Advances in the Treatment of Menopausal Symptoms

Abstract

Vasomotor symptoms and vaginal atrophy are both common menopausal symptoms. Hormone therapy is currently the only FDA-approved treatment for hot flashes. Current recommendations are to use the lowest dose of hormone therapy for the shortest period that will allow treatment goals to be met. Although the reanalysis of the WHI in 2007 by Roussow et al. provided evidence of coronary heart safety for users of hormone therapy under the age of 60 years and within 10 years of the onset of menopause, not all women desire or are candidates for hormone therapy. In this review we present an evidence-based discussion considering the effectiveness of hormonal and nonhormonal therapies for the relief of vasomotor symptoms and vaginal atrophy. Concern exists regarding systemic absorption of vaginal estrogen and possible adverse effects on the breast and uterus. Selective estrogen receptor modulators and estrogen agonists offer benefits through targeted estrogen agonist/antagonistic effects and are being evaluated with and without estrogen for symptomatic menopausal women. Centrally acting nonhormonal therapies that are effective for the relief of vasomotor symptoms include various antidepressants, gabapentin and clonidine. A limited number of clinical trials have been conducted with nonprescription remedies, including paced respiration, yoga, acupuncture, exercise, homeopathy and magnet therapy, and some, but not all of these, have been found to be more effective than placebo. Dietary herbal supplements, such as soy and black cohosh, have demonstrated mixed and inconclusive results in placebo-controlled trials. Potential therapies for vasomotor symptoms and vaginal atrophy require randomized, placebo-controlled trials of sufficient duration to establish efficacy and safety. Agents under investigation for vasomotor symptoms relief include neuroactive agents, such as gabapentin and desvenlafaxine; an estrogen receptor-β-targeted herbal therapy, MF-101; and the selective estrogen receptor modulator, bazedoxifene, paired with estrogen.

Continuing Medical Education Online

This activity has been planned and implemented in accordance with the Essential Areas and policies of the Accreditation Council for Continuing Medical Education through the joint sponsorship of MedscapeCME and Future Medicine Ltd. MedscapeCME is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians. MedscapeCME designates this educational activity for a maximum of 1.75 AMA PRA Category 1 Credits™. Physicians should only claim credit commensurate with the extent of their participation in the activity. All other clinicians completing this activity will be issued a certificate of participation. To participate in this journal CME activity: (1) review the learning objectives and author disclosures; (2) study the education content; (3) take the post-test and/or complete the evaluation at http://cme.medscape.com/CME/futuremedicine; (4) view/print certificate.

Learning objectives

Upon completion of this activity, participants should be able to:

• Diagnose menopausal symptoms accurately

• Distinguish the role of hormone therapy for menopausal symptoms

• List nonhormone prescription medications that may improve vasomotor symptoms of menopause

• Analyze nonprescription treatment for menopausal symptoms

Financial & competing interests disclosure

CME Author

Charles P Vega, MD, FAAFP, Associate Professor; Residency Director, Department of Family Medicine, University of California, Irvine, CA, USA. Disclosure: Charles P Vega, MD, FAAFP, has disclosed no relevant financial relationships.

Authors and Disclosures

JoAnn V Pinkerton, MD, Box 801104, University of Virginia Health System, Charlottesville, VA 22908, USA. Disclosure: JoAnn V Pinkerton has disclosed that she has received multicenter research grants from Wyeth Pharmaceuticals Inc.; Solvay Pharmaceuticals, Inc.; and Pfizer Inc. Pinkerton has also disclosed that she has served as a consultant to Wyeth Pharmaceuticals Inc.; Amgen Inc.; Eli Lilly and Company; Duramed Pharmaceuticals, Inc.; Novo Nordisk, and on the data safety monitoring board for Boehringer Ingelheim Pharmaceuticals, Inc.

Dale Stovall, MD, Department of Obstetrics & Gynecology, Divisions of Midlife Health & Reproductive Endocrinology, University of Virginia Health System, Charlottesville, VA 22908, USA. Disclosure: Dale Stovall has disclosed that he has served as a consultant to Amgen Inc. and on the speaker's bureau for Eli Lilly and Company.

Rebecca Kightlinger, Department of Obstetrics & Gynecology, Divisions of Midlife Health & Reproductive Endocrinology, University of Virginia Health System, Charlottesville, VA 22908, USA. Disclosure: Rebecca Kighlinger has disclosed no relevant financial relationships.

The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties. No writing assistance was utilized in the production of this manuscript.

Editor

Elisa Manzotti, Editorial Director, Future Science Group, London, UK. Disclosure: Elisa Manzotti has disclosed no relevant financial relationships.

Common menopausal symptoms include hot flashes, night sweats, mood and cognition disturbances, sleep issues, vaginal dryness and loss of libido, which may be severe enough to affect quality of life. Vasomotor symptoms (VMS) affect 75% of women (estimates range from 68 to 90%) and require office visits in approximately 25% of these women [1]. Hormone therapy (HT) is currently the only FDA-approved treatment for hot flashes and remains the gold standard for treatment. The re-analysis of the WHI in 2007 provided evidence of coronary heart safety for hormone users under the age of 60 years and within 10 years of the onset of menopause [2]. However, not all postmenopausal women with menopausal symptoms are candidates for HT [3].

Relevant placebo-controlled trials starting in 1992 were identified and evaluated, with a focus on the latest literature, through searches of PubMed and Medline databases using terms, such as ‘hot flashe’, ‘menopause’, ‘complementary therapie’, ‘estrogen’, ‘vaginal atrophy’, ‘ᵧ-aminobutyric acid’, ‘antidepressant agent’ and ‘selective estrogen receptor modulator’ (SERMs). Manual searches were performed by scanning bibliographies of all identified papers. Articles written in languages other than English were excluded.

Evidence-based information is provided to assist in decision making relating to estrogen and alternative therapies, recognizing that these decisions and options will change over time as new information becomes available. Lower dose estrogen products that may minimize safety concerns are currently in testing. In the future, centrally acting, nonhormonal therapies and targeted estrogen receptor (ER) selective agents, alone or in combination with estrogens, should be available for the treatment of VMS.

Vasomotor symptoms

The terms ‘hot flashe’, ‘hot flushe’, ‘night sweat’, ‘climacteric symptom’ and ‘VM’ are often used interchangeably. The most prominent VMS of menopause, the hot flash, is described as a progressive spreading of a feeling of heat through the upper body. Individual hot flashes often last between 1 and 5 min, although they can be shorter or last up to 15 min. Although no ‘universal syndrome’ exists, Avis et al. found that hot flashes and night sweats occur together [4]. Perspiration and palpitations may accompany hot flashes and contribute to the discomfort, inconvenience or anxiety associated with VMS. Fatigue may develop as a result of frequent night awakenings.

Multiple studies have demonstrated that these symptoms begin to increase before menopause, peak within 2–3 years of menopause onset then gradually taper-off for most women. The Study of Women's Health Across the Nation (SWAN), a community-based survey of 16,065 women aged 40–55 years, found that hot flashes and night sweats increased as women transitioned through menopause and varied by ethnicity, since 37% of women who complained of hot flashes or night sweats during the study were African–American and 24% were Caucasian [5,6]. SWAN also demonstrated that hot flashes are reported in up to 57% of women during perimenopause and 50% during postmenopause. In a survey of 501 self-selected postmenopausal women aged 29–82 years, Kronenberg et al. found that the majority (60%) experienced hot flashes for less than 7 years but a small minority (15%) continued with hot flashes for more than 15 years [1]. In a post-WHI survey of 151 women, Ness et al. found that 25% of women restarted or continued HT after completion of the WHI study owing to persistent, severe menopausal symptoms [7]. Risk factors [8–10] for hot flashes include surgical menopause (resulting in an abrupt decline of estrogen levels), increased BMI, cigarette smoking (past and current) and strenuous physical activity [11].

Pathophysiology of hot flashes

Hot flashes are associated with peripheral dilation with increased skin temperature and blood flow, typically within the first few seconds of the flash. Freedman found that sweating and skin conductance are increased with hot flashes and that sternal skin conductance was the best objective marker of menopausal hot flashes [12]. Freedman's theory is that hot flashes are triggered by small elevations in core body temperature in symptomatic postmenopausal women, acting within a greatly reduced thermoneutral zone. If the core temperature crosses the upper threshold, a hot flash with sweating and peripheral vasodilation occurs. If the lower threshold is crossed, shivering occurs. Core temperature elevations occur in both symptomatic and asymptomatic women. The difference for symptomatic women is that the thermoneutral zone is narrowed; in asymptomatic women, it is not [13,14].

Studies have demonstrated that neither luteinizing hormone pulses nor changes in endogenous opiates cause hot flashes. Although estrogen levels have been found to be similar in both symptomatic and asymptomatic women, hot flashes are clearly associated with estrogen withdrawal, and hormone therapies have been demonstrated to eliminate or reduce their severity. Freedman has demonstrated that elevated sympathetic activation acting through central α-2 adrenergic receptors contributes to the initiation of hot flashes, possibly narrowing the thermoneutral zone in symptomatic women [15,16]. Changes in neurotransmitter levels, such as serotonin and norepinephrine [17,18], are associated with fluctuations in circulating estrogens and are believed to contribute to the etiology and pathophysiology of hot flashes. Elevated brain norepinephrine has been found to narrow the thermoneutral zone. Estrogen level fluctuations may increase the sensitivity of hypothalamic 5-HT2A receptors. The changes in levels of β-endorphins and other neurotransmitters affects the thermoregulatory center in the hypothalamus [16]. Alternative therapies may affect these central neuromodulators.

The role of estrogen depletion in the development of VMS is supported by the occurrence of rebound symptoms when HT is discontinued. In a cross-sectional survey of 8405 women on study medication or placebo in July 2002 when the WHI study was halted, (surveys were mailed 8–12 months later, with 90% response rate), Ockene et al. found that 21.2% of women in the HT group and 4.8% of women in the placebo group developed moderate-to-severe VMS after stopping therapy [19]. Of the patients who had VMS before enrolling in the HT arm, 55.5% had a recurrence of these symptoms after terminating treatment.

Box 1.

Approach to the treatment of menopausal symptoms

Hormone therapy for moderate-to-severe symptoms

• Hormone therapy:

Vaginal atrophy

• Topical estrogen therapy

• Vaginal moisturizers or lubricants

Alternatives to hormone therapy

• For moderate-to-severe vasomotor symptoms:

• Possibly helpful for mild vasomotor symptoms:

Those at risk of breast cancer

• Selective estrogen receptor modulators

• Selective estrogen receptor modulators in combination with topical estrogen therapy

Future

• Neuroactive agents: gabapentin and desvenlaxfaxine

• Selective estrogen receptor modulators in combination with estrogen

• Estrogen receptor agonist-β MF-101

Treatment for vasomotor symptoms

Hormone therapy

Hormone therapy is approved for the treatment of moderate-to-severe VMS associated with menopause, treatment of vulvar and vaginal atrophy and prevention of osteoporosis ( Box 1 ). HT is recognized as the most effective form of treatment of menopausal symptoms with 80–90% of women obtaining substantial relief of symptoms [20]. The principal results from the WHI for estrogen plus progestin released after an average duration of 5 years revealed a slightly increased risk for breast cancer, coronary heart disease, stroke and venous thromboembolic events (VTEs) and a decreased incidence of fractures and colon cancer [21]. The estrogen-alone arm demonstrated no increased risk of heart events or breast cancer. Stroke, VTE and dementia (in women over the age of 65 years) were increased at 6.7 years [22]. The risk of breast cancer appears to be associated with the duration and possibly with the type of exposure to estrogen and progestins, with some studies suggesting less risk with progesterone compared with progestins [23].

In 2007, Rossouw et al. released a re-analysis of the WHI, which revealed coronary heart safety for those aged under 60 years and within 10 years of menopause onset with a decrease in mortality found for those aged between 50 and 59 years when compared with those aged 70–79 years [2]. Women who began estrogen-only therapy within 10 years of menopause onset or before the age of 60 years achieved a cardioprotective effect [24,25].

The American College of Obstetricians and Gynecologists (ACOG) [26] and the North American Menopause Society [27] continue to support the use of HT as appropriate for the relief of VMS after risk/benefit analysis and with periodic reviews of the decision to continue therapy. Estrogen reduces fractures in women without bone loss and may be an appropriate first choice of treatment for osteoporosis in women with menopausal symptoms. Extended use of estrogen alone or estrogen plus progestin may be appropriate [27] after informed discussion with supervision for persistent VMS, particularly after failing an attempt to withdraw from HT, or for women with moderate-to-severe VMS at high risk for osteoporotic fracture or for whom alternative therapies are not appropriate. Over time, attempts should be made to lower the dose or cease HT and introduce an alternative therapy if appropriate.

Lower doses of hormone therapy

Lower-than-standard doses of estrogen and progestin have been found [28,29] to relieve VMS and vaginal atrophy, protect the endometrium and prevent early postmenopausal bone loss [30]. Using dose-related estimations, Ettinger suggests that after 12 weeks, VMS would be reduced by approximately 38% for placebo, 63% for low-dose estrogen and 83% for standard-dose estrogen [31]. Lower doses may take longer (8–12 weeks) for maximal VMS relief compared with 4 weeks for standard-dose estrogen, and persistence of VMS may occur in up to a third of patients. Low doses of hormones produce approximately half of the bleeding or breast tenderness side effects compared with standard doses, and may require less progestogen for uterine protection [31].

For women aged over 60 years who have osteopenia, even lower doses of HT with increases in serum estradiol from only 4.7–8.5 pg/ml were shown to be effective at preventing bone loss in a 2-year study of women aged 60–80 years [32], although it did not improve postmenopausal symptoms [33]. However, low doses of estradiol in combination with norethindrone have been demonstrated to result in an improvement in menopausal symptom relief [34]. Concern exists regarding the initiation of estrogen therapy in women aged over 60 years owing to increasing adverse events (AEs) observed with age in the WHI [27].

Progesterone-only therapy

Although effective at relieving VMS, progestin-only products have not been widely used owing to concerns regarding the risk of thromboembolic disease and the effect on breast cancer risk [35]. Loprinzi et al. confirmed the effectiveness of medroxyprogesterone acetate (MPA) 400 mg administered intramuscularly compared with venlafaxine 37.5 mg/day for 1 week, followed by 75 mg/day in a randomized, controlled trial (RCT; n = 109) [36]. At 6 weeks, hot flash scores were reduced by 55% in the venlafaxine arm versus 79% in the MPA arm (p < 0.0001). Less toxicity was reported in the MPA arm. Although effective for VMS relief, the relationship between progestins and breast cancer risk remains controversial [37].

A new approach to menopausal therapy is the tissue-selective estrogen complex. This complex pairs a SERM with an estrogen. Clinical profile will depend on the tissue-selective activities of the SERM and the estrogen. The goal is to provide targeted benefits of the estrogen and SERM on the relief of VMS and vaginal atrophy while preventing bone loss and not stimulating either the breast or the endometrium. Currently, a blended tissue-selective estrogen complex, bazedoxifene (BZA), with conjugated estrogen (CE), is in development in Phase III clinical trials [38].

Discontinuation of menopausal hormone therapy

For many women, VMS improve or resolve spontaneously over time, allowing many women to stop HT. Up to 25% of postmenopausal women will resume HT owing to persistent VMS [39]. Experts disagree on the optimal approach to discontinuation of HT – whether it is better to stop ‘cold turkey’ (i.e., abrupt, immediate cessation) or through tapering with either a slow decrease in dose over a period of 6–12 months or a slow decrease in the number of days per week that HT is used. Tapering is often continued until mild symptoms develop and then that particular dose is maintained until symptoms improve [39]. Discontinuation of HT may be associated with reoccurrence of VMS, significant bone loss or the development of genitourinary atrophy.

Health outcomes following discontinuation of the estrogen plus progestin arm in a WHI trial were evaluated at 3 years (mean 2.4 years of follow-up) [40]. No persistent increased coronary heart risks were found. A 12% higher global index was found with a greater risk of fatal and nonfatal malignancies (more lung cancers in the estrogen plus progestin group) and with loss of fracture protection. Chlebowski et al. reported that the increased risk of breast cancer associated with the use of estrogen plus progestin markedly declined soon after discontinuation of combined HT [41].

Bioidentical hormone therapy

Bioidentical HT refers to therapy similar in chemical composition to that made in the human body. There are many different types, doses and formulations of FDA-approved bioidentical products containing estradiol and progesterone available in a variety of routes of administration that provide constant, low levels of hormones. These products are regulated and monitored by the FDA and have proven efficacy at relieving menopausal symptoms and published endometrial safety profiles. Non-FDA-approved options of compounded bioidentical products are available by prescription therapy for those who cannot tolerate FDA-approved products; however, these products have not undergone rigorous RCTs of safety or efficacy [42,43]. Patient education is needed regarding potential risks and benefits of these products [44]. Adequate endometrial protection is needed if compounded hormones are prescribed.

Transdermal progesterone has demonstrated mixed results in randomized, placebo-controlled, hot-flash trials [45,46]. Wild yam cream contains progesterone precursors, which humans lack the enzyme to metabolize, and, therefore, US Pharmacopeia progesterone is needed. Over-the-counter progesterone cream may contain up to 450 mg progesterone per ounce of cream while compounding pharmacies can formulate concentrations of up to 900 mg/oz. The amount of drug delivered varies depending on the type and dose of cream used, as well as absorption factors [42]. Three studies have demonstrated that serum levels of transdermal progesterone are insufficient to protect against estrogenic stimulation of the endometrium [47–49]. In contrast to progesterone creams, FDA-approved progesterone gels are water soluble and appear to enter the microcirculation rapidly, thus giving rise to elevated serum progesterone levels comparable to those used in oral products [50].

Salivary testing

Although there is significant media hype regarding salivary testing for menopausal HT measurement, there are no reference standards available for salivary hormone testing. A recent population-based study involving older patients validated sex hormone measurements, but no direct correlations were performed [51]. Inter- and intra-patient variability has been found as well as a lack of correlation with serum estradiol levels [42]. According to Boothby et al., the data from salivary testing does not give information regarding what is occurring in the target tissue [42,43]. There is no evidence to suggest that individualized estrogen or progesterone regimens based on these tests increase efficacy or improve safety [42,44].

Vaginal atrophy

Following hot flashes, a second, common, bothersome menopausal complaint is vaginal atrophy, which may present with vaginal or vulvar dryness, dyspareunia, multicoital bleeding, vaginismus or itching. The percentage of women with superficial dyspareunia or signs of vaginal atrophy increases with age [52]. If untreated, vaginal atrophy may progressively worsen over time with superficial vulvovaginal fissures and petecheia owing to thinner and more fragile vaginal epithelium. A decrease in superficial epithelial cells occurs along with loss of subcutaneous fat in the labia majora, which leads to shrinkage and retraction of clitoral prepuce and the urethra, fusion of the labia minora and introital narrowing and then stenosis [53]. Vaginal pH increases from normal values of 3.8–4.5 to 6.0–7.0 [54]. This increasingly alkaline vaginal environment increases vulnerability to opportunistic organisms [55].

The approach to vaginal atrophy depends upon the pattern and the severity of symptoms. Nonhormonal, water-soluble vaginal lubricants and moisturizers can decrease friction during sexual activity, and may relieve mild vaginal dryness [56] as well as improve maturation of the vaginal epithelium [57,58]. Several studies have documented improvement in the maturation of the vaginal epithelium [59] with the use of Replens® (Lil’ Drug Store Products, IA, USA) equivalent to that provided by estrogen creams. with significant decreases of vaginal dryness of up to 60%. In a meta-analysis, Suckling et al. reviewed 16 trials (n = 2129) and found that estrogen products were more favorable than placebo or nonhormonal gels [60].

Estrogen therapy: systemic & vaginal

Estrogen therapy reverses vaginal atrophy, promotes cell growth and cellular maturation, enhances blood flow in vaginal tissue, improves pH towards premenopausal levels and improves the thickness and elasticity of the vaginal tissues [61]. All estrogen formulations are FDA approved for the treatment of atrophic vaginitis. Oral doses of CE as low as 0.3 mg daily [28], and transdermal doses of 12.5 mcg daily [62], demonstrated improvement in vaginal maturation index, although symptomatic relief was mixed. Vaginal estrogen is recommended if systemic estrogen is not needed for other menopause-related symptoms [26,63]. Estrogen creams, rings and tablets have all been demonstrated to be effective for atrophic vaginitis [64–66].

Lower doses of vaginal estrogen

Lower doses of vaginal estrogen compared with standard vaginal therapy are effective at relieving vaginal atrophy. A randomized, placebo-controlled trial by Simon et al. (n = 310; mean age: 58.6 years) found significant improvement in vaginal maturation index, vaginal pH and severity of the most bothersome symptom from baseline to the end of the 12-week treatment compared with placebo [67]. Although the usual dose of estradiol vaginal cream is 100–400 mcg/day, a pilot study (n = 7) used lower doses of 10 mcg/day, which relieved more than 80% of symptoms, improved vaginal and urethral cytology, and lowered vaginal pH without significantly increasing serum levels of estrogen or causing endometrial hyperplasia [68].

Two studies have demonstrated that lower doses of vaginal estradiol tablets (10 μg dose compared with 25 μg) are effective. In a study by Simon et al. (n = 309), significant improvements for both 10 and 25 μg vaginal estradiol tablets over placebo in vaginal cytology, pH and most bothersome symptoms score were demonstrated [69]. Similarly, Bachmann et al. found that both doses provided relief of vaginal symptoms, improved urogenital atrophy, decreased vaginal pH and increased maturation of the vaginal and urethral epithelium [70]. Greater improvement was observed with the 25 mcg dose; however, both were effective in the treatment of atrophic vaginitis.

Local vaginal estrogen therapy affects ERs in urethral mucosa and smooth muscle. Although some studies have demonstrated a worsening of incontinence with oral therapies, vaginal estrogen has reduced the frequency of urinary tract infections in menopausal women [71].

Potential systemic effects of vaginal estrogen

Vaginal estrogen delivery reduces the first-pass effect of oral delivery such that the equivalent vaginal dose is ten-times less than oral doses. However, in doses sufficient to relieve atrophic symptoms, vaginal estrogens vary according to the amount absorbed systemically as well as their ability to stimulate the endometrium. Labrie et al. found that vaginal estradiol tablets and conjugated equine estrogens vaginal cream, after 1 week of daily treatment, caused an approximate fivefold increase in serum estradiol in postmenopausal women with atrophy [72]. The vaginal estrogen 2 mg ring released 7.5 μg/day of 17β-estradiol for up to 90 days. After the first day following ring insertion, an estrogen surge was measurable before dropping to levels ranging from 7.0 to 8.1 pg/ml (normal menopausal levels: <30 pg/ml) [73]. Newer techniques will allow better detection of systemic absorption at these low estrogen levels.

Endometrial effects

Direct transport mechanisms between the vagina and uterus have been demonstrated with preferential uterine delivery of hormones administered vaginally [74]. A small crossover trial (n = 10) suggested that local effects of vaginal delivery varied depending on whether the tablet was placed in the outer or inner third of the vagina. Doppler flow studies demonstrated an increased flow in the uterine arteries when the estradiol tablet was applied near the upper third of the vagina compared with little change when it was applied more distally [75]. Preferential distribution for the lower third was toward the periurethral area. Serum estradiol levels were slightly elevated 3 h after placement and were found to be similar regardless of the delivery location [75].

In the Cochrane review, no cases of endometrial cancer were reported [60]. Endometrial stimulation was found with 2% simple hyperplasia with vaginal ring compared with conjugated equine estrogens cream and 4% hyperplasia (one simple and one complex) in the conjugated equine estrogens cream compared with estradiol tablets. Randomized studies with 25 μg 17β-estradiol vaginal tablets found no increase in endometrial thickness [66,76]. No endometrial proliferation was noted after 1 year of use with the 2 mg vaginal ring [73,77]. Dugal et al. compared estradiol vaginal tablets (25 mg 17β-estradiol) or estriol vagitories (0.5 mg estriol) daily for 2 weeks, then twice weekly for a total duration of 24 weeks [78]. Endometrial thickness increased (1.1 and 0.5 mm with estradiol tablets and estriol cream, respectively) in both treatment groups during the first 2 weeks, but returned to baseline levels when the frequency was reduced to twice-weekly.

Although the risk of endometrial stimulation with vaginal administration of unopposed estrogen in women with an intact uterus appears to be minimal, low doses of vaginal estrogen could lead to endometrial proliferation, hyperplasia or carcinoma [79]. There are currently no evidence-based recommendations for endometrial monitoring or progestin dosing for women using unopposed low-dose vaginal estrogen therapy [80]. Evaluation should be performed for women who experience postmenopausal bleeding while using topical vaginal estrogen.

Effect on the breast

Few studies have investigated the impact of using topical vaginal estrogen preparations on breast cancer recurrence and mortality and the reports are conflicting regarding breast safety. Rosenberg et al. found no associations between nonexclusive or exclusive use of local low-potency estrogen therapy and either ductal or lobular breast cancer [81]; however, Kendall et al. found that the vaginal estradiol tablet significantly raised systemic estradiol levels greater than 19 pmol/l over low baseline estradiol levels and suggested that this could affect the estradiol suppression achieved by aromatase inhibitors in women with breast cancer [82].

Selective estrogen receptor modulators with & without estrogen & the associated menopausal symptoms

Estrogen receptor agonist/antagonists, also referred to as SERMs, are compounds that vary in regard to their specific tissue-selective actions. Beneficial effects of SERMs include providing estrogen agonist effects in bone and the brain, but estrogen antagonists effects in tissues where long-term estrogen actions would be deleterious, such as the endometrium and breast. Compounds in this class that are FDA approved or in clinical testing include tamoxifen, raloxifene, BZA, lasofoxifene and arzoxifene.

Currently approved SERMs demonstrate estrogen antagonism in the hypothalamus with increased hot flash incidence. A possible solution to the problem of hot flashes with SERMs is to administer them in combination with estrogen, which could provide relief of hot flashes and additive increases in BMD. The primary concerns include potential additive effects on the risk for VTEs and endometrial proliferation, as well as a reversal or mitigation of the estrogen antagonistic effects on breast tissue. The potential risks and benefits would potentially vary with the time since menopause onset, the route and type of estrogen administration as well as the type of estrogen agonist/antagonist utilized.

Tamoxifen

Tamoxifen is approved as an adjuvant therapy for breast cancer, for the prevention of breast cancer in women at high risk for breast cancer (defined as a 5-year absolute risk of greater than 1.67%) and in women with a prior history of breast cancer to prevent recurrence and/or disease in the contralateral breast. Negative estrogen agonist effects of tamoxifen include effects on the endometrium and in the liver with a two- to three-fold increase in deep venous thrombosis and pulmonary embolism and a twofold increase in endometrial cancer [83]. The rate of endometrial carcinoma was increased among tamoxifen users relative to nonusers (relative risk [RR]: 2.53; 95% CI: 1.35–4.97), primarily among women aged 50 years or older whose tumors were classified low grade and stage I. In analyses by age, women aged 49 years or younger were not found to be at increased risk for stroke if they had received tamoxifen (RR: 1.13; 95% CI: 0.39–3.36). However, for women aged 50 years or older, there was some evidence that tamoxifen increased the risk of stroke, deep vein thrombosis and pulmonary embolism [83]. The usefulness of tamoxifen as a preventative agent, even in high-risk individuals, appears to be limited by the induction of hot flashes and other adverse effects.

Tamoxifen with estrogen

Tamoxifen has been studied in combination with estrogen, primarily in Europe, to relieve hot flashes. In the International Breast Cancer Intervention Study (IBIS), a 5-year RCT of women at high risk of developing breast cancer [84], HT was permitted but limited to the duration required to control severe VMS. Approximately 70% of women in the tamoxifen-only arm reported hot flashes compared with 53% in the placebo group with severe hot flashes more strongly related to tamoxifen. For those receiving tamoxifen, 60.8% taking HT at entry experienced hot flashes in the first 6 months compared with 49.2% for those not receiving HT (p = 0.09). Fewer women who were on placebo who received HT at entry experienced hot flashes compared with women who stopped HT (22.9 vs 34.3%, respectively; p = 0.03). HT was ineffective at decreasing tamoxifen-related hot flashes; HT was more effective in controlling hot flashes in months 6–12 in the placebo arm (47.9 vs 20.4% women taking HT/tamoxifen and HT/placebo, respectively) than in the tamoxifen arm (51.4 vs 39.0% women taking no HT/tamoxifen and no HT/placebo, respectively). In conclusion, HT begun either during initiation or after initiation of tamoxifen was not as effective in alleviating hot flashes. A recent study of ER genotypes demonstrated that the incidence of hot flashes induced by tamoxifen may vary with the genotype present [85]. A total of 180 women evaluated with either tamoxifen alone or in combination with estrogen found the combination of tamoxifen and estrogen to be associated with the greatest increase in hot flash scores, although not significantly different than the increase with tamoxifen alone [86]. The addition of HT to tamoxifen is unlikely to improve compliance as vasomotor instability does not appear to be lessened with the addition of supplemental hormones. Concern exists that addition of estrogen may limit the chemoprevention efficacy of tamoxifen.

Raloxifene

Raloxifene is an estrogen agonist in the bone and liver and an estrogen antagonist in breast and endometrial tissue that has been demonstrated to increase VMS. Raloxifene is approved for use in postmenopausal women for the prevention and treatment of osteoporosis and for the reduction of the risk of invasive breast cancer in postmenopausal women with osteoporosis and at high risk for invasive breast cancer. Fracture efficacy for vertebral fractures has been demonstrated with a relative-risk reduction (RRR) of 55% for osteoporotic postmenopausal women without a prior fracture and relative-risk reduction of 30% for those with a prior vertebral fracture [87]. Reduction of breast cancer has been found with raloxifene in multiple trials. The Study of Tamoxifen and Raloxifene (STAR), a RCT comparing raloxifene and tamoxifen (n = 19,747) of high-risk women for an average of 3.9 years, demonstrated no significant difference between tamoxifen and raloxifene in the rate of invasive breast cancers or endometrial cancer [88]. Previous studies have demonstrated increased hot flashes and a twofold elevated RR for VTE with raloxifene, similar to that observed with HT. In the STAR trial, when compared with tamoxifen, raloxifene resulted in 29% fewer VTEs and 36% fewer uterine cancers [88]. Women on raloxifene at high risk for coronary heart events in the Raloxifene Use for the Heart (RUTH) trial [89] had an increased risk of death from stroke, but no increase in overall mortality, risk for heart attack or risk for stroke. In the 8-year Continuing Outcomes Relative to Evista (CORE) trial, there was no increase in the incidence of uterine cancer, endometrial hyperplasia or postmenopausal bleeding when compared with placebo [90]. The effect of raloxifene on hot flashes varied depending on prior HT use and patient age, with 15% VMS in the Multiple Outcomes of Raloxifene Evaluation (MORE) trial for those aged under 65 years, but only 6% for those over the age of 65 years [91]. In a study by Gordon et al. (n = 83), the transition to raloxifene from HT did not change the frequency of VMS [92].

Raloxifene with estrogen

Raloxifene 60 mg daily was compared with raloxifene and oral estradiol 1 m daily in a small RCT over 52 weeks [93]. The combination of raloxifene–estradiol significantly reduced the frequency of VMS compared with baseline and the raloxifene-only group. The study was not powered to assess effects on bone or fracture risk, VTEs, breast cancer or the endometrium. The raloxifene–estradiol-treated group demonstrated an increased endometrial thickness of 0.74 mm on transvaginal ultrasound, compared with raloxifene-only group. Although the clinical significance of this small difference is unknown, two of the women in the combination-therapy group developed endometrial hyperplasia, one with atypia, in comparison with no cases of endometrial hyperplasia in the raloxifene-only group.

Raloxifene was tested in a 6-month trial (n = 91) in combination with a 17β-estradiol-releasing vaginal ring [94]. Similar efficacy in resolution of vaginal atrophy symptoms in the combination arm was found compared with those treated with the vaginal ring alone, with no signs of endometrial proliferation observed in either group. Transdermal 17β-estradiol 0.5 mg/day was studied in combination with raloxifene compared with raloxifene alone for the relief of VMS over a 3-month interval [95]. Findings included a significant reduction in VMS in the combination-therapy group compared with raloxifene-only group. No endometrial changes were noted with either hysteroscopy or endometrial biopsy. Short-term use of raloxifene with vaginal estrogen appears safe, but has only been tested for 6 months. However, the combination of oral estradiol and raloxifene cannot be recommended.

Bazedoxifene

Bazedoxifene binds to both ERs with high affinity and acts as an agonist on skeletal tissue, with bone turnover reduced by 20–25% with doses of 20 or 40 mg daily, as well as an antagonist on breast tissue and uterine tissue [96,97].

Bazedoxifene & estrogen

In a 2-year osteoporosis prevention RCT [98], the percentage of women with either worsening of hot flashes or new hot flashes was higher in the BZA group than in the placebo, and, similar to the raloxifene group, with few withdrawals owing to hot flashes.

In the Selective Estrogen Menopause and Response to Therapy (SMART-2) study, a 12-week, double-blind, randomized, placebo-controlled trial of BZA 20 mg/CE 0.625 mg and BZA 20 mg/ CE 0.45 mg (n = 332; aged 40–65 years; ≥seven hot flashes per day or 50 per week) [99] of reduction of hot flashes in postmenopausal women revealed, after 12 weeks, those on BZA 20 mg/CE 0.625 mg had 80% fewer hot flashes and those on BZA 20 mg/CE 0.45 mg had 74% fewer hot flashes compared with placebo response of 51%. Median time for 50% reduction in hot flashes for BZA and either dose of CE was 14–15 days compared with 30 days for placebo. More subjects reported at least a 75% reduction in the number of hot flashes at week 12 in the BZA 20 mg/CE 0.45 mg and the BZA 20 mg/CE 0.625 mg treatments groups (61 and 73%, respectively) compared with the placebo group (27%; p < 0.001). Safety profile was similar between treatment groups and placebo. A similar 5% incidence was found for uterine bleeding and breast pain within the BZA/CE groups and placebo. Thus, BZA 20 mg paired with CE 0.45 or 0.625 mg was found to be effective at relieving VMS in postmenopausal women with no short-term safety concerns. Larger Phase III RCTs are in progress to evaluate the effects of BZA combined with estrogen on bone and longer-term endometrial safety.

Selective estrogen receptor-β agonist, MF-101

Selective ER agonists can stimulate either ER-α or ER-β and induce tissue-specific estrogen-like effects, thus providing a safer alternative to conventional HT. MF-101 is derived from 22 herbs that are traditionally used in Chinese medicine for the treatment of menopausal symptoms. Studies with the MF-101-isolated active compounds, liquiritigen and chalcone, demonstrated selectivity for ER-β, with no induction of proliferative events. MF-101 did not promote the growth of breast cancer cells or stimulate uterine growth in preclinical studies and, in a Phase II trial, was demonstrated to be safe and more effective in reducing the frequency and severity of hot flashes in postmenopausal women compared with placebo [100]. In order to confirm the safety and efficacy of MF-101, larger Phase III trials have been planned for 2009.

Nonhormonal treatment options for vasomotor symptoms

For those who cannot use estrogen or choose not to use it, yet have persistent moderate-to-severe menopausal symptoms, treatment options for VMS include nonhormonal prescription therapy, such as centrally acting antidepressants, gabapentin and clonidine ( Box 2 ). For those with milder symptoms, options include lifestyle changes and complementary modalities, such as paced breathing, yoga, acupuncture and exercise. Dietary supplements, such as phytoestrogens, black cohosh or other herbal products, have had mixed reports.

Box 2.

Approach to management of menopausal symptoms

Neuroactive agents, such as selective serotonin reuptake inhibitors, serotonin norepinephrine reuptake inhibitors and gabapentin, appear to have efficacy, but larger, long-term, randomized, placebo-controlled trials are needed.

• The most promising agents appear to be venlafaxine, desvenlafaxine, paroxetine extended release and clonidine (oral or transdermal).

• Nonprescription remedies, such as black cohosh or soy, have been tested short term with little-to-mild efficacy over placebo. Trials are limited by variation in products, menopause populations and inclusion criteria.

• Homeopathy, magnetic therapy, reflexology, dong quai, ginseng, evening primrose oil and vitamin E have not been demonstrated to be clinically significant compared with placebo.

Nonhormonal options

• Lifestyle modifications: layering clothing and avoiding triggers

• Paced respiration: randomized, controlled trials show benefit

• Yoga: pilot trials show benefit

• Exercise: no conclusive evidence; sweating increases vasomotor symptoms

• Homeopathy: no benefit over placebo

• Magnetic therapy: no benefit

• Reflexology: no benefit

Nonhormonal prescription options demonstrating the benefits over placebo

• Venlafaxine: benefit over placebo

• Densvenlafaxine: benefit over placebo

• Paroxetine and other selective serotonin reuptake inhibitors: benefit over placebo

• Gabapentin: benefit over placebo

• Clonidine: benefit over placebo

• Progesterone: benefit over placebo

Hormonal therapy

• Standard, low and ultra-low dose: benefit over placebo depending on degree of symptoms, age and time since menopause.

Future

• FDA-approved nonhormonal medications, such as desvenlafaxine and gabapentin

• Lower dose vaginal estrogen

• Selective estrogen receptor modulators in combination with estrogen

• Estrogen receptor-β agonist MF-101

Review of alternative and complementary and nonhormonal therapies reveals that adequate testing with randomized, placebo-controlled trials of adequate numbers and duration are lacking and none are as effective as estrogen. The 50% placebo response in hormone trials [20] on VMS relief highlights the need for rigorous RCTs with strictly defined vasomotor inclusion criteria, adequate subject numbers and duration of the trials [101]. Clear descriptions of products tested and outcome measures are needed in order to compare trials and outcomes [102].

Nonhormonal prescription therapy

Neuroactive agents have been studied mostly in women with breast cancer or who are at risk for breast cancer for their effectiveness at relieving VMS. Increased serotonergic activity within the CNS appears to reduce hot flashes, possibly through an effect on attenuation of central opioid peptide withdrawal or through reducing noradrenergic activity. Studies have included selective serotonin reuptake inhibitors (SSRIs; fluoxetine, paroxetine, citalopram and mirtazapine), serotonin norepinephrine reuptake inhibitor (SNRI; venlafaxine and desvenlafaxine) and gabapentin. Most studies of nonhormonal prescription therapies have required women to have two hot flashes per day or 14 per week at baseline prior to study enrolment, which differ from the FDA industry requirement of seven hot flashes per day or 50–60 per week. Antihypertensives have been studied, including clonidine (patch or pill) and methyldopa. Sedatives, such as Bellergal, are rarely recommended because of side effects.

Of those studied, venlafaxine and its metabolite desvenlafaxine, controlled release (CR) paroxetine and clonidine appear to provide the best relief and reduce VMS by 50–67%. Loprinzi et al. reviewed hot flash score data from placebo-controlled trials with newer antidepressants and gabapentin and found effects apparent within 4 weeks of therapy and stable response for up to 12 weeks [103]. However, the extent of relief provided over time, the dosages needed to achieve and maintain maximal benefit and whether the impact varies by specific drug were not clear [103].

Contraindications to SSRIs include concomitant use of monoamine oxidase inhibitors, thioridazine or warfarin. Adverse effects in depression trials have included asthenia, sweating, nausea, decreased appetite, somnolence, insomnia and dizziness.

Casper recommends caution in using SSRIs, particularly paroxetine, in women on adjuvant breast cancer therapy, such as tamoxifen, since SSRIs have been found to reduce the metabolism of tamoxifen to its most active form, endoxifen, by 60% through inhibition of the cytochrome P450 enzyme, CYP2D6 [104]. Casper ranks SSRI potency CYP2D6 inhibition as paroxetine (highest), fluoxetine, sertraline, citalopram and venlafaxine (lowest) [104]. Tamoxifen's anti-estrogenic activity on the breast could be affected by a reduced endoxifen concentration.

SSNR: venlafaxine

Positive pilot studies led Loprinzi et al. to undertake a double-blind, randomized, placebo-controlled trial (n = 191; efficacy evaluable; 4 weeks) of venlafaxine 37.5, 75 or 150 mg in women with a history of breast cancer or a reluctance to undergo HT [105]. After 4 weeks of treatment, median hot flash scores were reduced from baseline by 27% for placebo, 37% for 75 mg, and 61% for both 75 and 150 mg venlafaxine. The higher 150 mg dose was no more effective than the 75 mg dose but it was associated with an increase in mouth dryness, decreased appetite, nausea and constipation. An open-label continuation study (n = 102) by Barton et al. reported a maintenance in the reduction in hot flashes without increased toxicity [106]. A RCT by Evans et al. (n = 80; 12 weeks) of postmenopausal women with more than 14 hot flashes per week who received 37.5 mg venlafaxine per week followed by 75 mg for 11 weeks found that hot flash severity scores based on daily dairies were lower but not statistically significant [107]. However, the effects of hot flash symptoms on daily living were significantly improved (p < 0.001). Dry mouth, sleeplessness and decreased appetite were increased in the treatment group. The ‘patient perceived’ hot flash score was significantly improved. Most importantly, 93% of patients in the venlafaxine group chose to continue treatment. The efficacy of venlafaxine for relief of VMS must be balanced against the potential side effects [108].

Desvenlafaxine

Desvenlafaxine, a metabolite of venlafaxine, was demonstrated in a RCT [109,110] (n = 458; postmenopausal women; ≥50 moderate-to-severe hot flashes per week) at 100 and 150 mg/day to significantly reduce the number of hot flashes compared with placebo at weeks 4 and 12 (all p ≤ 0.012) with 65.4 and 66.6% reduction from baseline at week 12, respectively, compared with placebo at 50.8%. Hot flashes severity and number of night-time awakenings were significantly reduced at both time points (p ≤ 0.048). More AEs were reported during week 1 than with placebo, with no difference in discontinuations. Dose titration appears to improve initial tolerability and decrease AE reporting.

SSRIs

Fluoxetine 20 mg was studied in a double-blind, crossover RCT (n = 81; 4 weeks per period; ≥14 hot flashes per week; tamoxifen or raloxifene allowed; history of breast cancer or a concern regarding estrogen) [111]. Hot flash scores (frequency × average severity) decreased by 50% in the fluoxetine arm versus 36% in the placebo arm. Crossover analysis demonstrated significantly greater hot flash score improvement with fluoxetine compared with placebo (p < 0.02). The fluoxetine appeared to modestly improve hot flashes without significant side effects.

Citalopram and fluoxetine were studied (n = 150; 9 months) in a randomized, placebo-controlled trial [112] starting at 10 mg, increasing to 20 mg at 4 weeks and 30 mg at 6 months. No significant differences in VMS relief were found between the treatment groups and placebo, although a positive tendency was found for the SSRIs. Insomnia improved significantly in the citalopram group.

A multicenter, double-blind, placebo-controlled, parallel group RCT by Stearns and colleagues (n = 165; 6 weeks; 2–3 hot flashes per day; no medications for 6 weeks; tamoxifen, raloxifene and vitamin E allowed) demonstrated a 62.2% hot flash reduction for those receiving paroxetine CR 12.5 mg and 64.6% for those receiving paroxetine 25 mg, compared with 37.8% reduction in the placebo group [113].

In a blinded, crossover, randomized, placebo-controlled trial (n = 151; postmenopausal women; 14 hot flashes per week), Stearns found that paroxetine, in doses of 10 or 20 mg significantly decreased hot flash frequency and composite scores compared with placebo [114]. Similar efficacy was found with 10 and 20 mg doses of paroxetine, which was significant when compared with placebo; however, paroxetine 10 mg resulted in a significant improvement in sleep and less discontinuation.

In a RCT by Soares et al. (n = 50; 17 hot flashes per week on average), paroxetine CR out-performed placebo at reducing VMS (mean reduction of 6.1 vs 2.8 hot flashes per week, respectively; p = 0.03) and improving depressive symptoms [115].

Antihypertensives

Clonidine has been studied in oral and transdermal forms [116–118]. In one randomized, placebo-controlled trial [116] using clonidine 0.1 mg/day (n = 194; 8 weeks) in breast cancer patients on tamoxifen, hot flash frequency decreased by 38% compared with 20% for placebo. Side effects included dry mouth, drowsiness, constipation and dizziness. Transdermal patches with a slowly escalating dose may avoid some of these side effects. In a small, double-blind RCT of breast cancer patients with at least two hot flashes per day treated with clonidine (0.075 mg twice per day; n = 33) and venlafaxine (37.5 mg twice per day; n = 31), nine patients stopped early (owing to AEs) and seven withdrew consent [119]. At the end of week 4, the median hot flash frequency decreased by 7.6 hot flashes per day for venlafaxine and 4.85 hot flashes per day for clonidine (p = 0.025). Opposite results were found in a double-blind, nonplacebo-controlled, crossover study by Buijs et al. (n = 60; breast cancer patients with hot flashes) where patients were randomized to 8 weeks venlafaxine followed by no medications for 2 weeks and then 8 weeks of clonidine, or vice versa (30 in each group) [120]. Both were effective with regard to median reduction in hot flash score of 49% for venlafaxine and 55% for clonidine (not statistically different). Premature discontinuation for AEs was reported in 14 out of 59 patients during venlafaxine and five out of 53 during clonidine (p = 0.038) with venlafaxine inducing more side effects. Thus, clonidine and venlafaxine both improved hot flashes in mildly symptomatic breast cancer survivors. Methyldopa has also been tested in a small pilot study (n = 0) [4,5] demonstrating a modest improvement but significant side effects [121].

Gabapentin

Gabapentin is approved as an anticonvulsant and for the treatment of postherpetic neuralgia. Gabapentin serves as a ᵧ-aminobutyric acid agonist and is hypothesized to affect temperature regulation through its binding to calcium channels. RCTs of gabapentin for hot flashes suggest efficacy superior to placebo. Guttoso et al. (n = 59; ≥ seven hot flashes per day; anti-estrogens, clonidine or antidepressants allowed) found that gabapentin 900 mg daily reduced hot flash frequency by 54% compared with 31% for placebo (p = 0.01) [122]. Open-label gabapentin dosing greater than 900 mg/day was associated with higher reductions; however, 50% of recipients reported at least one AE compared with 27.6% on placebo, and 13.3% on gabapentin withdrew for somnolence, dizziness or rash. In a larger RCT of 300 and 900 mg of gabapentin daily by Pandya et al. (n = 420 with 371 evaluable; prior breast cancer; ≥14 hot flashes per week) hot flash severity score was reduced by 21% (week 4) and 15% (week 8) for placebo, 33 and 31%, respectively for gabapentin 300 mg, and 49 and 46% respectively for gabapentin 900 mg [123]. The 900 mg dose was significant for decreased hot flash frequency and severity. A small, double-blind RCT by Reddy et al. (n = 60; moderate-to-severe hot flashes; 12 week duration) compared CEs 0.625 mg/day, placebo or gabapentin titrated to 2400 mg/day [124]. Intention-to-treat analysis revealed a similar reduction in the hot flash composite score for estrogen (72%; p = 0.016) and gabapentin (71%; p = 0.004) with both resulting in a greater reduction than placebo (54%) after 12 weeks. Overall, approximately 70% of those who received higher dose gabapentin (2400 mg daily) reported a greater than 50% improvement in hot flashes, similar to standard estrogen therapy and significantly greater than placebo. Loprinzi et al. demonstrated in a small, nonplacebo-controlled trial that, regardless of continued antidepressant use, gabapentin reduced hot flashes by 50% [125]. In 2008, Butt et al. (n = 200; evaluable 197; ≥14 hot flashes per week; 4-week trial), evaluated gabapentin 300 mg and placebo three-times daily. Hot flash scores decreased by 51% for gabapentin (95% CI: 43–58) compared with 26% for placebo (95% CI: 18–35; p < 0.001). AEs included dizziness (18%), unsteadiness (14%), and drowsiness (12%) at week 1 compared with placebo; however, these symptoms improved by week 2 and returned to baseline by week 4. Although gabapentin appears effective, side effects are significant and the longest published trial to date is only 12 weeks in duration.

Lifestyle changes, mind–body interventions & complementary modalities

For women with mild VMS, measures with limited effectiveness suggested by the North American Menopause Society (NAMS) [126] include:

• Cooling body core temperature through the use of fans or dressing in layers;

• Exercise;

• Avoiding hot and spicy foods;

• Relaxing therapies including yoga, massage, meditation, slow breathing or leisurely baths;

• Mind–body therapies, including relaxation, biofeedback, participation in group education, paced diaphragmatic respiration and hypnosis, which may decrease norepinephrine levels and, thus, widen the thermoregulatory zone leading to a decrease in frequency or intensity of hot flashes.

Paced respiration/relaxation

Small RCTs by Freedman, Irvin and colleagues considering paced respirations (i.e., slow, controlled diaphragmatic breathing) have demonstrated up to 50% efficacy using objective laboratory recorded symptoms and diary-recorded hot flashes compared with α-wave electroencephalographic biofeedback or muscle relaxation [127–130]. Diaphragmatic breathing is performed for 15 min twice daily using a pattern of 5 s inhale, 5 s exhale. Strengths of these RCTs include use of an ambulatory monitoring device (sternal skin conductance) to provide objective measures of hot flash frequency. Limitations include small numbers of participants with varying degrees of hot flash symptomatology. While small, early trials suggest that paced respiration may be effective, larger studies of longer duration are required.

Hypnosis

Case studies of hypnosis wave suggested that hypnosis might reduce the frequency and intensity of hot flashes. Younus et al. studied self-selected women (n = 14; ≥ five hot flashes per week; tamoxifen allowed) who participated in hypnosis classes and found a decrease in frequency, duration and severity of hot flashes during the training period and reported improved quality of life and reduced fatigue and insomnia [131]. Elkins et al. (n = 51) randomly assigned women to complete five weekly hypnotic sessions or serve as the study control [132]. Hot flash scores (frequency × average severity) decreased by 68% from baseline to end point (p < 0.001). Significant improvements were also found in self-reported anxiety, depression, interference of hot flashes on daily activities and sleep for treated subjects (p < 0.005) compared with controls. Larger RCTs with improved placebo groups are required.

Yoga

Yoga has been hypothesized to decrease sympathetic arousal. Early pilot studies suggested potential benefit. Booth-LaForce et al. studied 11 participants in a 10-week yoga program, comprising breathing techniques, postures and relaxation poses, and found significant improvement compared with pretreatment assessments for severity of total menopausal symptoms, hot-flash daily interference and sleep efficiency, disturbances and quality [133]. Chattha et al. published a single-blind, prospective RCT (n = 120; age 40–55 years) randomized to practising integrated yoga or physical exercises for 8 weeks [134]. Yoga decreased climacteric symptoms, perceived stress and neuroticism in perimenopausal women more so than physical exercise at 8 weeks. Cohen summarized yoga trials that included 8 weeks of focused time for breathing exercises and relaxation as a promising complementary and alternative medicine therapy [135]. Lee et al. stated that, as yet, there was insufficient evidence to recommend yoga as an effective intervention for menopause; more RCTs are needed [136].

Exercise

In regard to exercise, research findings are mixed. Observational studies initially suggested that physically active women reported fewer, less intense hot flashes than age-matched sedentary controls [11,137]. Stadbert et al. [138] and Ivarrson et al. [139] found that perimenopausal women who exercised regularly were more likely to be symptom free than women undertaking little or no exercise (p < 0.05). However, Freedman et al. found that strenuous exercise associated with perspiration could trigger hot flashes in symptomatic women [140]. In a RCT by Aiello et al. (n = 173; overweight women; no HT for 6 months), a significant increase in hot flashes at 1 year in a small number of the moderate-intensity exercisers compared with the control group of women undertaking stretching exercises was found [141]. Guthrie et al. reported a 9-year longitudinal study of Australian-born women in which mild exercise was associated with bothersome hot flashes (p < 0.05) [142]. Daley et al. found one small RCT that met inclusion criteria where the HT group had fewer hot flashes compared with the exercise group; however, the reviewers felt that this one RCT did not provide enough evidence regarding the effectiveness of exercise in reducing VMS [143]. Whitcomb et al. reported a population-based study where frequent physical activity was associated with greater severity and frequency of VMS [144]. In summary, exercise is hypothesized to positively affect central thermoregulation unless associated with obesity, sweating or severe VMS. Further RCTs are required to clarify these findings.

Acupuncture

Studies of acupuncture and hot flashes have yielded mixed results. In a study by Wyon et al., 45 postmenopausal women were randomized to electroacupuncture, superficial needle insertion, exercise, oral estrogen treatment or relaxation [145]. The effect on mean number of hot flashes per 24 h were as follows:

• The acupuncture and hormone group: decreased from 7.3 to 3.5 (p < 0.001);

• Electroacupuncture group: decreased from 8.1 to 3.8 (p < 0.001 in the superficial needle insertion [control group]);

• Estrogen: decreased from 8.4 to 0.8 (p < 0.001).

All groups demonstrated persistent effects over the 24-week follow-up. Estrogen reduced hot flashes more effectively than electroacupuncture (no p-value given). Both the superficial (control group) and deep acupuncture group demonstrated decreased hot flashes.

Avis et al. reported on a three-group design, 8-week RCT (n = 56 women; age 44–55 years; no menses for 3 months, ≥ four or more hot flashes per day) [146]. Subjects were randomized to twice weekly treatments as follows: usual care (19), sham acupuncture (18) and standardized individual acupuncture based on traditional Chinese medicine (19).

All groups demonstrated a significant decrease in mean frequency of hot flashes (p = 0.01), with no significant differences between study groups. Similar to Wyon, the two acupuncture groups (sham and traditional Chinese medicine) demonstrated a significantly greater decrease than the usual-care group (p < 0.05), but no significant difference from each other for hot flash index score, hot flash interference, sleep, mood, health-related quality of life or psychological well-being. These and other recent studies [147,148] suggest that traditional and sham acupuncture may reduce hot flash frequency; more studies with adequate control groups are required.

Homeopathy, magnet therapy & foot reflexology

In RCTs, neither homeopathic remedies, magnet therapy, nor foot reflexology out-performed placebo in relieving menopausal symptoms [149–153].

Nonprescription (over-the-counter) therapies

Surveys of menopausal women demonstrated increasing interest in alternative therapies, fueled by aggressive marketing and post-WHI fear of hormones. Education of patients is needed regarding side effects and limited available safety and efficacy data.

Ang-Lee et al. reported bleeding from garlic, ginkgo and ginseng, coronary heart instability from ephedra and hypoglycemia from ginseng [154].

Herb–drug interactions include: kava and valerian, potentiation of the sedative effect of anesthetics with St John's Wort, increased metabolism of preoperative drugs and Black cohosh, and Kava kava and liver toxicity [155,156]. Kava kava is banned in the UK owing to reports of associated liver damage [157].

The safety of alternative therapies in patients with medical problems, such as venous thrombolic disease, breast or endometrial cancer, stroke or coronary artery disease, is unknown.

Phytoestrogens

Isoflavones or phytoestrogens are plant-derived diphenolic compounds that exhibit both hormonal and nonhormonal properties [126]. Two common sources are soy and red clover. Dog et al. reported that hot flashes were only slightly reduced in women who consume soy-derived isoflavones compared with controls [102]. The high placebo rate consistently observed in hot flash trials makes it more difficult to determine differences and efficacy of mildly effective treatments. Comparisons between trials are difficult owing to variations in products, dosages, inclusion criteria for number of hot flashes, outcome measures and menopausal status [102].

A systematic review of 25 RCTs [158] found negative results in seven of eight soy food trials and in three of five soy extract trials. In a large, double-blind, placebo-controlled crossover trial (n = 177; symptomatic breast cancer survivors), soy tablets (total 150 mg isoflavones per day) were no more effective than placebo in preventing or reducing hot flashes [159]. No toxicity was observed. Similar lack of efficacy on hot flash frequency was found in a double-blind, randomized, placebo-controlled trial of soy protein with isoflavones 42 and 58 mg/day [160].

By contrast, a RCT by Han et al. found that 100 mg/day of soy isoflavone supplementation produced a significant decrease in vasomotor complaints (on Kupperman's Index); however, there was a slight worsening of hot flash symptoms in the placebo group [161]. Smaller RCTs of soy protein and soy flour have demonstrated mixed results. The safety of long-term higher dose isoflavone extracts has been questioned. In a double-blind, randomized, placebo-controlled trial (n = 376), six cases (3.8%) of endometrial hyperplasia were found in women who received 150 mg/day soy isoflavone extract for 5 years versus no cases in the placebo group (p < 0.05). However, the 41 unassessable placebo specimens may have affected the results [162].

A meta-analysis by Lethaby et al. of 30 RCTs of at least 12 weeks duration using high levels of phytoestrogens revealed that some trials demonstrated relief in frequency and severity of menopausal symptoms compared with placebo [163]. The trials were mostly of low quality and underpowered with a variable placebo effect (1–59%). There was no evidence that phytoestrogens were clearly effective for the alleviation of hot flashes; in addition, there was no evidence of estrogenic proliferation in patients who had taken phytoestrogens for up to 2 years.

Red clover extracts (trifolium pretense)

Randomized, controlled trials of red clover leaf extracts containing isoflavones similar to soy have demonstrated mixed results. Two systematic reviews [102,158] suggested that red clover extract did not significantly improve hot flashes over placebo. Two small studies using Promensil® (Novogen, NSW, Australia) in postmenopausal women found no difference between groups in the frequency of hot flashes [164,165]. The largest trial [166] found no significant difference in hot flashes for either Promensil (82 mg/day isoflavones) or Rimostil® (Novogen, NSW, Australia; 57 mg/day isoflavones) compared with placebo, although a greater reduction in hot flashes was found for women with above-average BMI.

Combination therapy: mixed results

Studies of herbal combinations have demonstrated mixed results. Kupfersztain et al. found a 90% decline in hot flashes using a natural plant extract of Angelica sinensis and Matricaria chamomilla (Climex, The Netherlands) compared with 15% reduction with placebo [167]. Verhoeven et al. found a combination of isoflavones and Actaea racemosa Linnaeus (formerly black cohosh) no more effective than placebo in reducing the daily number and severity of hot flashes [168].

Black cohosh (Cimicifuga racemosa, Actaea racemosa Linnaeus)

Black cohosh has been found to bind to the serotonin receptor, but does not bind to the ER [169,170]. In a 1-year, double-blind, randomized, placebo-controlled trial, Reed et al. (n = 351; age 45–55 years; ≥ two hot flashes per day) found that black cohosh, alone or as part of a multibotanical product with or without soy dietary changes, had no effects on vaginal epithelium, endometrium or reproductive hormones [171]. Four RCTs of black cohosh (total of 354 menopausal women) using placebo and/or estrogen treatment arm reviewed by Huntley found that three of the four trials found black cohosh to be beneficial (although only one of the three was placebo controlled) [172]. Dog et al. summarized five RCTs as demonstrating a reduction in menopausal symptoms, but with limitations owing to methodological short comings and variations in product and dosage [155]. The most studied product, Remifemin® (Enzymatic Therapy, Inc., WI, USA), is standardized to 1 mg of triterpenes per 20 mg tablet with recommended dose 20 mg twice daily. Davis et al. reported increased breast metastasis from breast to lung in mice who received black cohosh [173]; no other animal studies have suggested effect on mammary tumors [174–176]. Safety reviews of black cohosh in published clinical trials have found a few gastrointestinal AEs when taken for up to 6 months [156,177]. Case reports of acute liver disease have been reported [156,177]. Mello, in an editorial in Menopause, stated that since no objective effect on estrogen levels, luteinizing hormone pulse frequency or hot flashes have been found with black cohosh, the potential risks with long-term use appear to outweigh the benefits for treatment of menopausal symptoms [178].

Probably nonefficacious products for vasomotor symptoms relief

A limited number of RCTs suggest that dong quai, evening primrose oil, ginseng extract, Kava kava and vitamin E are either ineffective or not clinically significant in relieving VMS [179–183]. Side effects of evening primrose oil include nausea, flatulence and bloating, along with a possible aggravation of temporal lobe epilepsy [155]. Chaste tree berries (Vitex agnus castus) may have progestin-like effects but have not been well studied. Small studies of essential oils from the leaf or berry have demonstrated a reduction in menopausal symptoms [184]. Side effects include gastrointestinal symptoms, headache, rash, itching and increased menstrual flow. No estrogen activity has been found [185].

Conclusion

Efficacy and safety studies suggest that menopausal HT is safe and effective for relief of menopausal symptoms, vaginal atrophy and prevention of osteoporosis. Concern exists with longer use regarding risk of breast cancer, heart events, dementia, stroke and VTEs. Lower doses administered both systemically and vaginally appear to be as effective as standard doses, although data is limited regarding long- or short-term safety. Local vaginal estrogen is recommended if treatment is only aimed at relief of vaginal atrophy.

Estrogen receptor agonists/antagonists vary in regard to their specific tissue-selective actions with beneficial estrogen-agonist effects in bone and the brain and estrogen-antagonists effects in tissues where long-term estrogen actions would be deleterious, such as the endometrium and breast. For patients at high risk for breast cancer, tamoxifen is approved for the prevention of invasive breast cancer and raloxifene is approved for the prevention and treatment of osteoporosis and the prevention of invasive breast cancer. These agents do not relieve VMS or vaginal atrophy. They do appear to be safe in combination with topical vaginal estrogen, but concerns have been raised regarding their use with systemic products. BZA is undergoing large multicenter trials in combination with estrogen for the prevention of osteoporosis and relief of VMS with no increased risk of breast or endometrial cancer observed to date. Selective ER agonists can stimulate either ER-α or ER-β and induce tissue-specific estrogen-like effects. One such agent, MF-101, which appears to be a ER-β selective agonist, is currently being studied to determine whether it can serve as a safer alternative to conventional HT.

Alternatives to HT exist and can be tailored to the individual. For moderate-to-severe VMS, HT is currently the only FDA-approved treatment, with lower doses being effective. Neuroactive agents, such as SSRIs, SSNRs and gabapentin, appear to have efficacy, but larger, long-term, randomized, placebo-controlled trials are needed. The most promising agents appear to be venlafaxine, desvenlafaxine, paroxetine extended release, gabapentin and clonidine (oral or transdermal). For vaginal atrophy, therapies include local topical or vaginal estrogen for the relief of vaginal atrophy, with reduced effectiveness when nonhormonal vaginal moisturizers or lubricants are used. Lower doses appear effective. For mild hot flashes, lifestyle changes, including paced respiration, acupuncture and yoga, may provide limited relief. Nonprescription remedies, such as black cohosh or soy, have little-to-mild efficacy over placebo, tested in short-term trials with limitations owing to variation in products, menopause populations and inclusion criteria. Clinical trials have not demonstrated significant clinical efficacy in the treatment of VMS for homeopathy, magnetic therapy, reflexology, dong quai, ginseng, evening primrose oil and vitamin E.

Executive summary

Hormonal options

• Treatment of menopausal symptoms requires an evidence-based approach.

• Hormone therapy remains the gold standard for treatment of menopausal symptoms.

• For moderate-to-severe vasomotor symptoms (VMS), hormone therapy is currently the only FDA-approved treatment.

• The WHI changed the approach to menopause. The 2007 re-analysis provides evidence-based data relating to women under the age of 60 years and within 10 years of menopause onset.

• There are many different types and formulations of hormonal therapy available.

• Compounded bioidentical hormones have not been proven to be safer or more efficacious; therefore, owing to FDA regulation and oversight, FDA-approved products are recommended.

Nonhormonal options

• Neuroactive agents, such as selective serotonin reuptake inhibitors, serotonin norepinephrine reuptake inhibitors and gabapentin, appear effective, but larger, long-term, randomized, placebo-controlled trials are needed.

• The most promising therapies appear to be venlafaxine, desvenlafaxine, paroxetine extended release and clonidine (oral or transdermal).

• Nonprescription remedies, such as black cohosh or soy, have demonstrated little efficacy over placebo. The mostly short-term trials were limited by variation in products, menopause populations and inclusion criteria.

• Homeopathy, magnetic therapy, reflexology, dong quai, ginseng, evening primrose oil and vitamin E have not demonstrated significant clinical effectiveness over placebo in trials.

Vaginal atrophy

• Thinning of vaginal epithelium leads to symptoms of vaginal dryness, itching and dyspareunia. Systemic and topical vaginal estrogen products provide relief of vaginal atrophy. Vaginal moisturizers and lubricants provide symptomatic relief but are not as effective as estrogen.

• Topical estrogen may prevent urinary tract infection and help urinary incontinence.

• Long-term safety has not been demonstrated in randomized, controlled trials. Therefore, concern exists regarding systemic absorption and potential adverse effects on the uterus or breasts.

Selective estrogen receptor modulators alone or in combination with estrogen

• Raloxifene is FDA approved for the prevention and treatment of postmenopausal osteoporosis and for the prevention of invasive estrogen-receptor positive breast cancer in postmenopausal women at high risk for breast cancer.

• Raloxifene can be combined with vaginal estrogen, but there are safety concerns if it is combined with systemic estrogen.

• Bazedoxifene is being tested alone and in combination with oral estrogen for the prevention and/or treatment of menopausal symptoms and prevention of bone loss.

• MF-101, a selective herbal estrogen receptor-β agonist for the treatment of VMS, is in clinical testing.

Future perspective

Hormone therapy is effective at relieving VMS. Lower doses of systemic and vaginal estrogen are currently in testing. For women who cannot or will not take HT, safe and effective alternatives are needed. SERMs provide potential benefits of estrogen agonist and antagonistic activity; however, as a class, they have been found to increase VMS. A new paradigm in testing is a tissue-selective estrogen complex, BZA, that pairs CE with a new SERM. This combination offers estrogen agonist relief of VMS as well as estrogen antagonistic effects on the uterus and breast, obviating the need for progestogens. Additional SERMs are currently in testing with agonist effects on bone and antagonistic effects on uterus, although none appear to relieve hot flashes. Promising nonhormonal centrally active therapies in testing for the treatment of VMS include gabapentin, venlafaxine and its variant, desvenlafaxine, and various SSRI antidepressants. Although herbal products have not been demonstrated to be as effective as HT or better than placebo in most well-designed trials, a targeted herbal ER-β agonist, MF-101, is in clinical testing for hot flash relief.