Index
Contributors
Search

Whether to recommend hormone replacement therapy (HRT) to postmenopausal patients is a challenge facing many physicians. Equally difficult is an individual woman's decision to take or not to take HRT. This conflict will become more burdensome for physicians as the population ages. According to the US Census Bureau, in 1990, 18 million women were over the age of 65. By the year 2020, this number will increase to more than 25 million. The 3.5 million American women who become menopausal each year have an average life expectancy of an additional 30 years.

As life expectancy increases, women can expect to live a considerable portion of their lives after menopause. How well they live may depend on how their clinicians manage their hormone deficiency. However, many women obtain more information about menopause from the lay press than from their doctors. A survey conducted by the American College of Obstetricians and Gynecologists showed that many women are unaware of the increased risks of heart disease and osteoporosis after menopause1.

In part, this lack of information may explain the low rate of compliance with HRT. Up to 30% of women who are prescribed HRT fail to fill their prescriptions. Among women who do fill their prescriptions, only 40% continue taking their medication longer than one year2.

The findings of the Women’s Health Initiative study, a prospective, randomized trial of more than 16,000 healthy, post-menopausal women, published in July 2002, have thrown the use of HRT into question in both the medical and lay communities. The estrogen plus progestin arm of the study was recently halted because there was a small, increased risk of invasive breast cancer among women receiving the combined therapy, as well as heart attacks, stroke and clotting. These risks were not offset by the benefits: a decrease in colon cancer, and hip fractures (Reference 66).

However, the average age of the women in the WHI was 63.2 years, and does not reflect normal clinical practice where replacement is used mainly for symptoms including hot flashes, in women 10 to 30 years younger. Women in the WHI also had an average BMI of 28, one-third had hypertension, and one-half had a history of smoking (66). Thus, at the present time, the relative risk to benefit of using HRT in younger, healthier women is largely unknown, and physicians cannot make all clinical decisions by the WHI study, as it appears to apply specifically to the population studied. Hormone replacement is still an important therapeutic modality for women for symptoms and quality of life issues which deserves further study, and should be considered by physicians with their patients on an individual basis.

A more recent study on the effects of HRT is the Million Women’s Study (MWS), a British study that was published in Lancet in September 2003 [67]. The principal findings of the study were that current use of HRT, but not past use, is associated with an increase risk of breast cancer. The breast cancer increase with regard to ET alone was much less, and opens the door for consideration of this therapy as a meaningful alternative to EPT. This study, however, which focuses exclusively on breast cancer, does not address other effects of HRT including possible benefits.

This article will review the current state of knowledge concerning menopause and hormone replacement therapy. What is the effect of HRT on hot flashes, genitourinary tract atrophy, and other symptoms of the menopausal syndrome? Does HRT reduce a woman's risk of osteoporosis, cardiovascular disease, or cancer of the breast, endometrium, or colon? Can HRT slow the decline of cognitive function and prevent Alzheimer's disease? What recommendations should women be given in light of the Women’s Health Initiative findings?

THE MENOPAUSAL SYNDROME

Estrogen production during natural menopause does not stop abruptly. For five to seven years before the onset of the last menstrual period, ovarian function begins to diminish. This usually occurs when a woman is in her mid to late 40s and is characterized by irregular periods. While age at menopause ranges from 47 to 55 years (median age, 51 years), cigarette smoking or exposure to toxic chemicals (including chemotherapy) can induce menopause earlier in life.

After the last menstrual period, the ovary ceases to secrete estradiol. However, smaller amounts of a weaker estrogen, estrone, are still synthesized from androstenodione in the cortex of the adrenal gland and in the interstitial ovarian cells (in minor amounts)3. Small amounts of this estrone can be transformed into estradiol.

Body mass is directly correlated with the rate of peripheral production of estrone and estradiol in postmenopausal women. Estrogen synthesis takes place largely in adipose tissue. Therefore, an obese woman may produce twice as much estrone and estradiol as a thin woman. This may help explain the increased prevalence of hypoestrogenemia symptoms and the higher risk of osteoporosis observed in thin women4.

HOT FLASHES

Perhaps 75% of perimenopausal women experience hot flashes5. They occur with greatest frequency in the first two years after menopause and continue to decrease over time. However, for some women, the onset of hot flashes occurs before menopause, and for others they continue for 10 to 40 years after6. This, however, is uncommon.

Hot flashes occur with greater frequency in women who undergo surgical menopause than in those who experience natural menopause. They are also more common at night, which often results in sleep disturbances, fatigue and depression. Additional factors including warm environments, consumption of alcohol or caffeine, and stress can exacerbate hot flash occurrence.

A variety of studies are showing a decrease in the frequency and severity of hot flashes compared to placebo using the standard dose of conjugated equine estrogen (0.625mg/d CEE) as well as lower doses. A recent study, The Women’s Health, Osteoporosis, Progestin and Estrogen (HOPE) study, tested various doses and after 3 weeks, there was an improvement in the number of hot flashes in all HRT groups compared to placebo (68). Oral and transdermal therapy regimens have also proved successful in alleviating motor symptoms. In one comparative 12-week trial of 204 postmenopausal women (average age, 52 years), oral CEE 0.625 mg and transdermal estradiol 50 mg/day provided similar symptom relief (69).

HRT for relief of hot flashes is effective in up to 90% of menopausal women. However, many women still experience sleep disturbances. While some researchers attribute this complaint to hot flashes or night sweats, others have found that sleep problems persist after the hot flashes have been relieved. A double-blind, crossover study in hypogonadal postmenopausal women compared the effects of CEE (0.625 mg/d) and placebo on sleep patterns7. Results showed no relation between hot flashes or night sweats and sleep disturbances. For the women on estrogen, however, sleep quality improved, along with length of REM sleep and sleep latency.

With regard to mood, the data supports the theory that symptoms of depression may be alleviated with the use of ET/HRT. In a double-blind placebo controlled trial of perimenopausal women, Soares et al found that depressive symptoms were significantly relieved in women receiving estradiol compared to placebo (70). In addition, estrogen has been shown to improve mood in post menopausal women without clinical depression (71,72). Further research is still needed in this area but the preliminary data suggest that estrogen may be a possible treatment method for some depression symptoms.

GENITOURINARY TRACT ATROPHY

Large numbers of estrogen receptors are found in the vagina, vulva, urethra, and trigone of the bladder. Thus, atrophy of the genitourinary tract can occur as estrogen levels diminish.

After menopause, the vaginal walls thin and lose their elasticity. They also produce fewer secretions and lose much of their lubricating ability in response to sexual stimuli. The vulva becomes flattened and thin as a result of the loss of collagen, adipose tissue and the ability to retain water8. The urethra also becomes thinner and less efficient, with detrusor pressure at the urethral opening decreasing, both during and after voiding. Estrogen deficiency also leads to an increase in fibrosis of the bladder neck, reduced collagen in surrounding tissues, and a decrease in the number and diameter of the muscle fibers in the pelvic floor.

These changes increase a woman's risk of vaginal and urinary tract infection. Atrophic genitourinary tissues are also at increased risk of injury by trauma. Estrogen replacement therapy can significantly lessen these problems.

SEXUAL DYSFUNCTION

All of the changes to the genitourinary tract can result in dyspareunia, leading to a decreased interest in sexual intercourse. Fatigue and depression brought on by the vasomotor symptoms and sleep disturbances of menopause can exacerbate this lack of interest in coitus.

Decreased levels of endogenous testosterone, both in women who have undergone surgical menopause, as well as in those who experience natural menopause, may cause decreased libido9. Women who complain of lack of sex drive may be candidates for androgen replacement, as well as estrogen. In general, androgen levels do not decrease abruptly at menopause but decrease gradually as women age so that decreased libido may be a problem of older postmenopausal women.

OSTEOPOROSIS

The loss of ovarian hormone production after menopause puts women at increased risk for osteoporosis. Without estrogen, osteoclast activity and bone resorption are increased, and bone formation decreases. The consequent reduction in skeletal mass and the microarchitectural skeletal deterioration increase the risk of fracture. At age 50, a Caucasian woman has a 15% lifetime risk for hip fracture. By the age of 85, 33% of white women will have fractured a hip.

Bone loss, however, begins before menopause. Children and young women gradually increase skeletal mass until it peaks around the age of 30, when it begins to decrease10. Thus, by the time a woman reaches menopause, she may have already lost up to 30% of her bone mass11. The rate of bone loss increases to 4% to 5% per year, for 5 to 10 years postmenopause. Over the subsequent 10 to 20 years, it begins to slow to the baseline rate.

The first-line therapy for postmenopausal osteoporosis prophylaxis is estrogen. Studies have shown it to not only prevent bone loss (by decreasing osteoclastic activity), but also to reduce fracture rates by as much as 65%12. Women in the Women’s Health Initiative study receiving estrogen plus progestin suffered 5 fewer hip fractures per 10,000 compared to women on placebo. (66) Prophylactic benefit increases when estrogen replacement is begun as soon after menopause as possible. However, there is no upper age limit for starting replacement therapy; women 70 years or old may stand to benefit13.

Because bone loss continues as soon as estrogen replacement is stopped, treatment should be continued indefinitely so as to maintain the positive effects on bone metabolism. However, the longer a woman has been taking estrogen, the more bone she will have when treatment is stopped and bone loss resumes.

While estrogen prevents bone loss in most postmenopausal women, some continue to lose bone mass despite the therapy because of a genetic predisposition or environmental factors. Bone density studies should be conducted during the perimenopausal period and then repeated as needed to assess the status of bone loss14.

As an adjunct to estrogen therapy, all postmenopausal women should receive calcium (1500 mg) and vitamin D (400 IU) supplementation. This can help compensate for poor dietary intake of calcium and inefficient vitamin D synthesis. Note that because calcium carbonate requires acid for absorption, women taking acid-suppressing drugs, or with atrophic gastritis should take calcium citrate, which does not require gastric acid for absorption.

CARDIOVASCULAR DISEASE

Cardiovascular disease accounted for 30.7% of deaths in American women in 1999. It surpasses cancer, cerebrovascular disease, lung disorders, infectious disease, diabetes, suicide, and renal disease as the leading cause of death in women today15. A woman has about 10 times the lifetime risk of dying of ischemic heart disease than of breast cancer, reproductive cancer, or osteoporotic fracture.

An acceleration of heart disease occurs after age 50 and approximately one third of the women who die of cardiovascular disease every year are under 65 years old (more than 100,000). This suggests that menopause (whether surgical, premature, or natural) may be a risk factor for heart disease16. Because premenopausal women have lower incidences of cardiovascular disease than men and lose this advantage after menopause, it is logical to conclude that estrogen has a cardioprotective effect. It is thought that estrogen deficiency is at least partially responsible for the increased risk of developing heart disease after menopause. Treatment with estrogen replacement therapy reduces the relative risk of cardiovascular disease in women after menopause by about 50%.

Considerable controversy and confusion has recently erupted over the role of estrogen replacement therapy in preventing cardiovascular disease. A number of trials reported an increased risk of ischemic events when hormone therapy was started in older women with a history of heart disease17-19. In response, the American Heart Association recommended that hormone replacement therapy not be used for primary prevention of cardiovascular disease. The results of the estrogen-progestin arm of the WHI showed similar results. Women on estrogen-progestin therapy suffered 7 more CHD events per 10,000 women than women on placebo. They also suffered 8 more strokes per 10,000 women than those taking placebo. (66)

However, there is very plausible biological evidence that hormones do prevent heart disease. Emerging evidence suggests hormone therapy is most effective in protecting women whose hearts are not yet compromised from future cardiovascular disease as seen in a recent study by Hodis et al.20. Researchers randomized 222 postmenopausal women with no history of cardiovascular disease, stroke, or cancer who had high levels of LDL (= 130 mg/dL) to receive either 1 mg unopposed 17-ß estradiol or placebo. After two years, women on estrogen had significantly less thickening of the inner carotid artery wall. However, at the present time, HRT should not be recommended for the prevention of heart disease.

ALTERED LIPOPROTEIN PROFILES

One mechanism which may explain the increased risk of cardiovascular disease after menopause are the atherogenic changes in plasma lipoprotein levels associated with estrogen deficiency. At menopause, plasma levels of low-density lipoprotein (LDL) increase by 10% to 15%. This increase can be prevented with estrogen replacement therapy. Plasma levels of high-density lipoprotein (HDL) increase by 10% to 15% with estrogen therapy and may be an important factor in the cardioprotective effect of estrogen.

The use of progestins, however, in conjunction with estrogen seems to attenuate these beneficial effects on plasma lipoprotein levels to some extent. Data from the Nurses' Health Study showed that women who took estrogen and progestin in combination had the same apparent protection from coronary events as did the women who took estrogen alone19.

However, as mentioned above, a randomized trial showed that HRT (with 0.625 mg/d of conjugated equine estrogen and 2.5 mg/d of medroxyprogesterone acetate) increased the risk of coronary events in women with a mean age of 65 who had established cardiovascular disease21. This effect was noted during the first year of HRT use. Following the second year, a progressive protective trend was found with HRT, although there was no overall beneficial effect in the study as a whole. Another study examined the effect of HRT/ERT women with ERT22. Again, no benefit was seen. However, these women had proven heart disease and were, on average, 65 years of age. This is considerably older than the age when HRT is usually started. These data suggest that HRT may need to be started prior to the development of cardiovascular disease.

Progestins may have variable effects on lipoproteins based on their androgenicity. More androgenic progestins tend to lower HDL levels to a greater degree than do the less androgenic progestins23. The two types of progestins most commonly used for hormone replacement therapy are those derived from 19-norestosterone and 17-hydroxyprogesterone. The former are the more androgenic, while the latter have a little androgenicity. Medroxyprogesterone is the most commonly prescribed progestin in the United States and is derived from 17-hydroxyprogesterone.

More recently, micronized progesterone has become available. The Postmenopausal Estrogen/Progestin Interventions Trial (PEPI) showed that micronized progesterone, used with conjugated equine estrogen, had less attenuation of the favorable lipid profile induced by estrogen than medroxyprogesterone acetate24.

VASODILATION

As important as estrogen's effects on lipid metabolism may be its vasodilatory properties. It appears to potentiate the effects of endothelium-derived relaxing factor (EDRF) in the coronary arteries. It also may affect vasodilation through an endothelium-independent pathway in the peripheral vasculature.

One study looking at postmenopausal women with angina and normal coronary arteries (syndrome X) saw diminished vasodilation before initiation of estrogen therapy and normalized hyperemic response after two months of treatment25. Vasodilation was measured by testing hyperemic response to forearm blood flow occlusion. Chest pain either improved markedly, or resolved, in 19 of the 20 subjects. This improvement in angina symptoms suggests that the impaired vasodilatory response to an EDRF/nitric oxide stimulus may be systemic.

An additional study reported a beneficial effect for sublingual estradiol in reducing symptoms of exercise-induced myocardial ischemia in 11postmenopausal women with coronary artery disease26. These results suggest both a reduction in peripheral vascular resistance and a direct vasodilatory effect in the coronary arteries.

OTHER EFFECTS

Additional studies have found an association between HRT and a marked reduction in the pulsatility index of the internal carotid and middle cerebral arteries with HRT27. This finding may help explain the reduction in stroke risk and the improvement in cognitive function seen with HRT using estrogen plus progesterone. According to recent data from the Nurses' Health Study, this effect is seen at low doses only (0.3 mg conjugated equine estrogen).

A recent study also looked at the effects of estrogen replacement in women who had recently suffered ischemic stroke or transient ischemic attacks and found no reduced mortality or recurrence prevention with 1.0 mg estradiol compared to placebo28. These findings discourage the use of HRT for secondary stroke prevention.

Other factors associated with estrogen-use which could lower the risk for cardiovascular disease include decreases in levels of the proatherosclerotic factor, lipoprotein(a), the procoagulant factor, fibrinogen, and increases in levels of factor 11 (prothrombin).

BREAST CANCER

One in 8 women will be diagnosed with breast cancer and risk increases with age29. In 2001, approximately 40,200 women died of breast cancer, although survival rates have been increasing. The five-year survival rate for women with localized breast cancer has risen from 72% in the 1940s to 97% today. This high survival rate, however, decreases to 77% if the cancer has spread regionally, and to 21% if it has spread distantly30.

Estrogen, a trophic growth hormone, may promote the growth of preexisting breast cancer. It is still unknown whether it may also induce the growth of new cancers. Use of estrogen alone for at least five years, may be associated with a slightly increased risk of breast cancer according to the Nurses' Health Study. However, in the recent report from the Women’s Health Initiative study, women on estrogen only showed no increased incidences of breast cancer compared to women on placebo (REFERENCE 66). The study is ongoing and future reports should clarify this issue.

Many studies have not shown an increased risk of breast cancer with estrogen use. A large meta-analysis of 51 epidemiologic studies (involving more than 160,000 women from 21 countries) showed that HRT increases the risk of breast cancer and that risk increases with longer use31. That is, for every 1,000 women who began using HRT at age 50 and continued using it for 5, 10, or 15 years, an additional 2, 6, or 12 cases of breast cancer would be expected to occur. However, another review showed that at doses of 0.625 mg/d conjugated estrogens, there was no increased risk of breast cancer32.

Recent data from the Iowa Women's Health Study showed no increased risk of breast cancer in women who had used HRT versus those who had not taken hormones33. Additionally, when researchers went back and analyzed data from women who had developed breast cancer, they found that HRT, in a very small number of women, was associated with cancer with a favorable prognosis34. This finding is supported by other studies which have shown that women who use HRT are less likely to have metastatic disease, and have a longer life expectancy than women who haven't used HRT35. The findings of these studies suggest that rather than acting as a carcinogen, estrogen may act as a mitogen. However, one possible explanation for these findings is that women on HRT are more likely to be see a doctor regularly and to undergo regular breast examinations and mammograms.

Data from the Nurses' Health Study showed a survival advantage for women taking estrogen at the time their breast cancer was diagnosed. The increased survival rate was associated with a lower frequency of late-stage disease and undoubtedly reflects earlier diagnosis in estrogen users36,37. However, other evidence suggests that estrogen users develop better differentiated tumors and that surveillance or detection bias is not the only explanation for better survival38.

A number of recent studies have aroused concern over the effect of hormone replacement therapy on breast tissue density. In women not on hormone replacement therapy, breast density has been found to be an independent risk factor for breast cancer39. Hormone replacement therapy has been found to increase breast density, with the greatest increase in women on conjugated estrogen and progesterone40.

Although the association between breast density and breast cancer has not been seen in women on HRT, there has been some concern that mammograms may be less effective in women on HRT with greater breast density. However, recently Rutter et al. showed that two weeks after discontinuing HRT, women's breast density returned to normal41. Therefore, until this issue is better understood, it may be advisable for women to discontinue HRT for two weeks before a mammogram exam, especially in the case of problematic mammograms.

Recent evidence suggests, however, that estrogen plus progestin may have an impact on breast cancer. In July 2002, the estrogen plus progestin arm of the Women’s Health Initiative study was stopped due to a small increase in the incidence of breast cancer among women taking this combination. This risk amounted to approximately 8 more women per 10,000 being diagnosed with breast cancer compared to those on placebo. (66) It is important to note, however, that the average age of women in this study was 63.2 years, and does not reflect women on HRT in normal clinical practice. In addition, 50% of the women in WHI were either current or former smokers, they had an average BMI of 28 (well-above normal), and 1/3 suffered from hypertension.

In the recent MWS, the large British study, women on HRT followed for 2.6 years were found to have an increased risk of breast cancer (RR 1.66) [67]. Various hormone preparations were tested in this trial and similar risks were reported for all types, suggesting that risks are not confined to the standard CEE/MPA dose used in WHI. It is important to note though, that women taking estrogen only had a significantly lower increase in risk compared with women taking HRT. While these statistics are more compelling than those of WHI, they are observational studies and hence have a larger potential area for error.

Although there is some evidence that combination therapy may increase risk of breast cancer above that of estrogen alone, neither a protective, nor a detrimental effect has been demonstrated convincingly, particularly for younger, healthier women. One study interviewed nearly 4000 women with and without breast cancer and found a significant correlation between use of continuous combined replacement therapy and breast cancer42. However, the risks were higher in thin women than in heavier women which may confound the results. Also, it is possible that the use of cyclic therapy could provide the additional risk, and it was generally given at higher doses rarely used today.

While there has been little consistency among the findings of the various studies on the effects of ET/HRT on breast cancer, one issue that is consistent in the literature is the observation that mortality from breast cancer is decreased among ET/HRT users. A summary of the literature from 1990-2001 shows the RR of mortality consistently to be <1.0 with ET/HRT use [75-80]. One hypothesis to explain this observation is that ET/HRT may promote the development of slow-growing tumors or discourage the development of more aggressive tumors. Holli et al, reported that tumors in women taking ET/HRT were smaller, had a better histologic differentiation, an a lower cell-proliferation rate compared to nonusers[81]. It has also been posited that better screening of these women leads to lower mortality rates.

The argument that postmenopausal HRT should not be given to women who have a personal history of breast cancer may seem reasonable based on evidence that breast cancer is a hormone responsive tumor. However, while women with a first-degree relative (mother, sister, or daughter) who has or had premenopausal breast cancer are at increased risk by virtue of their family history alone, their risk of breast cancer is not thought to be increased further by HRT use. Eighty percent of women who develop breast cancer do not have a family history. Sellers et al, examined ET/HRT use and breast cancer risk in women with a family history of breast cancer and found no statistically significant increase in risk in past or current users, regardless of duration of use [82]. This is supported by the findings of Rebbeck et al, who studied women who were carriers of the BRCA1 gene mutation [83]. Bilateral prophylactic oophorectomy was associated with a 47% reduction in breast cancer risk in this population. HRT use did not negate the observed reduction in cancer risk. Interestingly, studies of breast cancer survivors showed that women using ET/HRT had a lower risk of recurrence compared to survivors not using ET/HRT [84,85].

Breast cancer prognosis does not appear to be influenced by the high hormone levels during pregnancy, nor has oral contraceptive use been shown to increase breast cancer risk. These observations may allay some of the fear regarding the use of exogenous hormones after menopause.

ENDOMETRIAL CANCER

In 2001, 38,300 cases of endometrial cancer were diagnosed, and 6,600 women died of the disease30. The mean age at diagnosis is 61 years, with most cases occurring in women 50 to 59 years old.

Estrogen alone, causes endometrial hyperplasia and a two to three-fold increase in the risk of endometrial cancer. However, the addition of progesterone reduces this risk to lower levels than those seen in women not on HRT43,44. Thus, the addition of a progestational agent to postmenopausal estrogen therapy is now standard for women with an intact uterus. While there have been some reports that the risk of endometrial cancer may be slightly increased even with the combined therapy, most studies have not confirmed this. Women in the WHI study on combined therapy showed no difference in endometrial cancer rates compared to women on placebo (66). Recent research has focused on the use of lower doses of estrogen and a progestogen in HRT to reduce the risk of endometrial cancer45.

The dose of progestogen given depends on several factors, including the number of days given each month, the amount of estrogen given, the individual needs of the patient, and her ability to tolerate the medication. Side effects of progestogen can include anxiety, irritability, depressed mood, acne, bloating, fluid retention, headaches, breast tenderness, and bleeding problems. The inability to tolerate these effects is the main reason for poor compliance or discontinuation of HRT.

The potential association between ovarian cancer and HRT is currently unclear. While WHI researchers reported an increased risk of ovarian cancer (HR 1.58), it did not reach statistical significance [86]. Other studies too, including HERS [87] and a meta-analysis of 15 case-controlled studies [88], found no significant association.

COLON CANCER

Despite being one of the major causes of cancer-related mortality in women, colon cancer is often overlooked by patients in their risk assessment of HRT. Case-controlled and cohort studies have both found a 50% decrease in relative risk of colon cancer in women who are current or long-term HRT users compared to women not on HRT. In addition, reports from the WHI study showed that the combined estrogen plus progestin therapy was associated with a decrease in the incidences of colon cancer compared to women on placebo (6 fewer cases per 10,000 women on HRT)(66). Although the exact mechanism of estrogen’s protective effect on the colon is unclear, it has been suggested that estrogen acts to decrease bile acids which are thought to be carcinogenic. At present, however, although the evidence that HRT may be beneficial in reducing the risk of colon cancer should be considered, there is insufficient evidence to warrant recommending long-term HRT solely for this purpose.

NEUROLOGIC FUNCTION

Cognition

The existence of estrogen receptors in the hippocampus, a part of the brain essential to learning and memory, has been known for some time46. Several mechanisms may account for the effects of estrogen on the brain. Firstly, estrogen increases levels of choline O-acetyl-transferase, the enzyme needed to synthesize acetylcholine, a neurotransmitter thought to be critical for memory47. Studies on healthy middle-aged and elderly postmenopausal women have supported the theory that estrogen may help to maintain aspects of cognitive function48,49. Data also suggest that estrogen replacement therapy may enhance short- and long-term memory50,51. Additional effects of estrogen on neural function include: protecting neurons from oxidative stress and glutamate toxicity [89,90], increasing glucose transport and cerebral blood flow [91], and stimulating the branching of neurites [91].

Alzheimer's Disease

For every five years after the age of 65, the prevalence of Alzheimer's disease doubles in the population. Nearly 50% of women over the age of 75 may suffer from the condition52. As the population ages over the next 20 years, these numbers are expected to increase.

According to epidemiologic evidence, there is reason to believe that estrogen deficiency may contribute to Alzheimer's disease. Low body weight is associated with low levels of circulating estrogens in postmenopausal women. Women who suffer from Alzheimer's disease tend to have lower body weights than women without the disorder53. Incidences of Alzheimer's disease are low or its expression is delayed in postmenopausal women with high levels of endogenous estrogenic steroids or those receiving long-term HRT.

One explanation for estrogen's apparent protective effect may involve neurotransmission. Estrogen acts as a trophic factor for cholinergic neurons in vitro. Cholinergic depletion is the most prominent neurotransmitter deficit in Alzheimer's disease.

With regard to the association between risk of Alzheimer’s Disease and ET/HRT use, however, there is little consistency in the literature. However, while HRT does show promise in preventing or delaying the onset of the disease, a recent study showed no benefit of either 0.625 mg/d or 1.25 mg/d of estrogen on Alzheimer's progression54. Most likely, estrogen may merely delay the deterioration seen in Alzheimer's patients. Paganini-Hill and Henderson [92] reported a 35% decrease in risk for ET users compared to placebo and Zandi et al [93] reported a 41% reduced risk for ever users of HRT. However, the results from the WHIMS, the Women’s Initiative Memory Study, a substudy of WHI, reported that while HRT did not significantly increase the risk of mild cognitive impairment (HR 1.07), it did increase the risk of probable dementia (HR 2.05) [94]. The effect of HRT on different subtypes of dementia could not be determined because the number of cases was too small. It must be noted, however, that because the WHIMS participants were all 65 or older, these results may not apply to women who initiate HRT at a younger age.

The results of the Cache County Study [95] serve to further confuse the issue. In this prospective study of incident dementia in older women (mean age 74.5 years), the risk of AD was increased in current HRT users with 10 or fewer years of therapy (HR 2.41 for fewer than 3 years of therapy, 2.12 for 3-10 years). For current users with more than 10 years of therapy the HR was .55, indicating a decrease in risk, but this value did not reach statistical significance. Interestingly, in past users, reductions were present in all age groups and showed a duration effect (HR .58 for fewer than 3 years, .32 for 3-10 years, and .17 for more than 10 years).

OTHER POSSIBLE RISKS

Thromboembolic disease

The Nurses' Health Study showed a twofold increase in the risk of pulmonary embolism among postmenopausal women who were current estrogen users. The recent findings of the WHI study confirmed these findings for women on combined estrogen plus progestin therapy. Women on this treatment suffered 8 more pulmonary emboli per 10,000 than women on placebo. (66) Although estrogen use has been associated with an increase in the relative risk of venous thromboembolism (VTE), the absolute risk remains low, as VTE occurs infrequently in this setting. Women on combined estrogen-progestin therapy in the WHI study suffered 18 more venous thromboembolims than women on placebo. (66) However, when considered against a 50% reduction in cardiovascular disease risk, the increased risk of VTE does not contraindicate estrogen replacement. It does, however, show that patients should be screened for a history of idiopathic thrombosis55.

Gallbladder disease

Some epidemiologic studies have found an increased risk of gallstones among women who use HRT. Estrogen has been shown to increase cholesterol saturation of bile, alter bile acid composition, and decrease bile flow. Each of these effects can enhance gallstone formation. Data from the Nurses' Health Study (54,845 postmenopausal women monitored for eight years) showed that current HRT users were more likely to have undergone cholecystectomy than nonusers (relative risk, 2.1). This risk tends to increase with long-term therapy and with high doses of estrogen56.

Weight gain

Because many women gain weight as they age, a common fear is that HRT will exacerbate this problem. However, this is unconfirmed by prospective studies. The PEPI trial showed that women on HRT gained less weight than women not taking hormones25. Attention to diet (with reduced fat intake) and regular aerobic exercise for weight maintenance should be recommended to all postmenopausal women.

OTHER POSSIBLE BENEFITS

Age-related macular degeneration

About 35% of patients over the age of 75 are affected by macular degeneration, the leading cause of severe vision loss in the elderly. One study showed that women who experienced menopause earlier in life had a 90% increased risk of developing symptoms of age-related macular degeneration later in life as compared to women who underwent menopause at an older age.(57) Some studies have shown a small reduction in the incidence of this eye disorder among users of HRT58,59.

Skin

It is thought that skin may be an important target organ for reproductive hormones. In postmenopausal women, dermal collagen decreases, and skin becomes thinner60. Applying estrogen cream to the skin after menopause improves the external appearance of facial skin. In addition, systemic HRT increases dermal collagen and limits age-related skin extensibility61. To date, of the eleven clinical trials that examined the effect of HRT on collagen levels, only one failed to demonstrate efficacy [96]. Furthermore, results from a recent study indicates that estrogen also increases skin thickness [97].

HRT has also been shown to accelerate cutaneous wound healing, both microscopically and macroscopically, in postmenopausal women 62. This study also showed delayed repair of acute incisional wounds in ovariectomized young female rodents; the delay was reversed by the topical application of estrogen.

Tooth loss

The risk of tooth loss increases after menopause. Osteoporosis, as well as estrogen deficiency, could both be contributing to this effect63. Data from the Nurses' Health Study indicate that the risk of tooth loss may be decreased in women with a history of estrogen therapy 64.

The results of the Women's Health Initiative Study are expected to answer many of the outstanding questions about HRT for postmenopausal women. However, they will not be available until 2006 or later 65.

TREATMENT

When assessing risk vs. benefit for long-term risks, the following from the WHI should be taken into consideration:

TREATMENT GUIDELINES  

When assessing risk vs. benefit for long-term risks, the following conclusions from the WHI should be taken into consideration (66):