Results of literature search and review. We searched MEDLINE (1966 to March 2005), PsychINFO (1974 to March 2005), the Cochrane Library database (1966 to March 2005), AMED (1953 to November 2004), and MANTIS (1880 to November 2004).
Nedrow A, Miller J, Walker M, Nygren P, Huffman LH, Nelson HD. Complementary and Alternative Therapies for the Management of Menopause-Related SymptomsA Systematic Evidence Review. Arch Intern Med. 2006;166(14):1453-1465. doi:10.1001/archinte.166.14.1453
Nearly half of adults in the United States use complementary and alternative therapies each year for a variety of reasons. These therapies are increasingly popular among women seeking alternatives to treatment with estrogen for managing menopausal symptoms. The objective of this review was to assess the effectiveness of complementary and alternative therapies in the management of menopausal symptoms.
MEDLINE, PsychINFO, Cochrane Library database, MANTIS, and AMED.
Full-text, English-language, randomized controlled trials and meta-analyses comparing a complementary or alternative therapy with placebo or control for treatment of menopausal symptoms.
All eligible trials were reviewed, abstracted into evidence tables, and rated for quality.
Seventy randomized controlled trials met inclusion criteria. Forty-eight studies of phytoestrogens and other biologically based agents showed mixed results. Smaller numbers of studies using mind-body, energy, manipulative, and body-based therapies and whole medical systems showed little benefit in treating menopausal symptoms.
Although individual trials suggest benefits from certain therapies, data are insufficient to support the effectiveness of any complementary and alternative therapy in this review for the management of menopausal symptoms. Many of these potential therapies warrant further study in trials with rigorous scientific designs to determine benefit and safety.
Approximately 40% of women seek medical advice for the management of menopausal symptoms that include hot flashes, night sweats, vaginal dryness, and sleep disorders.1 The Women's Health Initiative influenced many women to discontinue estrogen therapy,2 leading more health care professionals and their patients to consider alternatives to estrogen.1 The 25 million women who will move through menopause during the next decade face an increasingly complex array of alternative therapies for their symptoms.3
Previous surveys4 indicate that 42% of US adults have used alternative medicine during the past year (excluding prayer). Several hundred menopausal women in Seattle, Wash, took part in a survey that listed stress management, herbal remedies, and massage as popular approaches.5 Seventy percent of them did not reveal to their physicians the use of dietary supplements for menopausal symptoms.6 Millions of dollars are spent on over-the-counter products for menopausal symptoms by women who have little knowledge of their quality, safety, or effectiveness. The National Institutes of Health invests over $200 million annually in research on alternative therapies.7
The National Center for Complementary and Alternative Medicine has divided complementary and alternative medicine into the following 5 categories: biologically based, mind-body, energy, manipulative and body-based therapies, and whole medical systems.7
Biologically based therapies, purchased over the counter or through specialized health professionals, include botanicals, animal-derived extracts, vitamins, minerals, fatty acids, amino acids, proteins, probiotics, whole diets, and functional foods. Mind-body therapies focus on the interactions among the mind, body, and behavior and the ways in which emotional, mental, social, spiritual, and behavioral factors can directly affect health. The mind-body approach respects and enhances each person's capacity for self-knowledge and self-care, and it emphasizes techniques that are grounded in this approach. Examples include abdominal breathing, guided imagery, and meditation.
Energy therapies address veritable and putative energy fields. Veritable energies employ mechanical vibrations, such as sound and electromagnetic forces. Electromagnetic forces use specific, measurable wavelengths and frequencies to treat patients, as found in transcutaneous electrical nerve stimulation (TENS) units or magnets. In contrast, to date, putative energy fields have defied measurement by reproducible methods. Therapies involving these energy fields are based on the concept that human beings are infused with a subtle form of energy. Therapists claim that they can see and work with this energy, using it to effect changes in the physical body to influence health. Popular examples include therapeutic touch and Reiki.
Manipulative and body-based therapies include chiropractic and osteopathic manipulation, massage, and techniques such as the Feldenkrais method and Rolfing.
Whole medical systems involve complete systems of theory and practice that have evolved independently from allopathic medicine and that are often culturally based. Eastern whole medical systems include those from China (traditional Chinese medicine) and India (ayurvedic medicine). Western whole medical systems include homeopathy and naturopathy. Additional examples can be found in Native American, African, Middle Eastern, Tibetan, and Central and South American cultures.
This systematic review evaluates alternative therapies for the management of menopausal symptoms classified using these categories. A summary of the complete review has been published separately.8
We identified studies from multiple searches of MEDLINE (1966 to March 2005), PsychINFO (1974 to March 2005), the Cochrane Library database (1966 to March 2005), systematic reviews, reference lists, experts, and Web sites.8 Searches of MANTIS and AMED revealed no relevant studies. Articles were selected based on predetermined inclusion and exclusion criteria.8 Randomized placebo-controlled trials and meta-analyses published in English, using alternative therapies categorized by the National Center for Complementary and Alternative Medicine, were included. Trials enrolling women with breast cancer were included; trials of nonmenopausal women and studies using animals were excluded.
Data were abstracted into evidence tables and summarized descriptively; we did not perform statistical analysis because of the heterogeneity of trials. Outcomes included hot flash frequency and severity, sleep disturbance, vaginal dryness, vaginal bleeding, urinary frequency or incontinence, quality-of-life changes, depression, anxiety, sexual dysfunction, and cognitive function, most frequently measured by the Kupperman Index9 and the Greene Climacteric Scale.10 The Kupperman Index measures the self-reported severity of hot flashes on a scale of 0 to 3 and 10 other symptoms: paresthesias, insomnia, nervousness, melancholia, vertigo, weakness, arthralgia or myalgia, headache, palpitations, and formication. The Greene Climacteric Scale assesses the severity of 21 self-reported symptoms on a 4-point scale and includes psychological, somatic, vasomotor, and sexual dysfunction symptoms.
A best-evidence approach was applied to eligible trials; trials with the highest-quality and most rigorous design are emphasized.11 Two reviewers (A.N. and J.M.) independently rated the quality and external validity of trials using criteria specific to randomized controlled trials developed by the US Preventive Services Task Force (see the Appendix12 available online at www.ohsu.edu/cam). When reviewers disagreed, a final rating was reached through consensus. Characteristics of poor-quality trials include enrollment of fewer than 20 subjects, study duration of less than 12 weeks, and not reporting group differences or ages of subjects.
From 1432 identified abstracts, 70 randomized controlled trials met inclusion criteria: 48 biologically based therapies, 9 mind-body therapies, 1 manipulative or body-based therapy, 2 energy therapies, and 10 whole medical systems (Figure).
Table 1 shows results from trials of phytoestrogens. Included in the table are dietary soy isoflavones (eg, flour,19,21 beverage,14,15,20,22,23 and powder13,16- 18 forms; supplement and extract forms of soy isoflavones; red clover phytoestrogens; and other phytoestrogens).
A good-quality study enrolling breast cancer survivors compared 56 patients ingesting 90 mg/d of isoflavone soy drink with 55 patients who took placebo, with no differences reported between the 2 groups in hot flash frequency or intensity, yet the study found improvement in both groups from baseline.13 The largest fair-quality trial of women without breast cancer enrolled 241 women for 24 months and compared 2 doses of isoflavones in a soy drink (58 mg/d vs 42 mg/d) with placebo.16 No between-group differences in frequency or severity of hot flashes or nights sweats were observed. Four additional trials failed to show benefit compared with placebo.18- 21 Albertazzi and coauthors14,15 reported mixed results in hot flash reduction in 2 fair-quality studies. Three studies were rated as being of poor quality owing to short duration23 or failure to report between-group differences.17,22
The 8 trials of soy isoflavone supplements24- 31 included 2 studies of women with breast cancer.28,29 Doses ranged from 36 to 150 mg/d of isoflavones, averaging 50 mg/d. One fair-quality trial of women with breast cancer showed improvement in hot flash severity yet no improvement in frequency of night sweats.29 A similar trial showed reduction in hot flash frequency but not severity compared with placebo.28 A trial of 75 women comparing the effectiveness of soy isoflavone with that of placebo showed significant improvement in hot flash frequency; however, no effect was seen on other menopausal symptoms.25 Poor-quality studies showed variable results.30,31 Results were mixed in 2 fair-quality studies on cognitive memory testing among women using soy supplements.24,26
Red clover was the active compound examined in 6 trials.32- 37 A good-quality 12-week trial32 comparing red clover isoflavone tablets (82 mg/d vs 57 mg/d) with placebo in 252 women experiencing hot flashes showed no difference in any of the 3 groups as measured by number of hot flashes per day (P<.001). Three other trials, 1 of which enrolled 205 women,33 reported similar results.34,37
Seven studies, 2 enrolling women with breast cancer,39,41 used other phytoestrogen preparations.38- 44 The longest study included 90 women and compared 54 mg/d of genistein isoflavone with placebo or 1 mg/d of estradiol and norethisterone for 1 year.38 Genistein decreased the number of hot flashes by 22% compared with placebo at 3 months, and this effect persisted at 12 months (P<.01). Estrogen had a superior effect compared with genistein, decreasing the number of hot flashes by 53% vs placebo at 3 months, and this effect persisted at 6 and 12 months. Two studies using phytoestrogen topical creams failed to show any benefit.43,44 In the 2 trials including women with breast cancer, neither reported benefits compared with placebo on the Kupperman Index39 or the Greene Climacteric Scale.41
Other biologically based studies (Table 2) included black cohosh, dehydroepiandrosterone, vitamin E, kava, phospholipid liposomes, s-adenosyl-L-methionine, melatonin, guar gum, vaginal lubricants, and combination therapies.
Four trials of black cohosh (Cimicifuga racemosa) met our criteria45- 48; 2 included women with breast cancer who were taking tamoxifen.45,48 The most recent and largest trial enrolled 304 women randomized to either 40 mg/d of a black cohosh supplement or placebo for 12 weeks.46 Improvement was observed in the treatment group for a variety of menopausal complaints as measured by the Menopause Rating Scale98 and included mood, sleep disorders, sexual disorders, sweating, and hot flashes. These results are in contrast to a smaller study that used a different botanical formulation and that did not show significant reduction in hot flashes compared with placebo.47 One poor-quality study45 and one fair-quality study48 suggested no improvement in hot flashes in breast cancer survivors.48,45
The 2 dehydroepiandrosterone studies were of fair and poor quality.49,50 The fair-quality trial did not demonstrate a significant benefit of dehydroepiandrosterone compared with placebo for menopausal symptoms or quality of life,49 and the poor-quality trial did not report between-group differences.50
Eleven trials of other agents met our criteria51- 61; 8 were of poor quality. One fair-quality trial enrolled 125 women with prior breast cancer51 and compared the effectiveness of 800 IU/d of vitamin E with placebo. There were no between-group differences in hot flash frequency or severity.51 Two fair-quality studies, 1 of a trial with kava and 1 with phospholipid liposome injections, showed measurable improvement in anxiety over the placebo.52,53 Women using phospholipid liposomes, but not kava, also had improvement in menopausal symptoms.
Nine trials (Table 3) met our criteria in the mind-body and behavioral therapy category and included exercise,62,65,69 relaxation breathing,66,68 progressive muscle relaxation,67 audiotape relaxation,70 stress management and menopause education,64 and counseling support for women with breast cancer.63
A fair-quality trial randomized 173 women to aerobic exercise therapy 225 min/wk compared with controls undertaking stretching 45 min/wk. There were no between-group differences in hot flash frequency, sleep disturbance, mood, or cognitive function.62 Another fair-quality trial compared women undertaking aerobic exercise 3 times per week with women under usual care that included menopause hormonal therapy for 6 weeks. The exercise group showed improved quality of life compared with the control group.65
Four poor-quality studies evaluated benefits of relaxation therapy.66- 68,70 Two studies trained participants in breathing therapies66,68 but did not report between-group differences. Six months of progressive muscle relaxation suggested a delayed onset of hot flashes in the treatment group compared with a control biofeedback group.67 A study involving 60 minutes per week of audiotape sound wave therapy had no impact on menopausal, somatic, or psychological symptom outcomes.70
A stress and menopause management education intervention was compared with usual care in 86 women older than 50 years and demonstrated that the control group more commonly attributed pain to menopause than the education group; however, no other benefit for menopausal symptoms was reported.64 In another fair-quality study that used counseling and emotional support in 76 women with breast cancer, an adjusted mean change in both menopausal symptoms and sexual function in the intervention group compared with the usual care group was reported.63
Table 4 lists body-based therapies. The only eligible study incorporating body-based techniques evaluated low-force osteopathic manipulation of the pelvis, spine, and cranium compared with sham low-force touch for 30 menopausal women.71 Investigators reported improvement in hot flashes, night sweats, urinary frequency, depression, and insomnia with the treatment group compared with the sham touch group.71
Two studies examined the energy therapies of reflexology72 and magnetic devices.73 A comparison of standard foot reflexology and routine foot massage reported no differences between study groups in hot flashes, night sweats, or quality-of-life measures.72 A trial enrolling 15 women with breast cancer evaluated benefits of 6 magnetic devices applied over classic Chinese acupressure points used to treat hot flashes.73 Hot flash frequency improved in the placebo group compared with the group using magnets (P = .02).
For whole medical systems (Table 5), 4 trials of acupuncture74- 77 and 6 trials78- 83 of traditional Chinese medicinal herbs met our criteria. All trials of acupuncture used 6 to 14 acupuncture sessions specific to menopausal symptoms over an 8- to 12-week period. Sham acupuncture using superficial needle insertion was used as the control in 3 trials,74,76,77 and acupuncture intended for general well-being was used as the control in the fourth trial.75 In the 3 trials reporting between-group differences, none showed improved hot flash frequency with acupuncture compared with sham acupuncture,74,76,77 although mood was improved in the treatment group in 1 poor-quality trial.76 A fair-quality trial of 45 women compared electroacupuncture, sham acupuncture, and a conjugated dosage of 0.625 mg of estrogen per day. Although all 3 groups reported improvement in self-reported symptom diaries over 12 weeks, only the estrogen group had significant improvement compared with the other groups (P<.001).74
Six trials78- 83 of traditional Chinese medicinal herbs (half using combination therapies) met our criteria.80- 82 None showed a significant benefit over controls for menopausal symptoms. Therapy with standardized (ie, quality control of what is believed to be the active ingredient to ensure consistent dosage between samples) ginseng in 384 women showed improvement in depression, well-being, and health scores compared with placebo over 16 weeks, but there were no between-group differences in hot flashes.79
Adverse effects and safety data from these and other studies are limited, and most studies lacked consistent or clear reporting. A 5-year follow-up study of soy indicated that users were at a significantly increased risk for endometrial hyperplasia compared with placebo.84 Hepatotoxicity has been associated with both black cohosh85- 87 and kava.88 An increased number of breast cancer metastasis as studied in MMTV-neu transgenic mice receiving black cohosh has been observed.89
Individual trials suggest a benefit for certain therapies, yet data are insufficient to recommend any complementary and alternative therapy as effective for the management of menopausal symptoms. Of the 15 fair- to good-quality trials of phytoestrogens, only 4 showed a benefit. This is consistent with a recent systematic review of phytoestrogens as treatment for hot flashes that included 25 trials and 2348 participants.90
A single large trial46 showed a benefit for vasomotor symptoms with the black cohosh formulation Remifemin (Enzymatic Therapy Inc, Green Bay, Wis) and is consistent with older German studies.91- 93 However, the remaining 3 studies showed no benefit of black cohosh. Two45,48 were confounded by concurrent use of tamoxifen, known to exacerbate hot flashes, and 1 study47 used a more unusual (research) preparation, BNO1055.
Limitations to this review are the exclusion of many relevant articles in the field of alternative medicine written in non-English languages, difficulty in interpretation of the data owing to large placebo effects, nonstandardized outcome measures, nonstandardized biologically based therapies, variable definitions of the menopausal transition, and use of inconsistently defined populations. The large placebo effect is consistent with preexisting work of menopausal hormonal therapies. A study of estrogen compared with placebo reported a 50% improvement in frequency of hot flashes in the placebo group.94 The placebo effect likely plays an important role in the expanding number of dietary supplements marketed to menopausal women.
The variety of outcome measures may have contributed to the mixed results and range from sophisticated objective sternal conductance tests66- 68 to subjective measures through daily self-reported symptom diaries. For menopause, as for pain management, subjective relief of symptoms may be the most meaningful data.95 The Kupperman Index and the Greene Climacteric Scale cannot be compared because they measure different symptoms.
Another limitation to interpreting trials of therapies is the variability of nomenclature describing the menopausal transition. Progress is under way in defining standard nomenclature for the menopause transition,4 helping to define optimal treatment for various phases.
The populations enrolled in trials varied from random samples of menopausal women to women specifically recruited through menopause or gynecological clinics. Results may not be generalizable. Clinicians caring for menopausal women recognize different symptoms in women undergoing abrupt surgical menopause, premature ovarian failure, or the natural menopausal transition at midlife. In most of the studies included in this review, the cause of menopause was not correlated with therapy outcomes.
Data interpretation is further complicated by the nature of research in alternative medicine. Current published trials are generally small, of short duration, and have inadequate methods, such as not reporting between-group differences. Standardization of biologic products is poor, making direct comparison difficult. Nearly half of the studies in this review were rated as being of poor quality owing to these and other methodological shortcomings. Observational data confirm that different ethnic groups of women seem to experience menopause differently.96 Ethnicity of the women included in trials was not routinely recorded, and, when it was, therapeutic responses were not correlated.
Many women report the onset of hot flashes, night sweats, or sleep disturbance on cessation of any therapy for their menopausal symptoms. This lack of curative benefit complicates the risk-benefit ratio of recommending therapies to women.97
The most important thing that the health professionals can do for symptomatic menopausal women is to encourage open communication that allows patients to disclose treatments they are using. Women value partnership, choice, and shared decision making. Because there is no universal menopausal presentation or treatment, it is essential that health care professionals provide accurate information and options for midlife women.
The next decade will see an increasing number of women transition through menopause. Currently, patients and health care professionals are more thoughtful in considering menopausal hormonal therapy for symptom management because of safety concerns. Such caution will increase the demand for alternatives that are effective and safe. Future research should focus on large, rigorously designed trials that ensure reliable comparisons between studies, distinguishing cause and phase of menopause and ethnic differences. Lifestyle modification and mind-body techniques may have high safety profiles and result in additional health benefits. Many alternative therapies used by menopausal women, such as massage, aromatherapy, yoga, and ayurvedic therapy, need to be studied in randomized, controlled trials. It is imperative that future research be directed toward safety and efficacy of the common modalities used by women to treat their menopausal symptoms.
Correspondence: Anne Nedrow, MD, Oregon Health and Science University, Mail Code L466, 3181 SW Sam Jackson Park Rd, Portland, OR 97239 (email@example.com).
Accepted for Publication: April 21, 2006.
Financial Disclosure: None reported.
Funding/Support: This study was funded in part by the Oregon Evidence-based Practice Center under contract to the Agency for Healthcare Research and Quality, contract No. 290-02-0024, task order No. 5, Rockville, Md. Additional support came from National Institutes of Health (NIH) grant AT01173-03 and the Portland Veterans Affairs Medical Center Women's Health Fellowship. The NIH Office of Medical Applications of Research funded this research through the Agency for Healthcare Research and Quality Evidence-based Practice Centers Program for the NIH-sponsored State of the Science Conference on Managing Menopause-related Symptoms.
Acknowledgment: We thank the institute and agency staff and content experts who reviewed interim reports. We thank Linda Humphrey, MD, Christina Nicolaidis, MD, Elizabeth Haney, MD, Kimberly Vesco, MD, and Christina Bougatsos, BS, for their contributions on this project. We also thank our research librarian, Andrew Hamilton, MLS, MS, for conducting the literature searches.