CHAPTER 22: The Mature Woman

The typical “mature woman” is aged 40 years or older and has completed childbearing. During their late 40s, most women enter menopausal transition. Detailed in Chapter 21, this period of physiologic change due to ovarian senescence and estrogen decline is usually completed between ages 51 and 56. Menopause marks a defining point in this transition and is defined as the point in time of permanent menstruation cessation due to loss of ovarian function. Clinically, the menopause refers to a point in time that follows 1 year after cessation of menstruation.

With ovarian senescence, declining hormone estrogen levels have specific effects. Some lead to physical symptoms, such as vasomotor symptoms and vaginal dryness, whereas others are metabolic and structural changes. These include bone loss, skin thinning, fatty replacement of the breast, lipoprotein changes, and genitourinary atrophy. As a result, postmenopausal women have unique issues associated with aging and estrogen loss that may negatively affect their individual health.

For many years, menopause was seen as a “deficiency disease” of estrogen, progesterone, and testosterone. For this reason, hormone replacement therapy has been used in one form or another for more than 100 years. The history surrounding this treatment is discussed here, as are current recommendations for the treatment of menopausal symptoms.

Clinicians ideally provide evidence-based medicine, and seldom is a single study relied on solely to guide practice (Lobo, 2008). Thus, providers should understand the weaknesses and strengths of clinical trials to accurately counsel their patients. As one example, hormone treatment (HT) was widely prescribed to menopausal women, in good faith, based on initial observational studies. The general medical consensus was that HT, in addition to its prevention and treatment of osteoporosis, could protect against cardiovascular disease, stroke, and dementia. Subsequently, prospective, randomized controlled trials (RCTs) challenged the validity of these earlier observational studies. However, in this evaluation, clinicians must appreciate the type of population studied, the ages and risk factors of participating women, and the hormone regimens tested.

Early Estrogen Administration Trends

Estrogen treatment (ET) for menopausal symptom relief gained popularity in the 1960s and 1970s. Proponents promoted ET for its “preservation of youth” and prevention of chronic disease. By the mid-1970s, more than 30 million prescriptions were written for estrogen each year, and half of all menopausal women were using ET for a median of 5 years. Premarin (conjugated equine estrogen) was the fifth most prescribed drug in the marketplace.

In 1975, a study revealed a connection between endometrial cancer and ET. Investigators found a 4.5 times greater risk of this cancer in those using estrogen (Smith, 1975). As a result, the Food and Drug Administration (FDA) ordered labeling changes to state this higher risk. Progestins were then added in the 1980s to therapy regimens to significantly reduce endometrial cancer risks.

During that same time, estrogens were documented by several studies to prevent bone loss (Gambrell, 1983). Additionally, an expanding literature provided a robust affirmation of the effectiveness of menopausal HT in reducing vasomotor symptoms, maintaining bone mineral density (BMD), and preventing and treating vulvovaginal atrophy (Shulman, 2010). Several observational studies also suggested that estrogens prevented development of coronary heart disease (CHD) and other conditions, such as Alzheimer disease. However, in 1985, conflicting reports from two studies were published.

First, the Framingham Heart Study, an observational study of 1234 women, showed that those who took hormones had a 50-percent elevated risk of cardiac morbidity and more than a twofold risk for cerebrovascular disease (Wilson, 1985). In the same edition of the New England Journal of Medicine, a much larger observational trial, The Nurses’ Health Study, with 121,964 women, found significantly lower rates of heart disease in postmenopausal women taking estrogen compared with a similar cohort not taking estrogen (Stampfer, 1985). Numerous subsequent articles reported the protective effects that combination HT provided menopausal women against cardiovascular disease and osteoporosis.

Current thinking is that that these early nonrandomized, unblinded, observational studies included samples of women who were not necessarily representative of the entire postmenopausal population. These hormone users tended to have superior access to care and to be thinner, wealthier, and healthier (Grodstein, 2003; Prentice, 2006). An additional source of confounding and possible selection bias is suggested to be the timing of HT initiation relative to the underlying state of the vasculature. Some investigators suggest that estrogen may delay the onset of the earliest stages of atherosclerosis, which are more likely to be present in younger women. However, estrogen may be ineffective or even trigger events in advanced lesions that are found in older women (Mendelsohn, 2005). This potential “window of opportunity” is supported by animal and laboratory studies (Grodstein, 2003). In sum, these study biases may have contributed to favorable outcomes attributed to estrogen in observational trials. When these biases are eliminated and the data reanalyzed, the results of earlier observational trials and later RCTs are remarkably similar.

Postmenopausal Estrogen/Progestin Interventions Trial

Because of data available in the late 1980s, estrogens were prescribed, not only for vasomotor symptom relief, but also for prevention of other conditions. In 1995, The Writing Group for the Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial published results that suggested benefit for CHD risk. In this study, menopausal women with a mean age of 56 years were randomly allocated to one of five treatments: (1) placebo, (2) estrogen alone, (3) estrogen plus cyclic medroxyprogesterone acetate (MPA), (4) estrogen plus cyclic micronized progesterone, or (5) estrogen plus continuous MPA. Primary outcomes studied in the 875 women evaluated during 3 years included measurement of systolic blood pressure and of serum lipid, insulin, and fibrinogen levels. The PEPI trial documented that low-density lipoprotein (LDL) cholesterol levels declined and high-density lipoprotein (HDL) levels were increased in the four treatment groups receiving estrogens. Levels were most substantially improved in women solely given estrogen. An intermediate effect was noted in those prescribed conjugated equine estrogen (CEE) and micronized progesterone, whereas the smallest change followed CEE and MPA administration. Fibrinogen levels were increased in the placebo group compared with groups given hormones. However, no differences were identified in systolic blood pressure or glucose-challenged insulin levels. Clinical outcomes were also reported, and complications were few. Of these, all occurred in the HT-treated groups and included one cardiac arrest, two myocardial infarctions (MIs), and two cerebrovascular events.

Heart and Estrogen/Progestin Replacement Study

With results published in 1998, the Heart and Estrogen/Progestin Replacement Study (HERS) described cardiac morbidity in 2763 women with preexisting heart disease (Hulley, 1998). These women received estrogen as secondary prevention for further cardiac disease progression. First-year findings showed an increase in MIs in women who received CEE with continuous MPA. However, after average treatment duration of 4 years, rates of cardiovascular death or nonfatal MI did not differ between treatment groups.

This trial represented the first RCT at variance with prior data and created confusion for both clinicians and their patients. Beliefs that hormones prevented heart disease persisted, but the HERS data caused many investigators to question the cardioprotective effects of hormones. Subsequent HERS II results also showed that HT was not beneficial in the secondary prevention of heart disease even after 6.8 years (Grady, 2002). One reanalysis of the Nurses’ Health Study focused on early hazard among women initiating HT during the monitoring period and showed a similar time trend, with early harm (Grodstein, 2001).

Women’s Health Initiative

After an unsuccessful effort in 1990 to obtain FDA approval for HT as prevention for CHD, the need for RCTs to demonstrate conclusive benefit was widely acknowledged. As a result, before the results from the PEPI trial and HERS trials were available, the National Institutes of Health (NIH) launched the Women’s Health Initiative (WHI) in 1993. This major study was to evaluate the putative protective effects of HT on common chronic diseases of aging. The WHI examined the effect of a single combined CEE and MPA drug compared with placebo in 16,608 healthy postmenopausal women aged 50 to 79 years (mean of 63.3 years) who had not undergone hysterectomy (Rossouw, 2002). Specific end points were evaluated: CHD, venous thromboembolism (VTE), breast cancer, colon cancer, and bone fractures. Concurrently, the study also compared CEE with placebo in postmenopausal women without a uterus (the estrogen-only arm).

As part of the original WHI study design, investigators predetermined targets for CHD (anticipated benefit) and breast cancer (anticipated risk) as primary disease end points. This design dictated that if the incidence of an end point was exceeded within a given period, the study would be terminated. Moreover, combined end points were weighted into a “global index,” which if exceeded, would result in study termination. After a mean 5.2 years of monitoring, the estrogen and progestin arm of WHI was halted early upon recommendation of its Data and Safety Monitoring Board because overall risks exceeded the benefits. In July 2002, results were released to the media. This preceded journal publication of the data and timely education of health care providers. Chaos ensued while physicians and patients evaluated research facts and before recommendations could be made.

In a subsequent detailed analysis of cardiovascular end points, the hazard for cardiovascular death or nonfatal MI was 1.24. This translated into 188 actual cases in the hormone group and 147 in the placebo group (Anderson, 2004). However, there were no significant differences in coronary revascularization, hospitalization for angina, confirmed angina, acute coronary syndrome, or congestive heart failure.

To explore the timing of HT initiation, Rossouw and colleagues (2007) performed a secondary analysis of the WHI data. They looked specifically at the effect of HT on stroke and CHD rates across categories of age and years since menopause in the combined trial. Women who initiated HT closer to menopause tended to have reduced CHD risk compared with the increase in CHD risk among women more distant from menopause. In women with less than 10 years since their menopause began, the hazard ratio for CHD was 0.76. In those with 10 to 20 years since menopause, the ratio was 1.10; and with 20 or more years, 1.28.

In evaluating data by age, lower risk was found for young women and higher risk for older patients. Specifically, for the age group of 50 to 59 years, the hazard ratio for CHD was 0.93, or two fewer events per 10,000 person years; for the age group 60 to 69 years, 0.98 or 1 fewer event per 10,000 person years; and for those 70 to 79 years, 1.26 or 19 extra events per 10,000 person years. HT increased the stroke risk—the hazard ratio was 1.32—and this risk did not vary significantly by age or time since menopause. Rossouw and colleagues concluded that women who initiated HT closer to menopause tended to have reduced CHD risk compared with the increase in CHD risk among women more distant from menopause. Whether CEE or combined CEE and MPA administration improves the cardiovascular health of women who recently experienced menopause remains to be determined. Presently, evidence is insufficient to suggest that either of these regimens should be initiated or continued for primary or secondary CHD prevention (American College of Obstetricians and Gynecologists, 2013). Although this was the principal analysis conclusion, the results led providers and patients to limited HT use, even for healthy women with bothersome vasomotor symptoms during menopausal transition.

Concurrent with the WHI, a similarly constructed study, the Women’s International Study of Long Duration Oestrogen after Menopause (WISDOM), began enrollment in 1999. This trial was stopped following halting of the WHI. Analyzing data collected from this study, Vickers and coworkers (2007) found that HT increases cardiovascular and thromboembolic risk when started many years after the menopause.

With this background in mind, data from no single trial can be extrapolated to all women. In the North American Menopause Society’s (2012) hormone therapy position statement, they note that while the WHI is, for some outcomes, the only large long-term RCT of postmenopausal women using HT, the trial has several characteristics that limit generalization to all postmenopausal women. These include the use of only one formulation of estrogen and only one progestin. Unlike most HT studies that focused on symptomatic, recently postmenopausal women, the WHI enrolled generally healthy but older postmenopausal women. These parameters should be accounted for when applying the WHI findings to clinical practice.

Risks and Benefits

As a result of these and other studies, most clinicians agree that HT is associated with an increased risk of CHD in older menopausal women, and an increased risk of breast cancer, stroke, VTE, and cholecystitis. Breast cancer appears only to be a risk factor with long-term use (>5 years). Two studies have shown an increase in ovarian cancer risk with long-term use (>10 years) but not with short-term use (<5 years) (Danforth, 2007; Lacey, 2006). However, other studies have not confirmed this risk (Noller, 2002).

In contrast, several long-term benefits are noted with HT. These include increased BMD and decreased rates of fracture and colorectal cancer. HT’s effects on mortality rates have also been examined. A metaanalysis by Salpeter and associates (2004) pooled data from 26,708 participants to reveal that the mortality rate associated with HT was 0.98. Of note, HT reduced mortality rates in women younger than 60 years but not in women older than 60. These investigators suggest that once CHD is established, HT has no effect in reversing disease progression. Moreover, the incidence of cardiovascular events can potentially increase in older groups due to an increased risk for blood clots. Similarly, Rossouw and colleagues (2007) showed a nonsignificant tendency for the effects of HT on mortality rates to be more favorable in younger than older women.

Indications and Contraindications

Based on current literature, HT is indicated only for treatment of vasomotor symptoms and vaginal atrophy and for osteoporosis prevention or treatment. Current guidelines recommend reevaluation of the need for HT at 6- to 12-month intervals. HT is prescribed in the lowest effective dose for the shortest period of time. Accordingly, bone-specific agents would likely be more appropriate in women requiring long-term osteoporosis prevention or treatment.

For women with a uterus, a progestin is combined with an estrogen to lower risks of endometrial cancer. Progestins may be prescribed daily with estrogen, and this dosing is termed continuous therapy. Another suitable continuous oral agent is the combination of CEE plus the selective estrogen-receptor modulator (SERM) bazedoxifene. This is marketed as the drug Duavee.

Instead, HT may be provided in a cyclic regimen. For this, estrogen is administered for 25 days each month and a progestin added for the final 10 days. Drugs are withdrawn for 5 days and endometrial sloughing and bleeding follows. Another common regimen includes treatment with estrogen continuously with a progestin administered for the first 10 days of each month. Such cyclic therapy is most often used in those during menopausal transition, whereas continuous therapy is usually selected for women following menopause.

Oral progestins are most commonly prescribed, although a progestin-releasing intrauterine device (Mirena) provides another promising option for localized rather than systemic progesterone administration in postmenopausal women (Peled, 2007). In addition, combined estrogen and progestin products are available for either oral or transdermal use. Low-dose combination oral contraceptives are effective in the young perimenopausal woman and have the additional benefit of pregnancy prevention.

Importantly, estrogen is contraindicated or evaluated in women who exhibit conditions found in Table 22-1. Ultimately, the decision regarding whether or not to begin HT is a personal one, to be decided by the patient with guidance from her health care provider.

Common early symptoms of menopause include hot flushes, insomnia, irritability, and mood disorders, which can be caused by vasomotor instability. Physical changes include vaginal atrophy, stress urinary incontinence, and skin atrophy. Long-term health risks that have been attributed to the hormonal changes from menopause include osteoporosis, cardiovascular disease, and, in some studies, Alzheimer disease, macular degeneration, and stroke.

Of these, the most frequent complaint of the menopausal transition is vasomotor symptoms. Also known as hot flushes or hot flashes, their physiology is discussed in Chapter 21. Following menopause, hot flushes are still pervasive and are experienced by 50 to 85 percent of postmenopausal women. Significant distress results for approximately 25 percent of women. Sleep disturbances can lead to lethargy and depressed mood.

The frequency of hot flushes does decrease with time. In the PEPI trial, the percentage of women taking placebo who experienced vasomotor symptoms declined from 56 percent at their entry into the study to 30 percent by their third year in the trial (Greendale, 1998). Freeman and coworkers (2014) found that moderate to severe hot flushes continue on average for nearly 5 years after menopause, and more than one third of women observed for 10 years or more after menopause have these. Fifteen years after menopause, approximately 3 percent of women report frequent hot flushes, and 12 percent report moderate to severe vasomotor symptoms (Barnabei, 2002; Hays, 2003).

Several treatment options are discussed subsequently, and three have FDA approval for this indication. These are systemic estrogen, one selective serotonin-reuptake inhibitor (SSRI), and the combined CEE plus bazedoxifene agent. When deciding among the available interventions for vasomotor symptoms, the safest options are encouraged first, such as lifestyle changes, but may proceed to prescription treatments, as needed. Patient preference, symptom severity, side effects, and the presence of other comorbid conditions will influence treatment options.

Hormonal Therapy

Estrogen

Systemic ET is the most effective treatment for vasomotor symptoms, and the value of such treatment has been demonstrated in numerous RCTs (Nelson, 2004). MacLennan and associates (2004) performed a systematic review of 24 RCTs involving 3329 women who had moderate to severe hot flushes. These investigators found that HT reduced the frequency of hot flushes by approximately 18 events per week, that is, approximately 75 percent compared with placebo. The severity of vasomotor symptoms also was reduced significantly. Moreover, in the PEPI trial, all treatment arms were more effective than placebo in reducing vasomotor symptoms. There were no significant differences between specific hormone regimens (Greendale, 1998).

Estrogen can be administered by oral, parenteral, topical, or transdermal routes with similar effects (Table 22-2). Within these groups, several different formulation choices are available. Continuous estrogen therapy is recommended, although doses and route of administration can be changed relative to patient preference. In the United States, oral estrogens are the most popular. Transdermal estrogen patches avoid the liver’s first-pass effect and offer the convenience of less frequent administration (once or twice weekly). The lowest effective dose and duration of therapy are unknown, but this “mantra” is cited by most major menopause organizations for ensuring safety. For treatment of vasomotor symptoms, the FDA has approved all oral estrogen formulations, most transdermal patch formulations, a topical gel, and one intravaginal estrogen product. All systemic HT products except for the ultra-low-dose estradiol transdermal patch (Menostar) have FDA approval for this indication (North American Menopause Society, 2012).

Progestins

These alone are somewhat effective for daily treatment of hot flushes in women for whom estrogen is contraindicated. However, adverse effects that include vaginal bleeding and weight gain may limit their use. Beyond mild reduction in number of hot flushes, progestins used as agents in combined HT offer only one additional benefit. Namely, they provide essential protection against estrogen-induced endometrial hyperplasia and endometrial cancer in women with a uterus. Clinical trials have shown that progestins provide no meaningful increase in estrogen’s benefits to bone. Moreover, progestins may attenuate estrogen’s beneficial effects on lipids and blood flow and increase the risk for breast cancer.

Bazedoxifene

This SERM combined with CEE was FDA-approved in 2014 for the management of menopausal vasomotor symptoms in women with an intact uterus (Lobo, 2009). As with other SERMs, bazedoxifene (BZA) acts as an estrogen agonist or antagonist, depending on the tissue affected. This combination is called a tissue-selective estrogen complex (TSEC), and its advantage is to combine the benefits of CEE with the SERM’s ability to offset estrogen stimulation of the endometrium and breast. Thus, a progestin is not required. By itself, BZA reduces BMD loss, increases the number of hot flushes, and raises VTE risks. The combination of BZA 20 mg plus CEE 0.45 mg reduces BMD loss and incidence of hot flushes but does not increase VTE rates compared with CEE and MPA. Notably, BZA acts as an antagonist in uterine tissue to prevent the endometrial hyperplasia commonly associated with unopposed estrogen (Pinkerton, 2014).

In the randomized Selective estrogens, Menopause, and Response to Therapy (SMART-2) trial, BZA plus CEE reduced hot flush frequency by 74 percent at week 12 compared with a 51-percent reduction from placebo (Pinkerton, 2009). A secondary efficacy analysis of SMART-2 data assessed sleep parameters and health-related quality of life in 318 women experiencing >7 moderate to severe hot flushes daily. Time to fall asleep, sleep disturbance, and sleep adequacy were improved with BZA plus CEE compared with placebo. This dose also significantly improved the vasomotor function score and the Menopause-Specific Quality of Life (MENQOL) questionnaire score compared with placebo (Utian, 2009). Overall the BZA plus CEE combined agent is well tolerated and effective in treated hot flushes, but its effectiveness relative to HT remains uncertain.

Bioidentical Hormones

Some women believe that conventional FDA-regulated pharmaceutical estrogens and progestins hold a clear and present danger. Ironically, more is known about the absolute risks and benefits of HT than almost any other class of drugs. Although prescriptions for estrogen and progesterone have declined significantly since publication of the WHI results, the use of “bioidentical hormones” has increased. This term invented by marketers is often used to describe custom-compounded HT products but has no clear scientific meaning (Shifren, 2014). The term now usually refers to compounds that have the same chemical and molecular structure as hormones that are produced in the body.

Custom compounding of HT may combine several hormones (e.g., estradiol, estrone, and estriol) and use nonstandard routes of administration (e.g., subdermal implants). Some of the hormones are not FDA approved (estriol) or monitored, and some compounded therapies contain nonhormonal ingredients (e.g., dyes, preservatives) that some women cannot tolerate. Moreover, custom-compounded formulations have not been tested for efficacy or safety; product information is not consistently provided to women along with their prescription, as is required with commercially available HT; and batch standardization and purity may be uncertain. Thus, these hormones cannot be assumed to be safer than conventional pharmaceutical estrogen or progestins. Several organizations note that conventional FDA-approved HT products are preferred to custom-compounded formulations (American College of Obstetricians and Gynecologists, 2014a; North American Menopause Society, 2012). In conjunction with bioidentical hormones, salivary hormone testing to assist with hormone level adjustment is inaccurate and unreliable (Lewis, 2002; Zava, 1998).

Central Nervous System Agents

Several nonhormonal therapies are available to treat vasomotor symptoms (Table 22-3). Alternatively, women who are principally bothered by night sweats and sleep disruption may benefit from a trial of sleep medication, and options are listed in Table 1-16. Of vasomotor symptom therapies, SSRIs, selective norepinephrine-reuptake inhibitors (SNRIs), clonidine, and gabapentin have demonstrated benefits compared with placebo (American College of Obstetricians and Gynecologists, 2014b). Of these, only the SSRI paroxetine mesylate (Brisdelle) is FDA-approved for vasomotor symptom treatment.

For many, the side effects or relative ineffectiveness of these agents compared with estrogen therapy can limit their routine use for this indication. Moreover, long-term studies with any of these agents for vasomotor symptom treatment are not available.

Serotonin and Norepinephrine

As noted in Chapter 21, both serotonin and norepinephrine are implicated in modulation of the thermoregulatory setpoint, which is integral to hot flushes. Accordingly, SSRIs have been studied, and in two RCTs, the low-dose mesylate salt of paroxetine (LDMP) was given to postmenopausal women experiencing daily, moderate to severe hot flushes. One study lasted 12 weeks, and the other 24 weeks (Simon, 2013). In both trials, LDMP reduced the weekly frequency of hot flushes compared with placebo, but persistent declines in hot flush severity were less consistent (Carris, 2014). Following initiation, frequency reduction begins as early as 1 week after therapy, and for severity reduction, as early as 2 weeks.

Other SSRIs have also been studied. With venlafaxine extended release (Effexor XR), Evans and coworkers (2005) noted 51-percent fewer hot flushes compared with placebo. Several other trials have compared this same SSRI with gabapentin or with clonidine in breast cancer survivors with hot flushes and also showed symptom benefit from the SSRI (Boekhout, 2011; Bordeleau, 2010; Loprinzi, 2000). Joffe and associates (2014) compared venlafaxine and low-dose estrogen and noted that estradiol reduced hot flushes by 2.3 events daily compared with 1.8 for the SSRI. Both outperformed placebo.

With paroxetine (Paxil) modest improvement in hot flushes are seen compared with placebo. In one RCT, a 20-mg dose lowered hot flush frequency by 51 percent. Stearns and coworkers (2003) evaluated paroxetine CR, 12.5 mg/d and 25 mg/d dosages, compared with placebo. At both dosages, paroxetine led to approximately three fewer hot flushes per day compared with 1.8 fewer per day with placebo.

With citalopram (Celexa), mean hot flush scores were lowered by 37 to 50 percent (Barton, 2010; Kalay, 2007). With escitalopram (Lexapro), two RCTs noted significant improvement compared with placebo (Carpenter, 2012; Freeman, 2011). Of the SSRIs, fluoxetine (Prozac, Sarafem) and sertraline (Zoloft) appear to be less effective (Gordon, 2006; Grady, 2007; Loprinzi, 2002; Suvanto-Luukkonen, 2005).

Of the SNRIs, desvenlafaxine (Pristiq) significantly reduces the number of hot flushes by approximately 60 percent and their severity by 25 percent compared with placebo (Archer, 2009a,b; Speroff, 2008). A 1-year study noted that these improvements persisted (Pinkerton, 2013).

Importantly, benefits of SSRIs are balanced against drug side effects, which can include nausea, headache, diarrhea, insomnia, jitteriness, fatigue, and sexual dysfunction. With the SNRIs, hypertension may also be aggravated (Handley, 2015).

Clonidine

The centrally active α₂-adrenergic receptor agonist clonidine (Catapres) has been effective in some clinical trials (Nagamani, 1987). Many of these RCTs specifically evaluated benefits for breast cancer survivors and showed improvement in vasomotor symptoms (Boekhout, 2011; Buijs, 2009; Loibl, 2007). Notably, hypotension, dry mouth, dizziness, constipation, and sedation may limit its use. For many women, low-dose clonidine is ineffective, and thus adequate therapy may require substantially higher doses that may magnify side effects.

Gabapentin

Gabapentin (Neurontin) is structurally related to the neurotransmitter γ-aminobutyric acid (GABA), but its exact mechanism of action is unknown. Currently, gabapentin is FDA-approved to treat seizures and neuropathic pain. However, it has extensive off-label use for various other neurologic conditions.

In 2003, Guttuso and colleagues evaluated the use of gabapentin, 900 mg daily, for treatment of vasomotor symptoms. They found a 45-percent reduction in hot flush frequency compared with a 29-percent reduction with placebo. Moreover, Reddy and coworkers (2006) conducted an RCT in which 60 postmenopausal women received gabapentin, 2400 mg/d; oral CEE, 0.625 mg/d; or placebo for 12 weeks. The reductions in the hot flush composite scores for both estrogen (72 percent) and gabapentin (71 percent) were greater than that associated with placebo (54 percent). However, headache, dizziness, and disorientation occurred in almost 25 percent of the women treated with gabapentin.

Complementary and Alternative Medicine

Phytoestrogens

Phytoestrogens (isoflavones) are plant-derived compounds that bind to estrogen receptors and have both estrogen agonist and antagonist properties. They are found in soy products and red clover. In sum and outlined subsequently, data supporting their effectiveness for vasomotor symptom treatment show no conclusive efficacy (Lethaby, 2013).

With soy products, although the mechanisms of action are not fully understood, they appear to bind to the estrogen receptor. For this reason, one should not assume these dietary supplements are safe for women with estrogen-dependent cancers. That said, sonographic surveillance has not shown increased endometrial thickness or altered lipid profiles in women taking isoflavones (Palacios, 2010; Quaas, 2013; Ye, 2012). For treatment of hot flushes, data supporting isoflavone efficacy are mixed (Albertazzi, 1998; Cheng, 2007; Liu, 2014; Quella, 2000). Moreover, the effects of soy protein found in various food preparations are not bioequivalent. Even soy foods are not necessarily reliable sources of biologically active isoflavones.

Flaxseed or flaxseed oil (Linum usitatissimum) is rich in α-linolenic acid, a form of omega-3 fatty acid. Also known as linseed, flaxseed is touted to reduce vasomotor symptoms. However, data regarding its efficacy for this indication are contradictory (Lemay, 2002; Lewis, 2006; Pruthi, 2012).

Red clover (Trifolium pretense) is a member of the legume family. It contains at least four estrogenic isoflavones and is therefore marketed as a phytoestrogen source. Several studies and analyses, however, have failed to demonstrate an effect over placebo in the treatment of menopausal symptoms (Geller, 2009; Nelson, 2006; Tice, 2003).

Dong quai, also translated as don kwai, dang gui, and tang kuei, is a Chinese herbal medicine derived from the root of Angelica sinensis and is the most commonly prescribed Chinese herbal medicine for “female problems.” Within traditional Chinese medicine practice, dong quai is suggested to exert estrogenic activity. In most studies, however, its benefit cannot be substantiated (Haines, 2008; Hirata, 1997). Notably, dong quai contains numerous coumarin-like derivatives and may cause excessive bleeding or interactions with other anticoagulants. This herbal agent also contains psoralens and is potentially photosensitizing.

Black cohosh (Cimifuga racemosa) is also thought to have estrogenic properties, although its mechanism of action is unknown. In their Cochrane Review, Leach and Moore (2012) found insufficient evidence to support its use for vasomotor symptom relief. Although few adverse effects have been reported, the long-term safety of this product is unknown.

Phytoprogestins

Extracts, tablets, and creams derived from yams are claimed to be progesterone substitutes. Specifically, claims are made that the plant sterol dioscorea is converted into progesterone in the body and alleviates “estrogen dominance.” However, there is no human biochemical pathway for bioconversion of dioscorea to progesterone in vivo.

In contrast, Mexican yam extract contains considerable diosgenin, an estrogen-like substance found in plants. Some estrogen effects might be expected from eating these yam species, but only if large quantities of raw yams are consumed. Yams from the grocery store generally are not the varieties known to contain significant amounts of dioscorea or diosgenin.

Based on the lack of bioavailability, the hormones in wild and Mexican yam would not be expected to have efficacy. Wild yam extracts are neither estrogenic nor progestational, and although many yam extract products contain no yam, some are laced with progestins. There are no published reports demonstrating the effectiveness of wild yam cream for postmenopausal symptoms. Moreover, oral ingestion does not produce serum levels.

Vitamin E

In a few studies in breast cancer survivors, vitamin E provided minimal or no vasomotor symptom improvement (Biglia, 2009; Rada, 2010).

Lifestyle Changes

Practices that lower core body temperature such as cooling the room, dressing in layers, and consuming cool drinks may temporarily help with night sweats and flushing (American College of Obstetricians and Gynecologists, 2014b). However, behavioral efforts to curb the frequency of hot flushes have no firm support in RCTs. For example, little evidence demonstrates the efficacy of relaxation techniques, acupuncture, exercise, and yoga to control vasomotor complaints (Daley, 2014; Dodin, 2013; Newton, 2014; Saensak, 2014).

Indications

The primary goal of osteoporosis treatment is fracture prevention in women who have low BMD or additional risk factors for fracture (Fig. 22-1). Toward this end, therapy aims to stabilize or increase BMD. Treatment includes lifestyle changes and often pharmacologic therapy. The key lifestyle modifications that can decrease fracture risks of in postmenopausal women include regular weight-bearing exercise and a balanced diet with adequate calcium and vitamin D. Factors for patients to avoid include smoking, an excessively low body weight, excessive alcohol intake, and fall risks at home (Christiansen, 2013).

With pharmacologic therapy, several organizations offer guidelines. The National Osteoporosis Foundation (NOF) (2014), American Association of Clinical Endocrinologists (AACE) (Watts, 2010), and North American Menopause Society (NAMS) (2010) recommend starting therapy for: (1) all postmenopausal women with total hip, femoral neck, or spine T-scores at or below –2.5; (2) for those with an osteoporotic vertebral or hip fracture; and (3) all postmenopausal women with total hip or spine T-scores from –1.0 to –2.5 who have one or more additional risk factors for fracture, such as those in Table 21-8.

Drugs prescribed for fracture prevention attempt to restore and balance bone remodeling by: (1) reducing bone resorption, termed antiresorptive agents, or (2) stimulating bone formation, termed anabolic agents. Most of the bone-active agents currently available in the United States inhibit bone resorption. Estrogen, SERMs, bisphosphonates, denosumab, calcitonin, and vitamin D each have antiresorptive properties (Table 22-4). All have been shown to halt bone loss, and most also increase BMD. Improvement in BMD with therapeutic intervention varies according to bone composition. Therapies that prevent bone resorption will act most quickly on bone that has high trabecular content and rapid turnover, such as the vertebrae. Alternatively, the impact of drug therapies on the hip may be delayed because the hip is composed of approximately 50 percent trabecular and 50 percent cortical bone (Fig. 21-5). In contrast to these antiresorptive agents, recombinant parathyroid hormone (PTH 1–34), known as teriparatide, is anabolic and leads to BMD increases.

For osteoporosis, each agent differs in its indication for prevention, for treatment, or both. HT is indicated for the prevention of osteoporosis. Bisphosphonates and SERMs can be used for prevention and for treatment. Calcitonin and teriparatide are approved for treatment. Denosumab is a monoclonal antibody to the receptor activator of nuclear factor (RANK) ligand and is approved for osteoporosis treatment.

Hormonal Therapy

Estrogen and Progesterone Replacement

As estrogen levels decline, bone-remodeling rates increase and favor bone resorption over bone formation. In observational studies, HT reduces osteoporosis-related fractures by approximately 50 percent when started soon after menopause. Continued long term, HT significantly decreases fracture rates in women with established disease (Tosteson, 2008). Results from 57 randomized, placebo-controlled trials show that HT reduces the rate of bone resorption and results in an increase in BMD (Wells, 2002). The WHI controlled trials confirmed a significant 33-percent reduction in hip fractures in healthy postmenopausal women receiving HT after an average follow-up of 5.6 years. Notably, hip fracture reduction was not limited to women with osteoporosis, as in trials of other pharmacologic agents (The Women’s Health Initiative Steering Committee, 2004). Even very low estrogen doses combined with calcium and vitamin D produce significant increases in BMD compared with placebo (Ettinger, 2004; Prestwood, 2003). Specifically, oral daily 0.2-mg doses of estradiol; 0.3-mg doses of CEE; or transdermal estradiol patch delivering 0.014 mg/d are suitable.

Unfortunately, this preventive effect is lost rapidly following HT discontinuation (Barrett-Connor, 2003). Women participating in the National Osteoporosis Risk Assessment (NORA) trial who had discontinued estrogen therapy within the 5 years preceding the study demonstrated a significantly higher hip fracture risk than did women who had never received estrogen therapy. In addition, current HT users in the NORA trial had a 40-percent reduction in hip fractures, which was lost by past users. Thus, fracture risk and the potential need for an alternative therapy is ideally assessed when women discontinue HT.

Bazedoxifene

The drug that combines CEE with the SERM bazedoxifene is indicated to alleviate vasomotor symptoms and prevent postmenopausal bone loss. In the randomized SMART trials noted earlier, investigators found a significantly increased adjusted mean percentage BMD change in the lumbar spine from baseline to 24 months in those taking BZA plus CEE compared with women using placebo (Lindsay, 2009). Findings were similar for total hip BMD. Moreover, the FDA-approved dose of CEE (0.45 mg) plus BZA (20 mg) does not cause a change in breast density or endometrial thickness compared with placebo (Pinkerton, 2014).

Raloxifene

This SERM is approved for the prevention and treatment of osteoporosis and to reduce the risk of invasive breast cancer. It activates estrogen receptors in the bone but does not appear to activate those in the breast or uterus.

Of bone loss sites, raloxifene may be most appropriate for prevention and treatment of vertebral disease. For example, raloxifene prevented vertebral fractures in the Multiple Outcomes of Raloxifene Evaluation (MORE) trial, which enrolled 7705 postmenopausal women with osteoporosis. The beneficial effects of oral raloxifene, 60 mg/d, appeared rapidly, and clinical vertebral fracture risk was reduced by 68 percent following the first year of therapy. In addition, this effect was sustained over time. At 4 years of treatment, dosages of 60 mg daily led to a 36-percent reduction in fractures, and 120 mg each day produced a 43-percent decline (Delmas, 2002; Ettinger, 1999). However, in the MORE trial, Ettinger and associates (1999) reported that raloxifene therapy compared with placebo was not associated with significant reductions in nonvertebral fracture risks at 3 and 4 years. A more recent retrospective cohort study showed that patients treated with alendronate and raloxifene had similar adjusted fracture rates in up to 8 years of compliant treatment (Foster, 2013).

In 2012, Chung and colleagues showed that raloxifene, but not bisphosphonates, significantly suppressed circulating concentration of sclerostin, an inhibitor of bone formation, suggesting that sclerostin may in part mediate the action of estrogen on bone metabolism.

In addition to its bone effects, raloxifene may protect against breast cancer, as suggested by observational studies of various clinical trials (Barrett-Connor, 2006). The incidence of breast cancer was evaluated as a secondary end point in the MORE trial. Investigators found that raloxifene was associated with a 65-percent relative risk reduction in all breast cancers. Of specific breast cancer subtypes, they noted a 90-percent reduction in estrogen receptor–positive cancers, a 12-percent reduction in estrogen receptor–negative breast cancers, and a 76-percent relative risk reduction in invasive breast cancer.

Raloxifene may not have the same increased cardiovascular risk profile as estrogen. In a MORE post hoc analysis, 4 years of raloxifene therapy had no adverse effect on cardiovascular events in the overall cohort. Advantageously, it did result in a significant 40-percent reduction in the incidence of cardiovascular events among a subgroup of women with increased cardiovascular risk (Barrett-Connor, 2002).

Of side effects, hot flushes are associated with raloxifene therapy, although the incidence is low (Cohen, 2000). In addition, raloxifene, 60 mg daily for 4 years, has been associated with an increased risk of thromboembolic events. In one study, the relative risk associated with any dosage of raloxifene was 2.76 for deep-vein thrombosis, 2.76 for pulmonary embolism, and 0.50 for retinal vein thrombosis (Delmas, 2002). It is therefore contraindicated in patients with a history of VTE and is used with caution in women with hepatic or moderate to severe renal impairment.

Nonhormonal Antiresorptive Agents

Bisphosphonates

Four bisphosphonates are currently available and include alendronate (Fosamax, Binosto), risedronate (Actonel, Atelvia), ibandronate (Boniva), and zoledronate (Reclast). The first three are taken orally, but zoledronate is an intravenous infusion. These agents chemically bind to calcium hydroxyapatite in bone (Figs. 22-2 and 22-3). They decrease bone resorption by blocking the function and survival but not the formation of osteoclasts (Diab, 2012; Russell, 2008).

The oral bisphosphonates display poor bioavailability and therefore are taken on an empty stomach with adequate water for proper dissolution and absorption. In general, these agents have a favorable overall safety profile, and adverse event rates are comparable with placebo (Black, 1996; Harris, 1999). However, bisphosphonates may cause upper gastrointestinal inflammation, ulceration, and bleeding (Lanza, 2000). Thus, to aid delivery to the stomach and reduce the risk of esophageal irritation, dosing instructions are reinforced with each patient. First, bisphosphonates are taken in the morning with a full glass of water. During the 30 minutes following administration, no other food or beverages are consumed. Finally, women must remain upright (sitting or standing) for at least 30 minutes after ingesting the drug.

In terms of long-term safety, concerns about two uncommon adverse events have emerged: osteonecrosis of the jaw (ONJ) and atypical femur fractures (AFFs) (Diab, 2013). ONJ is defined as exposed necrotic bone in the maxillofacial region that fails to heal after 8 weeks. It can be associated with pain, paresthesia, soft tissue ulceration and swelling, and loosening of teeth. This can develop spontaneously but is generally associated with invasive dental procedures. ONJ has been described in patients receiving chronic bisphosphonate therapy for osteoporosis treatment but appears to be much more common in cancer patients receiving higher-dose bisphosphonates (Khosia, 2007).

AFFs are stress fractures that are frequently bilateral, are typically associated with minimal or no trauma, and are heralded by prodromal pain in the fracture region (Dell, 2012). One case-controlled study found that longer bisphosphonate use (5 to 9 years) was associated with a greater risk of AFFs compared with shorter use (<2 years) (Meier, 2012). One review showed that bisphosphonate exposure was associated with an increased risk of AFF and an adjusted risk ration of 1.70 (Gedmintas, 2013). However, in this analysis, the large variation in relative risks reported in the different studies suggests significant heterogeneity in the patient populations.

No causal relationship has been established between prolonged bisphosphonate exposure and either of these outcomes. Even though the risks of ONJ and AFF may increase after 5 years of bisphosphonate therapy, the likelihood remains low. The FDA suggests reevaluating individually the need for continuing bisphosphonate therapy beyond 3 to 5 years (Whitaker, 2012). A “drug holiday” may be considered because of the unique pharmacokinetics of bisphosphonate. These agents accumulate in the skeleton to create a reservoir that continues to be released for months or years after treatment. Although this provides some residual benefit in terms of fracture reduction after an initial 3- to 5-year course of therapy, continuing treatment for 10 years seems to be a better choice for high-risk patients. If a drug holiday is advised, risk is reevaluated sooner for drugs with lower skeletal affinity. Thus, reassessment may be prudent after 1 year for risedronate, 1 to 2 years for alendronate, and 2 to 3 years for zoledronate (Diab, 2014).

Alendronate

This bisphosphonate is approved for the treatment and prevention of osteoporosis. Alendronate has been shown to reduce the risk of vertebral fractures in postmenopausal women with low BMD or osteoporosis, either with or without existing vertebral fractures (Black, 1996). Alendronate also reduces nonvertebral fracture risk in women with osteoporosis. Among women with osteoporosis who participated in the Fracture Intervention Trial (FIT), the risk of nonvertebral fractures was reduced by month 24. In addition, the effects of alendronate are sustained. For example, women who used alendronate for 5 years and then discontinued use for a subsequent 5 years had nonvertebral fracture rates comparable to those of women using the drug for 10 years (Black, 2006; Bone, 2004).

Risedronate

This bisphosphonate is an effective agent in the prevention and treatment of postmenopausal osteoporosis. The strongest data supporting its efficacy stem from the Vertebral Efficacy with Risedronate Therapy (VERT) trials, conducted multinationally and also in North America. In the VERT multinational trial, Reginster and coworkers (2000) showed that risedronate reduced the risk of new vertebral fractures by 61 percent at 1 year and by 49 percent at 3 years of use. Moreover, both VERT trials found significant reductions in vertebral fractures as early as 6 months after initiation of risedronate therapy (Roux, 2004). Two extensions of these trials have provided evidence of sustained efficacy. The continuation of risedronate therapy for 2 additional years (5 years total) in the multinational VERT study was associated with a 59-percent reduction in new vertebral fractures compared with placebo.

Ibandronate

This bisphosphonate is approved for the prevention and treatment of postmenopausal osteoporosis. Ibandronate is an effective agent, and data from the Oral Ibandronate Osteoporosis Vertebral Fracture Trial in North America and Europe (BONE) trial showed that daily ibandronate lowered incident vertebral fracture risk by 62 percent (Chesnut, 2004). To improve compliance, this drug was evaluated as a once-monthly therapy. Once-monthly oral ibandronate is at least as effective and well tolerated as daily treatment (Miller, 2005; Reginster, 2006). Moreover, once-monthly administration may be more convenient and thereby improve compliance rates.

Zoledronate

This bisphosphonate is approved for the prevention and treatment of postmenopausal osteoporosis. Zoledronate is effective, and data from the Health Outcomes and Reduced Incidence with Zoledronic Acid Once Yearly (HORIZON) Pivotal Fracture Trial showed that yearly zoledronate infusion lowered incident vertebral fracture risk by 70 percent compared with placebo (Black, 2007).

Denosumab

This is a fully human monoclonal antibody to RANK ligand, which is described and illustrated in Chapter 21. To summarize, denosumab’s binding to RANK ligand inhibits osteoclast development and activity. This in turn decreases bone resorption and increases BMD. In a manufacturer-sponsored trial, 7868 women randomly received either denosumab or placebo subcutaneously every 6 months for 3 years (Cummings, 2009). Relative risk for new radiographically diagnosed vertebral fractures was 68-percent lower in the denosumab group. The risk for hip fractures was 40-percent lower in the denosumab group. Overall incidence of significant adverse events was similar between groups.

Calcitonin

The polypeptide hormone calcitonin decreases the rate of bone absorption by inhibiting resorptive activity in osteoclasts. Calcitonin is a protein, and as such, oral administration leads to its digestion. For this reason, it is delivered as an injection or nasal spray (Fortical, Miacalcin). Salmon calcitonin nasal spray has been associated with a reduction in vertebral fracture risk among postmenopausal women with osteoporosis. In the Prevent Recurrence of Osteoporotic Fractures (PROOF) study, calcitonin nasal spray, 200 IU administered daily for up to 5 years reduced the risk of vertebral fractures by 33 percent compared with placebo. However, vertebral fracture reduction was not seen at lower (100 IU/d) or higher (400 IU/d) dosages (Chesnut, 2000). Moreover, in this study, calcitonin failed to produce significant reductions in nonvertebral fracture.

Some observational data suggest that calcitonin has an analgesic effect independent of its effect on bone (Häuselmann, 2003; Ofluoglu, 2007). This analgesic effect may make this agent particularly useful as an adjunct to other therapies for osteoporosis in women with painful, symptomatic fracture (Blau, 2003). Injectable or intranasal calcitonin is associated with an 8- to 10-percent incidence of nausea or gastric discomfort and a 10-percent incidence of local site reactions. These symptoms tend to decrease in severity with continued use. Nasal symptoms such as rhinitis occur in 3 percent of patients treated with intranasal calcitonin (Cranney, 2002).

Parathyroid Hormone

Recombinant parathyroid hormone (PTH 1–34), known as teriparatide, is given by daily subcutaneous injection and is approved by the FDA for the treatment of postmenopausal women with established osteoporosis, who are at high risk for fracture. Teriparatide (Forteo) increases osteoblast numbers and activity by recruiting new cells and reducing apoptosis of differentiated osteoblasts. At low daily doses of teriparatide, the anabolic effects of PTH predominate. This is in contrast to the catabolic effects generally associated with long-term, higher-dose, and chronic exposure to PTH.

Clinical studies indicate that teriparatide increases bone quality by increasing bone density, turnover, and size (Rubin, 2002). Moreover, improvements in microarchitectural elements are evident in both cancellous and cortical regions. In women with postmenopausal osteoporosis, teriparatide, 20 or 40 μg/d, administered for approximately 21 months, was associated with 65-percent and 69-percent reductions in vertebral fractures, and 35-percent and 40-percent reductions in nonvertebral fractures, respectively (Neer, 2001).

Similar findings were reported in a study of 52 women treated with concomitant teriparatide and HT compared with HT alone (Lindsay, 1997). In this study, at the end of 3 years, increases in vertebral, total hip, and total body BMD were 13.4 percent, 4.4 percent, and 3.7 percent, respectively, in the combined treatment group. The addition of alendronate to teriparatide, however, does not appear to enhance effects on BMD (Gasser, 2000). The effects of combination use of PTH with other bisphosphonates are not known.

In general, PTH is safe and well tolerated, although additional data from long-term studies are needed. The most frequent treatment-related adverse events in clinical trials of teriparatide were dizziness, leg cramps, nausea, and headache. Toxicity studies with rats have shown an increased risk of osteosarcoma. However, there are significant differences in bone metabolism between rats and humans, and the applicability of rat data to humans is unclear. That said, a black box warning has been included on the product labeling in the United States, and teriparatide use is avoided by patients at increased risk for skeletal malignancy. Use for more than 2 years is not recommended due to the potential for side effects (Tashjian, 2002).

In sum, denosumab seems to be as effective as teriparatide and zoledronate. Osteonecrosis of the jaw and fragility fractures associated with long-term bisphosphonate use are unlikely to be linked to short-acting agents such as denosumab. Because denosumab is an antibody, its potential to affect the immune system requires scrutiny. Long-term adherence to oral bisphosphonate therapy is often poor, making the relative ease of biannual injections attractive. Although teriparatide and intravenous bisphosphonates are expensive, weekly oral alendronate is available at low cost as a generic. Clinically, cost will likely play a central role in determining how these agents are selected.

Nonpharmacologic Therapy

Calcium

Nonpharmacologic interventions are cornerstones of osteoporosis prevention and include dietary modifications, exercise programs, fall prevention strategies, and education. Of these, adequate daily calcium intake is essential for bone maintenance. For women aged 31 to 50 years, the recommended dietary reference intake (DRI) is 1000 mg daily, and for those 51 years and older, 1200 mg is recommend daily. Ideally, these DRIs are obtained through diet alone, but supplementation is used to attain these levels if needed (Institute of Medicine, 2010; Prentice, 2013). Few women meet these goals, and calcium deficiency is widespread. For example, more than 90 percent of women fail to take in enough dietary calcium to meet DRIs put forth by the Institute of Medicine. Although poor calcium intake is observed at all ages, it appears to be most common among older individuals. Specifically, fewer than 1 percent of women 71 years or older actually meet recommended goals.

Calcium supplementation combined with vitamin D administration has been associated with reduced bone loss and decreased risk for fractures in several prospective studies (Chapuy, 1992; Dawson-Hughes, 1997; Larsen, 2004). Prentice and associates (2013) examined the health benefits and risk of calcium and vitamin D supplementation using data from the WHI. This showed a substantial reduction in hip fracture risk. Notably, supplementation must be continued long term for efficacy to be sustained.

Vitamin D

The DRI of vitamin D is 600 IU daily for a postmenopausal woman who is not at high risk for fractures or falls. For persons who have a high risk of osteoporosis or who are older than 70, 800 IU daily is recommended (Institute of Medicine, 2010). As with calcium, the prevalence of vitamin D deficiency is high, especially in the elderly. Deficiency leads to poor calcium absorption, secondary hyperparathyroidism, increased bone turnover, increased rates of bone loss, and, if severe, impaired bone mineralization. In addition, vitamin D deficiency causes muscle weakness and is associated with an increase in rates of falls. Vitamin D supplementation can reverse many of these effects and significantly reduce falls and hip fractures (Dawson-Hughes, 1997).

The metabolite 25-hydroxyvitamin D is considered to be the best clinical measure of vitamin D stores (Rosen, 2011). Vitamin D deficiency is defined as a 25-hydroxyvitamin D serum level below 10 ng/mL. Vitamin D insufficiency is a serum level of 25-hydroxyvitamin D between 10 and 30 ng/mL.

Diet

A relationship between protein intake and BMD has been reported, but a relationship with fractures has not been described. Using data from the Third National Health and Nutrition Examination Survey (NHANES III), Kerstetter and colleagues (2000) demonstrated a significant association between protein intake and total femur BMD among non-Hispanic white women aged 50 years and older. Moreover, protein supplementation (20 g/d) five times weekly for 6 months following hip fracture was associated with a 50-percent reduction in femoral bone loss at 1 year compared with placebo.

Although no specific recommendations regarding protein intake can be made based on the limited data available, it seems prudent for clinicians to ensure that their patients eat healthy diets that provide the daily DRI of protein. As put forth by the Institute of Medicine, diets ideally contain at least 46 g/d for women (Dawson-Hughes, 2002). There may be upper limits for desirable protein intake as well. Excess urinary calcium excretion has been observed in association with the large acid loads delivered by very-high-protein diets (Barzel, 1998).

Caffeine consumption does not appear to influence bone health in postmenopausal women who maintain an adequate daily intake of calcium and vitamin D. However, one longitudinal study showed that even moderate amounts of caffeine (two to three servings of coffee daily) may lead to bone loss in women with low calcium intake (<800 mg/d) (Harris, 1994).

Calcium reabsorption is directly proportional to sodium reabsorption in the renal tubule. Accordingly, increases in dietary sodium have been observed to cause increases in urinary calcium excretion and corresponding increases in biochemical markers of bone turnover. Specifically, a relationship between high sodium intake (>1768 mg/d) and lower BMD has been described (Sellmeyer, 2002). This sodium effect appears to be independent of calcium intake and activity levels. As with caffeine, sodium intake moderation is a reasonable precautionary measure until this relationship is fully understood.

Physical Preventions

Small but statistically significant increases in BMD have been observed in postmenopausal women participating in exercise programs, including aerobic exercise and resistance training (heavy weight with few repetitions). One metaanalysis of 43 RCTs concluded that lower limb resistance exercise was most effective for femur neck BMD, and combination exercise programs were most effective for vertebral BMD (Howe, 2011). Another analysis that focused on walking as exercise showed a benefit for femur neck BMD (Ma, 2013).

Although an increase in bone density may occur, especially at the sites at which the exercise is directed, benefits of exercise may likely also be due to factors other than greater BMD (Carter, 2002). For example, an association between exercise and reduced falls has been reported. Improvements in balance, stronger muscles, better muscle tone, and stronger, more flexible bone all undoubtedly contribute to fracture reduction.

Falls are responsible for more than 90 percent of hip fractures (Carter, 2002). Sideways falls appear to be the most detrimental and were independently associated with hip fracture in a study by Greenspan and associates (1998). Therefore, fall prevention is essential for women with osteopenia or osteoporosis. Living conditions are modified to minimize falls by reducing clutter and implementing nonslip tiles, rugs with nonskid backing, and night lights.

Hip protector padding was also initially thought to reduce hip fractures in elderly adults. However, one Cochrane database analysis indicates that the effect of hip protectors to lower hip fracture risk is small, and compliance is low (Santesso, 2014). Falls and fractures often occur at night, when women are likely to have taken off their hip protectors. This may result from the bulkiness of hip protectors, which are uncomfortable to wear while sleeping (van Schoor, 2003).

Dyspareunia

Estrogen Replacement

Low estradiol levels commonly lead to the genitourinary syndrome of menopause. This syndrome may include but is not limited to: (1) genital symptoms of dryness, burning, and irritation; (2) sexual symptoms of lack of lubrication, discomfort or pain, and impaired function; and (3) urinary symptoms of urgency, dysuria and recurrent urinary tract infections (Portman, 2014). Data from the Yale Midlife Study showed a close relationship between serum estradiol levels and sexual problems. In this study, significantly more women with estradiol levels less than 50 pg/mL reported vaginal dryness, dyspareunia, and pain compared with women whose estradiol levels were greater than 50 pg/mL (Sarrel, 1998). Prospective records of coital behavior and concomitant steroid analysis revealed that women with estradiol levels less than 35 pg/mL reported significant reductions in coital activity.

Estrogen replacement effectively reverses atrophic changes. Vaginal atrophy and diminished vaginal mucosal elasticity, vaginal fluid secretion levels, blood flow, and sensorimotor responses are improved by either topical or systemic estrogen (Dennerstein, 2002). In one metaanalysis evaluating RCTs from 1969 to 1995, investigators found that compared with placebo, oral or vaginal estrogens significantly improved vaginal atrophy symptoms, dyspareunia, and vaginal pH (Cardozo, 1998). If oral and vaginal estrogens were compared, vaginal products had greater patient acceptance, lower systemic estradiol concentrations, yet significant improvement of dyspareunia and pH changes.

Of vaginal topical agents, available forms include creams, continuous-release rings, and tablets (Table 22-5). In comparing types during a 12-week study period, Ayton and colleagues (1996) found that a continuous low-dose estradiol-releasing vaginal ring (Estring) provided comparable relief to CEE vaginal cream used for 12 weeks. In addition, study patients found the vaginal ring significantly more acceptable than the cream. The ring is prescribed as a single unit. Each unit contains 2 mg of estradiol and is worn vaginally for 90 days and then replaced.

Alternatively, a 10-μg 17β-estradiol tablet (Vagifem) is available for vaginal application. One tablet is inserted daily for an initial 2 weeks of treatment and is followed by twice-weekly application. These tablets and CEE vaginal cream have been found to be equivalent in relieving symptoms of atrophic vaginitis (Rioux, 2000). Advantageously, women using vaginal tablets had less endometrial proliferation or hyperplasia than those using cream. Additionally, tablets were rated significantly more favorably than the cream, and their use was associated with fewer patient withdrawals from the study.

Studies of the vaginal tablets and ring have confirmed endometrial safety at 1 year, but studies of the long-term effects of low-dose vaginal ET on the endometrium are lacking. Women using vaginal ET should report any vaginal bleeding, and this bleeding should be evaluated thoroughly. Progestins typically are not prescribed to women using only low-dose vaginal estrogen products. If a woman is at high risk of endometrial cancer or is using a higher dose of vaginal ET, transvaginal ultrasound or intermittent progestogen therapy may be considered (North American Menopause Society, 2013).

Ospemifene

Marketed as Osphena, this SERM is taken orally daily in a 60-mg dose to treat moderate to severe dyspareunia from menopause-associated vulvar and vaginal atrophy. In one RCT with 826 women, ospemifene was significantly superior to placebo in improving the vaginal pH, the self-assessed severity of most bothersome symptoms, and the percentage of superficial vaginal epithelial cells (Bachmann, 2010). As noted in Chapter 21, superficial cells are increased relative to parabasal cells in estrogen-sufficient vaginal epithelia. Ospemifene was shown to be safe and well tolerated.

Hot flushes, a common adverse event associated with SERM use, were more frequently reported in the ospemifene groups. These did not raise discontinuation rates and were generally mild. Urinary tract infections were slightly more prevalent. The endometrial effects of ospemifene were negligible. There were no cases of endometrial hyperplasia or carcinoma and no reports of vaginal bleeding/spotting in this 3-month trial.

Vaginal Lubricants and Moisturizers

Various water-soluble vaginal lubricants are available over the counter for treatment of vaginal dryness with coitus. Most commonly used water-based lubricants include K-Y Jelly, Astroglide, and Slippery Stuff. They can be applied before intercourse to the vaginal introitus. Alternatively, a polycarbophil-based gel (Replens) offers a more sustained correction of vaginal dryness. This gel is an acidic hydrophilic insoluble polymer, which can hold water to act as a vaginal moisturizer. The polymer binds to the vaginal epithelium and is sloughed with epithelial layer turnover. In addition, the acidity of the gel helps to lower the vaginal pH to that found in premenopausal women. This vaginal moisturizer is used three times weekly, but the schedule can be individualized.

As a group, these agents are an effective first-line tool for vaginal atrophy symptoms. From their review of 44 RCTs and prospective comparative trials, Rahn and associates (2014) suggest vaginal moisturizers/lubricants or vaginal estrogen for relief in women with a sole complaint of vaginal dryness, dyspareunia, itching or burning, dysuria, or urinary urgency. For those with multiples of these, they favored vaginal estrogen therapy.

Libido

A decline in libido is common during menopause (Sarrel, 1990). As described in Chapter 21, the cause is no doubt multifactorial. Estrogen therapy can be helpful for GSM symptoms, as described in the last section. However, HT is not recommended as the sole treatment of other problems of sexual function, including diminished libido (North American Menopause Society, 2012).

Flibanserin (Addyi) is an agent that has both serotonin-receptor agonist and antagonist activity. In 2015, this agent was approved by the FDA to treat female hypoactive sexual desire disorder (HSDD). For postmenopausal women, data derive mainly from the SNOWDROP trial (Simon, 2014). In this study, flibanserin, compared with placebo, provided a significant increase in the number of satisfying sexual events (1.0 versus 0.6) and in the Female Sexual Function Index (FSFI) desire domain score (0.7 versus 0.4). Notably, these gains are measured against drug side effects that can include hypotension and syncope, especially with alcohol use. To emphasize this interaction, as part of the FDA-approval, prescribing clinicians must participate in a risk evaluation and mitigation strategy (REMS) program. An additional black box warning describes hypotension and syncope when flibanserin is administered with moderate or strong CYP3A4 inhibitors or in patients with hepatic impairment (Sprout Pharmaceuticals, 2015).

Androgen replacement in women with HSDD is a controversial topic. Although some studies have documented an association between androgen replacement and improved sexual desire, large quality trials with long-term follow-up are needed (Lobo, 2003; Pauls, 2005; Shifren, 2000). Dehydroepiandrosterone (DHEA) is an androgen precursor, and reviews of RCTs show no or only slightly improved sexual functioning in supplemented postmenopausal women (Elraiyah, 2014; Scheffers, 2015).

Symptoms of androgen insufficiency include diminished sense of well-being, persistent fatigue, sexual function changes, and low levels of serum free testosterone. Women with these findings may be offered replacement. Importantly, candidates are counseled that androgen replacement therapy is off-label and not FDA-approved. Therapy should be performed under close clinician supervision. Early effects of androgen therapy include acne and hirsutism, with one study reporting a 3-percent increased rate of acne in testosterone-therapy groups (Lobo, 2003). Long-term side effects such as male pattern baldness, voice deepening, and clitoral hypertrophy are infrequent within normal androgen levels. Androgen therapy may adversely affect the lipid profile, and long-term effects on cardiovascular and breast cancer risks are unknown (Braunstein, 2007; Davis, 2012, 2013).

Depression in women is common, with a lifetime prevalence of approximately 18 percent (Chaps. 13). Antidepressant medications together with psychotherapy and counseling are the principal therapeutic interventions for women with depression.

HT is not considered treatment for depression, but lesser mood symptoms may be improved concurrent with resolution of hot flushes and disrupted sleep. For example, in the prospective Massachusetts Women’s Health Study, hot flushes and trouble sleeping were highly related to depression and provide support for the “domino” hypothesis that menopausal symptoms are a cause of increased depressed mood at this life stage (Avis, 2001). Several controlled studies have demonstrated that HT can improve depression in perimenopausal women. Notably, most studies involved women with vasomotor symptoms, so it is likely that the depressed mood improvement was predominantly linked to resolution of bothersome hot flushes and sleep disruption and improved quality of life (Soares, 2001; Zweifel, 1997). Women who present with bothersome vasomotor symptoms and associated disordered mood at the time of the menopausal transition may elect a trial of HT for symptom relief. Namely, consideration may be given to those who fail to respond to a conventional first-line intervention, those who refuse to take psychotropic agents, or those who will begin HT for other acute menopausal symptoms and who could delay antidepressant therapy until determining whether estrogen treatment is sufficient.

As people age, their skin loses elasticity, collagen fibers, vascularity, and moisture. As a result, the skin lies more loosely, and lines appear where the facial muscles attach to the skin’s undersurface. Many factors play a role in the rate and degree of aging. First and foremost is genetics. People with thin, dry, fair skin realize signs earlier. In addition, overexposure to sunlight and excessive use of tobacco and alcohol accelerate skin aging. Thus, prevention of skin aging includes protection from ultraviolet light, avoidance of tobacco, and limited alcohol intake.

Skin is a hormonally sensitive structure, and both estrogen and androgen receptors have been localized to skin (Hasselquist, 1980). However, it is difficult to separate hormonal deficiency from chronological skin aging and age-related environmental insults such a smoking or photo-aging from sun exposure.

The predominant evidence for an estrogen effect on skin has been derived from observational studies using various estrogen preparations with or without cyclic progestin. Thus, it is difficult to clearly separate the effects of estrogen from estrogen and progestin in many of the studies. Results of RCTs show improvement in certain skin parameters, but these were inconsistent among the trials (Maheux, 1994; Sator, 2007; Sauerbronn, 2000). In the largest RCT, Phillips and associates (2008) noted that low-dose HT did not significantly alter age-related facial skin changes in 320 women compared with placebo. Thus in balancing the risk and benefits, there is currently insufficient evidence to recommend HT to improve age-related skin characteristics.

Leading causes of mortality for two age groups of women are found in Table 22-6. Testing and prevention strategies are aimed at reducing the incidence and effects of these causes. In addition to testing, illness prevention requires patient education to enable women to play an active role in maintaining their own health. Through dialogue and counseling, clinicians and their actively participating patients can reap the benefits of preventive care. Prevention recommendations for many of these causes of morbidity are reviewed in Chapter 1, but a select few found commonly in older populations are discussed below.

Alzheimer Senile Dementia

Dementia is defined as a progressive decline in intellectual and cognitive function. Its causes can be categorized into three broad groups: (1) cases in which the brain is the target of a systemic illness; (2) primary structural causes such as tumor; and (3) primary degenerative diseases of the nervous system, such as senile dementia of the Alzheimer type (SDAT). It is estimated that up to 50 percent of women aged 85 years or older may suffer from senile dementia or SDAT. Early signs of dementia may be subtle, and testing strategies are found in Chapter 1.

The role of estrogen in the prevention of dementia is controversial. Several epidemiologic studies have suggested that HT prevents development of SDAT. Moreover, metaanalyses of observational studies found that HT was associated with a decreased risk of dementia, but it does not seem to improve established disease (Yaffe, 1998; Zandi, 2002). However, data from a large RCT found negative findings for a preventive role. Women enrolled in the Women’s Health Initiative Memory Study (WHIMS), an ancillary study of the WHI, were noted to have increased rates of dementia compared with those given placebo (Shumaker, 2003, 2004). Although this increased risk was statistically significant only in the group of women >75 years, the observation nonetheless is a cause for concern in terms of its long-term implications for HT in older postmenopausal women who are well advanced in menopause. It is unclear whether, similar to CHD, concepts of critical window and timing hypotheses or if the duration of HT have an effect in the prevention of SDAT. Unfortunately, these mixed findings leave unanswered questions regarding HT’s efficacy in preventing dementia in postmenopausal women. Currently, HT is not recommended for this indication.

Urogynecologic Disease

The development of pelvic organ prolapse and urinary incontinence is multifactorial. Thus, the effectiveness of preventive measures such as cesarean delivery, pelvic floor muscle training (Kegel exercises), and estrogen therapy is unclear. Estrogen receptors are found throughout the lower urinary and reproductive tracts. In these areas, hypoestrogenism is associated with collagen changes and diminished vascularity of the urethral subepithelial plexus. However, separating the effects of hypoestrogenism from aging in the genesis of pelvic organ prolapse and urinary incontinence is problematic and discussed in Chapters 23 and 24.

For a woman with obvious lower reproductive tract atrophic changes, a trial of vaginal estrogen treatment for urinary incontinence is reasonable. Vaginal ET reduces irritative urinary symptoms, such as frequency and urgency, and has been demonstrated to reduce the likelihood of recurrent urinary tract infections in postmenopausal women (Eriksen, 1999). However, several other studies evaluating effects of estrogen have noted either de novo development or worsening of incontinence in women using HT (Hendrix, 2005; Jackson, 2006). Accordingly, there is no current indication for the use of HT for the prevention of pelvic organ prolapse or incontinence.

In sum, the body of current evidence shows that HT prescribing is complex and should be tailored to the individual woman’s risk/benefit profile. Thus dose, type, and route of administration are carefully evaluated. While clinicians spout the mantra of “lower doses for shorter periods of time,” there actually are no arbitrary time limits regarding the duration of HT use in the symptomatic woman. It can be used for as long as the woman feels the benefits outweigh the risks for her. Annual or semiannual visits to reevaluate symptoms, side effects, risks, and benefits are tailored to the individual patient.