Does scientific evidence support the use of non-prescription supplements for treatment of acute menopausal symptoms such as hot flushes?

Abstract

The objective of the present review was to critically evaluate the scientific evidence for efficacy of non-prescription supplements (NPS) available for treatment of acute menopausal symptoms; and where available, to identify constituents, specify dosage, propose mechanisms and indicate safety concerns. A Medline search identified 24 clinical trials assessing efficacy of seven NPS for treatment of hot flushes in symptomatic menopausal women; 19 were randomised placebo-controlled trials. Articles were located on black cohosh, dong quai, evening primrose oil, ginseng, isoflavonoid phytoestrogens (from red clover or soy) and sage. Evidence for each NPS was tabulated according to study design and menopausal outcomes. Findings from 13 randomised placebo-controlled trials, using isoflavonoids from red clover or soy, demonstrated significant efficacy for six studies (46%) with reductions in moderate to severe flushes by 10-44% compared with placebo; seven studies (54%) reported no significant improvements. In vitro studies indicate that isoflavonoids have significant oestrogenic activity, providing a supporting mechanism. One controlled trial reported that an extract of sage and alfalfa significantly reduced severe hot flushes by 60% compared with placebo. Two controlled trials reported that black cohosh improved overall menopause symptom scores; however, no data were available specifically on hot flushes. In conclusion, evidence from a small number of placebo-controlled trials suggested that further research is warranted to test efficacy of red clover, soy, sage and black cohosh for treatment of hot flushes in menopausal women. To date, controlled trials do not support use of dong quai, evening primrose oil and ginseng for treatment of hot flushes.

Key words: complementary and alternative medicines, hot flush, isoflavonoids, menopause, soy protein, women's health

(Nutr Diet 2005;62:138-151)

Introduction

Controversies surrounding the safety of hormone therapy (HT) have prompted increased interest in non-prescription supplements (NPS) available for treatment of menopausal symptoms. In contrast to HT, NPS are often perceived as natural and, therefore, may represent a safer and more appropriate alternative to HT. However, little is known about their efficacy and the scientific rationale for their use. In view of the increasing community interest and marketing of these supplements, and their traditional use in some cultures, assessment of the scientific evidence supporting NPS is warranted. In the present review, the term NPS includes all tablets, powders or tinctures purchased without a doctor's prescription that are taken to treat menopausal symptoms. Other terms frequently used include herbal remedies, HT alternatives and complementary and alternative medicines.

The present review focuses on evidence supporting the use of NPS for the treatment of acute symptoms of menopause, particularly hot flushes. Although assessed in some of the trials, efficacy for other symptoms less specifically associated with menopause, such as anxiety and muscle aches, is beyond the scope of the present review. Hot flushes are one of the most common symptoms reported by women undergoing the menopausal transition. The underlying pathophysiology of hot flushes is not well understood; however, a relationship with declining oestrogen levels is assumed; hot flushes peak in the perimenopausal period, when oestrogen levels are fluctuating, and usually subside with age, suggesting sensitivity to relative rather than absolute oestrogen concentration. (1) Many NPS are designed to provide an ingredient with purported oestrogenic activity.

Previous published reviews have assessed the evidence supporting use of NPS for treatment of both acute symptoms associated with menopause, such as hot flushes, and chronic conditions, such as cardiovascular disease and osteoporosis. (2-9) The present review expands upon previous work by focusing in greater depth on acute menopausal symptoms and assessing a range of NPS including black cohosh, dong quai, evening primrose oil, ginseng, sage and phytoestrogens. Investigation of alternatives to HT for treatment of symptomatic menopausal women is a rapidly evolving area. Five reviews were identified that focused on acute symptoms, of which four did not include studies published after 2002. (10-13) The most recently published study investigated phytoestrogens alone. (14) Authors concluded that there was insufficient evidence to support the use of phytoestrogens for treatment of hot flushes and other menopausal symptoms; however, one limitation was that conclusions were drawn from studies that included phytoestrogen-rich diets, for which it is not possible to blind subjects to placebo or intervention status.

The aim of the present review is to provide health professionals with a critical evaluation of the scientific evidence for efficacy of the NPS available for treatment of acute symptoms of menopause, particularly hot flushes; and where available, to identify constituents, specify dosage, propose mechanisms and indicate safety concerns.

Methods

Search strategy for identification of studies

The search strategy identified relevant studies published from 1940 up to March 2004. The following databases were searched: MEDLINE (ScienceDirect), PUBMED, PROQUEST and the Cochrane Controlled Trials Register. Keywords used were: black cohosh, dong quai, evening primrose oil, ginseng, liquorice, red clover, sage, soy, kudzu, hormone replacement therapy and alternatives, phytoestrogens, isoflavonoids, menopause, hot flush, menopause supplement and non-prescription supplement. The literature review identified intervention studies on the efficacy of seven types of NPS for treatment of acute symptoms of menopause: black cohosh, dong quai, evening primrose oil, ginseng, red clover and soy isoflavonoids, and sage.

Criteria for selection of studies

Selection criteria for types of studies were non-restrictive because of the small number of randomised placebo-controlled trials for all supplements, apart from isoflavonoids. Consequently, non-randomised and uncontrolled intervention trials were included. Studies were included if subjects were otherwise-healthy women with menopausal symptoms; women with hormone-sensitive conditions such as breast cancer were not included. In total, 19 randomised controlled trials were identified. Interventions to test the efficacy of NPS for menopause have measured a range of outcomes such as indices of general or psychological health and markers of hormonal status. However, the present review focused only on outcomes specifically associated with menopause, including frequency and severity of hot flushes or night sweats, and scales designed to measure overall menopausal symptoms, such as the Kupperman Menopause Index and the Greene Menopause Score. These scales attempt to quantify symptoms of menopause and incorporate measures of vasomotor (hot flushes), somatic and psychological symptoms. (15,16)

Results and discussion

The review of the literature identified two lines of evidence pertaining to NPS for treatment of menopausal symptoms. The first approach included investigation of the mechanism of action to explain the use of the NPS, most commonly the ability to bind to oestrogen receptors and exert an oestrogenic response. The second line of inquiry was through clinical intervention studies involving symptomatic women, and examined the effects of NPS on hot flushes. As few clinical studies specifically addressed safety issues of NPS, additional information was located in the literature and included in a separate section. The strengths and limitations of the studies are discussed, enabling evaluation of the evidence regarding the efficacy of a particular NPS for treatment of hot flushes.

Black cohosh

Black cohosh (Cimicifuga racemosa) was one of the more extensively studied menopausal NPS. Traditionally used in North America, Europe and China for gynaecological conditions, black cohosh has been used in Germany since the 1940s to treat disorders of menstruation and menopausal symptoms. Biological activity is attributed to the presence of triterpene glycosides, including actein, cimicifugoside, deoxyacetylacteol and 27-deoxyactein. (17) The mechanism proposed to explain the effect of black cohosh has been the possession of oestrogenic components, (18-20) although this has not been supported by in vitro studies investigating the ability of the triterpenes to bind to the oestrogen receptor and exert an oestrogenic response. (21-23) Authors have postulated that black cohosh may possess alternative mechanisms of action by activating receptors for pituitary hormones that regulate oestrogen synthesis, such as luteinising hormone or follicle stimulating hormone, or that black cohosh components influence enzymes required in the synthesis of endogenous oestrogen. (20,21,23) However, these possibilities have not been tested.

Seven clinical trials between 1982 and the present have investigated the efficacy of black cohosh for relief of symptoms in menopausal women (Table 1); two of these trials were placebo-controlled. (27,30) The majority used Remifemin, a standardised extract of Cimicifuga racemosa root containing 1 mg of the triterpene 27-deoxyacetein per tablet. (17) The current recommended dose of Remifemin is two tablets/day (40 mg black cohosh extract/day). This is lower than the dose of four tablets/day (or 80 drops of liquid extract/day) used in trials prior to 1996, because of an improvement in the extraction process; however, the quantity of active constituent was not altered. (17) Of the seven clinical trials, two were conducted after 1996, using the current recommended dose; (29,30) there was no observed difference in efficacy between studies conducted either before or after introduction of the new formulation.

In a placebo-controlled trial with 80 women, Stoll reported that Remifemin reduced mean overall symptom scores by 59% compared with baseline and responses were significantly greater than the reduction by 29% with placebo. (27) Similar results were obtained in a second placebo-controlled trial by Wuttke et al. (30) Both studies were conducted over 12-week periods among German women with moderately severe symptoms of at least three hot flushes/day. They featured a parallel design that compared the effect of black cohosh with conjugated oestrogen (0.625 mg/day) or placebo on a measure of overall menopausal symptom scores. Although both studies reported a decrease in overall menopausal symptom scores, the effect on the hot flush component was not described (27) or was not significant. (30) Remifemin compared with placebo also resulted in a significant increase in the degree of proliferation of vaginal epithelial cells, suggestive of an oestrogenic effect. (27)

Five uncontrolled trials have also reported consistently significant improvements in overall menopausal symptoms compared with baseline after 12-24 weeks of treatment with Remifemin. (24-26,28,29) None of the studies reported separately on the frequency of hot flushes. In one study the improvement in symptoms was accompanied by increased proliferation of vaginal epithelial cells, (26) although another study found no changes in vaginal epithelium. (29) However, in the absence of a placebo control, the results of these studies are not definitive.

Dong quai

Extracted from the root of Angelica sinesis, dong quai or tang-kuei has been used in Chinese herbal medicine to treat disorders of menstruation and menopausal symptoms. Dong quai contains various coumarins and a phenolic constituent, ferulic acid. The mechanism of action is uncertain, although it has been proposed that the ferulic acid could provide oestrogenic activity. (31) However, results of in vitro studies investigating the ability of dong quai extracts to activate the oestrogen receptor are inconsistent, with findings of no oestrogenic activity, (20,21) an antagonist effect, (22) or a weak oestrogenic effect. (23)

One randomised placebo-controlled trial has investigated the use of dong quai root in 71 symptomatic postmenopausal American women (Table 2). (31) Following a 24-week intervention, no significant differences between treatment and placebo groups were observed in frequency of hot flushes or overall menopausal symptom scores. A power-based estimate of sample size indicated that subject numbers were adequate to detect a difference of 11 hot flushes per week. Although dong quai root given alone did not demonstrate an effect in this group of women, practitioners of Chinese medicine who have reported positive outcomes typically use dong quai in combination with at least four other herbs, including Radix paeoniae lactiflorae, Rhizoma ligustici, Rhizoma atractylodes, Rhizoma alismatis, Sclerotium poriae. (31) These herbs could potentially affect hot flushes either alone or in combination with dong quai but these possibilities have not been tested in placebo-controlled studies.

Evening primrose oil

Evening primrose oil has been perceived by many women to be effective for the control of acute menopausal symptoms. (32) Gamma-linolenic acid, an omega-6 polyunsaturated fatty acid, is considered to be the active component. (35) Gamma-linolenic acid is the precursor of other omega-6 fatty acids including dihomogamma-linolenic acid and prostaglandins of the '1' series, which possess anti-inflammatory and antithrombotic properties. (36,37) Evening primrose oil also contains arachidonic acid, which gives rise to the '2' series of prostaglandins, which are pro-inflammatory. (38) On the basis of anti-inflammatory and antithrombotic properties, evening primrose oil has been promoted for treatment of a range of health problems including cardiovascular disease and arthritis, although there is no scientific rationale to support the use of evening primrose oil for treatment of hot flushes.

One placebo-controlled trial has investigated the effect of evening primrose oil on hot flushes in 56 English women (Table 2) and reported no significant differences between treatment and placebo groups. (32) The sample size was adequate to detect a 13% treatment difference in hot flush frequency; however, one limitation was the high withdrawal rate, with only 63% of women completing the trial, attributed to a lack of efficacy of the treatment. Based on the lack of observed efficacy and a plausible rationale, authors of the study concluded that evening primrose oil should not be promoted for treatment of hot flushes.

Ginseng

Ginseng (Panex ginseng), a traditional herbal medicine used in China, Japan and Korea, is widely valued as a stimulant. (39) The biologically active ingredients are considered to be a series of saponin glycosides known as ginsenosides or panoxosides. (40) Oestrogenic activity of the ginseng saponins has been suggested as a mechanism of action to support the use of ginseng in treatment of menopausal symptoms. (39) In one study, which used a methanolic extract of ginseng, significant binding to oestrogen receptors in a human myometrial tissue assay was observed; (39) however, two assays reported no significant oestrogen receptor binding affinity. (22,23)

One randomised placebo-controlled trial has been conducted to test the efficacy of ginseng in 384 postmenopausal Swedish women (Table 2). (33) After 16 weeks, ginseng, compared with the placebo, had no significant effect on the frequency of hot flushes, with similar reductions in frequency observed in both treatment (30%) and placebo (24%) groups. The placebo response of 24% was typical for studies of hot flush frequency. The sample size was calculated to detect a change in five units of the Patient Generated Wellbeing Index, rather than the frequency of hot flushes. However, the large sample, with approximately 190 women per group, and high completion rate of 98%. suggest that the numbers would have been sufficient to detect possible changes in hot flush frequency. Results of that study indicate that ginseng is not useful for the treatment of hot flushes.

Sage

Sage (Salvia officinalis) contains the phenolic constituent ferulic acid, which has the potential to bind to the oestrogen receptor, (41) although the specific binding affinity of sage has not been tested. De Leo et al. undertook a study to test the efficacy of a product based on an extract of sage and alfalfa in a group of 30 Italian women with an average of 14 hot flushes/day (Table 2). (34) Following 12 weeks of therapy the occurrence of hot flushes ceased in 20 women (67%); was reduced in frequency to between one and three flushes/day in four women (13%), and to between four and six flushes/day in six women (20%). The mean reduction by 85% in the frequency of hot flushes in the treatment group was significantly greater than the reduction by 25% in the placebo group, and comparable in magnitude to the effect of HT. (42) De Leo et al. concluded that the product was an effective agent for treatment of hot flushes and without side-effects. (34) However, the mechanism is not clear and it cannot be assumed that sage alone was the effective component because alfalfa (Medicago sativa) is a significant source of the isoflavonoid, coumestrol, which could have contributed to the observed efficacy. (43) Further studies using sage and/or alfalfa in treatment of menopausal symptoms are required to confirm findings of that study and to clarify a mechanism.

Phytoestrogens

Phytoestrogens were the most extensively studied NPS for the treatment of acute menopausal symptoms. The main supplemental sources of phytoestrogens used in the clinical trials were isoflavonoids derived from either red clover or soy. These are discussed separately as their isoflavonoid profiles differ. Red clover (Trifolium pratense), a forage legume, contains genistein, daidzein and their methylated precursors, biochanin A and formononetin, respectively. Soybean (Glycine max) contains predominantly genistein and daidzein, with smaller quantities of glycitein. (44,45)

The primary mechanism proposed to explain the effect of isoflavonoids in the treatment of hot flushes pertains to weak oestrogen agonist effects. Isoflavonoids are structurally similar to the mammalian oestrogen, 17-[beta] oestradiol, and in vitro studies using tissue assays have demonstrated their capacity to bind to oestrogen receptors (ER). (22,23,43,46-48) Although molar binding affinities for isoflavonoids to the classic ER-[alpha] are 100- to 1000-fold weaker than that of oestradiol, certain isoflavonoids, notably genistein and coumestrol, have a relatively high binding affinity for the ER-[beta] subtype. For example, the binding affinity of genistein compared with oestradiol is 4% for ER-[alpha] and 87% for ER-[beta]. (48) Isoflavonoids can also stimulate transcriptional activity in both ER-[alpha] and ER-[beta] subtypes at physiological concentrations, suggesting they have the potential to exert an oestrogenic response. (48,49) These data support a potential mechanism for the possible efficacy of phytoestrogens in the treatment of hot flushes.

Isoflavonoids from red clover

Four randomised placebo-controlled trials have been conducted to assess the efficacy of red clover isoflavonoids in the treatment of hot flushes (Table 3). (50-53) Following adjustment for the placebo, the effect of isoflavonoids remained significant in one study, which reported a 44% decrease in the frequency of hot flushes. (52)

The report of a significant 44% reduction compared with the placebo in the frequency of hot flushes in the study by van de Weijer and Barentsen supports a role for red clover in the treatment of hot flushes. (52) However, this conclusion is disputed by the results of the three other trials, (50,51,53) including a study by Tice et al. with a larger sample size (252 women with 84 per group), high response rate (98%), equally stringent selection criteria in terms of symptom severity (>five flushes/day), and the same duration, dose and form of supplement (80 mg isoflavonoids/day for 12 weeks). (53) Notably, both studies showed a similar reduction in the frequency of symptoms, by 41% (53) or 44% (52) compared with baseline. However, a major difference was the magnitude of the placebo response compared with baseline, with a 36% decrease in the Tice study (53) compared with a 0% decrease in the van de Weijer and Barentsen study, (52) which allowed a significant treatment effect to emerge in the latter but not the former study. In a power-based estimate of the sample size, Tice et al. designed their study to detect at least a 15% greater reduction in hot flush frequency in the treatment compared with the placebo arms, and assumed a placebo response of 25% compared with the baseline. Therefore, the higher than expected placebo response (36%) would have contributed to the lack of overall treatment effect. (53) Interestingly, although these investigators reported no effect of isoflavonoids in the treatment group as a whole, they did observe a greater reduction of hot flushes in women above the median body mass index (BMI) (>25.1 kg/[m.sup.2]) compared with those below the median BMI (<25.1 kg/[m.sup.2]); the BMI trends were seen within the two isoflavonoid treatment groups but not within the placebo group (Table 3). Further studies are required to investigate the effect of BMI and adiposity on individual responsiveness to isoflavonoid treatment.

Isoflavonoids in the soy protein matrix or purified soy isoflavonoids

Nine randomised placebo-controlled trials have been conducted to investigate isoflavonoids within a soy protein matrix (54,55,60,61) or in purified form (56-59,62) (Table 3). Taken together, the reduction in the number or severity of hot flushes ranged from 28% to 61% (median 45%) in treatment groups compared with baseline and 18-77% (median 30%) in the placebo groups. After accounting for the placebo, the effect of isoflavonoids was significant in five of the nine studies. (54,55,57-59)

Power-based estimates of the sample size in two of the studies indicated that 30-35 subjects per group were required to detect a treatment difference of three hot flushes per day. (54,58) The majority of studies had a sample sizes between 27 and 89 women per group, although two studies had smaller numbers of 17 (55) or 20 per group, (59) which could limit the interpretation of their findings. Inclusion criteria for the frequency of hot flushes varied from low, more than one or two hot flushes/day (55,60,61) to moderate, more than five hot flushes/day (57) or high, more than seven to 10 hot flushes/day. (54,58,62) With the exception of one study, (62) significant improvements compared with the placebo were demonstrated only in studies that selected women with at least moderate to high flushing frequency; (54,57,58) no significant effects were observed in women with milder symptoms of one to two flushes/day; (56-59,60-62) one study did not specify the frequency of hot flushes in the inclusion criteria. (59)

There were no apparent dose-dependent effects, although in one study, a split dose designed to provide more consistent blood levels during the day was more effective than a single daily dose. (55) Efficacy was reported at isoflavonoid doses between 34 and 76 mg/day, (54,55,57-59) whereas in studies with no significant improvements, doses ranged from 42 to 118 mg/day. (56,60-62) The effective doses were within the physiological range of intakes achievable by regular consumption of phytoestrogen-rich foods, for example, mean isoflavonoid intakes obtained in communities with traditional consumption of soy foods have ranged from 15 to 54 mg isoflavonoids/day. (63-65)

Another possible source of variation relates to the reproducibility of the isoflavonoid content of supplements. An analysis of the isoflavonoid content of commercially available supplements, reported that Promensil (Novogen, Sydney, NSW, Australia), the supplement used in all studies with red clover isoflavonoids, provides a reproducible dose of isoflavonoids. (66) However, the accuracy of the isoflavonoid content of soy supplements cannot be confirmed from the data provided.

There was no clear preference between the provision of isoflavonoids in a soy protein matrix and as a purified supplement, with significant effects reported in two of four studies using soy protein (54,55) and three of five studies using purified soy isoflavonoids. (57-59) Efficacy was not consistently related to the duration of the study, with improvements noted in trials ranging from 6 to 16 weeks (54,55,57-59) and no improvements in studies from 12 to 104 weeks. (56,60-62)

A striking source of variation in the studies with either red clover or soy isoflavonoids was the placebo response, which ranged in magnitude from 0 to 70% compared with the baseline. The larger placebo responses in some studies could reflect the high motivation of the participants and belief in the benefits of isoflavonoids or inadvertent consumption of isoflavonoid-containing foods, such as soy and other legumes, during the study, as confirmed by increased isoflavonoid excretion in the placebo groups in two of the clinical trials. (50,51) This highlights the challenge of conducting such studies alongside growing community awareness of the possible benefits of soy and isoflavonoids and ready availability of products in pharmacies and supermarkets. The study that showed the greatest reduction in hot flushes, by 44% with treatment compared with placebo, implemented procedures to minimise the placebo effect by using a four-week observation period prior to intervention to identify susceptible individuals. (52) That study also involved Dutch participants whose diets are traditionally low in isoflavonoids, and participants were further instructed not to increase their intake of isoflavonoid-rich foods during the study.

Responsiveness to treatment could also depend on the individual's capacity to convert the isoflavonoid daidzein to equol, a secondary metabolite, which has greater oestrogenic potency than the original isoflavonoid. Not all individuals have the capacity to produce equol, with observed prevalence rates from 20% to 36% in different studies. (67) This factor could contribute to some of the unexplained variation in responses to the same supplement, with greater efficacy in equol producers. Differences in dietary macronutrient intake could also influence the absorption and hence efficacy of isoflavonoids. (68) None of the studies reviewed specifically addressed these issues.

Isoflavonoids from Kudzu

Kudzu (Pueraria lobata), or Japanese arrowroot, is used as a herb in traditional Chinese medicine. It has a unique profile of isoflavonoids with high concentrations of daidzein, exceeding those found in the soybean, and small quantities of genistein, formononetin, biochanin A and coumestrol. (44) No clinical trials were located on the efficacy of this herb in treatment of menopausal hot flushes.

Safety

Although NPS are considered to represent a lower risk than pharmaceutical goods, a range of safety concerns have been identified in the literature. The studies detailed in Tables 1-3 focused on the efficacy rather than the safety of supplements. Although mild adverse events were reported in trials of black cohosh, dong quai, evening primrose oil, red clover, sage and soy, these did not differ significantly between placebo and intervention groups. (30-32,34,52,54,56-58) Of the studies that specified the nature of adverse events, the most common symptoms related to gastrointestinal effects, such as bloating, nausea and constipation. (54,56-58) However, as the duration of the clinical trials was 12-24 weeks in most studies, long-term safety cannot be assumed and interactions with pharmaceutical drugs may be possible.

Additional information on the safety of NPS was identified in the literature, although many reports of adverse events were isolated and the dose responsible was not specified. Other adverse effects included an association between dong quai and increased skin photosensitivity; (41) and between ginseng and increased hypertension, jitteriness, confusion, depression and insomnia, and risk of hypotension on withdrawal. (69) Dong quai contains coumarins and ferulic acid that can exert antithrombotic effects, which could potentiate the effect of anticoagulant drugs such as warfarin and increase the risk of bleeding in susceptible individuals; (70) increased prothrombin time was observed in a patient who had taken a 565 mg tablet of dong quai one to two times/day for 4 weeks. (71) Consequently, women with clotting disorders or on anticoagulant therapy should not use dong quai.

Conclusions

Results from two randomised placebo-controlled trials suggested that black cohosh compared with the placebo was beneficial for relief of overall symptoms of menopause; (27,30) however, further trials are required to specifically measure the frequency and severity of hot flushes. One preliminary placebo-controlled trial reported that an extract of sage and alfalfa significantly reduced the frequency of severe hot flushes by 60% compared with the placebo. (34) To date, placebo-controlled trials do not support the use of dong quai, (31) evening primrose oil (32) or ginseng (33) for treatment of hot flushes. There was no conclusive evidence for the presence of oestrogenic components in any of these supplements.

The findings from 13 randomised placebo-controlled trials using isoflavonoids from red clover or soy, demonstrated significant efficacy in six studies (46%), with reductions in moderate to severe hot flushes by 10-44% compared with placebo; (52,54,55,57-59) seven studies (54%) reported no significant improvements. (50,51,53,56,60-62) If there was no relationship between isoflavonoids and hot flushes, the chance of a statistically significant treatment effect would be less than one in 20, or 5% of studies. The fact that significant findings were reported by one in two studies (46%), conducted by 13 independent investigators, suggests that these findings may not have arisen by chance. Therefore, it is recommended that further studies are conducted with appropriate design, including an adequate sample size, attention given to minimise inadvertent consumption of isoflavonoids by the placebo group, and evaluation of possible differences in individual responsiveness to treatment, including equol-producer status, (67,68) background diet (67,68) and BMI. (53) In vitro studies have demonstrated that isoflavonoids can bind to oestrogen receptors and exert an oestrogenic response, (20,22,23,46) providing a mechanism to support their use.

The present review provides evidence for health professionals that can be used to inform their advice to women regarding the efficacy of NPS for treatment of acute menopausal symptoms. Studies on the use of dong quai, evening primrose oil and ginseng suggest that they are unlikely to be useful for treatment of hot flushes. Evidence from a small number of short-term, randomised control trials of 12- to 24-week duration indicates that further research is warranted on black cohosh, red clover, soy, sage and alfalfa, involving studies with appropriate design and outcome measures, to establish any evidence of efficacy and evaluate optimal dosage, elucidate precise mechanisms of action and identify any long-term adverse effects.

References

1. Walsh B, Schiff I. Vasomotor flushes. In: Flint MKF, Utian W, eds. Multidisciplinary Perspectives on Menopause. New York: New York Academy of Sciences, 1989; 346-56.

2. Knight D, Eden J. A review of the clinical effects of phytoestrogens. Obstet Gynaecol 1996; 87: 897-904.

3. Knight D, Eden J. Phytoestrogens--a short review. Maturitas 1995; 22: 167-75.

4. Kurzer MS, Xu X. Dietary phytoestrogens. Annu Rev Nutr 1997; 17: 353-81.

5. Glazier G, Bowman M. A review of the evidence for the use of phytoestrogens as a replacement for traditional estrogen replacement therapy. Arch Int Med 2001; 161: 1161-72.

6. Kass-Annese B. Alternative therapies for menopause. Clin Obster Gynecol 2000; 43: 162-83.

7. Keller C, Fullerton J, Mobley C. Supplemental and complementary alternatives to hormone replacement therapy. J Am Acad Nurse Pract 1999; 11: 187-98.

8. Kang H, Ansbacher R, Hammoud M. Use of alternative and complementary medicine in menopause. Internat J Gynecol Obstet 2002; 79: 195-207.

9. Seidl M, Stewart D. Alternative treatments for menopausal symptoms. Systematic review of scientific and lay literature. Can Fam Physician 1998; 44: 1299-308.

10. Kronenberg F, Fugh-Berman A. Complementary and alternative medicine for menopausal symptoms: a review of randomized, controlled trials. Ann Int Med 2002; 137: 805-13.

11. Huntley A, Ernst E. Soy for the treatment of perimenopausal symptoms--a systematic review. Maturitas 2004; 47: 1-9.

12. Huntley A, Ernst E. A systematic review of herbal medicinal products for the treatment of menopausal symptoms. Menopause 2003; 10: 465-76.

13. Philp H. Hot flashes--a review of the literature on alternative and complementary treatment approaches. Altern Med Rev 2003; 8: 284-302.

14. Krebs E, Ensrud K, MacDonald R, Wilt T. Phytoestrogens for treatment of menopausal symptoms: a systematic review. Obster Gynaecol 2004; 104: 824-36.

15. Alder E. The Blatt-Kupperman menopausal index: a critique. Maturitas 1998; 29: 19-24.

16. Greene J. Constructing a standard climacteric scale. Maturitas 1998; 29: 25-31.

17. Pepping J. Black cohosh: Cimicifuga racemosa. Am J Health Syst Pharm 1999; 56: 1400-402.

18. Lieberman S. A review of the effectiveness of Cimicifuga racemosa (black cohosh) for the symptoms of menopause. J Womens Health 1998; 7: 525-9.

19. Shao Y, Harris A, Wang M et al. Triterpene glycosides from Cimicifuga racemosa. J Nat Prod 2000; 63: 905-10.

20. Onorato J, Henion J. Evaluation of triterpine glycoside estrogenic activity using LC/MS and immunoaffinity extraction. Anal Chem 2001; 73: 4704-10.

21. Zava D, Dollbaum C, Blen M. Estrogen and progestin bioactivity of foods, herbs, and spices. Proc Soc Exp Biol Medical 1998; 217: 369-79.

22. Rosenberg Zand R, Jenkins D, Diamandis E. Effects of natural products and nutraceuticals on steroid hormone-regulated gene expression. Clin Chim Acta 2001; 312: 213-19.

23. Liu J, Burdette J, Xu H et al. Evaluation of estrogenic activity of plant extracts for the potential treatment of menopausal symptoms. J Agric Food Chem 2001; 49: 2472-9.

24. Daiber W. Klimakterische beschwerden: ohne hormone zum erfolg/Menopause symptoms: success without hormones. Arztl Prax 1983; 35: 1946.

25. Vorberg G. Treatment of menopause symptoms/Therapie klimakterischer beschwerden. Zeitschrift Allege Meimedizin 1984; 60: 626-9.

26. Warnecke G. Using phyto-treatment to influence menopause symptoms. Med Welt 1985; 36: 87.

27. Stoll W. Phytotherapy influences atrophic vaginal epithelium. Therapeutikon 1987; 1: 23.

28. Lehmann-Willenbrock E, Riedel H. Clinical and endocrinologic examinations concerning therapy of climacteric symptoms following hysterectomy with remaining ovaries. Zen Bl Gynakol 1988; 110: 61.

29. Liske E, Hanggi W, Henneicke-Von Zepelin H, Boblitz N, Wustenberg P, Rahlfs V. Physiological investigation of a unique extract of black cohosh (Cimicifugae racemosae rhizoma): a 6-month clinical study demonstrates no systemic estrogenic effect. J Womens Health Gend Based Med 2002; 11: 163-74.

30. Wuttke W, Seidlova-Wuttke D, Gorkow C. The Cimicifuga preparation BNO 1055 vs conjugated estrogens in a double-blind placebo-controlled study: effects on menopause symptoms and bone markers. Maturitas 2003; 44 (Suppl): S67-77.

31. Hirata J, Small R, Swiersz L, Ettinger B, Zell B. Does dong quai have estrogenic effects in postmenopausal women? A double-blind, placebo-controlled trial. Fertil Steril 1997; 68: 981-6.

32. Chenoy R, Hussain S, Tayob Y, O'Brien P, Moss M, Morse P. Effect of oral gamolenic acid from evening primrose oil on menopausal flushing. Br Med J 1994; 308: 501-3.

33. Wiklund I, Mattsson L, Lindgren R, Limoni C. Effects of a standardized ginseng extract on quality of life and physiological parameters in symptomatic postmenopausal women: a double-blind, placebo-controlled trial. Int J Clin Pharmacol Res 1999; 19: 89-99.

34. De Leo V, Lanzetta D, Cazzavacca R, Morgante G. Treatment of menopausal symptoms with non hormonal drug therapy. Minerva Ginecol 1998; 50: 207-11.

35. Bender D, Bender A. Nutrition: A Reference Handbook. New York: Oxford University Press, 1997; 131-3.

36. Fan Y, Chapkin R. Importance of dietary [gamma] linolenic acid in human health and nutrition. J Nutr 1998; 128: 1411-14.

37. Galperin C, German C, Gershwin E. Nutrition and diet in rheumatic diseases. In: Shils M, Olson J, Shike M, Ross A, eds. Modern Nutrition in Health and Disease. Baltimore: Williams and Wilkins. 1999; 1339-51.

38. Belch J, Hill A. Evening primrose oil and borage oil in rheumatologic conditions. Am J Clin Nutr 2000; 71 (Suppl): 352S-6S.

39. Punnonen R, Lukola A. Oestrogen-like effect of ginseng. Br Med J 1980; 281: 1110.

40. Merck Index. Merck Index, 13th edn. Whitehouse Station, NJ: Merck & Co, Inc., 2001.

41. Fetrow C, Avila J. Professionals Handbook of Complementary and Alternative Medicines, 2nd edn. Springhouse, PA: Springhouse Corporation, 2001; 272-4.

42. Steingold K, Laufer L, Chetkowski R et al. Treatment of hot flushes with transdermal estradiol administration. J Clin Endocrinol Metab 1985; 61: 627-32.

43. Franke A, Custer L, Cerna C, Narala K. Quantitation of phytoestrogens in legumes by HPLC. J Agrie Food Chem 1994; 42: 1905-13.

44. Mazur W, Duke J, Rasku S, Adlercreutz H. Isoflavonoids and lignans in legumes; nutritional and health aspects in humans. J Nutr Biochem 1998; 9: 193-200.

45. Eldridge A. Determination of isoflavones in soybean flours, protein concentrates, and isolates. J Agric Food Chem 1982; 30: 353-5.

46. Dornstauder E, Jisa E, Unterrieder I, Krenn L, Kubelka W, Jungbauer A. Estrogenic activity of two standardized red clover extracts (Menoflavon) intended for large scale use in hormone replacement therapy. J Steroid Biochem Mol Biol 2001; 78: 67-75.

47. Kuiper G, Carlsson B, Grandien K et al. Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinology 1997; 138: 863-70.

48. Kuiper G, Lemmen J, Carlsson B et al. Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor beta. Endocrinology 1998; 139: 4252-63.

49. Miksicek R. Interaction of naturally occurring nonsteroidal estrogens with expressed recombinant human estrogen receptor. J Steroid Biochem Mol Biol 1994; 49: 153-60.

50. Baber R, Templeman C, Morton T, Kelly G, West L. Randomized placebo-controlled trial of an isoflavone supplement and menopausal symptoms in women. Climacteric 1999; 2: 85-92.

51. Knight D, Howes J, Eden J. The effect of Promensil, an isoflavone extract, on menopausal symptoms. Climacteric 1999; 2: 79-84.

52. van de Weijer P, Barentsen R. Isoflavones from red clover (Promensil) significantly reduce menopausal hot flush symptoms compared with placebo. Maturitas 2002; 42: 187-93.

53. Tice J, Ettinger B, Ensrud K, Wallace R, Blackwell T, Cummings S. Phytoestrogen supplements for the treatment of hot flashes: the Isoflavone Clover Extract (ICE) study. J Am Med Assoc 2003; 290: 207-14.

54. Albertazzi P, Pansini F, Bonaccorsi G, Zanotti L, Forini E, De Aloysio D. The effect of dietary soy supplementation on hot flushes. Obstet Gynecol 1998; 91: 6-11.

55. Washburn S, Burke G, Morgan T, Anthony M. Effect of soy protein supplementation on serum lipoproteins, blood pressure, and menopausal symptoms in perimenopausal women. Menopause 1999; 6: 7-13.

56. Kotsopoulos D, Dalais F, Liang Y, McGrath B, Teede H. The effects of soy protein containing phytoestrogens on menopausal symptoms in postmenopausal women. Climacteric 2000; 3: 161-7.

57. Upmalis D, Lobo R, Bradley L, Warren M, Cone F, Lamia C. Vasomotor symptom relief by soy isoflavone exctract tablets in postmenopausal women: a multicenter, double-blind, randomized, placebo-controlled study. Menopause 2000; 7: 236-42.

58. Faure E, Chantre P, Mares P. Effects of a standardized soy extract on hot flushes: a multicenter, double-blind, randomized, placebo-controlled study. Menopause 2002; 9: 329-34.

59. Scambia G, Mango D, Signorile P et al. Clinical effects of a standardized soy extract in postmenopausal women: a pilot study. Menopause 2000; 7: 105-11.

60. St Germain A, Peterson C, Robinson J, Alekel D. Isoflavone-rich or isoflavone-poor soy protein does not reduce menopausal symptoms during 24 weeks of treatment. Menopause 2001; 8: 17-26.

61. Burke G, Legault C, Anthony M et al. Soy protein and isoflavone effects on vasomotor symptoms in peri- and postmenopausal women: the Soy Estrogen Alternative Study. Menopause 2003; 10: 147-53.

62. Penotti M, Fabio E, Modena A, Rinaldi M, Omodei U, Vigano P. Effect of soy-derived isoflavones on hot flushes, endometrial thickness, and the pulsatility index of the uterine and cerebral arteries. Fertil Steril 2003; 79: 1112-17.

63. Kim J, Kwon C. Estimated dietary isoflavone intake of Korean population based on National Nutrition Survey. Nutr Res 2001; 21: 947-53.

64. Somekawa N, Chiguchi M, Ishibashi T, Aso P. Soy intake related to menopausal symptoms, serum lipids, and bone mineral density in postmenopausal Japanese Women. Obstet Gynaecol 2001; 97: 109-15.

65. Arai Y, Uehara M, Sato Y et al. Comparison of isoflavones among dietary intake, plasma concentration and urinary excretion for accurate estimation of phytoestrogen intake. J Epidemiol 2000; 10: 127-35.

66. Setchell K, Brown N, Desai P, Zimmer-Nechemias L. Bioavailability of pure isoflavones in healthy humans and analysis of commercial soy isoflavone supplements. J Nutr 2001; 131: 1362S-75S.

67. Duncan A, Merz-Demlow B, Xu X, Phipps W, Kurzer M. Premenopausal equol excretors show plasma profiles associated with lowered risk of breast cancer. Cancer Epidemiol Biomarkers Prev 2000; 9: 581-6.

68. Blakesmith S, Lyons-Wall P, Joannou G, Petocz P, Samman S. Urinary isoflavonoid excretion is inversely associated with the ratio of protein to dietary fibre intake in young women. Eur J Clin Nutr 2005; 59: 284-90.

69. Sheehan D. Herbal medicines, phytoestrogens and toxicity: risk: benefit considerations. Proc Soc Exp Biol Med 1998; 217: 379-85.

70. Fugh-Berman A. Herb-drug interactions. Lancet 2000; 355: 134-8.

71. Page R, Lawrence J. Potentiation of warfarin by dong quai. Pharmacotherapy 1999; 19: 870-76.

School of Public Health, Queensland University of Technology, Brisbane

K. Hanna, BHealthSci (NutrDiet)

A. Day, BMedSci, MNutrDiet

C. Patterson, PhD, MSc, GradDipBusAdmin

P. Lyons-Wall, PhD, GradDipNutrDiet

Betty Byrne Henderson Centre, Royal Brisbane and Women's Hospital and The University of Queensland, Brisbane

S. O'Neill, MBBCh, BAO

Correspondence: P. Lyons-Wall, School of Public Health, Queensland University of Technology, Victoria Park Road, Kelvin Grove, Qld 4059, Australia. Email: [e-mail]

Table 1. Clinical intervention studies with black cohosh for treatment
of acute menopausal symptoms

Study          Subjects                      Intervention

Daiber (24)    36 German women with          Uncontrolled trial,
               climacteric symptoms, 45-62   parallel: Remifemin twice/
               years                         day
Vorberg (25)   50 German women with          Uncontrolled trial,
               climacteric symptoms, 45-60   parallel: Remifemin 40
               years                         drops twice/day (liquid
                                             extract). No control group
Warnecke (26)  60 German women with          Uncontrolled trial,
               climacteric symptoms, 45-60   parallel:
               years                         1. Remifemin 40 drops
                                             twice/day (liquid extract)
                                             2. Conjugated oestrogen
                                             0.625 mg/day
                                             3. Diazepam 2 mg/day
Stoll (27)     80 German women with          Placebo-controlled trial,
               [greater than or equal to]3   randomised double-blind,
               hot flushes/day and           parallel:
               psychological symptoms (e.g.  1. Remifemin 8 mg/day (4
               anxiety, vaginal symptoms),   tablets/day)
               46-58 years. Power estimate   2. Conjugated oestrogen
               of sample size not reported   0.625 mg/day
                                             3. Placebo
Lehmann-       60 German women,              Uncontrolled trial,
Willenbrock    hysterectomised with at       randomised, parallel:
and            least one intact ovary and    1. Oestriol 1 mg/day
Riedel (28)    climacteric symptoms, <40     2. Oestrogen 1.25 mg/day
               years                         3. Oestrogen/gestagen
                                             4. Remifemin 8 mg/day (4
                                             tablets/day)
Liske          150 Polish perimenopausal     Uncontrolled trial,
et al. (29)    and postmenopausal women      randomised double-blind,
               with KMI score                parallel:
               [greater than or equal to]20  1. Remifemin standard dose
               (moderate severity), 42-60    39 mg extract/day (no. of
               years. Sample size            tablets not stated)
               calculated to detect          2. Remifemin high dose 127
               standardised difference of    mg extract/day
               0.5 units, [alpha] = 0.05.
               Power not stated
Wuttke         62 German women with =3 hot   Placebo-controlled trial,
et al. (30)    flushes/day, 40-60 years.     randomised, double-blind,
               Power estimate of sample      parallel:
               size not reported             1. Black cohosh ((a)) 40 mg
                                             extract/day (2 capsules/
                                             day)
                                             2. Conjugated oestrogen 0.6
                                             mg/day
                                             3. Placebo

               Duration
Study          (weeks)      Results

Daiber (24)    12           * [down arrow] KMI (P < 0.001)
Vorberg (25)   12           Remifemin at 12 weeks compared with
                            baseline:
                            * [down arrow] KMI (P < 0.001)
Warnecke (26)  12           Remifemin at 12 weeks compared with
                            baseline:
                            * [down arrow] KMI
Stoll (27)     12           Remifemin at 4, 8 and 12 weeks compared with
                            oestrogen or placebo:
                            * [down arrow] KMI (P < 0.001)
Lehmann-       24           All groups at 8, 12 and 24 weeks compared
Willenbrock                 with baseline:
and                         * [down arrow] KMI (P < 0.01)
Riedel (28)
Liske          12           Standard or high-dose Remifemin at 12 and 24
et al. (29)    extended to  weeks compared with baseline:
               24           * [down arrow] KMI (change from moderate to
                            normal range symptoms) in 70-72% of subjects
Wuttke         12           Black cohosh compared with placebo at 12
et al. (30)                 weeks:
                            * [down arrow] MRS (P = 0.05)
                            * NS change in hot flush scale of MRS

Study          Comments

Daiber (24)    Mean KMI improved from moderate-severe to light-moderate
               symptoms at 12 weeks
Vorberg (25)   Mean KMI improved from moderate to light symptoms at 12
               weeks
Warnecke (26)  P-values not cited
Stoll (27)     [down arrow] 48-59% in KMI scales with Remifemin,
               [down arrow] 24-29% with placebo and low-dose oestrogen,
               compared with baseline. Oestrogen dose was too low for
               efficacy. 86% completed trial
Lehmann-       [down arrow] 43% in KMI with Remifemin, [down arrow]
Willenbrock    47-62% with oestrogen groups, compared with baseline
and
Riedel (28)
Liske          No difference between standard and high dose. NS
et al. (29)    side-effects at either dose of Remifemin. 82% completed
               12-week trial. 77% completed 24-week trial
Wuttke         [down arrow] 1.8 MRS points with black cohosh versus
et al. (30)    [down arrow] 1.5 MRS points with placebo, compared with
               baseline. Specific values not presented. 97% completed
               trial

(a) Cimicifuga racemosa preparation CR BNO 1055 (Klimadynon/Menofem).
KMI = Kupperman Menopause Index; MRS = Menopause Rating Scale; NS =
non-significant.

Table 2. Clinical intervention studies with dong quai, evening primrose
oil, ginseng and sage for treatment of acute menopausal symptoms

Study        Subjects                     Intervention

Hirata       71 US postmenopausal women   Placebo-controlled trial,
et al. (31)  with >14 hot flushes any     randomised, double-blind,
             severity/week or >5 hot      parallel:
             flushes moderate to          1. Dong quai root 1.5 g (3
             severe/week, mean age;       capsules/day)
             placebo group, 52.6          2. Placebo
             [+ or -] 6 years; treatment
             group, 52.2 [+ or -] 4
             years. Sample size
             calculated to detect
             difference of 11 vasomotor
             episodes/week, 90% power,
             [alpha] = 0.05
Chenoy       56 UK menopausal women with  Placebo-controlled trial,
et al. (32)  [greater than or equal to]3  randomised, double-blind,
             hot flushes/day, 45-67       parallel:
             years. Sample size to        1. Evening primrose oil 500 mg
             detect 13% difference in     /day + Vit E (8 capsules/day)
             hot flush frequency, 90%     2. Placebo
             power, [alpha] = 0.05

Wiklund      384 Swedish postmenopausal   Placebo-controlled trial,
et al. (33)  women with =6 hot flushes,   randomised, double-blind:
             at least 3/7 days, 45-65     1. 100 mg ginseng extract (2
             years. Sample size to        capsules/day)
             detect difference of 5       2. Placebo
             units in Patient Generated
             Wellbeing Index, 80% power,
             [alpha] = 0.05
De Leo       30 Italian postmenopausal    Placebo-controlled trial:
et al. (34)  women, with                  1. 120 mg sage + 60 mg alfalfa
             [greater than or equal to]   n = 30 (tablet no. not stated)
             10 hot flushes/day, 46-52    2. Placebo, n = 12
             years. Power estimate of
             sample size not reported

Study        Duration  Results             Comments

Hirata       24 weeks  Dong quai compared  [down arrow] 35% in hot flush
et al. (31)            with placebo at 24  frequency with treatment,
                       weeks:              [down arrow] 19% in hot flush
                       * NS change in hot  frequency with placebo,
                       flush frequency     compared with baseline. 86%
                       * NS change in      completed trial
                       Kupperman
                       Menopausal Index
Chenoy       24 weeks  Evening primrose    [down arrow] 11% in day
et al. (32)            oil compared with   frequency and [down arrow]
                       placebo at 24       19% in night frequency in
                       weeks:              treatment versus [down arrow]
                       * NS change in hot  37% in day frequency and
                       flush frequency     [down arrow] 29% in night
                       or intensity        frequency in placebo,
                                           compared with baseline.
                                           63% completed trial
Wiklund      2 weeks   Ginseng group       [down arrow] 30% in hot flush
et al. (33)  run in    compared with       frequency with treatment,
             16 weeks  placebo at 16       [down arrow] 24% in placebo,
                       weeks:              compared with baseline.
                       * NS change in hot  99% completed trial
                       flush frequency
De Leo       3 months  Sage and alfalfa    [down arrow] 85% in hot flush
et al. (34)            compared with       frequency with treatment,
                       placebo at 3        [down arrow] 25% in placebo,
                       months:             compared with baseline
                       * [down arrow] 60%
                       in hot flush
                       frequency (P <
                       0.01). Sage and
                       alfalfa compared
                       with baseline at 3
                       months

NS = non-significant.

Table 3. Clinical intervention studies with phytoestrogens for treatment
of acute menopausal symptoms

Study           Subjects                     Intervention

Red clover
Baber           51 Australian                Placebo-controlled trial,
et al. (50)     postmenopausal women,        randomised, double-blind,
                [greater than or equal to]   crossover:
                3 hot flushes/day, 45-65     1. Promensil 40 mg
                years. Power estimate of     isoflavonoids (1 tablet/
                sample size not reported     day)
                                             2. Placebo
Knight          37 Australian                Placebo-controlled trial,
et al. (51)     postmenopausal women with    randomised, double-blind,
                [greater than or equal to]   parallel:
                3 hot flushes/day, 40-65     1. Promensil 40 mg
                years. Power estimate of     isoflavonoids/day (1 tablet
                sample size not reported     /day)
                                             2. Promensil 160 mg
                                             isoflavonoids/day (4
                                             tablets/day)
                                             3. Placebo
Van de          30 Dutch postmenopausal      Placebo-controlled trial,
Weijer and      women with >5 hot            randomised, double-blind,
Barentsen (52)  flushes/day, mean age:       parallel:
                52.5 [+ or -] 5.2 years      1. Promensil isoflavonoids
                (placebo), 54.2 [+ or -]     80 mg/day (2 tablets/day)
                7.4 years (treatment).       2. Placebo
                Power-based estimate of
                sample size not reported
Tice            252 US menopausal women,     Placebo-controlled trial,
et al. (53)     [greater than or equal to]   randomised double-blind,
                35 hot flushes/week, 45-60   parallel:
                years. Sample size to        1. Promensil ((a))
                detect 15% decrease in hot   isoflavonoids 82 mg/day (2
                flush frequency, 90% power,  tablets/day)
                [alpha] = 0.05               2. Rimostil ((a))
                                             isoflavonoids 57 mg/day (2
                                             tablets/day)
                                             3. Placebo

Soy isoflavonoids and soy protein
Albertazzi      104 Italian                  Placebo-controlled trial,
et al. (54)     postmenopausal women,        randomised, double-blind,
                [greater than or equal to]   parallel:
                7 hot flushes/day, 48-61     1. Soy protein ~76 mg
                years. Sample size to        aglycone isoflavonoids
                detect difference of 3 hot   (powder)
                flushes/day, 90% power,      2. Placebo
                [alpha] = 0.05
Washburn        51 US perimenopausal women,  Placebo-controlled trial,
et al. (55)     [greater than or equal to]   randomised, double-blind,
                1 hot flush/night sweats/    crossover:
                day, 45-55 years. Power      1. Soy protein 34 mg
                estimate of sample size not  isoflavonoids single dose
                reported                     (powder)
                                             2. Soy protein 34 mg
                                             isoflavonoids, split dose
                                             3. Placebo
Kotsopoulos     94 Australian                Placebo-controlled trial,
et al. (56)     postmenopausal women with    randomised, double-blind,
                mild menopausal symptoms,    parallel:
                hot flush number not         1. Soy isoflavonoids 118 mg
                specified, 50-75 years.      /day (soy powder beverage)
                Power estimate of sample     2. Placebo
                size not reported
Upmalis         177 US postmenopausal        Placebo-controlled trial,
et al. (57)     women,                       randomised, double-blind,
                [greater than or equal to]   parallel:
                5 hot flushes/day, (mean     1. Soy isoflavonoid extract
                age 55 years). Power         50 mg isoflavonoids/day (2
                estimate of sample size      tablets/day)
                not reported                 2. Placebo
Scambia         39 Italian postmenopausal    Placebo-controlled trial,
et al. (59)     women, hot flush number      randomised, double-blind,
                not specified, 29-63 years.  parallel:
                Power estimate of sample     1. Soy isoflavonoids 50 mg/
                size not reported            day (tablet no. not
                                             reported)
                                             2. Placebo Conjugated
                                             equine oestrogens (CEE)
                                             give to all participants
                                             after week 6
St Germain      69 US perimenopausal women,  Placebo-controlled trial,
et al. (60)     [greater than or equal to]   randomised, double-blind,
                10 hot flushes/night         parallel:
                sweats/week. Power estimate  1. Soy protein
                of sample size not reported  isoflavonoids 80.4 mg/day
                                             (aglycone) (powder +
                                             muffin)
                                             2. Soy protein
                                             isoflavonoids 4.4 mg/day
                                             (powder + muffin)
                                             3. Placebo
Faure           75 French                    Placebo-controlled trial,
et al. (58)     postmenopausal women,        randomised, double-blind,
                [greater than or equal to]   parallel:
                7 hot flushes/day, mean age  1. Soy isoflavonoids 70 mg
                53 years. Sample size        glycones (4 capsules/day)
                calculated to detect         2. Placebo
                difference of 3 hot
                flushes/day, 90% power,
                [alpha] = 0.05
Burke           241 US perimenopausal        Placebo-controlled trial,
et al. (61)     women, mild symptoms         randomised, double-blind,
                [greater than or equal to]   parallel:
                1 hot flush/day, 45-55       1. Soy protein 42 mg/day
                years. Power estimate of     isoflavonoids (SP ((b))
                sample size not reported     beverage)
                                             2. Soy protein 58 mg/day
                                             isoflavonoids (SP beverage)
                                             3. Soy protein
                                             [greater than or equal to]
                                             4 mg/day isoflavonoids
Penotti         62 Italian                   Placebo-controlled trial,
et al. (62)     postmenopausal women,        randomised, double-blind,
                [greater than or equal to]   parallel:
                7 hot flushes/day, 49-58     1. Soy isoflavonoids 72 mg/
                years. Power estimate of     day (2 tablets/day)
                sample size not reported     2. Placebo (2 tablets/day)

                Duration
Study           (weeks)     Results

Red clover
Baber           2 X 12 4    Promensil compared with placebo at 12 weeks
et al. (50)     wash out    (combined data):
                            * NS change in Greene Menopause Score
                            * NS change in hot flush frequency
Knight          12          Promensil compared with placebo at 12 weeks:
et al. (51)                 * NS change Greene Menopause Score
                            * NS change in hot flush frequency
Van de          12 plus 4   Promensil compared with placebo at 12 weeks:
Weijer and      run in      * [down arrow] 44% hot flush frequency
Barentsen (52)              (P = 0.02)
                            * NS change in Greene Menopause Score
Tice            12          At 12 weeks:
et al. (53)                 Promensil/Rimostil compared with placebo:
                            * NS change in Greene Menopause Score
                            * NS change in hot flush frequency
                            Subgroup with BMI > 25.1 versus subgroup
                            with BMI <25 kg/[m.sup.2];
                            * [down arrow] 19% in hot flush frequency
                            (Promensil) (P = 0.09)
                            * [down arrow] 23% in hot flush frequency
                            (Rimostil) (P < 0.02)

Soy isoflavonoids and soy protein
Albertazzi      12          Soy protein group compared with placebo at
et al. (54)                 12 weeks:
                            * [down arrow] 15% in hot flush frequency
                            (P = 0.01)
Washburn        6 per       Soy protein split-dose group compared with
et al. (55)     treatment,  placebo:
                18 total    * [down arrow] 23% in hot flush severity
                            (P < 0.05)
                            * [down arrow] 11% in night sweat severity
                            (NS)
                            * [up arrow] 4% in hot flush frequency (NS)
                            * [up arrow] 4% in night sweat frequency
                            (NS)
Kotsopoulos     12          Soy isoflavonoids compared with placebo at
et al. (56)                 week 12:
                            * NS change in menopause questionnaire
                            * NS change in hot flush component of
                            menopausal symptoms questionnaire
Upmalis         12          Soy extract group compared with placebo at
et al. (57)                 12 weeks:
                            * [down arrow] 10% in hot flush severity
                            score (P = 0.01)
                            * [down arrow] 9% in hot flush frequency
                            (trend, P = 0.08)
                            * [down arrow] 28% in night sweat frequency
                            (P = 0.04 at 6 weeks, NS at 12 weeks)
Scambia         12          Soy isoflavonoids compared with placebo at
et al. (59)     treatment   week 6:
                and         * [down arrow] 20% in hot flush frequency
                placebo     (P < 0.01) Soy isoflavonoids compared with
                given to    placebo at week 12 (both groups taking CEE):
                week 10     * NS change hot flush frequency
                only
St Germain      24          High isoflavonoid soy compared with
et al. (60)                 placebo at 24 weeks
                            * NS change in hot flush frequency
                            * NS change in night sweat frequency
Faure           16          Soy isoflavonoids compared with placebo at
et al. (58)                 16 weeks:
                            * [down arrow] 40% in hot flush frequency
                            (P = 0.01)
Burke           104         Soy isoflavonoids compared with placebo at
et al. (61)                 12 or 104 weeks:
                            * NS change in hot flush frequency
Penotti         24          Soy isoflavonoids compared with placebo at
et al. (62)                 12 or 24 weeks:
                            * NS change in hot flush frequency

Study           Comments

Red clover
Baber           [down arrow] 22% in hot flush frequency with Promensil,
et al. (50)     [down arrow] 32% with placebo, compared with baseline.
                84% completed trial
Knight          [down arrow] 29% in hot flush frequency with 40 mg dose,
et al. (51)     [down arrow] 34% with 160 mg dose, [down arrow] 33% with
                placebo, compared with baseline. 95% completed trial
Van de          [down arrow] 44% in hot flush frequency with Promensil,
Weijer and      [down arrow] 0% with placebo, compared with baseline.
Barentsen (52)  87% completed trial
Tice            Hot flush frequency compared with baseline: whole group:
et al. (53)     [down arrow] 41% with Promensil, [down arrow] 34% with
                Rimostil, [down arrow] 36% with placebo. Subgroup BMI >
                25.1 versus BMI <25 kg/[m.sup.2]:
                Promensil: [down arrow] 49% versus [down arrow] 30%;
                Rimostil: [down arrow] 45% versus [down arrow] 22%;
                Placebo: [down arrow] 32% versus [down arrow] 40%.
                98% completed trial

Soy isoflavonoids and soy protein
Albertazzi      [down arrow] 45% in hot flush frequency with soy
et al. (54)     protein, [down arrow] 30% with placebo, compared with
                baseline. 76% completed trial
Washburn        Original scores compared with baseline not reported. No
et al. (55)     significant changes in soy protein single-dose group,
                compared with placebo. 82% completed trial
Kotsopoulos     Hot flush scores not reported. 80% completed trial
et al. (56)
Upmalis         [down arrow] 28% in hot flush severity with soy versus
et al. (57)     [down arrow] 18% in placebo; [down arrow] 28% hot flush
                frequency with soy, [down arrow] 19% in placebo;
                [down arrow] 62% in night sweat frequency with soy,
                [down arrow] 34% in placebo, compared with baseline. 69%
                completed trial
Scambia         [down arrow] 45% hot flush frequency with isoflavonoids,
et al. (59)     [down arrow] 25% with placebo, compared with baseline.
                Withdrawal rate not specified
St Germain      [down arrow] 45% in hot flush frequency with 80.4 mg
et al. (60)     isoflavonoids, [up arrow] 3% with 4.4 mg isoflavonoids,
                [down arrow] 45% with placebo, compared with baseline.
                99% completed trial
Faure           [down arrow] 61% in hot flush frequency with soy
et al. (58)     isoflavonoids, [down arrow] 21% with placebo, compared
                with baseline. 84% completed trial in soy group. 60%
                completed trial in placebo group
Burke           Hot flush frequency at 12 or 104 weeks compared with
et al. (61)     baseline, in whole group:
                1. [down arrow] 19/42%, (42 mg isoflavonoids)
                2. [down arrow] 50/59% (58 mg isoflavonoids)
                3. [down arrow] 51/77% with placebo.
                88% completed trial
Penotti         Hot flush frequency at 12 or 24 weeks, compared with
et al. (62)     baseline:
                1. [down arrow] 44/53% with soy isoflavonoids
                2. [down arrow] 53/53% with placebo.
                79% completed study

(a) Promensil contains a higher proportion of biochanin A and genistein;
Rimostil contains a higher proportion of formononetin and daidzein.
(b) Soy protein.
BMI = body mass index; NS = non-significant.