Estrogen deficiency in the menopause and the role of hormone therapy: integrating the findings of basic science research with clinical trials.

The predominant symptom experienced by both perimenopausal and postmenopausal women is the hot flash. Data from clinical trials indicate that 40% to 80% of postmenopausal women report that they had experienced hot flashes. 20 , 21 Other symptoms experienced by transitional and postmenopausal women include sleep disturbances, mood changes, irritability, increased headaches, “the fog,” and accelerated weight gain. In most postmenopausal women, vasomotor complaints last for at least 1 to 2 years; however, in a significant number of women, the symptoms may last for several years. Hot flashes occur as a result of low or rapidly decreasing levels of endogenous estrogens. It is well established that estrogen replacement is the most effective treatment for hot flashes, and many estrogen products have FDA approval for this indication. 22 The precipitating event causing hot flashes is an abrupt estrogen withdrawal to the hypothalamus. Early evidence supporting this belief was the observation that there was an immediate development in intense and frequent hot flashes resulting from an abrupt loss of estrogen such as following premenopausal oophorectomy. 23 Additional support for the central role of estrogen withdrawal came from the observation that prepubertal girls and adult women with ovarian agenesis, who are not treated with estrogen and have low serum estrogen levels, typically do not experience hot flashes. Once exposed to estrogen replacement, these women, however, will often experience hot flashes following cessation of estrogen administration. 24 Although hot flashes occur concurrently with loss of estrogens, circulating estrogen levels do not correlate between women experiencing hot flashes and those not experiencing hot flashes. 25 The molecular role of estrogens in the initiation of hot flashes and the genes regulated by estrogen to prevent hot flashes is not fully known. The physiologic events of the hot flash include peripheral vasodilation, elevated skin blood flow, and an increase in skin temperature occurring on the face, arms, chest, fingers, back, and legs. The increased skin temperature leads to sweating, often to a fall in core temperature and, in some cases, chills and shivering. 26 Although the hot flash is the most common transitional and menopausal symptom, the other related symptoms are also very common and often result in additional and significant negative impacts on women’s lives. Sleep disturbance often means women do not sleep well and are “chronically tired.” This by it-self affects energy, happiness, attentiveness, work effort, interpersonal relationships, and mood. These symptoms are often further amplified by low circulating estrogen levels in those women. The changes in brain function, often referred to as “the fog,” include less clear thinking and short-term memory problems. This can affect a woman’s home and work productivity and performance. Treatment with estrogen hormone replacement is often associated with the statement “It changed my life.” Urogenital symptoms associated with the transition and the onset of menopause include vaginal dryness, vulvovaginal itching, more frequent urination, more frequent vaginitis and bladder infections, and dyspareunia. These symptoms result from a decrease in vaginal lubrication, vaginal and bladder atrophic changes, and a change in vaginal pH. These changes can be prevented or reversed by continual systemic or local estrogen administration. In the absence of estrogen replacement, later urogenital events include more severe vaginal changes such as atrophy; decreased vaginal connective tissue elasticity; loss of elasticity in the bladder and urethra; loss of urethral mucus; more urinary frequency, urgency, and incontinence; pale and thin vaginal mucosa; and shortening and narrowing of the vagina. As with other long-term estrogen deficiency tissue changes, estrogen replacement works best as a preventative measure and is more effective the earlier it is initiated and continued. Educating women about management of vulvovaginal atrophy and timely treatment can improve the sexual health and quality of life of perimenopausal and postmenopausal women. 27

Estrogen plays a central role in reproductive system functions and is involved in many pathologies impacting the female reproductive organs. As stated earlier, ERs are widely distributed throughout the ovaries and uterus, with an overall dominance of ERα ( Fig. 1 ). In the ovaries, ERβ is generally found in granulosa cells, whereas ERα is more abundant in theca cells. 28 In the uterus, both receptor types are found in epithelial, stromal, and muscle cell nuclei, although ERα is dominant. 28 Local estrogen concentrations in these tissues fluctuate with the menstrual cycle but are significantly higher than serum levels. 29 Therefore, estrogen deficiency can lead to significant changes in reproductive organ physiology and influence the course of reproductive system disorders. The ovaries are the main source of estrogen production prior to menopause and are also impacted by estrogen signaling themselves. In reproductive-aged women, the follicular phase of the menstrual cycle involves a steady rise in estrogen levels, which regulate pituitary secretion of luteinizing hormone and follicle-stimulating hormone (FSH) to induce egg release. Next, the corpus luteum in the luteal phase of the cycle continues to secrete estrogen and progesterone, supporting endometrial stability in anticipation of egg implantation. During the menopause transition, the number of follicles in the ovaries decreases, leading to a compensatory increase in FSH and an eventual decrease in E 2 production. 30 Studies of premature ovarian failure have found that unilateral oophorectomy, nulliparity, and lower body mass index were associated with earlier onset of menopause. 31 This suggests that overall estrogen status may impact ovarian function, although the number of ovulation cycles and follicular reserve may also play a role. One common disease impacting the ovaries is polycystic ovarian syndrome (PCOS), which is a complex disorder associated with anovulation, hormonal imbalances, and metabolic irregularities. Though PCOS typically arises earlier in life, postmenopausal women may continue to experience similar or worsening symptoms of the disease. 32 Notably, while E 2 falls, androgen production persists for many years after menopause. 32 This and other metabolic features of PCOS point to the need for continuous treatment and monitoring of metabolic and cardiovascular issues in postmenopausal women with PCOS. Another ovarian pathology that commonly affects postmenopausal women is the development of ovarian cysts. Although premenopausal and postmenopausal women experience similar rates of ovarian cysts, the postmenopausal group has a significantly higher risk that an ovarian cyst may be malignant. 33 Therefore, these ovarian cysts should be monitored carefully and surgically excised if suspicious in size or other characteristics. Though the impact of HT on ovarian cancer risk has been controversial, a meta-analysis including 12,110 postmenopausal women showed that the current or recent use of HT is significantly associated with an increased risk of serous and endometrioid ovarian cancers, and this risk increases with length of exposure. 34 Another meta-analysis that included more than 2 million women reported an overall increased risk of ovarian cancer and serous ovarian cancer among women using HT, regardless of type and regimen. 35 Within the uterus, the endometrial lining is regulated by estrogen fluctuations, and many endometrial pathologies are directly influenced by estrogen. Prior to menopause, endometrial estrogen levels are elevated during the proliferative phase of the menstrual cycle, causing the tissue to thicken in preparation for blastocyst implantation. After ovulation, progesterone acts to prevent endometrial hyperplasia from unopposed estrogen. ER expression has been shown to vary throughout the menstrual cycle, and increased ERα expression is thought to promote proliferation. 36 In postmenopausal women, ERβ is significantly up-regulated, possibly as a result of androgen exposure in the absence of estrogens. 37 Without sustained estrogen exposure, the postmenopausal uterus becomes atrophied and endometrial thickness decreases substantially. Although the premenopausal endometrium changes constantly, endometrial thickness becomes a useful diagnostic criterion in evaluating for possible malignancy in postmenopausal patients. 38 Many endometrial pathologies may be impacted by the onset of menopause as well as estrogen replacement. Abnormal expression of ERα, ERβ, and GPER1 has been correlated with abnormal endometrial changes, suggesting that these receptors may impact the manifestation of endometriosis, endometrial hyperplasia, and endometrial cancers. 36 Endometriosis is the growth of endometrial tissue outside of the uterus. Endometriotic tissue samples have markedly greater E 2 concentrations than in serum, with ovarian lesions having the greatest E 2 /E 1 ratios. 29 As a result, endometriosis typically regresses with menopause, although lesions may persist in some patients and should be considered for surgical removal due to risk of malignancy. 39 Uterine fibroids are myometrial growths that are similarly dependent on estrogen for growth or to prevent atrophy, and many patients experience symptomatic improvement with the onset of menopause. 40 On the other hand, fibroids can still develop in postmenopausal patients, particularly in patients with alternative sources of estrogen or those with fibroids with increased aromatase expression. 40 In considering the risks and benefits of HT, providers should monitor any changes in uterine fibroid size or symptoms and discontinue treatment if needed. Alternatively, patients with large or symptomatic fibroids can be treated with a lower dose of estrogen, along with adequate progestin options. Lastly, unopposed estrogen plays a direct role in stimulating endometrial hyperplasia and the development of endometrial cancers. Unopposed estrogen exposure is known to increase the risk of hyperplasia in postmenopausal patients. 36 For example, obesity is an increasingly common risk factor for endometrial hyperplasia and polyps, as adipose tissue contributes to peripheral estrogen synthesis and inflammation. 41 Studies show increased ERα and GPER1 expression in endometrial samples with glandular hyperplasia, and particularly complex hyperplasia. 36 Again, progesterone plays a crucial role in preventing endometrial hyperplasia and cancer in patients receiving HT. It is also crucial to monitor patients taking SERMs, as higher doses of tamoxifen have been found to increase ERα activity and endometrial cancer risk. The cervix undergoes structural and functional changes with the onset of menopause. The cervix shrinks in size and the squamocolumnar junction moves into the endocervix, making it difficult to see on a speculum exam. Also, due to lack of estrogen, the volume of cervical discharge decreases substantially. Lack of estrogen can cause atrophy of the vaginal mucosa, leading to vaginal dryness and increased susceptibility to vaginal infection. In vitro studies have shown that paracellular permeability in the cervix decreases after menopause and can be somewhat restored by estrogen administration. 42 On the other hand, aging leads to increased paracellular tight junction resistance, counteracting the effects of estrogen. 42 Another change that has been reported with menopause is the prevalence of cytological abnormalities of the cervix. Misra and coworkers 43 report an increase in squamous intraepithelial lesions (SIL) in the perimenopause stage, suggesting a potential adverse effect of estrogen withdrawal and stressing the importance of continued screening in this population. Estrogen status significantly alters the risk for cardiovascular disease in women and the mechanisms driving this relationship are multifaceted, including adverse effects on myocardial cells, atherosclerosis progression, and the coagulation cascade. An initial observation showed that premenopausal women have lower rates of coronary artery disease and myocardial infarction than men of the same age. 44 Estrogen status has been shown to beneficially impact recovery after cardiovascular injury. Recent publications of the American Heart Association affirm that beneficial outcomes and reductions in all-cause mortality may occur when HT is initiated in women <60 years of age or <10 years since menopause, whereas null or harmful effects may occur when HT is initiated at older ages. 45 Caution should be exercised, and transdermal HT may be considered when estrogen is initiated at older ages. Estrogen has direct beneficial effects on cardiac tissue. Thus, estrogen supplementation in menopause may impact the health of cardiomyocytes themselves after injury. Some studies have reported an increased risk of myocardial infarction in postmenopausal women, particularly after bilateral oophorectomy. 46 Another study showed that women receiving postmenopausal HT had a decreased risk of mortality after MI. 47 In a mouse model of acute myocardial infarction, female mice were found to have better short- and long-term outcomes than males. More specifically, females showed decreased rates of cardiac rupture and maladaptive chronic changes, such as left ventricular dilation. Similarly, in humans, women experience less cardiac remodeling after an acute ischemic event, although their risk of heart failure may be greater overall. 48 Sex-specific differences in the cardiac response to acute ischemia may be driven by estrogen signaling. ERα and ERβ are widely expressed in cardiac myocytes and actively regulate cellular functions such as gene expression, metabolism, and apoptosis. 49 In addition to systemic estrogen, aromatase expression in epicardial adipose tissue allows for local estrogen synthesis from androgens. 49 In studies using ER KO mice, ERα activity has been linked to cardiac growth. 50 Moreover, genetic polymorphisms of the ESR1 and ESR2 gene have been linked with cardiovascular disease risk as well as left ventricular structure and mass. 51 , 52 This suggests that estrogen deficiency in menopause may have an impact on cardiac cell metabolism through changes in ER activity in the heart. In the setting of hypoxic injury, ER activity impacts cardiac tissue response in a pro-survival manner. Increased ERα activity leads to the inhibition of lipopolysaccharide induced tumor necrosis factor (TNF) expression and apoptosis, facilitating the survival of cardiac myocytes. 53 ERβ signaling has also been shown to prevent cell death through Akt phosphorylation and Bcl-2 expression. Although hypoxia typically induces apoptosis in rat ventricular myocytes, ERβ overexpression led to the total suppression of autophagy proteins and mitochondrial apoptotic proteins. 54 Overall, these findings suggest that estrogen directly affects the heart both in a cardioprotective manner and in response to injury. For this reason, women may benefit from cardiac tissue estrogen exposure, and this is particularly pronounced in the event of ischemic injury. Another critical way estrogen impacts heart health is through its beneficial effects on circulating lipids and effects on preventing or decreasing atherosclerosis progression. Estrogen administered orally is associated with an increase in sex hormone–binding globulin, a decrease in low-density lipoprotein cholesterol, and increased high-density lipoprotein cholesterol. 55 But much of the debate around HT in postmenopausal women has centered around its efficacy in preventing the progression of atherosclerosis if initiated within the first 10 years of the menopause, or the menopause transition. The Early versus Late Intervention Trial with Estradiol (ELITE) demonstrated that HT led to decreased progression of atherosclerosis in early menopausal women but did not have an effect on late menopausal women. 56 In that study, atherosclerotic change was quantified through carotid intima-media thickness (CIMT) measurements. Posttrial analyses show that increased E 2 levels were associated with lower CIMT progression in early menopause but had no effect on CIMT progression if initiated in late menopause. 57 These observations appear to be related to the degree of atherosclerosis present prior to the initiation of HT. In mice, estrogen supplementation was found to prevent the formation of new atherosclerotic lesions but had little effect on the progression of established lesions or plaque stability in terms of intraplaque hemorrhage or medial erosion. 58 Another possible explanation is the variability of coronary arterial ER expression. Early studies of postmortem coronary artery samples showed that postmenopausal women not on HT have significantly lower ER expression than premenopausal women. 59 Furthermore, in samples collected from both women and men undergoing coronary artery bypass grafting, ERα promoter methylation was greater in coronary atherosclerotic plaques than in the normal proximal aorta or the internal mammary arteries. 60 On the other hand, ERβ is more widely expressed in the coronary arteries and has been positively associated with calcium content and plaque area in premenopausal and postmenopausal women. In one study, postmenopausal women receiving HT were found to have decreased ERβ expression, as well as decreased plaque area and calcification, though timing of HT initiation was not reported. 61 One major concern regarding HT is the effect of estrogen on the coagulation system. Estrogen given orally, particularly using combination oral contraceptives, is known to have various negative effects on procoagulant and anticoagulant factors with resultant net increases in venous thrombosis risk. This is particularly important for postmenopausal women who are more than 10 years postmenopausal and have never been on HT. This population is already at a heightened risk of venous thrombosis due to generally decreased mobility, greater prevalence of other medical conditions, and progressive atherosclerosis in the menopause transition without HT. 56 Thus, it is crucial to understand that protection from cardiovascular disease is limited to those women with minimal atherosclerosis having started estrogen therapy early during the menopause transition. In the REPLENISH trial, early and late menopausal women were started on different doses of oral E 2 and progesterone, and various coagulation metrics were obtained at multiple time points, including prothrombin time, activated partial thromboplastin time, fibrinogen, antithrombin, protein C, and protein S. 62 Higher doses of oral estrogen were significantly associated with decreased anticoagulant and increased procoagulant metrics. These effects were more pronounced in late menopausal women, suggesting that initiating HT later may also increase clotting risk through greater changes in coagulation risk with aging. 62 Some studies have suggested that the addition of a progestin may further impact the risk of thrombosis; however, others show no significant difference. The route of estrogen delivery appears to play an important role, as many studies report that transdermal estrogen does not lead to thrombosis risk as compared to oral estrogen that does increase clotting risk. 63 Both basic science studies and clinical research trials have demonstrated a significant and beneficial impact of estrogen replacement on brain function and cognition in postmenopausal women. Multiple neurological functions are dependent and functionally impacted by estrogen signaling. In many scientific studies, low estrogen status has been associated with adverse changes in brain structure, function, ER distribution, and the expression of various neuroprotective and inflammatory factors. 64 – 70 These changes are significant because they help to explain neurological processes occurring during menopause and aging, as well as the pathophysiology of common neurodegenerative diseases and psychiatric disorders. Furthering our understanding of these processes can help drive the development of new therapies and clinical recommendations. Imaging studies suggest that estrogen deficiency is associated with large-scale negative changes in brain structure and activity. 64 In comparison to age-matched males, females undergoing menopause showed significantly decreased gray matter volume in several cortical and subcortical regions. Notably, postmenopausal women had increased gray matter volume of the precuneus, which is a region involved in information processing and episodic memory retrieval. This increase was positively correlated with memory score, indicating that it may play some compensatory role. Perimenopausal and postmenopausal women also displayed decreases in white matter volume, although they showed higher measures of white matter integrity and connectivity in regions such as the corona radiata. Furthermore, other studies have found that HT significantly alters cortical blood flow and activation patterns throughout the brain. 65 Women who had received estrogen replacement showed increased activation of multiple brain regions, including the frontocingulate, medial temporal, and posterior parietal areas, during various cognitive tasks. Some studies focused on young women undergoing pharmacologically induced menopause. 66 This allowed researchers to isolate the effects of estrogen deficiency from those related to aging. They found that following gonadotropin-releasing hormone agonist treatment, these young women showed decreased activation of the left inferior frontal gyrus, which was reversible after restoration of physiologic estrogen production. Cerebral ischemia is a common cause of neuronal injury that is thought to be influenced by estrogen signaling. Zhang and coworkers 67 demonstrated that ovariectomized rats were more susceptible to ischemic damage of the CA-3 region of the hippocampus, which led to long-term cognitive deficits. Estrogen replacement at the time of ovariectomy, but not after 10 weeks, effectively prevented these effects, supporting the idea of a “critical period” for estrogen replacement. Other studies have also corroborated the neuroprotective effects of estrogen after ischemic injury, showing that ERα expression increases immediately after middle cerebral artery occlusion during the period immediately following ischemic brain injury. 68 This is thought to be partially mediated by interactions between estrogen and glial cells. 69 , 70 E 2 promotes remyelination through the regulation of Schwann cells and oligodendrocytes, and SERMs have also been shown to reduce axonal damage and demyelination. Additionally, E 2 acts on astrocytes to modulate the production of crucial signaling molecules in response to stress. For example, E 2 administration decreased the glial fibrillary acidic protein (GFAP) and vimentin expression typically seen in reactive gliosis after certain types of injury. Astrocyte heat shock protein production and glutamate uptake were also affected. Over the past few decades, estrogen deficiency has been implicated in various neurodegenerative disease processes, including Alzheimer’s disease (AD), Parkinson’s disease (PD), and other forms of dementia. Many of these diseases have sex-specific presentations and incidence rates; for instance, AD risk is greater in women and increases significantly after menopause. 71 Some studies have found that early surgical menopause and even unilateral oophorectomy are associated with greater risk of dementia, cognitive decline, and Alzheimer’s neuropathology later in life. 68 , 71 This effect may be mediated by hippocampal ERα expression, which has been shown to be decreased in patients with AD. 72 NF-κB signaling may also play a role, as one study found that in mice, E 2 inhibited NF-κB signaling and prevented Aβ accumulation and memory impairment. 73 Although many studies support the idea that estrogen replacement is protective against the development of dementia, there is one main consideration. The Women’s Health Initiative Memory Study showed that when women were started on a high dose of oral estrogen/progestin therapy after the age of 65 yr of age, they had a significantly increased risk of probable dementia, likely a result of clotting issues. 74 Imaging results after the study concluded showed greater brain atrophy in participants with dementia, particularly in the frontal lobe and hippocampus. Because these results were reported, many studies have suggested that, similar to effects of estrogen in progression of atherosclerosis, the age of initiation and route of administration of HT are crucial. Importantly, only women reporting initiating oral HT in later life seem to have an increased risk of dementia. 74 Although the majority of evidence supports the idea that estrogen has some neuroprotective effect against cognitive decline, more research is needed before formal clinical recommendations can be made. The other disease process that has been linked to estrogen deficiency is PD, which is known to be more prevalent in men and postmenopausal women. 75 Human and animal models of PD have suggested that estrogen may protect against neurological changes and disease progression. Multiple mechanisms of protection have been explored, including effects on dopamine (DA) signaling pathways, glial inflammation, and the renin-angiotensin system. Many studies in the past few decades have reported a pro-DA effect of estrogen in animal models of nigrostriatal degeneration. 75 , 76 ERs are now known to be closely associated with DA neurons, 77 and estrogen has an effect on multiple aspects of DA signaling, including DA receptors, DA transporters, and monoamine oxidase activity. 76 Additionally, more research has begun to focus on the role of glial cells on the neuroinflammatory process in PD. As discussed previously, estrogen is a major regulator of glial cell activity, and this has been shown to directly influence the survival of DA neurons. 69 , 78 Lastly, estrogen deficiency leads to increased activation of the brain’s renin-angiotensin system, which is thought to increase oxidative stress and DA neuron degeneration. 79 In one experiment, both estrogen replacement during a critical period and angiotensin antagonist administration facilitated neuroprotection. Recent research has also begun to link estrogen with psychiatric diseases. States of hormonal imbalance such as PCOS, the postpartum phase, and perimenopause have been associated with increased risk of depression and anxiety. ERs are found widely throughout the limbic system, impacting serotonin, and norepinephrine signaling pathways, 80 which are currently major targets for the treatment of mood disorders. In genetic animal models of depression, 5-hydroxytryptophan (5-HT) imbalance was largely reversed by estrogen. 80 E 2 supplementation has been trialed as a potential treatment for depressive symptoms in perimenopausal women as well, showing some sustained symptom improvement. 81 Estrogen deficiency has also been strongly linked to the development of schizophrenia and other psychotic disorders. It has long been described that women have a decreased risk of schizophrenia and typically present with different symptoms, later onset, and have lower disease severity than males. In women, the peak age of diagnosis coincides with the start of perimenopause (age 45–50 yr) and the age of menarche may even impact symptom severity. 82 For this reason, estrogen has been studied as a potentially protective factor. Trials of estrogen treatment in combination with antipsychotics have shown significantly improved management of positive symptoms. One major problem that begins with the onset of perimenopause and continues for the duration of menopause is a progressive loss of bone mineral density. Women have a significantly greater risk of osteoporosis than men, and estrogen deficiency is known to be directly associated with this disease process. Studies show that earlier age of menopause leads to lower bone density later in life, and complete surgical menopause is associated with an even more significant decline in bone density. 83 Bone mineral density of the spine declines most rapidly in the first 3 years surrounding the menopause transition, coinciding with changes in FSH and E 2 , and then continues to decrease steadily in the years after. 84 Trabecular bone is initially lost more rapidly due to its greater surface area to volume ratio, though the absolute volume of cortical bone in the body is larger. 85 Although trabecular bone loss slows after the first decade of menopause, cortical porosity continues to increase. These changes are clinically relevant due to the greater fracture risk in patients with osteoporosis that is further associated with higher mortality rates and decreased quality of life. 84 Other comorbidities such as chronic inflammatory disease, diabetes, and obesity additionally impact fracture risk and should be taken into consideration. Bone remodeling occurs continuously throughout life as regulated by the balance between osteoclasts, which facilitate bone resorption, and osteoblasts, which facilitate deposition of bone. 86 Unfortunately, estrogen deficiency during the perimenopause and then menopause leads to accelerated osteoclast activity and subsequent large increases in the rate of bone resorption. 85 Osteoblast deposition rates are unable to increase enough to offset bone resorption rates, thereby resulting in bone loss during the perimenopause transition, and menopause, if no estrogen replacement is initiated. ERs are found widely in osteoblasts, osteoclasts, and osteocytes, with greater ERα expression in cortical bone and ERβ in trabecular bone. In general, ERα and ERβ have a wide variety of actions and have been shown to have some antagonistic effects in bone. 87 For example, some mouse models suggest that ERβ KOs tend to have longer femur length than wild type, whereas ERα KOs have shorter femur length. Additionally, ERβ KO mice have a greater estrogenic response to mechanical bone loading, whereas ERα KOs have a diminished response. Estrogen deprivation is linked with increased osteoclast activity, as estrogen signaling leads to downstream osteoclast apoptosis. 86 One point of regulation is FasL, a signaling molecule produced by osteoblasts. E 2 up-regulates MMP3 via ERα, which cleaves full-length FasL into its soluble form, and then causes the apoptosis of osteoclast progenitors. 86 , 88 Ovariectomy and menopause are associated with decreased FasL activation and thus increased osteoclast proliferation. 89 One study found that IFN-gamma and TNF-α act synergistically to down-regulate FasL production in osteoblast progenitors. Administering neutralizing antibodies to either IFN-gamma or TNF-α resulted in increased bone density in ovariectomized mice. 90 This is consistent with the idea that estrogen suppresses the production of proinflammatory cytokines such as IL-1, IL-6, and TNF-α, many of which are regulated by transcription factor NF-κB. 91 , 92 Thus, NF-κB inhibitors are currently being explored as a potential therapy for menopause related osteoporosis. 93 , 94 Another pathway that regulates osteocyte maturation is the RANK/RANKL system. RANKL is produced by osteoblasts and binds to RANK, which then activates multiple pathways that promote osteoclast differentiation and proliferation. One downstream target of RANKL is transcription factor c-Jun, which promotes osteoclast development and is effectively blocked by E 2 as well as the SERMs tamoxifen and raloxifene. 95 Osteoprotegerin acts as a decoy receptor for RANKL, preventing it from binding to RANK, and is regulated by various signaling molecules, including hormones and cytokines. 86 In postmenopausal women, serum E 2 has been positively associated with osteoprotegerin levels. 86 , 96 Identification of the role that the RANK/RANKL system has on bone remodeling and resorption resulted in the introduction of a now popular drug used for treatment of osteoporosis. Denosumab is a medication that prevents RANKL from activating its receptor, RANK, that is located on the surface of osteoclasts. Additionally, estrogen has direct effects on osteoblasts. Studies show that estrogen supplementation can prevent osteoblast apoptosis through up-regulation of Bcl-2 97 and inhibition of apoptotic gene expression typically induced by TNF-α, such as ITPR1, which helps regulate calcium and apoptosis. 98 E 2 also promotes autophagy, which is a metabolic process crucial to cell survival, through ER-ERK-mTOR signaling 99 and up-regulation of RAB3GAP1. 100 Gavali and coworkers 100 found that RAB3GAP1 knockdown osteoblasts showed impaired mineralization and increased apoptosis even with nutrient-rich environments and estrogen supplementation. Conjugated equine estrogens, E 2 , and some SERMs are currently approved treatment options for preventing postmenopausal bone loss and osteoporosis. As previously mentioned, some experiments involving SERMs show that they potentially mimic the effects of estrogen on bone. 89 , 95 One study of tamoxifen and raloxifene showed that the two drugs were equally effective in fracture risk reduction. 101 The decision to try these medications is situation-dependent, and clinicians should weigh the risks and benefits. For example, raloxifene can be particularly helpful for women with increased risk of vertebral fracture and history of breast cancer. Lifestyle modifications such as smoking cessation, adequate calcium and vitamin D intake, and increased weight-bearing activity can help to counter bone loss, especially when used in combination with HT, started early during the menopause transition. Both ERα and ERβ have been identified in skin components including epidermal keratinocytes, dermal fibroblasts, and hair follicles. 102 The presence of ERs in skin serves as a molecular explanation for changes of skin condition after menopause when estrogen levels decline. 103 Aging skin has several structural changes such as thinning, dryness, scaliness, wrinkling, loss of elasticity, and pigmentation. In addition, there are also functional changes such as reduced barrier function, slow wound healing, reduced immunological responsiveness, and reduced thermoregulatory ability. 104 The process of skin aging involves both intrinsic (decreased epidermal proliferation, less dermal vasculature, reduced collagen content) and extrinsic (environmental exposure to light and chemicals) factors. 105 Estrogen deficiency during the menopause has been shown to be correlated with deterioration of skin extracellular matrix leading to overall structural skin aging. 106 During the early years after menopause, there is a 30% reduction of skin structural proteins such as collagen, 107 and in later years, the reduction of collagen content and decreased skin thickness continues to occur progressively. 108 Several studies reported on the effect of HT on improving skin aging in menopausal women by increasing skin thickness, preventing the loss of collagen, preventing wrinkles, and restoring skin hydration. 104 , 109 Although significant beneficial effects of HT on skin in menopausal women were reported in some observational studies, 110 – 115 the results could be subjected to selection bias due to the higher doses and longer duration of use. In contrast, the Kronos Early Estrogen Prevention Study (KEEPS), a 5-year, multicenter, double-blind, randomized placebo-controlled trial, reported significant changes in facial skin wrinkles or rigidity with race/ethnicity but not from use of systemic HT. 116 The nonsignificant effect of HT could be due to limited power, and/or administration of a relatively low HT dose. As the effect of low-dose systemic HT on skin is not well established, an alternative of local administration of other hormonal and cosmeceutical agents such as SERMs, phytoestrogens, and resveratrol has been demonstrated to improve estrogen deficiency in skin. 109 Postmenopausal women experience changes in immune processes in response to both aging and estrogen withdrawal that raise unique health concerns for this group. 117 Estrogen modulates immune function through several mechanisms, involving both the innate and adaptive immune systems. At the same time, immune dysfunction may contribute to changes in estrogen production. ERα and ERβ are found on immune cells, and cytokine receptors are found on estrogen-producing cells. 118 These pathways help to explain the sexually dimorphic expression of autoimmune disorders. 119 Sex hormone fluctuations have also been found to modulate the immune system in nonpathological states such as puberty and menopause. 117 These observations suggest that postmenopausal estrogen supplementation may have a significant beneficial effect on immune function. Studies have found that HT may influence B- and T-cell function, complement activity, and cytokine levels. 120 During normal development, immune system activity changes during the onset of puberty and during times of cyclical hormone fluctuations. Lamason and coworkers 121 categorized splenic gene expression in a mouse model in relation to puberty onset. They found postpubertal sex-specific differences, with males generally expressing more innate immunity genes and females more adaptive immunity genes, which was reflected by greater serum immunoglobulin levels in females. Immunoglobulin activity is mediated by the Fas/FasL pathway, which estrogen is known to modulate. 88 , 121 More precise changes in immune function have been mapped throughout the menstrual cycle. 117 For example, Faas and coworkers 122 found an increase in serum granulocyte, monocyte, and lymphocyte counts during the luteal phase, which is associated with greater estrogen production. Moreover, they concluded that the luteal phase is associated with a greater Th2 response, which is consistent with previous literature. 118 Local immune activity along the reproductive tract has also been found to be regulated by estrogen and progesterone levels, which may be crucial for the process of oocyte release or tolerance to implantation. 123 Sex-specific differences in autoimmune disease presentation provide additional information about the estrogen-immune system linkage. Overall, women are known to have a greater incidence of autoimmune conditions, including rheumatoid arthritis and systemic lupus erythematosus (SLE). Patients with rheumatoid arthritis produce abnormal levels of sex hormones such as estrogen as a result of cytokine regulation of aromatase activity. 119 SLE patients have been found to have abnormal hormone variations throughout pregnancy. 119 In postmenopausal patients with rheumatoid arthritis, HT has positive effects not only on prevention of bone loss and development of osteoporosis but also on immune-mediated inflammation. The normal aging process is associated with immunosenescence, during which the immune response to infection decreases and the body becomes more susceptible to some infectious agents. In females, menopause typically occurs around the same time. To separate the effects of estrogen deprivation and immunosenescence, Kumru and coworkers 124 studied the immune function of healthy perimenopausal women undergoing hysterectomy and bilateral oophorectomy. Following the surgery, CD8 cells were increased, whereas CD19 cells, IL-4, and IFN-γ concentrations were decreased. This was reversed after transdermal estrogen administration, suggesting that HT may help restore some aspects of immune function after menopause. 124 As with many other systems of the body, the digestive system has the capacity to interact with and respond to estrogen. ERα, ERβ, and GPER1 are found widely throughout the gastrointestinal (GI) tract. 125 Estrogen plays a role in normal GI function and has also been implicated in various GI diseases, including gastric and intestinal cancers, inflammatory bowel disease, and gastroesophageal reflux disease (GERD). 125 , 126 Furthermore, research also shows that estrogen has an effect on the composition of the gut microbiome, whereas the microbiome may affect estrogen levels as well. 127 , 128 Thus, it is crucial to consider the effects of estrogen on the digestive system when considering estrogen deficiency and supplementation. Overall, estrogen influences the composition and function of the gastrointestinal barrier and also impacts inflammatory processes within the digestive system. ERβ signaling has been shown to regulate the permeability of the intestinal barrier by increasing the integrity of tight junctions through the expression of occludin and junctional adhesion molecule A. 126 This impacts multiple areas of the GI tract, including the esophageal and colonic epithelia. The mucosal barrier is also an important layer of protection that helps to counteract the effects of digestive acids. In females, E 2 and the phytoestrogen, genistein, have been shown to increase duodenal mucosal bicarbonate secretion, which neutralizes gastric acids and prevents ulcer development. 126 , 129 Lastly, as with many other systems, estrogen signaling plays a significant role in regulating inflammatory processes. Although it is generally thought to have anti-inflammatory actions, the effects of different ERs are nuanced. 129 These functions are all crucial for the health of the GI tract and disease prevention. Estrogen has been implicated in common chronic GI disorders, including GERD and inflammatory bowel disease. GERD symptom severity is greater in males than females, with increased presentation in postmenopausal females. 125 , 130 This suggests that estrogen mediates the effects of gastric acid on mucosal and epithelial integrity. On the other hand, GERD symptoms appear in 30% to 80% of pregnant patients, despite the significant increase in estrogen levels. 131 Postmenopausal estrogen supplementation was also found to increase the risk of GERD in a cross-sectional study. One possible explanation for these findings is that estrogen induces relaxation of the lower esophageal sphincter through nitric oxide production. Overall, although estrogen may contribute to GERD onset through esophageal sphincter relaxation, it also protects against esophageal damage and reduces complications related to GERD such as esophagitis and metaplasia. 130 Inflammatory bowel diseases have varying sex-specific differences; ulcerative colitis is more common in males, whereas Crohn’s disease is more common in females. Although some research has shown a correlation between estrogen supplementation and ulcerative colitis or Crohn’s disease, the results have not been consistent across studies. 126 In multiple experimental models of colitis, estrogen has been shown to decrease intestinal inflammation and tissue damage. 126 Further studies have explored the relationship between the distributions of different ERs and inflammatory bowel disease. Pierdominici and coworkers 132 found that ERβ expression was decreased in peripheral T cells and colonic mucosa of patients with active disease. Moreover, plasma IL-6 was significantly inversely correlated with ERβ expression. Another study found abnormal levels of ERα, ERβ, and GPER1 in both males and females with inflammatory bowel disease, with an overall increase in ERα and GPER1. 133 Significant age- and sex-specific differences were also found, indicating that estrogen status has varied actions depending on these factors. Next, estrogen appears to play a protective role in the development of multiple GI cancers. Esophageal adenocarcinoma is commonly seen in the United States and disproportionally affects males, though prevalence increases in females after age 80. 125 As discussed previously, estrogen supplementation protects the esophageal epithelium from gastric acid damage and may lower esophageal adenocarcinoma risk. 126 Some studies also show that E 2 inhibits the proliferation of cancer cells in both squamous cell carcinoma and adenocarcinoma. 125 , 126 Similar to esophageal cancers, gastric cancers are also more prevalent in males and postmenopausal females, and various ER subtypes have been implicated in the disease process. 125 Multiple studies suggest that increased cancer cell ERα expression and decreased ERβ are correlated with poorer prognostic factors in gastric cancers, including metastasis and differentiation. 126 , 134 Regarding the role of estrogen in treatment, Lindblad and coworkers 135 reported a significant decrease in gastric cancer risk in men who had received estrogen for prostate cancer treatment. Qin and coworkers 136 found that in vitro exposure to estrogen induced apoptosis in gastric cancer cell lines. Another common GI cancer that estrogen has been implicated in is colorectal cancer, which again presents more commonly in men and postmenopausal women. 126 Estrogen is generally thought to have a protective role against colorectal cancer through ERβ-mediated pathways. 126 One aspect of the digestive system that has received more attention over the past few decades is the gut microbiome. There are sex-specific differences in microbiome composition that emerge during puberty and are further affected by changes in estrogen status, such as menopause or pregnancy. 127 In general, women who undergo menopause have been shown to have decreased microbiome diversity with an increase in similarity to the male microbiome. 127 Wu and coworkers 137 examined the gut microbiomes of women with premature ovarian insufficiency and found no changes in microbiome diversity, although they found compositional changes such as a decrease in Firmicutes and an increase in Bacteroidetes. While research is ongoing, some studies show similar trends in Firmicutes and Bacteroidetes, as well as other taxa, suggesting that microbiome changes due to estrogen deficiency may be impacting steroid hormone metabolism and other health factors. 127 Not only does estrogen status impact the microbiome, but residential microbes in the gut also impact estrogen levels. The “estrobolome” refers to the collection of bacterial genes that allow microbiota to metabolize estrogen and other steroids. 138 Many species produce β-glucuronidase, ß-glucosidase, and sulfatase enzymes, which deconjugate estrogen and extend its time in circulation. These changes in estrogen metabolism may have important implications for estrogen sensitive diseases, such as hormone receptor (HR)–positive breast cancer. 128 HR-positive breast cancer risk has been shown to decrease with increased ratios of certain estrogen metabolites of parent estrogens. Fuhrman and coworkers 139 found that greater fecal microbial diversity among 60 postmenopausal women was associated with an overall greater metabolite-to-parent estrogen ratio, suggesting lower HR positive breast cancer risk as well. Further research is needed to identify the roles of specific taxa and taxa interactions in modulating estrogen. Furthermore, factors such as diet, alcohol consumption, and adiposity must be considered due to their effects on both the microbiome and steroid hormone metabolism.

All three endogenous estrogens, as well as various synthetic ligands, including selective ER modulators (SERMs), bind to the ERs and mediate various downstream physiological processes. There are two primary types of ERs, ERα and ERβ, encoded by different genes on different chromosomes. ERα is encoded by the ESR1 gene on chromosome 6, whereas ERβ is encoded by the ESR2 gene on chromosome 14. 4 ERα and ERβ share a common structure typical of nuclear receptors, including DNA-binding, ligand-binding, and N-terminal and C-terminal domains. The two vary predominantly in their N-terminal domain and ligand-binding domain resulting in a smaller and narrower binding pocket in ERβ compared to ERα. 5 , 6 More recently, a third type of estrogen receptor has gained research interest: the G protein–coupled estrogen receptor 1 (GPER1, previously GPR30). 7 Unlike ERα and ERβ, GPER1 is a plasma membrane receptor with seven transmembrane domains that allows for rapid activation of downstream signaling cascades. GPER1 is expressed in many organ systems, including the reproductive organs, heart, brain, and skeletal muscles. 7 Some studies have found that GPER1 expression in some areas of the body vary with reproductive hormone fluctuations. The physiological effects of GPER1 activity are still under study; however, early studies involving GPER1 knockout (KO) mice have shown that this receptor may play a role in multiple physiological functions. 8 GPER1 KO mice not only maintain normal reproductive function but also display significant metabolic, cardiovascular, neurological, and immunological changes. 8 , 9 Although we focus on classical ERs throughout our review, we also note significant findings related to GPER1. Both ERα and ERβ are expressed widely as well, modulating a broad array of physiological functions in the reproductive, skeletal, cardiovascular, and central nervous systems. ERα is exclusively expressed in the mammary glands and liver and is the predominant receptor in the ovarian thecal cells, uterus, brain, heart, and bone ( Fig. 1 , red). 6 ERα plays a significant role in sexual and breast development, as well as bone health and metabolic regulation of cholesterol in the liver. Generally, ERα promotes cell proliferation and growth and has been implicated in cancer pathology. Conversely, ERβ is exclusively expressed in the lungs, adrenal glands, kidneys, colon, fallopian tubes, and bladder, and is the predominant receptor in ovarian granulosa cells, and the cardiovascular and central nervous systems ( Fig. 1 , blue). 6 ERβ provides cardiovascular protection, neuroprotection, and immune regulation of inflammation. In addition, ERβ has antiproliferative effects that counterbalance those of ERα. E 2 can bind both receptors with a slightly higher affinity for ERα. 10 E 1 and E 3 , the weaker and weakest estrogens, respectively, can also bind both receptors. SERMs target both receptor subtypes but display tissue-specific agonist and antagonist activities independent of ER subtypes. Much attention has been given to the development of ERβ agonists, which would minimize undesired proliferative changes in the breast and uterus mediated by ERα. In addition to the changes in secretion levels of the different types of estrogen during the transition from reproductive years through menopause and into postmenopause, there are also changes in expression and activity of ERs. During reproductive years, there is high receptor activity, particularly of ERα in the reproductive tract, breasts, and bones, responding to fluctuating levels of estrogen during menstrual cycles. As estrogen levels decline in menopause, studies report a decrease in sensitivity and number of ERα in certain tissues. 11 With consistently low estrogen levels postmenopause, there may be a shift in relative expression of ERα and ERβ; however, there are very few studies evaluating the actual changes in ERα and ERβ expression in different tissues in the premenopausal and postmenopausal years. One study investigating the abundance of ERs in premenopausal and postmenopausal women found a significant decline in levels of both receptor types measured in the skin. 12 In the hippocampus, which is crucial for memory and cognition, decreased E 2 during postmenopause has been associated with a subsequent decrease in ERα activation, leading to long-term reduction in ERα expression and downstream signaling. 13 In animal studies evaluating ERα and ERβ mRNA expression in different parts of the brain of young, middle-aged, and old female rats, old rats were found to have significantly decreased ERβ in the cerebral cortex and supraoptic nucleus and significantly decreased ERα in the periventricular preoptic nucleus. 14 Another study evaluating ERα and ERβ protein expression in the hippocampus of normal adult versus aged female rats reported a significant decrease in staining of both ERα and ERβ in CA1 and CA3 hippocampal neurons in the aged rat hippocampus. 15 To our knowledge, there are no studies directly comparing changes in ER and ER subtype levels in postmenopausal women receiving HT versus postmenopausal women not receiving HT. Such a study would provide valuable insight on the change in the distribution and biology of ER following HT. Another important question is whether or not we can disaggregate between endocrine-mediated and aging-mediated causes of the physiological changes typically associated with menopause. 16 In other words, are some aspects of menopause part of the normal aging process independent of estrogen deficiency? There are no good controls to parse through aging and menopause because the two processes are intricately linked. The best way to explore this question would be through longitudinal prospective studies evaluating changes in levels and distribution of estrogen and ERs over time (spanning reproductive years, perimenopause, and several time points postmenopause) in women receiving and not receiving postmenopausal HT. Such a study would necessitate large cohorts and would not only give insight on the changing biology of ER following HT, but could also elucidate the question of how well the physiological changes associated with aging can be effectively mediated with use of HT. 16

ERs can be regulated via genomic and nongenomic mechanisms. Genomic or classical mechanisms refer to the slow estrogen signaling pathway, which takes place over several hours or days. In this pathway, estrogen binds to the ER, which results in ER dimerization. The activated dimerized ER enters the nucleus and binds estrogen response elements (ERE) regulating transcription activity of specific genes ( Fig. 2 , left panel). 17 Activated ER can regulate transcription and inhibition of transcripts independent of ERE. Alternatively, ER can undergo ligand-independent activation. In this case, growth factors can activate the protein kinase cascade leading to the phosphorylation and activation of ER, which can in turn bind EREs and regulate gene transcription ( Fig. 2 , left panel). 17 Nongenomic mechanisms involve rapid estrogen signaling, taking place within seconds or minutes, via membrane-associated ERs or other cell surface receptors that can activate rapid downstream pathways including kinase signaling 17 ( Fig. 2 , right panel). To date, there are no explicit studies evaluating longitudinal changes in down-stream mechanisms by which ERs regulate physiological functions before and after menopause. Rather, existing studies have focused on evaluating downstream mechanisms of ERs in either premenopausal or postmenopausal states in relation to other aberrant physiological functions including cardiovascular disease and lipid metabolism. 18 , 19

In this conventional review, we report information from research studies and review articles published between 1972 and 2023, with the majority of literature produced in the 2000s and 2010s. We performed a literature search on PubMed using a list of search terms including “estrogen,” “estrogen deficiency,” and “menopause,” along with key terms in each organ system.

Menopause is a natural and inevitable part of aging for women. As women approach perimenopause and menopause, the dramatic fluctuations and eventual substantial decline in circulating estrogen levels affect almost all organ systems, including urogenital, reproductive, cardiovascular, neurologic, psychiatric, skeletal, dermatologic, immune, and digestive systems. These symptoms, first acute then chronic, have a profound and overall negative impact on the quality of life of women, and most symptoms can be alleviated with estrogen therapy. It is clear from basic science literature that ERs are found throughout the body and that they are associated with uncountable beneficial actions in both women and men, even as they age. In this review, we integrate understanding of basic science with the latest findings of clinical trials to offer an accurate representation of the risks and benefits of estrogen replacement. Education of these findings and an accurate representation of the risks and benefits of estrogen replacement to our aging patients is fundamental to improving their care, quality and even quantity of life.