrogen and other hormones, its effects on aging may differ significantly from hysterectomy without ovarian removal. The inability to differentiate between these groups limits our ability to determine whether the observed associations are due to hysterectomy itself or to ovarian removal in cases where oophorectomy was performed. Our findings indicate that hysterectomy, whether after normal-aged menopause or combined with early menopause, was significantly related to accelerated biological age based on 22 physiological measures. Upon the constraint on our lack of knowledge on ovarian removal, while alternative explanations, including preexisting health conditions that may have led to the hysterectomy, cannot be completely ruled out, we interpret with caution that hysterectomy appears to be associated with broader, more cumulative differences in accelerated biological age, suggesting that surgical menopause may be linked to systemic aging pathways.

On the other hand, early natural menopause (without hysterectomy) was not significantly linked to expanded biological age acceleration but was to accelerated epigenetic aging. This pattern indicates that the cellular and molecular processes measured by epigenetic clocks may be more sensitive to hormonal changes, while biological age, derived from clinical biomarkers, may capture later, cumulative physiological differences, particularly when surgical intervention is involved. These findings underscore the importance of considering the distinction between natural reproductive processes and a surgical factor when assessing accelerated aging and potential interventions.

Our results indicated that early menopause and/or hysterectomy are linked to higher levels of epigenetic aging. Epigenetic methylation is sensitive to various biological processes, including immune system dynamics.^28,31 Menopause has been associated with changes in immune system profiles, potentially due to related hormonal changes.^29,30 Short-term studies suggest that surgical menopause, characterized by abrupt estrogen deficiency, can result in reductions in B-cell populations and serum interferon-γ (INF-γ) levels, highlighting the regulatory role of estrogen in immune function.⁵⁹ This raises the possibility that long-term estrogen deficiency following hysterectomy may contribute to immune cell composition shifts, though direct evidence in postmenopausal women remains limited. While further research is needed to examine the long-term association of hysterectomy on immune cell compositions and their potential role in epigenetic aging, we ran a preliminary exploratory analysis to examine whether the observed differences among different menopause timing and hysterectomy groups in epigenetic aging could be related to changes in immune cell composition. After controlling for immune cell composition—including CD4⁺ and CD8⁺ T-lymphocytes, natural killer cells, B-lymphocytes, monocytes, and granulocytes—these associations of hysterectomy and/or early menopause, independently or in combination, with epigenetic aging, were reduced or no longer statistically significant (Supplemental Table 1, Supplemental Digital Content 1, http://links.lww.com/MENO/B366). This change implies that immune cell composition is related to the observed associations between menopausal factors and epigenetic aging. The adjustment for cell types suggests that immune cell dynamics, partially influenced by hormonal shifts due to menopausal/hysterectomy status, may be related to accelerated epigenetic aging observed in women with early menopause and hysterectomy. While immune cell composition is an essential component of biological aging, however, we did not include it in our primary model to allow for a broader assessment of how menopause timing and hysterectomy are directly related to epigenetic aging independent of specific immune parameters. Including cell types in the primary analysis could obscure the broader aging associations by isolating specific immune mechanisms, which are inherently linked not only to menopause/hysterectomy but also to other systemic aging processes. These findings highlight the relevance of immune system dynamics in understanding epigenetic aging and menopause, and suggest that future studies should further explore how menopause and hysterectomy are related to accelerated aging in conjunction with immune-related changes.

Because our study does not include direct measurements of hormone levels, we cannot confirm whether hormonal changes mediate the observed associations between hysterectomy and aging and, thus, future studies incorporating hormonal assays would be necessary to validate these hypothesized mechanisms. In addition, while we found associations between hysterectomy and biological aging, we cannot rule out the possibility that preexisting health conditions that led to the hysterectomy may have also contributed to accelerated aging. Women who undergo hysterectomy may have underlying gynecologic or metabolic conditions that predispose them to faster aging, independent of the surgery itself. Future research should explore whether hysterectomy is a causal factor in biological aging or whether it serves as a marker for preexisting health risks.

Our study advances the understanding of biological aging by examining the combined and independent associations of hysterectomy and menopause timing with 2 well-established biological aging measures that capture molecular and cellular aging and physiological dysregulation. While biological aging measures in this study were heavily derived from blood-based biomarkers with one non-blood based marker reflecting systemic physiological processes (systolic blood pressure), these measures reflect a conceptual and methodological breakthrough in aging research in recent years. The use of 2 measures of underlying “biological aging” rather than all physiological phenotypes associated with global aging is regarded as a plus among those who study aging and well-suited for the objectives of this study. Biological and epigenetic age measures are designed to reflect multisystem dysregulation and are seen as measures assessing underlying aging, which is generally upstream from diagnosed diseases, which are, in turn, upstream from disability and loss of functioning. These measures have been validated in prior research, used in clinical trials as reliable indicators of overall aging,^31,60-65 accepted as endpoints in numerous trials of interventions thought to influence aging.^66-68 A major plus of these measures is that they do not depend on an individual’s recall or the use of health services but are measured in the same way for all participants. These measures capture multisystem biological aging and have been related to a variety of major age-related health outcomes including mortality, multimorbidity, disability and cognitive loss.^31,48 Nonetheless, blood-based biomarkers, though highly informative and comprehensive, may not fully reflect localized aging processes in specific tissues such as the brain, bones, or reproductive system. Future research could complement these systemic aging measures with tissue-specific markers, such as neuroimaging for brain aging, bone mineral density for skeletal aging, and hormonal assays for endocrine aging, to provide a more specific indication of how reproductive history influences various aspects of biological aging.

Several limitations should be noted. First, the reliance on self-reported data for menopause and hysterectomy could introduce some degree of recall bias given an average gap of ∼21.6 years (SD=9.9) between menopausal age and reporting age. Some respondents reported a menopausal age older than their age at hysterectomy, suggesting potential misclassification or reporting errors. However, the mean age of menopause in our sample (50.4 y) and its median (50 y) are comparable to the reported average age of menopause in the United States (52 y), suggesting that recall bias might be minimal. Second, because our definition of early menopause is solely based on self-reported age at menopause, we were unable to differentiate between early menopause and primary ovarian insufficiency (POI). While this may introduce some heterogeneity among women with early menopause, the small proportion of POI cases within the early menopause group and the potential exclusion of menstruating POI cases likely minimize its impact on our findings. Future studies with clinical and hormonal data are needed to disentangle these distinct conditions and provide more targeted insights. Furthermore, the cross-sectional epigenetic and biological aging measures in our data limit our ability to establish causal relationships between the timing of menopause, hysterectomy, and accelerated aging. Longitudinal biomarker data would warrant the ability to examine the temporal associations between reproductive events and biological aging. Finally, oophorectomy (removal of ovaries) and differential medical reasons for hysterectomy across menopause timing/hysterectomy groups, which we do not have the information on, may further influence the hormonal changes and aging processes in women who undergo hysterectomy, therefore hysterectomy without ovarian status data should be interpreted with caution. Our study serves as a basis for more comprehensive models that incorporate additional factors such as genetic predisposition, lifestyle choices, and access to health care, which may interact with menopause and hysterectomy to influence aging outcomes.

CONCLUSIONS

Our findings highlight the clinical relevance of assessing both menopause timing and hysterectomy status when evaluating factors influencing the speed of aging in postmenopausal women. Women who experience early menopause or undergo hysterectomy, may face accelerated biological aging over the long term, that is, aging faster than their chronological age, which may predispose them to earlier onset of age-related diseases, such as cardiovascular disease,^1,2 diabetes,⁵ osteoporosis,^3,4 and cognitive decline.^6,7 Recognizing that women’s speed of aging may differ based on their reproductive history may be important for developing personalized care strategies that address these risks. For example, women with early menopause or hysterectomy may benefit from targeted early interventions—such as hormone therapy (HT),^69-71 lifestyle adjustments, and routine monitoring of health markers^72-74 to identify and manage early signs of dysregulation.

Supplementary Material

gme-32-692-s001.docx ^{(129.1KB, docx)}