Postmenopausal Hormone Replacement Therapy for Health Maintenance during Aging

Abstract

Several strategies have been proposed to prevent or reduce the rate of physiological decline in organ functions among aging postmenopausal women. These include increasing physical activity, improving nutrition, managing stress, and enhancing sleep quality. Although hormone replacement therapy (HRT) has been recommended, it remains a controversial topic, eliciting debate both in scientific circles and the public sphere. This book chapter aims to provide a comprehensive review of the current literature on the accelerated aging phenotype observed in postmenopausal women due to sex hormone deficiency. It will assess the efficacy and safety of HRT, offering a critical analysis of its benefits and risks. Moreover, the chapter will present a clinical perspective, suggesting practical advice for women approaching menopause. This guidance is intended for everyday clinical practice, aiming to support healthcare providers in offering informed, holistic care to this population.

1. Introduction

Menopause signifies more than just the cessation of a woman’s reproductive capabilities—it is a pivotal milestone in the female life cycle. This period heralds a range of significant physical transformations, often ushered in by challenging vasomotor symptoms (VMS) such as hot flashes and night sweats. Beyond these, a plethora of symptoms, including the genitourinary syndrome of menopause (GSM)—encompassing vaginal dryness, recurrent urinary infections, dysuria, and diminished libido—can severely affect a woman’s quality of life. Moreover, menopause coincides with other age-related shifts in female biology, manifesting as notable two changes in the cardiovascular system, brain, skin, and bone. Intriguingly, many women perceive these changes as signs of “accelerated aging,” a sentiment frequently echoed in clinical settings where women express feelings of suddenly “getting old.”

This study supports the clinical observation that postmenopausal women exhibit accelerated aging phenotype. In the following, we will discuss studies that examined postmenopausal biological changes in individual organs and the effects of hormone replacement therapy (HRT) on decreasing the “rate of aging.”

2. Search methods

A search was conducted on PubMed for randomized clinical trials (RCTs) using a combination of terms: “hormone replacement therapy” OR “HRT” OR “hormone therapy” OR “HT” AND “cardiovascular disease” OR “CVD” OR “coronary heart disease (CHD)” AND “postmenopausal women.” This search extended across various organ systems. Additionally, a manual review of references was performed, specifically those cited in the book “Estrogen Matters” (2018) by Avrum Bluming and Carol Tavris [1], to identify relevant studies. To be included in the analysis, studies had to meet several criteria: (1) utilize a randomized controlled trial (RCT) design; (2) compare HRT with a placebo control; and (3) report both the relative risk (RR) and the 95% confidence interval (CI) for the comparison between HRT and placebo, along with the total number of participants in each group. A summary of the RCTs discussed in this chapter is included in Table A1 in the Appendix.

3. Women exhibit accelerated aging during the postmenopausal period

3.1 Postmenopausal cardiovascular system and HRT

Cardiovascular disease (CVD) is the leading cause of death in both aged men and women [22]. Aging is associated with increased risk of atherosclerosis, hypertension, stroke, and atrial fibrillation [23]. Several studies have demonstrated that premenopausal women have a lower incidence of CVD when compared to age-matched males; however, the prevalence of CVD increases dramatically after menopause and women experience CVD 10 years later than men [22].

Murine animal studies have demonstrated that estrogen exerts cardioprotective effects including reduced oxidative stress, improved mitochondrial function, attenuation of cardiac hypertrophy, stimulation of angiogenesis, and vasodilation [24]. Estrogen deficiency leads to negative effects on the endothelium and vascular smooth muscle, decreases expression of nitric oxide (NO)-receptor soluble guanylate cyclase (sGC), decreases activation of vasodilator-stimulated phosphoprotein (VASP), enhanced sensitivity to vasoconstrictive agents and impaired vasorelaxation due to increased phenylephrine and angiotensin-II vasoconstriction effects through upregulation of angiotensin-II type I receptor, and changes in extracellular matrix (ECM) composition, all these changes lead to impaired vascular function and stiffening vessel walls as demonstrated in ovariectomized aging rats and that estrogen replacement attenuated these changes [25].

In humans, several RCTs have examined the role of HRT in the risk of CVD. In the following, we will discuss the main RCTs and contrast their findings. The most famous RCT is the Women’s Health Initiative (WHI) (1993) [2] which enrolled 27,347 postmenopausal women and aimed at examining the risks and benefits of HRT in postmenopausal women as a general “anti-aging” intervention measured by multiple outcomes (see Table A1, Appendix). The results of this trial have had enormous influence on clinical decision making regarding the use of HRT in postmenopausal women [2]. Specifically, postmenopausal women (age: 50–79 years and average 63 years) with a mean age of menopause of 12 years were randomized to receive in case of previous hysterectomy: conjugate equine estrogen (CEE) (oral dose 0.625 mg/day, N = 10,739) or placebo and in case of intact uterus: CEE in combination with medroxyprogesterone acetate (MPA) (oral 0.625 mg CEE + 2.5 mg of MPA/day, N = 16,608) or placebo. The “rate of aging” was determined by disease-specific endpoints: coronary heart disease (CHD) (primary outcome) and incidence of invasive breast cancer (primary safety outcome). Other outcome parameters measured included incidence of stroke, pulmonary embolism (PE), colorectal cancer, hip fracture, and death (measured by global index, GI) [2]. The details of inclusion and exclusion criteria are summarized in Table A1 in the Appendix. The CEE + MPA trial was terminated after 5.6 years and CEE-only after 7 years due to safety concerns [26]. The WHI found a significant increase in CHD events: 29% increase in the CEE + MPA compared to placebo group. The increase was in nonfatal myocardial infarction (MI), with no significant differences observed in CHD deaths. In the CEE-alone group, there was a non-significant 9% reduction in CHD rates in the CEE group compared with placebo. It should be noted that a large percentage of women included in the WHI trial had CVD and were on treatment for the disease: 70%were overweight or obese (average BMI 28.5), >50% were current or previous smokers, and 48% of the women enrolled were receiving treatment for hypertension [2].

The WHI trial was followed by a number of clinical trials with a smaller number of participants. The Heart and Estrogen-progestin Replacement Study (HERS) [5] and HERS II [6] (see Table A1, Appendix) were designed to assess the efficacy and safety of HRT for secondary prevention of CHD events (nonfatal MI or CHD deaths) in postmenopausal women. In this study, 2763 women were enrolled and randomized to receive either CEE + MPA or placebo for 4 years. The mean age of participants was 66.7 years (85% > 60 years, 27% > 70 years) and approximately 20 years postmenopause (mean age at menopause was reported 48.7 years). The prevalence of coronary risk factors among the participants was high: 62% were past or current smokers, 59% had hypertension, 90% had serum LDL-cholesterol of 100 mg/dL or higher, 23% had diabetes, and about 34% were obese or severely obese and almost 20% were taking medications for treatment of CHD or for established risk factors associated with CHD. The trial did not reveal statistically significant differences between the randomized groups, suggesting that HRT is not effective in preventing secondary cardiovascular events in women with pre-existing CHD.

A newer study named Early vs. Late Estrogen Intervention Trial (ELITE) [16] examined the “timing hypothesis” stratifying the participants as early (<6 years) and late (>10 years) postmenopausal women (see Table A1, Appendix) and the women were randomized to either oral 17β-estradiol 1 mg/day or placebo for 5 years. The rate of carotid intima-media thickness (CIMT) progression (a marker for the extent of carotid atherosclerotic vascular disease) was significantly reduced in the early postmenopausal group assigned to estradiol compared to placebo. In the late postmenopausal group, CIMT progression did not differ between groups.

Another trial, Kronos Early Estrogen Prevention Study (KEEPS) [11], was designed to determine the effects of HRT in postmenopausal women on the progression of diseases in different organ systems including cardiovascular system (see Table A1, Appendix). A total of 727 healthy (free of subclinical CHD) women were included in the RCT and randomized for either oral CEE 0.45 mg/day (N = 230), transdermal estrogen (t-E2) 50 μg/day (N = 222) or placebo (N = 275) for 4 years. The women were recent menopause and within 6 months to 3 years of natural menopause. The primary outcome was the progression of subclinical atherosclerosis measured by CIMT. The trial reported a similar increase in CIMT across all three groups.

The difference in outcome between ELITE [16] and KEEPS [11] trials may be related to the duration of the intervention, sample size, and duration of the follow-up period being too short to be able to detect between-group differences. Also, there were differences in the dose and type of estrogen used: 0.45 oral CEE (KEEPS) compared with oral 17β-estradiol 1 mg/day (ELITE). However, KEEPS and ELITE trials demonstrate the safety of using HRT in postmenopausal women when employed within 6 years of menopause, without pre-existing CVD risk factors, and with an average treatment duration of 5 years.

Another study, The Danish Osteoporosis Prevention Study (DOPS) [18], examined 1006 healthy, younger, early postmenopausal women (in average 0.7 years postmenopause) in a 10.1-year RCT and subsequent unblinded follow-up for 6 years in order to determine the long-term effects of HRT (see Table A1, Appendix) [18]. They found that long-term HRT significantly decreased heart failure, MI, and mortality, which was maintained during an additional 6 years of non-randomized follow-up (HR 0.48, 95% CI: 0.26–0.87). However, the trial is relatively small, with small number of outcome events.

One contentious issue related to the effect of HRT on cardiovascular health is the role of progesterone supplementation. Thus, the postmenopausal Estrogen/Progestin Interventions (PEPI) trial [9] assessed the effects of HRT on CVD risk factors using different forms of progesterone (see Table A1, Appendix). In this trial, 847 healthy postmenopausal women were randomly assigned to one of the following groups: (1) placebo; (2) CEE 0.625 mg daily; (3) CEE 0.625 daily plus MPA 10 mg, days 1–12; (4) CEE 0.625 daily plus MPA 2.5 mg daily; or (5) CEE 0.625 daily plus micronized progesterone (MP) 200 mg, days 1–12. The outcomes assessed were HDL-C, systolic blood pressure, serum insulin and fibrinogen, and thromboembolic events [9]. The investigators reported a significant increase in HDL-C and a decrease in LDL-C serum levels across all treatment groups compared to placebo. Interestingly, women who received CEEs or CEEs/MP had significantly increased serum levels of HDL-C compared to CEEs/MPA group. There were no significant differences in the observed changes in systolic blood pressure and serum insulin levels, but fasting glucose levels significantly decreased in all treatment groups. Women assigned to placebo had a greater increase in serum levels of fibrinogen compared to the active treatment group, but there was no increase in adverse thromboembolic effects [9]. This study suggests a protective role of progesterone on CVD risk factors by improving lipoprotein profile and that MP may have a greater benefit than MPA.

3.2 Postmenopausal brain and HRT

The most common neurodegenerative disease is Alzheimer’s disease (AD), and almost two-thirds of Americans with AD are women [27] and also the incidence rates of different forms of dementia and AD are greater in women than men [28, 29]. The increased risk of AD in postmenopausal women suggests that sex hormone deficiency is a significant risk factor [30]. This notion is supported by in vitro and in vivo studies.

In animal experiments, ovariectomy (OVX) leads to accelerated brain aging evidenced by reduced dendritic spines, decreased synaptic density, decreased number of synapses and a reduction in gray matter volume and the reverse effects with an increase in hippocampal dendritic spine and synapse density were observed following exogenous estrogen administration [31]. OVX studies in rodents have also shown that long-term estrogen deficiency causes impairment of memory and learning abilities [32, 33, 34] as well as declines in the performance of both cognitive and motor tasks, short- and long-term non-spatial and spatial memory, cognitive impairment, and increased anxiety [34, 35].

Several of the abovementioned clinical trials in humans (see Table A1, Appendix) examined the effects of HRT on brain and cognitive functions. The KEEPS study [11] included an ancillary study that examined the effects of HRT on brain functions (KEEPS-Cog) [14] and reported that 4 years of HRT did not exert harmful or beneficial effects compared to placebo. The WHI [2] performed an ancillary study, The Women’s Health Initiative Memory Study (WHIMS) [3], assessing the role of CEE or the combined CEE + MPA on the incidence of dementia and mild cognitive impairment (MCI) in older postmenopausal (aged 65–79) women (see Table A1, Appendix). After a mean follow-up of 4.2 years, 61 women in total were diagnosed with probable dementia: 40 among women assigned to CEE + MPA therapy relative to 21 in the placebo group. After restricting analyses to those adherents to therapy, the number of probable dementia cases was reduced to 21 in the CEE + MPA group; however, it still significantly increased relative to six in the placebo group. There were no statistically significant differences in MCI between the groups [3]. However, the increased risk of dementia was apparent already 1 year after randomization in both the CEE + MPA and placebo groups, suggesting that some participants had cognitive impairment at baseline and that CEE + MPA therapy may have accelerated pre-existing cognitive decline.

Another ancillary study to the WHI [2] was published in 2004 and is known as the Women’s Health Initiative Study of Cognitive Aging (WHISCA, see Table A1, Appendix) [4]. WHISCA assessed the efficacy of postmenopausal HRT on longitudinal age-related changes in specific cognitive functions that included memory, cognition (measured by language, attention, and spatial ability), motor function, and mood in 2302 women aged 66–84 years (a mean of 12 years since menopause) who did not meet criteria for dementia at baseline and received treatment period for 3 years [4]. Women assigned to CEE + MPA had a negative impact on verbal memory over time and CEE-alone was associated with lower spatial rotational ability compared to placebo [36].

The ELITE trial [16] also included an ancillary study, ELITE-Cog, examining the effect of HRT on cognitive functions (see Table A1, Appendix) [17]. A total of 455 postmenopausal women participated in the study for 5 years (80% adherence). Neither early (<6 years) nor late (>10 years past menopause) initiation of HRT altered verbal memory, executive functions, or global cognition as assessed after a mean treatment duration of 57 months of oral 17β-estradiol combined with progesterone. While the study did not reveal positive effects on cognitive functions, it supports the safety of HRT use on cognition as no adverse effects on cognitive abilities were observed during the 5-year period of the study.

3.3 Postmenopausal skin and HRT

The first signs of dermal aging (skin thinning, folds, or wrinkles) start around 30–35 years of age when estrogen levels start to decline. Several RCTs have reported that CEE and estradiol (E2) formulations increase the thickness of the skin and of the dermis and therefore decrease the rate of skin aging [37, 38]. Oral E2 has also been demonstrated to increase skin elasticity and hydration [38]. An ancillary skin study of the KEEPS trial [11] (see Table A1, Appendix) examined the effects of HRT on wrinkles (assessed on the face and neck) and skin rigidity (assessed in forehead and cheek) in postmenopausal women (age 42–58 and within 36 months of last menstrual period) of varying race/ethnicity [15]. Interestingly, there were no significant differences in the number and depth between women randomized to oral CEE, t-E2, or placebo during the 4-year follow-up period [15]. The timing of initiation of HRT seems to play an important role in preventing skin aging as it has been observed in a subgroup analysis that beneficial effects of HRT were observed when HRT was initiated within 24 months of the last menstruation [39]. Another study included 485 postmenopausal women in a 48-week RCT to assess the effects of low-dose HRT on postmenopausal skin changes, primarily facial wrinkling [39]. The women were randomly assigned to one of three interventions: placebo (165 subjects), 1 mg norethindrone/5 μg ethinyl estradiol (162 subjects), or a 1 mg norethindrone/10 μg ethinyl estradiol group (158 subjects) given orally. The participants were, on average, 53.6 years old and approximately 5 years postmenopausal. After 48 weeks, the investigators did not detect clinically significant differences between the groups; however, a subgroup analysis revealed marginal improvement in women who were less than 24 months in postmenopausal age [39].

3.4 Postmenopausal skeleton and HRT

A large body of literature in both animal models and humans has demonstrated that estrogen deficiency leads to accelerated bone loss, increased bone resorbing osteoclastic activity, and impaired bone formation [40]. Similar effects were observed in OVX mice [41] and sheep [42]. Also, in human observational studies, postmenopausal period results in accelerated bone loss in both trabecular and cortical bone, leading to an increased risk of fracture [43]. Multiple mechanisms have been identified for the positive effects of estrogen on the postmenopausal skeleton, including inhibition of bone resorption, enhancing bone formation, and exerting positive calcium balance on the kidney and intestine [44].

Several human studies have consistently demonstrated the efficacy of HRT in preventing postmenopausal hormone loss and decreased risk of osteoporotic fractures. In the following, we will provide some examples of the most important human trials.

A 2-year open trial [45] of young healthy non-obese postmenopausal women (mean age 56.6 years) received a low dose (LD) continuous combined HRT (containing 1 mg estradiol +0.5 mg norethisterone acetate (NETA), or 0.5 mg of 17β-estradiol and 0.25 mg of NETA (ultra-low dose, ultra-LD-HRT), along with 1000 mg calcium daily or control group that received 1000 mg calcium/day for 2 years. At the end of the trial, the control group experienced a significant decrease in BMD at the spine (−2.8 ± 0.2%) and femoral neck (−2.8 ± 0.7%), whereas in the LD-HRT treated group, bone mineral density (BMD) showed a significant increase at the spine (5.2 ± 0.7%) and femoral neck (2.8 ± 0.4%). In the ultra-LD-HRT treated women’s spine and femoral neck, BMD showed a significant increase (2.0 ± 0.3 and 1.8 ± 0.3%, respectively [45].

The PEPI trial [9] (see Table A1, Appendix) compared changes in BMD in women assigned to different treatment groups as described previously and reported a significantly increase in BMD in women assigned to HRT compared to placebo at 12 and 36 months. The participants assigned to placebo lost an average of 1.8% of spine BMD and 1.7% of hip BMD, while those assigned to HRT exhibited 3.5% to 5.0% increases in spinal BMD and a 1.7% increase in hip BMD [10].

The relative contribution of progesterone and estrogen to the observed positive effects of HRT on bone mass was examined in a systematic review and meta-analysis of five RCTs [46]. The analysis includes women (N = 1058) mid-50 years of age, early in menopause (<5 years except one study with average 12.8 years postmenopause) and found that CEE + MPA significantly increased annual percent spinal BMD gains (+0.68%/year; 95% CI: 0.38, 0.97%) relative to the same dose of CEE-alone suggesting beneficial additive effects of progesterone to estradiol in gain of spinal BMD [46]. Interestingly, one study examined dose–response effects of 10 or 20 mg of MPA 15 days/month together with estrogen and reported that a higher dose of MPA (20 mg) did not lead to further increase in spine and hip BMD after 1 year of treatment [47] suggesting that higher doses of MPA may not have an additive effect [46].

The Women’s Health, Osteoporosis, Progestin, Estrogen (HOPE) trial (N = 822) examined the effects of lower doses of CEE + MPA on BMD in women aged 40–65 years within 4 years of menopause [48]. Women were randomly assigned to receive one of eight treatment regimens: (1) CEE 0.625; (2) CEE 0.625 + MPA 2.5; (3) CEE 0.45; (4) CEE 0.45+ MPA 2.5; (5) CEE 0.45 + MPA 1.5; (6) CEE 0.3; (7) CEE 0.3 + MPA 1.5 (all mg/day); or (8) placebo (calcium 600 mg/day). Women assigned to all of the active treatment groups had significant gains in BMD from baseline in spine and hip BMD as well as total body BMD after 24 months and also when compared to the placebo group. CEE 0.625 + MPA 2.5. Also, the study reveals differences in the incremental increase in BMD among intervention groups, suggesting the existence of an optimal dose-dependent effect of HRT with respect to the prevention of postmenopausal bone loss. In the CEE-alone group, CEE 0.625 had the greatest increase in hip BMD; however, in the CEE + MPA group, the CEE 0.45+ MPA 2.5 had the greatest increase in hip BMD [48].

The Danish RCT DOPS [18] (see Table A1, Appendix), which also examined the fracture-reducing potential of long-term HRT in recent postmenopausal women in a primary preventive scenario, reported a significant reduction in both the overall fracture risk (HR 0.61; 95% CI: 0.39–0.97) and the risk of forearm fractures (HR 0.24; 95% CI: 0.09–0.69) [19].

Finally, the WHI trial [2] (see Table A1, Appendix) showed significantly reduced hip and vertebral fractures as well as total osteoporotic fractures in both CEE-alone and CEE + MPA groups.

3.5 HRT and breast cancer

One of the major concerns of postmenopausal women considered HRT is the perceived increased risk for developing breast cancer. This issue has been examined in several RCTs. The WHI trial [2] was halted 3 years prior to the scheduled end due to the evidence of increased risk for breast cancer. The estimated HR for breast cancer was 1.26 (95% CI: 1.00–1.59), with 290 cases equivalent to a non-significant 26% reported increase in breast cancer (38 vs. 30 per 10,000 person-years). Interestingly, the WHI CEE-alone arm exhibited a non-significant reduction in incidence of invasive breast cancer. The HR was 0.77 (95% CI: 0.59–1.01) with 218 cases (26 in CEE-alone arm vs. 33 in controls per 10,000 person-years) [2]. An overview of WHI is provided in Table A1 in the Appendix, and Table 1 illustrates the main findings related to risk of breast cancer.

Clinical trial	Outcomes assessed	Main outcomes	Conclusion
Women’s Health Initiative trial, CEE alone.; Stefanick et al. [49]	Breast cancer incidence among women assigned to CEE therapy relative to placebo.	All breast cancer incidents diagnosed through February 29, 2004, were included in a re-analysis with mean follow up of 7.1 years. Invasive breast cancer HR for women assigned to CEE compared with matched placebo was 0.8 (95%-CI: 0.61–1.04). After one year, 9.2% of women in the CEE had mammograms with abnormalities relative to 5–5.5% in the placebo group which persisted through the trials, however the differences were only requiring short term follow up and were not suspicious or abnormal findings.	Treatment with CEE alone does not increase incidence of breast cancer among postmenopausal women with prior hysterectomy.
Re-analysis of data from the combined WHI trials (N = 16,608) and a corresponding observational study cohort of women with an intact uterus who were either HRT users (N = 6,756) or non-users (N = 32,328).; Prentice et al. [50]	Relation between prior use of HRT and time from menopause to first use of HRT (“gap time”) to risk of invasive breast cancer among women assigned to CEE+MPA compared to placebo.	A small but statistically significant elevated risk of breast cancer was detected among women randomized to CEE+MPA treatment compared to placebo (HR 1.96 vs 1.02). For women initiating treatment <5 years after menopause: HR for CEE+MPA 2.06 versus 1.77 in placebo. For women initiating treatment ≥5 years after menopause: HR for CEE+MPA 1.30 versus placebo 0.99. HR for women initiating HRT immediately following menopause (gap time 0), breast cancer risk was elevated during the first two years of CEE+MPA use and decreased to HR 0.84 with a five-year increment in time.	The WHI clinical trial and observational study each reveal an increased breast cancer risk of daily 0.625 mg CEE + 2.5 mg MPA. Women who initiate treatment soon after menopause appear to have increased risk.
Extended follow-up (median 11.8 years) of the WHI CEE-alone trial.; Anderson et al. [51]	Breast cancer incidence and mortality among postmenopausal women with hysterectomy assigned to either CEE alone or placebo.	Women randomly allocated to CEE therapy (mean 5.9 years) had a lower incidence of invasive breast cancer (151 cases, 0.27% per year) than those receiving placebo (199 cases, 0.35% per year) with no differences between intervention and post-intervention phase effects. Performing subgroup analyses, the breast cancer risk was reduced in the CEE group in women without benign breast disease or family history of breast cancer. Fewer women died from breast cancer relative to controls (6 vs 16 deaths per year).	5-years of treatment with CEE therapy is associated with lower incidence of breast cancer and decreased mortality among women without benign breast disease or family history of breast cancer.
Extended follow-up (median 13 years through September 2010) of the two WHI trials.; Chlebowski et al. [52]	Influence of long-term HRT on breast cancer incidence rates and breast cancer characteristics during intervention and during early and late post intervention phases.	Treatment with CEE+MPA compared to placebo led to an increase in total (245 vs 185 cases, HR 1.24,) and invasive (199 vs 150 cases, HR 1.24) breast cancers. Short-term CEE+MPA use increases incident cancers which then substantially drops in risk in the early post intervention phase (within 2.75 years from intervention), however the risk remains elevated in the late post intervention follow-up (mean 5.5 years) and the HR for breast cancer risk remains elevated (HR 1.37; 95%-CI: 1.06–1.77).; The breast cancer risk reduction seen in the CEE alone trial during intervention was lower than 1 throughout the 7.2-year intervention phase and in the early postintervention phase the HR for the influence of CEE alone on breast cancer risk was substantially lower (HR 0.55; 95%-CI: 0.34–0.89).	With longer follow-up of the 2 WHI trials, a complex pattern of changing incidence of breast cancer was observed.; Combined CEE+MPA increased the risk of breast cancer during the initial intervention followed by a substantial drop in risk in the early postintervention period, but with sustained higher breast cancer risk during late postintervention phase.; Use of CEE alone reduced breast cancer risk through cumulative follow-up of 13 years.

Table 1.

Findings related to breast cancer risk as reported by the Women’s health initiative (WHI) trial.

CEE = conjugated equine estrogen; HR = hazard ratio; HRT = hormone replacement therapy; MPA = medroxyprogesterone acetate.

Several investigators were concerned that the presence of several confounders that affect the external validity of the WHI results as the study cohort comprised large percentage of women with chronic diseases and with a mean of 12 postmenopausal age and thus do not represent early and healthy postmenopausal women interested in HRT [50]. Also, it should be noted that the WHI trial reported a 32% reduction in breast cancer risk, after 7.2 years of follow-up in CEE-alone group when analyses are limited to a minimum of 80% compliance (HR 0.67; 95% CI, 0.47–0.97). Also, across all women, regardless of compliance, ductal carcinoma was significantly reduced by 29% in the CEE-only arm relative to placebo, and 18 years of cumulative, breast cancer mortality was significantly reduced by 45% (HR 0.55; 95% CI: 0.33–0.92) [50].

Another important confounder is the extraordinarily low breast cancer incidence in the placebo group that had prior HRT use, which arguably falsely gives the impression that the elevated HR of breast cancer risk was caused by CEE + MPA treatment [50]. The reason behind the low breast cancer incidence in the subgroup is not clear; however, it has been externally validated as an outlier with WHI observational data [50].

Interestingly, other interventional trials, PEPI [9], HERS [5, 6], Women’s Estrogen for Stroke Trial [51], ELITE [16], DOPS [18], and REPLENISH [52, 53], did not reveal an increase in breast cancer incidence (see Table A1, Appendix). In fact, DOPS [18] reported that after 10 years of HRT intervention and subsequent 6 years of unblinded follow-up, women aged 45–58 years (mean of 49.7 years) had significantly reduced risk of breast cancer relative to placebo with HR 0.49 (95% CI: 0.28–0.87) in women with intact uterus and HR 0.42 (95% CI: 0.18–0.97) in women who received estrogen alone [18]. Based on all these trials, it seems that HRT is generally safe with respect to breast cancer development when initiated in the early postmenopausal period and in healthy women.

4. A practical approach for hormone replacement therapy in the postmenopausal period

HRT is approved by US Food and Drug Administration (FDA) as an effective treatment for menopausal symptoms, for prevention of osteoporosis, and for treatment of premature ovarian failure [54]. Based on the results from RCTs discussed in the previous section, we will provide practical information regarding HRT use in postmenopausal women.

4.1 Translation of clinical evidence to practical advice

HRT is useful for treating the VMS associated with menopause, as supported by multiple studies [9, 52, 53]. However, advice regarding the long-term use of HRT for prevention of age-related diseases should be individualized, as shown in decision tree (Figure 1).

Figure 1.

Decision tree, a practical approach to initiating HRT in peri- and postmenopausal women.

HRT can be employed in women with increased risk for osteoporosis as a preventive measure for bone loss and also when other standard medications are not chosen, e.g., bisphosphonate. While it is possible that HRT initiated early in the postmenopausal period may prevent CHD and maintain brain health, HRT is not recommended as an adjuvant treatment to decrease the rate of complications in women suffering from CHD or cognitive impairment. It is recommended to identify the most appropriate regimen, dose, duration, and route of administration, with periodic reevaluation of the benefits and risks of continuing or discontinuing HRT, based on individual preferences. Women who are overweight, smokers, have hypertension, diabetes mellitus, or suffer from other comorbidities including vascular or liver disease must be closely assessed prior to initiating treatment, particularly as women with these risk factors are often excluded from participation in RCTs and therefore their benefits and risks from HRT are less known.

4.2 How do we inform women regarding the side effects of HRT?

Women have concerns when starting or continuing HRT regarding possible side effects and, most importantly, concerns related to the increased risk of developing breast cancer. We think that HRT should be offered for healthy women without a family history of breast cancer or a previous history of heart disease, stroke, or breast cancer. In these women, starting therapy early in the postmenopausal period (<5 years) is safe, as supported by findings from several trials: WHI [2], PEPI [9, 10], KEEPS [11], ELITE [16], and DOPS [18] that revealed no increase in thromboembolic disease, stroke, or breast cancer.

4.3 Duration of therapy? control during therapy?

The decision regarding the duration of treatment and when to stop HRT must be considered in the context of the individualized risk/benefit profile as well as personal preferences. It is not yet known if ongoing HRT by healthy women who initiate treatment in early postmenopausal years (<5–10 years) but are now older than 60 years carries the same risk as initiating HRT in women with more than 10 years of postmenopausal age or older than 60 years. HRT does not need to be routinely discontinued in women aged older than 60 or 65 years, especially in women with persistent VMS, quality-of-life issues, or receiving HRT for prevention of osteoporosis [55].

Regular reassessment of the woman’s health status (comorbidities, risk factors for CVD, and venous thromboembolism (VTE)) is necessary to ensure that the risks of continued treatment do not outweigh the benefits in each individual woman.

5. Conclusion and future directions

While several observational studies have reported the benefit of HRT in the prevention and treatment of chronic diseases such as CVD and dementia and in reducing all-cause mortality among postmenopausal women using HRT compared to nonusers [50], the results of RCTs did not provide unequivocal answers. However, the current literature regarding HRT in postmenopausal women has revealed the limitations of the benefits of using HRT as a general “anti-aging” strategy. HRT may be useful in a group of postmenopausal women with severe menopausal symptoms that reduce their quality of life and prevent postmenopausal bone loss. While more studies are needed to compare different routes of administration, regimens, and doses, physicians can currently provide postmenopausal women with balanced information regarding the risks and benefits of HRT to allow them to make an informed decision.

A. Appendix

See Table A1.

Clinical trial	Outcomes assessed	Inclusion criteria	Intervention	Main findings
Women’s Health Initiative (WHI) clinical trials; Writing Group for the Women’s Health Initiative Investigators, [2]	Two parallel, double-blinded RCT’s to determine whether HRT delays the rate of aging in postmenopausal women.; Primary outcome: Coronary heart disease (CHD) and incidence of invasive breast cancer (primary safety outcome).; Other outcomes include incidence of stroke, pulmonary embolism (PE), venous thromboembolism (VTE), colorectal cancer, hip fractures and death.	N = 27,357 postmenopausal women.; Age: 50–79 years (average 63 years), with a mean of 12 years since menopause.; 70% were overweight or obese (average BMI 28.5).; >50%: current or previous smokers (N=8,237).	Women with intact uterus: CEE+MPA; (1) 0.625 mg CEE+2.5mg MPA/day (N = 8,506); (2) placebo (N = 8,102).; Women with prior hysterectomy: (1) 0.625 mg CEE/day (N = 5,310); (2) placebo (N = 5,429).; July 2002: CEE+MPA study was halted 3 years early.; March 2004: CEE alone study was halted.	Breast cancer: HR was non-significant 0.77 (95%–CI: 0.59-1.01) with 218 cases (26 vs 33 per 10,000 person-years in CEE alone. HR for breast cancer were 1.26 (95%-CI: 1.00–1.59) with 290 cases equivalent to a 26% reported increase in breast cancer (38 vs 30 per 10,000 person-years) observed in the CEE+MPA. CHD: HR 1.29 (95%-CI: 1.02–1.63) with 286 cases with a significant increase in CHD events of 29% in the CEE+MPA relative to placebo group (37 vs 30 per 10,000 person-years). The increase was in nonfatal MI with no significant differences observed in CHD deaths. HR was 0.91 (95%-CI: 0.75–1.12) with 376 cases equivalent to a non-significant 9% reduction in CHD rates in the CEE group compared with placebo (49 vs 54 per 10,000 person-years). Stroke: The incidence in stroke was increased by 39% in CEE alone: HR: 1.39 (95%-CI: 1.10–1.77) with 276 cases with an absolute excess risk of 12 additional strokes per 10,000 person-years. Non-fatal stroke rates were higher in women receiving CEE+MPA (41% increase: 29 vs 21 per 10,000 person-years). PE/VTE: Women in the CEE+MPA group had 2-fold greater rates of PE, HR 2.13 (95%-CI: 1.39–3.25) with 101 cases. Rates of VTE were 34 and 16 per 10,000 person-years in the CEE+MPA and placebo groups, respectively. In the CEE alone only the risk of DVT was statistically significant increased. not in VTE or PE. Bones: CEE+MPA group had a significantly lower incidence of hip fractures (10 per 10,000 person-years vs 15 per 10,000 person-years). CEE+MPA reduced the observed hip and clinical vertebral fracture rates by one third. Reductions in other osteoporotic fractures was 23% and total fractures 24% decrease. CEE group had a reduction in hip fracture rates (11 vs 17 per 10,000 person-years,), clinical vertebral fractures (11 vs 17 per 10,000 person-years,) and total osteoporotic fractures (139 vs 195 per 10,000 person-years). Cancer: No differences were observed in the rates of colorectal or total cancer incidents except colorectal cancer rates were significantly reduced by 37% in CEE+MPA.
Women’s Health Initiative Memory study (WHIMS); Shumaker, [3]	To assess the role of CEE alone and combined CEE+MPA on 1) the incidence of probable dementia (primary outcome) and 2) mild cognitive decline (MCI) (secondary outcome) in older postmenopausal (aged 65–79) women.	See WHI above.	Data from the combined WHI trials. See WHI above.	N=47 were diagnosed with probable dementia in the CEE- trial; N=28 among women assigned to CEE and N=19 in the placebo group. During follow-up, the “49% higher” incidence in women assigned to CEE relative to placebo (37 vs 25 per 10,000 person-years) was not statistically significant.; In the CEE+MPA trial, the incidence was similar (45 vs 22 per 10,000 person-years). The HR associated with active treatment with CEE alone and CEE+MPA did not differ significantly even though the HR in the estrogen-alone trial was non-significant. When analyses were limited to data six months after first assessed nonadherence, the HR for probable dementia associated with CEE alone was 1.55 (95%-CI: 0.49–4.88).
The Women’s Health Initiative Study of Cognitive Aging (WHISCA); Resnick et al. [4]	To assess the efficacy of postmenopausal HRT on age related changes in specific cognitive functions including memory, cognition (measured by language, attention, spatial ability), motor function and mood.	N=2,302 postmenopausal women aged 66-84years (mean 73.9).	See WHI above.	Women assigned to CEE+MPA had a negative impact on verbal memory over time and a trend toward a positive influence on figural memory relative to placebo after long-term therapy (average 4-5 years) with no significant differences in positive affect, negative affect or self-reported depressive symptoms. CEE-alone was associated with lower spatial rotational ability relative to placebo at the initial assessment, however this diminished over time.
The Heart and Estrogen- progestin Replacement Study (HERS); Grady et al. [5]	Efficacy and safety of CEE+MPA treatment for secondary prevention of CHD in postmenopausal women.; Primary outcome: CHD events (nonfatal MI or CHD death).; Secondary cardiovascular outcomes: coronary artery bypass surgery, percutaneous coronary revascularization, hospitalization for unstable 2 angina, resuscitated cardiac arrest, congestive heart failure, stroke or transient ischemic attack and peripheral arterial disease.	N=2,763 postmenopausal women with intact uterus.; Age: 44-79 years of age (mean age 66 years).; Prior; history: ≥1 CVD event(s); MI, coronary artery bypass graft surgery, percutaneous transluminal coronary angioplasty, or other mechanical revascularization or ≥50% occlusion of a major coronary artery.	4-year RCT.; (1) 0.625 mg CEE+2.5mg MPA daily (N=1,380); (2) placebo (N=1,383).; Adherence: 79% in year 1 and 70% by year 3.	Primary and secondary cardiovascular outcomes: No statistically significant differences between the randomized groups.; Among women who stopped taking HERS medication, risk of primary CHD events was elevated in the first month after stopping; however, no differences were observed between the HRT therapy and placebo group.; Secondary outcomes: No differences in total mortality, rates of breast, endometrial or other cancers or other fractures in the hormone group relative to the placebo group. Increased risk of VTE (34 vs 12), DVT (25 vs 8) and PE (11 vs 4) in the hormone group relative to placebo, which authors suggest may be due to high age and multiple risk factors among HERS participants.; LDL-C was significantly reduced by 14% and 3% in the placebo group. HDL increased by 8% and decreased by 2% relative to baseline in the hormone and placebo group respectively.
HERS II; Grady et al. [6]	To determine if the risk reduction observed in the later years of HERS persisted and resulted in an overall reduced risk of CHD events with additional years of follow-up.	N=2,321 of the 2,763 women who participated in HERS.	6.8 years: 4.1-year RCT (HERS) and subsequent unblinded follow-up of 2.7 years (HERS II).; Was planned for 4-year follow-up but was terminated at the second annual review as the benefit in the group randomized to hormones relative to placebo were too low.; Adherence declined to 45% during year 6 of follow-up.	The lower rates of CHD events among women in the hormone group in the final years of HERS did not persist during the additional years of follow-up in HERS II. The HR for CHD events in the hormone group was higher in the first year of treatment however lower in the fourth year implying a decrease in risk during years 3-5 however no statistically significant trend over time.; Analyses adjusted for potential confounders (incl. age, ethnicity, smoking, BMI, diabetes, systolic blood pressure, creatinine clearance, exercise, history of congestive heart failure and MI, baseline use of aspirin, ACE- inhibitors and statins, and statin use during follow-up plus adherence >80%) did not alter these results.; It was examined whether the early increased risk in CHD events was limited to women at particularly high risk for events (age, prior manifestations of CHD, CHD risk factors, medication use); however, no clear evidence found that the early risk was limited to specific subgroups.
Ancillary study of the HERS; Simon et al. [7]	Risk of stroke among the participants enrolled in HERS.	See HERS above.	See HERS above.	149 women (5%) had one or more strokes resulting in 26 deaths. 85% were ischemic; 8% were hemorrhagic strokes, and 6% were not classified. 139 nonfatal strokes and 26 fatal strokes occurred over a mean follow-up of 4.1 years. The incidence rates of any stroke event were 15/1,000 person-years among women assigned to combined HRT and 12/1,000 person-years among women assigned to placebo. Thus, CEE+MPA therapy was not significantly associated with risk of nonfatal stroke (HR 1.18; 95%-CI: 0.83–1.66), fatal stroke (HR 1.61; 95%-CI: 0.73–3.55), or transient ischemic attack (HR 0.90; 95%-CI 0.57–1.42) compared to placebo in postmenopausal women with cardiovascular disease.
Women’s Estrogen for Stroke Trial; Viscoli et al. [8]	To examine the role of HRT for secondary prevention of CHD.; Primary outcome: fatal or nonfatal stroke.; Secondary outcome: transient cerebral ischemia or nonfatal myocardial infarction (MI).	N=664 postmenopausal women with cerebrovascular disease who had suffered a recent (<90 days) ischemic stroke or transient ischemic attack.; Age: 46–91 years of age (mean age 71 years).	2.8-year RCT.; (1) 17β-estradiol (N=337) 1 mg/d; (2) placebo (N=327).; When a nonfatal stroke occurred, the study drug was stopped and follow-up ended. Stroke followed by death within 30 days was classified as a fatal stroke.; Mean follow-up was 33 ± 17 months with a minimum follow-up of 12 months.	A total of 89 deaths and 103 nonfatal strokes occurred during the trial. There was no significant difference in incidence of deaths and nonfatal stroke, transient cerebral ischemia or nonfatal MI between women assigned to estrogen relative to placebo meaning HRT is ineffective as secondary CHD prevention in older postmenopausal women suffering recent cerebrovascular disease. The study found an increase in the risk of stroke early after E2 initiation; however, this was not statistically significant in women adherent to study medication.; Furthermore, there were no differences in rates of VTE, breast cancer or hospitalization for fractures among the two groups.
The Postmenopausal Estrogen/Progestin Intervention (PEPI); Miller, [9]	To assess HRT on selected; cardiovascular risk factors in healthy, young postmenopausal women.; Primary outcomes: HDL-C systolic blood pressure serum insulin fibrinogen Secondary outcomes: waist-hip ratio endometrial hyperplasia and cancer hysterectomies breast cancer and other cancers CVD events VTE	N=875 healthy young postmenopausal women.; Age: 45-64 years (mean age 56 years).	3-year RCT.; (1) placebo; (2) CEE 0.625 mg/d; (3) CEE 0.625 mg/d + MPA 10 mg/d, days 1–12; (4) CEE 0.625 mg/d + MPA 2.5 mg/d; (5) CEE 0.625 mg/d + MP 200 mg/d, days 1–12.; Adherence: 97%.	Women randomly assigned to any active treatment group had a significant increase in HDL-C relative to placebo. Average increases in HDL-C were similar in the CEE (5.6 mg/dl) and CEE + MP (4.1 mg/dl) group and significantly higher than observed in the CEE + cyclic or continuous MPA groups (1.6 mg/dl and 1.2 mg/dl respectively). LDL-C decreased comparably across all active treatment groups by an average of 15.9 mg/dl across, significantly different from placebo. Triglycerides increased comparably and significantly in all active treatment groups (mean increase to 111 mg/dl from 98.3 mg/dl) relative to placebo. No significant differences were observed in systolic blood pressure and serum insulin levels however fasting glucose levels decreased significantly in all active treatment groups. Women assigned to placebo had greater increases in fibrinogen relative to women assigned to active treatment (0.10 g/L vs -0.02–0.06 d/L respectively).
The Postmenopausal Estrogen/Progestin Intervention (PEPI) regarding bone mineral density; The Writing Group for the PEPI Trial, [10]	To assess the effects of HRT on BMD at the spine and hip in postmenopausal women.; Primary outcome: BMD at baseline, 12 months, and 36 months.	See PEPI above.	See PEPI above.	Women assigned to placebo lost an average of 1.8% of spine BMD and 1.7% of hip BMD by 36-months. Women assigned to active regimens gained BMD at both sites; ranging 3.5% to 5.0% mean total increases in spinal BMD and a mean total increase of 1.7% of BMD in the hip. Changes in BMD were significantly greater in women assigned to active treatment relative to placebo. Older women, women with low initial BMD, and those with no previous hormone use gained significantly more bone than younger women, women with higher initial BMD, and those who had used hormones previously.
Kronos Early Estrogen Prevention Trial (KEEPS); Miller et al. [11]	The effects of HRT in recently postmenopausal women.; Primary outcome: Progression of subclinical atherosclerosis (measured by carotid artery intima-media thickness, CIMT).; Secondary outcomes: Other risk factors for cardiovascular disease (LDL-C, HDL-C, CAC).	N=727 healthy (free of subclinical CVD).; Recent menopause: within 6 months-3 years of natural menopause (mean of 1.4 years).; Age 42-58 years (mean of 52.6 years).	4-year RCT; (1) o-CEE 0.45 mg/d (N=230); (2) t-E2 50 μg/d (N=222); (3) placebo (N=275).; All groups received cyclic 200 mg MP/day 12 days/month.	CIMT increased similarly across all three groups (0.007 mm/y after 48 months). There was a non- significant trend for reduced accumulation of CAC with o-CEE and t-E2 relative to placebo. No significant changes in blood pressure across groups. No severe adverse effects including venous thrombosis. Other benefits found in both active treatment groups include significantly improved sleep (measured as sleep satisfaction and sleep latency, sleep disturbances only improved by t-E2) and maintenance of BMD compared with placebo.
KEEPS ancillary menopausal symptoms study; Santoro et al. [12]	Assessing menopausal symptoms at 6, 12, 24, 36, and 48 months post-randomization.	See KEEPS above.	See KEEPS above.	Both o-CEE and t-E2 treatment significantly reduced hot flashes (44% at baseline to 28.3% placebo; 7.4% t-E2; 4.2% o-CEE) and night sweats (35% at baseline to 19% placebo, 5.3% for t-E2; 4.7% for o-CEE) after six months with no significant difference between active treatment arms.
KEEPS ancillary sexual function study; Taylor et al. [13]	Assessing sexual function measured by the Female Sexual Function Inventory (FSFI) questionnaire.	N=670 of the 727 who participated in KEEPS.	See KEEPS above.	Treatment with t-E2 improved FSFI overall score relative to placebo, however in the individual domains of sexual function both t-E2 and o-CEE differed significantly compared with placebo.
KEEPS ancillary cognition study (KEEPS-Cog); Gleason et al. [14]	Whether HRT in recently postmenopausal women modifies risk of Alzheimer’s Disease (AD).; Primary outcomes: Mini Mental State Examination (MMSE). Four cognitive factors: verbal learning/memory, auditory attention/working memory, visual attention/executive function and speeded language/mental flexibility Mood: The Profile of Mood States.	N=693 of the 727 women that participated in KEEPS.	Avg. follow-up period: 2.85 years for cognitive outcomes and 2.76 years for mood outcomes.	Aβ deposition was lower in women assigned to t-E2 therapy relative to placebo. Not among women assigned to o-CEE. Stratifying women by APOEε4 genotype, it was found that the reduced Aβ deposition in the t-E2 group was particularly among APOE ε4 allele carriers, who are known to be at increased risk of AD. For mood, model estimates found that women treated with o-CEE had fewer symptoms of depression and anxiety compared to women on placebo. Mood outcomes for women randomized to t-E2 were similar to those for women on placebo.
KEEPS ancillary skin study; Owen et al. [15]	Examined effects of HRT on wrinkles (assessed on the face and neck) and skin rigidity (assessed in forehead and cheek) in postmenopausal women.	N=116 of the 727 who participated in KEEPS.	See KEEPS above.	Neither total wrinkle score nor total rigidity score was significantly different at baseline or over the 4-year follow-up among patients randomized to CEE, E2, or placebo. Skin wrinkle and rigidity scores were primarily affected by race/ethnicity.
ELITE (The Early vs Late Intervention Trial with Estradiol (ELITE); Hodis et al. [16]	To examine “the timing hypothesis” on the risk of CVD.; Primary outcome: The rate of change in CIMT measured 6-monthly during trial.; Secondary outcomes: Measures of CAC measured by cardiac-CT.	N=642 healthy, postmenopausal women stratified in: 1) Early menopause <6 years (avg 3.5 yrs, median age 55.4 years old).; 2) Late menopause >10 years after menopause (avg 14.3 years, median age 65.4 years old).	5-year RCT.; (1) Oral 17β-estradiol 1 mg/d; (2) placebo; Women with intact uterus: + cyclic MP vaginal gel or placebo.; Adherence: 98%.	After 5-year intervention: rate of CIMT progression was significantly lower in the early menopause group (-0.0044 mm/year) among women assigned to estradiol relative to placebo (-0.0078 mm/year). In the late menopause group CIMT progression didn’t differ significantly among groups (-0.0100 mm/year vs -0.0088 mm/year respectively). Findings were similar when adjusting for 80% adherence as adherence rate was high (mean 98% across groups). Findings were independent of hysterectomy status and use of MP. The effect of HRT was significantly lower in the early vs late postmenopausal group assessing the absolute value of CIMT after the 5-year intervention.; The CT measurements of CAC, total stenosis, and plaque didn’t differ between the estradiol and placebo groups within either postmenopausal stratum.
ELITE-Cog; Henderson et al. [17]	Effects of HRT on cognitive abilities.; Primary outcome: change in verbal episodic memory assessed at 2.5 and 5 years and compared between treatment groups measured by standardized composite of neuropsychological test scores.; Secondary outcomes: executive functions and global cognition measured by cognitive tests.	N=567 after 2.5 years.; N=455 after 5 years.	See ELITE above.	For verbal memory, the mean estradiol minus placebo standardized difference in composite scores (HR −0.06; 95%-CI: −0.22–0.09) was not significant. Differences were similar in early and late postmenopause groups. Interactions between postmenopause groups and differences between treatment groups were not significant for executive functions or global cognition. Meaning estradiol neither benefits nor harms these cognitive abilities regardless of time since menopause.
The Danish Osteoporosis Prevention Study (DOPS); Schierbeck et al. [18]	To examine the long-term effects of HRT in healthy, recently postmenopausal women.; Primary outcomes: mortality admission to hospital, myocardial infarction, or heart failure. Secondary outcomes: admission to hospital for stroke.; Safety endpoints: death, diagnosis of breast cancer or other cancer, admission to hospital for pulmonary embolism or deep vein thrombosis.	N=1,006 healthy, recently postmenopausal women.; Age: 45–58 years (median 49.7 years).	10.1-year RCT and subsequent unblinded follow-up for 5.7 years. 15.8 years overall.; After a mean duration of 10.1 years, women randomized to treatment were encouraged to discontinue HRT August 1st, 2002, following adverse outcomes in the WHI.; 1) 2 mg 17β-estradiol/d (hysterectomy); 2) 2 mg 17β-estradiol/d 12 d/month, 2 mg estradiol/d + 1 mg NETA for 10 d/month and 1mg 17β-estradiol/d 6 days/month (intact uterus); 3) placebo	16 women in the treatment group suffered from primary composite endpoint compared to 33 in the control group (HR 0.48, 95%-CI: 0.26–0.87) and 15 died relative to 26 (HR 0.57, 95%-CI: 0.30–1.08). After 16 years of total follow-up, the reduction in the primary endpoint was still present. Fewer women died relative to placebo however this was not significant (HR 0.66, 95%-CI: 0.41–1.08).; There was no apparent increase in risk of cancer (any cancer incidence: 36 in treated group vs 39 in control group, HR 0.92; 95%-CI: 0.58–1.45; breast cancer: 10 in treated group vs 17 in control group, HR 0.58; 95%-CI: 0.27–1.27), VTE (HR (2 in treated group vs 1 in control group) was 2.01 (95%-CI: 0.18-22.16)), or stroke (HR 11 in treated group vs 14 in control group) was 0.77 (95%-CI: 0.35–1.70). Younger women (<50 years) receiving HRT had a significantly reduced risk of breast cancer relative to control group (HR 0.49, 95%-CI 0.28–0.87) and women undergoing hysterectomy (N=192) receiving estrogen alone (HR 0.42, 95%-CI 0.18–0.97).; There were no statistically significant differences in stroke rates, rates of DVT or PE. The incidents were generally low (9 in HRT-group and 4 in placebo respectively).
DOPS, bones; Mosekilde et al. [19]	To assess whether HRT in recently postmenopausal women can be used as primary prevention of fractures.; Primary outcomes: BMD at the lumbar spine (L2-L4), the femoral neck and the forearm at inclusion, after 6 months, 1, 2, 3 and 5 years. X-rays of the spine (Th4 and L5) at inclusion and after 5 years.	N=2,016 healthy women.; Age: 45-58 years (median 50.8 years).	See DOPS above.; The study was originally two-folded: 1) a randomized arm, randomized to HRT (N=502) and placebo (N=504); 2) a non-randomized arm, assigned to HRT (N=221) and no HRT (N=789).	Risk of forearm fracture was significantly reduced (HR 0,45, 95%-CI 0.22–0.90) in women assigned to active treatment with HRT. Restricting the analysis to women who adhered to their initially assigned group of wither HRT (N=395) or placebo/no HRT (N=977), there was a significant reduction in overall risk of fracture (HR 0.61, 95%-CI 0.39–0.97) and the risk of forearm fractures (HR 0.24, 95%-CI 0.09–0.69).
REPLENISH; Liu et al. [20]; Kaunitz et al. [21]	To assess the efficacy and safety of a combined 17-β-estradiol and natural progesterone capsule in various doses vs placebo for reducing the frequency and severity of moderate to severe menopause related VMS in postmenopausal women and to evaluate endometrial safety.; Primary outcomes: mean change in frequency of moderate to severe VMS.; Safety endpoints: endometrial hyperplasia and changes from baseline breast examinations and mammograms.	N=1,825 healthy postmenopausal women with intact uterus who are seeking treatment for menopause-related VMS.; Age: 40-65 years (mean age 55).; Mean time since menopause 6 years.	1-year RCT.; Four combinations of estradiol/progesterone: (1) 1.0 mg/100 mg; (2) 0.5 mg/100 mg; (3) 0.5 mg/50 mg; (4) 0.25 mg/50 mg; Or placebo.	For women who received the E2/P4 1/100 or 0.5/100 dose, ∼80% had at least a 50% decrease and 68% and 58%, respectively, had at least a 75% decrease in their moderate to severe VMS, all of which were significantly greater than with placebo. After 12 weeks of treatment, women who received the 1/100 dose or the 0.5/100 dose had 3 days in mean per week with no moderate to severe VMS and 56% and 47%, respectively, of the women no longer had any severe VMS. The incidence of endometrial hyperplasia or malignancy was 0% with all four doses and placebo after 12 months. No increasing incidence of abnormal mammograms after one year of treatment (all but 8 (0.4%) mammograms at screening were normal).

Table A1.

A summary of the clinical trials examining the benefits and risks of HRT on age-related diseases in postmenopausal women.