Fig. 9.1
The STRAW + 10 staging system . Reproduced with permission from Harlow SD, Gass M, Hall J, Lobo R, Maki P, Rebar RW, et al. Executive summary of stages of reproductive aging workshop + 10: addressing the unfinished agenda of staging reproductive aging. Menopause 2012;19(4):387–95
Conceptually, the adult life of a woman can be divided into three major periods: reproductive, menopausal transition, and postmenopause. As is noted in ◘ Fig. 9.1 the anchor for the staging system is the final menstrual period (FMP ). For clinical research purposes, five stages occur before and two after this anchor point. Stages −5 to −3 cover the reproductive period; stages −2 and −1 are the menopausal transition; and stages 1 and 2 are the postmenopause.
Among the most concrete achievements of the STRAW workshops was the development of clear and specific nomenclature that was previously vague and confusing in the literature. The authors of the STRAW workshops recognize that this is an evolving field and these concepts will change as the knowledge advances
9.3 Menopausal Transition
Stage −2 (early) and −1 (late) encompass the menopausal transition and are defined by menstrual cycle and endocrine changes. The menopausal transition begins with variation in menstrual cycle length in a woman who has a monotropic FSH rise and ends with the FMP (not able to be recognized until after 12 months of amenorrhea). For most women at these stages, day 3 follicular phase FSH levels are variable and anti-Müllerian hormone levels and antral follicle counts are low.
9.4 Postmenopause
Stage +1 (early) and +2 (late) encompass the postmenopause. The early postmenopause is defined as 5–6 years since the FMP . The participants agreed this interval is relevant because it encompasses a further dampening of ovarian hormone function to a permanent level as well as accelerated bone loss. Stage +1 was further subdivided into segment “a,” the first 12 months after the FMP , and segments “b and c,” the next 1–5 years. Stage +2 has a definite beginning, but its duration varies, since it ends with death. Further divisions may be warranted as women live longer and more information is accumulated.
9.5 Perimenopause
Perimenopause literally means “about or around the menopause.” It begins with stage −2 and ends 12 months after the FMP . “The climacteric” is a popular but vague term that we recommend be used synonymously with perimenopause. Generally speaking, the terms perimenopause and climacteric should be used only with patients and in the lay press and not in scientific papers.
9.6 Menopause Physiology and Pathophysiology
The supply of primordial follicles in the female gonad is predetermined before birth and diminishes with age until it is unable to provide enough mature follicles to sustain menstrual cyclicity [8]. The peak number of germ cell count is found at 20 weeks of gestation, with decreased numbers at birth and puberty. Based on the numbers of follicles at three successive stages of development which were obtained by counting follicles in histological sections of ovaries from 52 normal women, a mathematical model was developed to describe the rates of growth and death of ovarian follicles (atresia) in human ovaries between ages 19 and 50 [9]. While the number of oocytes dwindles throughout a woman’s life [10], there seems to be a transition at age 38 when the rate of follicle disappearance is augmented considerably with age. As a consequence, an estimated total of 1500 follicles remained at 50 years of age from the 300,000 present at age 19 years. This rate of decline in small follicles appears to be responsible for ovarian failure, and therefore menopause, and the transition to midlife in our species .
Recently, these concepts have been modified by Tilly and coworkers, who observed in the mouse model that there is a pool of germ cells that can proliferate and sustain oocyte and primordial follicle development in the postnatal mammalian ovary [11]. Moreover, in a chemotherapy-induced ovarian failure mouse model, his group was able to demonstrate that bone marrow transplantation is associated with a new population of immature oocytes [12]. Taken together, these new observations suggest that the control mechanism(s) of oocyte atresia can be potentially modified.
9.7 Age of Menopause
The mean age of menopause in normal women in the USA ranges between 50 and 52 years [1]. This number was based on a cross-sectional study, which is associated with recall bias. However, these findings have been confirmed in a large prospective cohort study of mid-aged women: The Study of Women’s Health Across the Nation (SWAN) cohort consisting of African-American (n = 916), Caucasians (n = 1533), Chinese (n = 248), Hispanic (n = 277), and Japanese (n = 279) women from premenopause to their final menstrual period. The median age of menopause was 52.54 years and was similar for all five ethnic groups [13].
In other parts of the world, population-based surveys have shown an earlier onset of menopause. For example, the median age of onset of menopause was 48 in a survey of 742 United Arab Emirates women [14]. In certain regions of India, such as the state of Himachal Pradesh, the mean age onset of menopause can be as low as 43.5 years, according to data from 500 postmenopausal women [15].
9.8 Factors Affecting the Onset of Menopause
◘ Table 9.1 includes several factors that have been linked with an earlier onset of menopause. Among these factors, smoking has been the most often linked environmental agent to early age onset of menopause. However, these studies have been inconclusive with regard to duration and intensity of smoking. Midgette et al. concluded after a review of 14 studies that the risk of being menopausal was approximately doubled for current smokers compared with nonsmokers among women 44–55 years of age [16]. However, in 2004 van Asselt et al., using data from a Dutch population-based cohort of 5544 women, assessed the effect of smoking duration and intensity on age at menopause correcting for the chronologic age-dependency of the variables concerned. After their modeling, they concluded, like previous researchers, that smoking lowers the menopausal age. However, the reduction in the menopausal age appears not to be dependent on smoking duration and that smoking cigarettes could have an effect only around the time of menopause itself [17]. In other words, the number of cigarettes smoked during perimenopause is apparently more significant than smoking history as the culprit of earlier age of onset of menopause among smoking women.
Table 9.1
Factors associated with earlier onset of menopause
Smoking |
Genetic factors |
Family history of early menopause |
Pelvic surgery |
Total abdominal hysterectomy |
Unilateral oophorectomy Ovarian cystectomy |
Metabolic factors |
Type 1 diabetes mellitus |
Galactose consumption |
Galactose-1-phosphate uridyl transferase deficiency |
Ovulation patterns |
Nulliparity |
Shorter menstrual cycles |
Non-use of birth control pills |
Women treated with total abdominal hysterectomy appear to be at risk of early menopause. Concurrent unilateral oophorectomy or ovarian cystectomy was associated with an even earlier onset. However, previous tubal ligation does not influence the age of menopause.
9.9 Time Course of Oocyte Pool Depletion
It is estimated that the entire process from initiation of the primordial follicle growth until complete maturation and finally ovulation is over 150 days [18]. The majority of follicles will experience atresia by apoptosis at some point in their developmental process [19]. Fifty percent of apoptosis occurs at the small antral follicle stage of 2.1–5 mm. On the other hand, the atresia of the resting follicles in the human fetus seems to be set off by a process of necrosis rather than apoptosis [20]. It seems clear now that the age-related decline in ovarian function in women is the result of the decline in both quantity and quality of the resting ovarian follicle pool. Recently, a total of 182 resting follicles from a young cadre of women (25–32 years) was compared with 81 resting follicles from the less young group (38–45 years) for signs of age-related changes by transmission-electron microscopy. De Bruin et al. concluded that, in resting follicles, the morphological changes with age are different from the changes seen in quality decline by atresia. The morphological changes with age specifically comprise the mitochondria, the dilated smooth endoplasmic reticulum, and the Golgi complex.
9.10 Genetic Contribution
The age of natural menopause is determined by the interplay of genetic and environmental factors [21]. There are some cross-sectional [22] and case–control [23] population studies suggesting the existence of genetic variability in the age of menopause to be as high as 70%. However, a study conducted in the Netherlands in 164 mother–daughter pairs with a natural menopausal age estimated a hereditability of 44% (95% CI 36%, 50%). The authors conclude that these estimates are more accurate because previous studies that were done in twins and siblings are overestimates because siblings shared many environmental factors [24].
9.11 Clinical Signs and Symptoms of Menopause
The diagnosis of menopause can be established when the absolute level of serum FSH is elevated. The threshold for the diagnosis of menopause will vary according to the assay employed. In any event, the level will be two standard deviations above from the normal value of a reproductive-age woman on cycle day 3. Anti-Müllerian hormone levels will typically be very low and may be undetectable. LH is of little value in the evaluation or diagnosis of menopause.
The clinical diagnosis of menopause is established retrospectively once a patient has had more than 12 months of amenorrhea in conjunction with vasomotor symptoms such as hot flashes and headaches. At this point, the patient has made the transition in the STRAW classification from −1 to +1. The range of symptoms due to immediate estrogen deficiency in women during STRAW stages −1 to +1 includes hot flashes and urogenital changes.
9.12 Hot Flashes
Vasomotor symptoms (hot flashes) are the most characteristic trait of estrogen deficiency. It is experienced at least once in 75% of women during menopause. It is also one of the most puzzling symptoms of menopause, because the etiology and physiology remain incompletely understood [25]. It is thought to be the result of a hypothalamic dysregulation from estrogen withdrawal that culminates in peripheral vasodilatation and increase in blood flow. This results in heat loss and a decrease in core body temperature. The hypothalamic dysfunction is also manifested by simultaneous pulse of LH and presumably GnRH that is coincident with the hot flash. “Hot flash,” “hot flush,” “night sweats,” and vasomotor symptoms are words frequently used to describe the same experience. Hot flashes are defined subjectively as the recurrent transient sensation of heat, and could be accompanied by palpitations, perspiration, chills, shivering, and feeling of anxiety. It is then followed by a heat dissipation response that habitually begins in the face, neck, chest, and often becomes generalized [26].
While menopause is the most common cause of hot flashes, there are other causes that should be considered. Fever by far is the most common cause of hot flashes, especially when coupled with a night sweat; thus, if during an episode of hot flashes the oral temperature is elevated, then the cause of the fever should be sought.
In general we can divide the potential causes of hot flashes into seven categories: systemic diseases, neurological, alcohol–medication interaction, drugs, food additives, eating, and miscellaneous [27] (◘ Fig. 9.2). However, it is important to emphasize that these other causes are much less common than those associated with low estrogen levels.
9.13 Systemic Diseases
The most common systemic disorders associated with hot flashes are carcinoid syndrome, mastocytosis, pheochromocytoma, medullary thyroid carcinoma, pancreatic carcinoma, and renal cell carcinoma.
9.14 Carcinoid Syndrome
These patients present with neuroendocrine tumors of the bowel. Carcinoid tumors can be localized to the bronchus, pancreatic islets, retroperitoneum, liver [28], and even in the ovary [29]. It is thought these tumors probably arise from gastrointestinal or bronchopulmonary pluripotential stem cells [30]. The carcinoid syndrome clinically has a classic triad of diarrhea, flushing, and valvular heart lesions. Skin flushing is the most common sign and is present in over 90% of patients. The mechanism of flushing is at least partially due to serotonin release, but other substances such as kinins, substance P, neurotensin, and prostaglandin may play a role [37].
9.15 Mastocytosis
Mast cell proliferations can be limited to the skin (cutaneous) or can spread to extracutaneous tissues (systematic). Vasomotor-like symptoms may be present in these patients because the mast cell granules contain a number of acid hydrolases, leukotrienes, histamine, heparin, and slow-reacting substance [31].
9.16 Pheochromocytoma
These tumors often arise from the adrenal medulla. Most of their clinical characteristics are due to the production, storage, and secretion of catecholamines. A key finding in these patients is hypertension, which is present in over 60% of patients. A significant number of them suffer from hot flashes. Documenting increased urinary catecholamines makes the diagnosis [32].
9.17 Medullary Carcinoma of the Thyroid
This is a malignant tumor that originates from the parafollicular or thyroid C cells. These tumor cells typically produce an early biochemical signal (hypersecretion of calcitonin) [33]. This cancer can occur sporadically, but many times is inherited in an autosomal-dominant pattern as a part of a syndrome: multiple endocrine neoplasia type 2 [34]. Other bioactive substances that can be secreted by medullary carcinomas and may be responsible for the vasomotor symptoms include ACTH, corticotropin-releasing hormone, and prostaglandins [34].
9.18 Neurological Flushing
Anxiety or emotional blushing, migraines, Parkinson’s disease, spinal cord lesions (autonomic hyporeflexia), and brain tumors can also be associated with hot flashes.
9.19 Alcohol–Medication Interaction
Alcohol and numerous drugs are associated with vasomotor symptoms. In some cases, the drug is not the real vasoactive agent, but rather, a metabolite or another mediator triggered by the drug ingested (i.e., histamine release). Some drugs will only create vasomotor symptoms when combined with alcohol [35]. Others, such as calcium channel blockers, have a direct impact in the vessels [36]. On the other hand, tamoxifen or bromocriptine produces hot flashes by triggering different mediators.
Vasodilators (nitroglycerin, prostaglandins), calcium channel blockers, nicotinic acid, opiates (i.e., morphine), amyl nitrite, cholinergic drugs, bromocriptine, thyrotropin-releasing hormone, tamoxifen, clomiphene, triamcinolone, and cyclosporine are the most commonly cited medications.
9.20 Food Additives and Eating Habits
Monosodium glutamate , sodium nitrite, and sulfites are the most common food additives associated with hot flashes. Hot beverages, auriculotemporal flushing (cheese, chocolate, lemon, highly spicy foods), gustatory flushing (chewing chili pepper), dumping syndrome (seen in patients after gastric surgery and triggered by a meal, hot fluid, or hypertonic glucose) are examples of symptoms associated with ingesting food or beverage.
The main limitation in our ability to assess the value of a treatment for hot flashes is the lack of an objective measure. This complex task is due to the current inability to reliably identify when a hot flash has taken place. The main objective method used today is sternal skin conductance monitoring, which has some limitations, but the main weakness is the failure of sternal skin conductance to provide any information on duration, intensity, and interference with patient activities. Therefore, all data are derived from imperfect methods [32].
9.21 Morbidity Associated with Hot Flashes
9.21.1 Sleep Disturbances
The relationship between hot flashes and interference with sleep is controversial. Multiple epidemiological studies have shown an association between awakening and arousal from sleep and hot flashes in menopausal women [37–39]. This has led to the commonly held conception that hot flashes and night sweats cause awakening, which subsequently create fatigue, and possibly decrease performance and quality of life [40]. The flaw in these studies is that these hypotheses have not been properly tested in controlled laboratory investigations. Also, neither of these investigations screened out patients with apnea and other sleep disturbances that are also prevalent in menopause and may represent a confounder factor.
Recently, Freedman et al. studied 31 patients between the ages of 46 and 51 who were classified into three groups: premenopausal asymptomatic (cycling), postmenopausal asymptomatic (asymptomatic), and postmenopausal symptomatic (symptomatic). They then assessed several outcome measures: sleep electroencephalogram recordings, sternal skin conductance to record hot flashes, multiple sleep latency test to assess sleepiness, simple and divided attention performance tests, sleep and fatigue questionnaires. There were no significant differences among the three groups on any sleep variable. Of the awakenings taking place within 2 min of a hot flash, 55.2% happened before the hot flash, 40.0% after the hot flash, and 5% simultaneously. Of arousals taking place within 2 min of a hot flash, 46.7% occurred before, 46.7% after, and 5.6% simultaneously. There were no significant group differences on any self-report measure or on any performance measure. They concluded that there is no evidence that hot flashes produce sleep disturbance in symptomatic postmenopausal women [41]. Although the use of estrogen will effectively relieve postmenopausal nocturnal vasomotor symptoms, sleep quality was not changed [42].
9.21.2 Migraines
However the relationships between menopause and migraine are still debated. Observational studies suggest that migraine worsens just before menopause and improves after cessation of menses in approximately two-thirds of cases.
Neri studied 556 postmenopausal women for the prevalence and characteristics of headaches and found that many had migraine with aura. Interestingly, women with prior migraine generally improved with the onset of spontaneous menopause. In contrast, women with bilateral oophorectomy usually experienced worsening of their migraines [45]. More recently, a cross-sectional community-based study of 1436 women using the 1988 International Headache Society Criteria showed the highest prevalence of migraines in the perimenopausal group (31%) and the lowest (7%) in the postmenopausal group [46].
9.21.3 Urogenital Changes
The lack of estrogen has been associated with the onset of vulvo-vaginal atrophy with thinning of the cells lining the vulva, urethra, and vagina. The terminology describing these constellations of changes has changed from “Vulvo-Vaginal Atrophy” to Genito- urinary Syndrome of Menopause (GSM) [47]. Epithelial secretions decline and with time lead to dryness of the vaginal tissues. The vaginal pH will increase and the bacterial flora will change. The percentage of superficial vaginal cells will decline and basal-appearing cells will predominate. The persistent dryness of the vaginal mucosal surfaces may lead to symptoms of vaginitis, pruritus, dyspareunia, and even stenosis. Other symptoms that may be related to estrogen deprivation in the urogenital tissues are dysuria, urgency incontinence, and urinary frequency. It is unclear whether all these symptoms are related to the lack of estrogen or whether they are part of a degenerative process of aging. It is postulated that changes in estrogen levels changes the composition of the collagen content and the connective tissues in the urogenital area [48].
Data suggest that lack of estrogen increases the likelihood of menopausal women to experience recurrent urinary tract infections (UTIs). A randomized, placebo-controlled trial of vaginal estrogen in 93 postmenopausal women demonstrated that patients being treated could reduce their number of UTIs per year. This study observed 0.5 episodes of UTI in the treatment groups vs. 5.9 episodes in the placebo group [49]. On the other hand, data from the Heart and Estrogen/Progestin Replacement Study (HERS) showed that UTI frequency was higher in the group randomized to hormone treatment, although the difference was not statistically significant (odds ratio 1.16, 95% confidence interval 0.99, 1.37) [50].
However conflicting these results are, from the clinician’s point of view it seems prudent to attempt a trial of vaginal estrogen therapy to address those postmenopausal with urogenital symptoms. It should be considered in the presence of a vaginal pH greater than 4.5. Like FSH, elevated vaginal pH appears to be a good prediction of estrogen status [51].
9.22 Long-Term Morbidity Associated with Postmenopausal Status
The two major long-term risks associated with menopause are osteoporosis and cardiovascular disease.
9.22.1 Osteoporosis
In this section, we will discuss its relationship as a long-term risk factor in menopausal women. The demographic changes and longevity increase described at the beginning of this chapter, coupled with the fact that osteoporosis rises dramatically with age, make [52] osteoporosis a serious economical burden for health care systems in our society. The estimated total direct expenditure (hospitals and nursing homes) for osteoporotic and associated fractures was $17 billion in 2001 ($47 million each day) [53]. The National Osteoporosis Foundation estimates that 50% of white women will suffer at least one osteoporosis-related fracture in their remaining lifetime. At least 90% of hip and spine fractures among elderly women can be attributed to osteoporosis [54]. These statistics are the result of an accelerated decline in bone-mass after menopause. However, this decline begins approximately at age 35, when bone resorption is greater than bone formation. After menopause there are two periods of net bone loss: An accelerated stage that begins with the onset of menopause (1–3 years) and continues for 5–8 years [55] (STRAW 0, 1, and 2) and a prolonged, slower stage of bone loss that remains throughout STRAW 2. The initial accelerated phase may account for bone loss of up to 30% [56].
The three most common osteoporotic-related fractures are hip, vertebral, and wrist. The most common are vertebral fractures, accounting for 700,000 cases a year in the USA. These should be suspected in postmenopausal women with back pain, loss of height, and kyphosis. In one observational study of 7223 postmenopausal women over 65 years of age, patients with radiographically detected vertebral fractures were found to have significantly more limited-activity days, whether they were symptomatic or not [57]. These data should raise awareness for the clinician of the decreased quality of life that menopausal patients may experience even with asymptomatic fractures.
The second most common fracture is hip fracture, accounting for 300,000 cases a year in the USA. These are without doubt the more serious consequence of osteoporosis in postmenopausal women. One in five women will die within 1 year postfracture and one in two will have permanent loss of function [58]. Lastly, distal forearm fractures occur in 250,000 patients a year in the USA. Only half of the patients that suffer these fractures recover full function of the arm in 6 months [59].
9.22.2 Cardiovascular Disease
The American Heart Association has designated cardiovascular disease a “silent epidemic.” Despite the overall decline in the mortality rate due to cardiovascular disease in the USA, the absolute number of deaths due to cardiovascular disease is actually increasing [60], in part due to the demographic changes in our society described in the introduction of this chapter. Cardiovascular disease that encompasses heart attacks and strokes combined are responsible for more deaths than all other causes combined in postmenopausal women [61, 67]. The burden and threat of this disease during menopause is in part due to the lack of perception of its magnitude by both physicians and patients. Nothing exemplifies this better than a 1995 Gallup survey, which revealed that four out of five women aged 45–75 were unaware that cardiovascular disease was the first cause of death for their age group. Instead, most of the women quoted cancer, specifically breast cancer, as their most probable cause of death. In reality this represents only 4% of the causes of death in this age group. The primary care physicians questioned did not do much better. Thirty-two percent were unaware that heart disease was the main cause of death in this age group of women [62].
The incidence of cardiovascular disease and particularly myocardial infarction dramatically increases after menopause and approximates the mortality of this entity in men [63, 64]. Furthermore, bilateral oophorectomy or premature ovarian failure increases the risk of cardiovascular disease beyond that of natural menopause [65]. Despite this seemingly logical association between estrogen cardioprotection and other coherent data from observational studies, the Vasomotor symptoms and the HERS have found no role for estrogen as a primary or secondary prevention for cardiovascular disease in postmenopausal women.
Substantive evidence from epidemiological studies and clinical research indicate that the best tools are preventive measures and lifestyle habit modifications: smoking cessation, blood pressure control, lowering cholesterol, and promoting exercise.
9.23 Medical Treatment of Menopause
The results of the Women’s Health Initiative study (WHI) have altered the principles of medical practice in menopausal women. We have changed from the concept of prevention of chronic diseases encompassed in the term “hormone replacement therapy” to the concept of “hormone therapy.” Thus, the US Food and Drug Administration (FDA) and professional organizations such as the American Congress of Obstetricians and Gynecologists recommend that the use of estrogenic-containing medications be restricted to the treatment of vasomotor and vaginal symptoms. They also affirm that the lowest effective dose be prescribed for the shortest duration of time [66–68].
More detailed analysis of the WHI results has suggested a more individualized approach to hormone therapy. For this reason, the North American Menopause Society and the Endocrine Society have modified their guidelines for hormone therapy in 2012 [69]. Current evidence suggests that the absolute risks of estrogen therapy between the ages of 50 and 59 are low and the individual benefit may favor a longer duration of therapy. In contrast, the duration of use of estrogen with progestin should be limited because of the increased risk of breast cancer associated with 3–5 years of use.
9.24 Principles of Hormone Therapy
During the last 3 decades, the clinical opinion on the use of estrogen during menopause has changed drastically. Initially, estrogen was recommended as a short-term treatment for menopausal symptoms. Later, on the basis of observational studies, estrogen was given for long-term prevention of heart disease and an improved quality of life. The Women’s Health Initiative Hormone Therapy trial, however, demonstrated that estrogen was not effective for the prevention of cardiovascular disease.
9.25 Key Findings from the Women’s Health Initiative
The WHI was a group of clinical trials designed to examine the impact of hormone therapy on cardiovascular disease and breast cancer, the effect of low-fat diet on breast and colon cancer, and the impact of vitamin D in calcium supplementation on fractures and colon cancer [70].
These trials included:
A randomized controlled trial of 16,608 asymptomatic postmenopausal women ages 50–79 years with a uterus comparing conjugated estrogens (0.625 g) and a progestin, medroxyprogesterone (MPA) (2.5), daily vs. placebo. The primary outcome measure of this trial was coronary heart disease (CHD) and breast cancer. The secondary outcome measures were stroke, congestive heart failure, angina, peripheral vascular disease, coronary revascularization, pulmonary embolism, deep venous thrombosis, ovarian cancer, endometrial cancer, hip fractures, diabetes mellitus requiring therapy, death from any cause, and quality-of-life measures
A randomized controlled trial on 10,739 asymptomatic postmenopausal women 50–79 years without a uterus (hysterectomized) comparing conjugated estrogens (0.625 mg/day) vs. placebo
A dietary modification randomized controlled trial of 48,837 postmenopausal women 50–79 years to either sustained low fat (20%) or self-determined diet. The primary outcome measures were breast and colorectal cancer. The secondary outcome measures included stroke, congestive heart failure, angina, peripheral vascular disease, coronary revascularization, ovarian cancer, endometrial cancer, hip fractures, diabetes mellitus requiring therapy, and death from any cause
A calcium/vitamin D supplementation diet trial of 38,282 postmenopausal women in which the primary outcome measure was hip fractures and the secondary outcome measures were death from any cause, breast and colon cancer
A cohort observation group of 93,676 postmenopausal patients
In May of 2002 the clinical trial that aimed to assess the cardiovascular effects of estrogen and progestin therapy in postmenopausal women with intact uterus was halted. The Data and Safety Monitoring Board reported that the estrogen/progestin treatment group had an increased risk in cardiovascular disease, thromboembolism, and breast cancer after 5.2 years of follow-up [71]. In 2004, after 6.8 years of follow-up, the estrogen-alone trial was halted [72]. In this clinical trial, estrogen-only treatment demonstrated an increase risk in strokes similar to the one found in the estrogen–progestin clinical trial previously halted. There was also reported a lack of benefit on cardiovascular disease incidents and a probable increase in dementia. Surprisingly, the breast cancer risks in the estrogen treatment group were lower than the placebo. The risks and benefits findings of the WHI are summarized in ◘ Table 9.2 [73].
Table 9.2
Women’s Health Initiative Findings: outcomes associated with use of combined estrogen and progestin and estrogen alone in healthy postmenopausal women, aged 50–79 years
Outcome | Estrogen and progestin | Average absolute risk differenceb | Estrogena | Average absolute risk differenceb |
|---|---|---|---|---|
RR (95% CI) | RR (95% CI) | |||
Cardiovascular | ||||
Deep venous thrombosis | 2.07 (1.49–2.87) | 13 | 1.47 (1.04–2.08) | 6 |
Pulmonary embolism | 2.13 (1.39–3.25) | 8 | 1.34 (0.87–2.06) | 11 |
Coronary heart disease | 1.24 (1.00–1.54) | 7 | 0.91 (0.75–1.12) | −5 |
Ischemic stroke | 1.44 (1.09–1.90) | 8 | 1.39 (1.10–1.77) | 12 |
Cancer | ||||
Breast | 1.24 (1.02–1.50) | 8 | 0.77 (0.59–1.01) | −7 |
Colorectal | 0.63 (0.43–0.92) | −6 | 1.08 (0.75–1.55) | 1 |
Ovarian | 1.58 (0.77–3.24) | 8 | NYR | NYR |
Endometrial | 0.81 (0.48–1.36) | −4 | N/A | N/A |
Other | ||||
Probable dementiac | 2.05 (1.21–3.48) | 23 | NYR | 12 |
All fractures | 0.76 (0.69–0.83) | −44 | 0.70 (0.63–0.79) | −56 |
Hip fractures | 0.67 (0.47–0.96) | −5 | 0.61 (0.41–0.91) | −6 |
Mortality | 0.98 (0.82–1.18) | −1 | 1.04 (0.88–1.22) | +3 |
It should be emphasized that the WHI trial did not intend to evaluate the effects of estrogen or estrogen–progesterone on vasomotor symptoms because hot flashes were not the major complaint among the majority of subjects. Therefore, these results must be translated into the specific needs of our patients when they request relief for hot flashes or other postmenopausal symptoms. It is also critical to point out that the serious adverse events in patients treated with estrogen therapy is low and calculated to be 2 out of 1000 women treated per year [74]. Health care providers and patients must balance the benefits of estrogen treatment vs. the absolute risk for adverse events. Today, more than ever, the concept of individualized menopausal care should be applied in the clinical setting.
The WHI investigators have performed a secondary, stratified analysis of the cardiovascular risk by age. These data suggest that estrogen or estrogen with progestin have potential cardiovascular benefits if started early in the menopause (aged 50–55 or less than 10 years from menopause), whereas those started on hormone therapy after 60 or more than 10 years from menopause are at increased risk (◘ Fig. 9.3). These newer findings are in keeping with observational studies and other cardiovascular preventive trials. Taken together, these data suggest that there is a “window or time” of opportunity for use of estrogen therapy and it has been termed the “timing hypothesis” [75, 76]. Studies to test this hypothesis are ongoing.
Fig. 9.3
Hazard ratio of cardiovascular risk for estrogen alone and estrogen with progestin relative to placebo controls in the WHI. Created with data from [75]
The WHI hormone trial is the first randomized controlled trial that demonstrates that estrogen actually decreases the risk of fractures in a low-risk population. However, when all the risks and benefits are weighted, it can be concluded in this study that estrogens are not indicated as an overall preventive measure in postmenopausal women and the potential harm outweighs the potential long-term benefit. Thus, at the present time, use of estrogen should be limited to the treatment of symptomatic menopausal women for the shortest possible time at the lowest effective dose. ◘ Table 9.3 shows a comparison between the WHI and HERS trials .
Table 9.3
Impact of estrogen–progestin treatment on cardiac events in a healthy menopausal population (WHI) vs. a population with a prior history of cardiac events (HERS)
WHI-CHD by year of follow-up | HERS-risk of cardiac events | |||||
|---|---|---|---|---|---|---|
Year | Estrogen–progestin | Placebo | Hazard ratio and confidence interval | Estrogen–progestin | Placebo | Relative hazard (risk) and Confidence Interval |
n = 8000+ | n = 1383 | n = 1380 | ||||
1 | 42 cases | 23 cases | 1.81 (1.09–3.01) | 57 cases | 38 cases | 1.52 (1.01–2.29) |
2 | 38 | 28 | 1.34 (0.82–2.18) | 47 | 48 | 1.00 (0.67–1.49) |
3 | 19 | 15 | 1.27 (0.64–2.50) | 35 | 41 | 0.87 (0.55–1.37) |
4 | 32 | 25 | 1.25 (0.74–2.12) | 33 | 49 | 0.67 (0.43–1.04) |
5 | 29 | 19 | 1.45 (0.81–2.59) | 1.06 (0.69–1.62) | ||
>6 | 28 | 37 | 0.70 (0.42–1.14) | 0.98 (0.72–1.34) | ||
9.26 Key Findings of the Heart and Estrogen/Progestin Replacement Study
While the WHI aimed to test the hypothesis that hormone therapy prevented cardiovascular disease in healthy postmenopausal women (primary prevention), the HERS intended to evaluate whether hormone therapy decreased the risk of CHD in postmenopausal women with established coronary disease. In this randomized trial, all the 2763 postmenopausal women had uteruses, and were allocated to either placebo (n = 1383) or 0.625 mg of conjugated equine estrogens plus 2.5 mg of MPA daily (n = 1380) [77, 78]. The primary outcome measures were (1) nonfatal myocardial infarction and (2) CHD death.
The results of HERS trial have been reported in two publications, HERS and HERS II. The HERS report is the result of randomized, blinded, placebo-controlled trial for 4.1 years and the HERS II reflects the unblinded follow-up for 2.7 more years [84, 85]. Both of the studies demonstrated that in patients with established heart disease, the use of estrogen and progestin does not prevent additional cardiovascular events. There were no differences in the primary or secondary outcomes of patients in the placebo or the treatment group.
The conclusion of HERS and HERS II studies is that postmenopausal hormone therapy should not be recommended for the purpose of reducing the risk of cardiovascular events.
9.27 Key Findings from the ELITE and KEEPS Randomized Trials
To examine the possibility that HRT was beneficial in early menopausal women and to test the “timing hypothesis” suggested by the WHI data, two small focused randomized trials were conducted.
The first randomized trial to study the “timing hypothesis” was the Early versus Late Intervention Trial with Estradiol (ELITE) [79]. The trial enrolled 643 postmenopausal women in two strata: those that were <6 years or >10 years postmenopause. Each group received either placebo or oral estradiol 1 mg/day with progesterone gel (45 mg/day) given for 10 days every 30 days for endometrial protection. The primary outcome was the rate of change in carotid-artery intima-media thickness (CIMT). This is a sensitive, reproducible biomarker for atherosclerosis progression. The key finding from this trial demonstrated that early menopausal women had a slower progression of CIMT increase 0.0044 mm/year versus placebo 0.0078 mm/year, p = 0.008. Whereas the late menopause women had similar rates of progression to placebo 0.0088 and 0.0100 mm/year, p = 0.29).
The Kronos Early Estrogen Prevention Study (KEEPS) was a four-year randomized trial involving more than 700 early menopausal women given either a low-dose oral conjugated equine estrogen (CEE) 0.45 mg/day, transdermal estradiol 50 μg/day, or placebo with 200 mg of micronized progesterone days 1–12 each month. Assessment of progression of atherosclerosis by carotid artery intima-media thickening demonstrated similar rates of progression of arterial wall thickness among the three groups [80]. KEEPS also evaluated the impact of estrogen on cognitive function. Treatment with CEE or transdermal estradiol did not alter cognitive performance. However, improvements in depression and anxiety symptoms were noted in women receiving CEE but not transdermal estradiol or placebo groups [81].
Collectively, these studies support the “timing hypothesis” that HRT given to early menopausal women may be beneficial for reducing cardiovascular risks whereas late postmenopausal use of HRT has no additional cardiovascular benefits.
9.28 Key Findings the Million Women Study
The Million Women Study is a prospective observational study that included 1,084,110 British women aged 50–64 recruited between 1996 and 2001. This study was undertaken by the UK National Health Service Breast Screening Programme targeting women ages 50–64 undergoing routine screening once every 3 years. Because approximately only half of these patients had ever taken estrogen after menopause, the aim of the study was to investigate the relation between various combinations of hormone therapy and two main outcomes: breast cancer and mortality [82]. All patients filled out a questionnaire and were monitored in this manner. This questionnaire is available at ► http://www.millionwomenstudy.org. The major strength of this study was the unparalleled database size, which was sufficiently powered to quantify absolute and relative risks and enabled researchers to discern the effects among different preparations of hormones in use among postmenopausal women. One weakness of the study was that hormone use vs. non-use was determined on admission to the study and was not modified during follow-up, even though there were potential multiple crossover treatments in some of the subjects. The authors reached these conclusions:
Current use of hormone therapy is associated with an increased risk of incident and lethal breast cancer
The risk is substantially greater for estrogen–progestin combinations of postmenopausal hormone therapy
The mortality rate due to breast cancer was 27% less in hormone users than in non-hormone users. This could be explained in part by more frequent medical care and early detection
The relative risks for invasive breast cancer in relation to current use of hormone therapy and type of hormone preparations are illustrated in ◘ Figs. 9.4 and 9.5.
Fig. 9.4
Relative risk of incident invasive breast cancer in relation to recency of use of HRT. FCI floated CI. *Relative to never users, stratified by age at first birth, family history, of breast cancer, body-mass index, region, and deprivation index. Reprinted from The Lancet, 362, Breast cancer and hormone-replacement therapy in the Million Women Study, 419–27, Copyright 2003, with permission from Elsevier
Fig. 9.5
Relative risk of incident invasive breast cancer in relation to recency and type of HRT used. FCI floated CI. *Relative to never users, stratified by age at first birth, family history, of breast cancer, body-mass index, region, and deprivation index. Reprinted from The Lancet, 362, Breast cancer and hormone-replacement therapy in the Million Women Study, 419–27, Copyright 2003, with permission from Elsevier
9.29 Candidates for Hormone Therapy during Menopause
It is clear that the treatment paradigm has shifted when it comes to whom should we recommend the use of estrogens during menopause. After reviewing the evidence in the randomized controlled trials, it seems prudent to state that estrogen should not be used to prevent long-term chronic diseases such as cardiovascular disease and dementia.
Nevertheless, patients with moderate to severe vasomotor symptoms should weigh the risk and benefits and might consider using estrogen. It seems that the group of patients with vasomotor symptoms and osteoporosis or at risk for osteoporosis are good candidates to consider prescribing estrogen. Estrogen is the most effective treatment to alleviate vasomotor symptoms and has been shown to reduce the fracture risk in postmenopausal women.
For these reasons, medical management of menopausal symptoms has evolved to a personalized approach. Personal choice is important and some will never take hormone therapy under any circumstances. The physician must consider many variables before offering hormone therapy in light of individual cardiovascular, osteoporosis, and breast cancer risk and the severity of estrogen-deficient symptoms.
9.30 General Principles of Drug Therapy
For patients on hormone replacement therapy, there are two important concepts that must be discussed. First, is the projected duration of therapy. Professional societies such as the North American Menopause Society and the Endocrine Society suggest that the lowest effective dose be utilized consistent with the treatment goals. If it is for the treatment of vasomotor symptoms, it could be for a limited period of time. The duration of therapy may vary depending on the individual. Those begun on hormone therapy earlier in the menopause (prior to age 55) appear to have a cardiovascular benefit. Secondly, the effectiveness of therapy should be quantified. Typically, this consists of assessing symptoms and signs of hypoestrogenemia and monitoring bone mineral density.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
