Isoflavone effects against menopausal complaints

Whereas 70 to 80 percent of European women complain about menopausal symptoms (mainly hot flushes, but also osteoporosis), the prevalence of such effects is only 10 to 20 percent in Asian countries. This observation was correlated with the higher dietary intake of soy products in Asia, especially with isoflavones (Ikeda et al. 2006; Kim et al. 2006; McCarty 2006; Messina et al. 2006a; Messina et al. 2006b; Nagata et al. 1999; Nagata et al. 2001; Qin et al. 2006).

In accordance with the most recent results from research and a German expert opinion the International Menopause Society has only recently given a recommendation for the application of isoflavones as “an efficacious alternative and addition to the treatment of symptoms of menopause and premenopause” (Anon 2007; Schindler 2006). The effects of isoflavones against vasomotor menopausal symptoms (hot flushes and profuse sweating) have been observed in a multitude of clinical trials, and have been confirmed in reviews and metaanalyses (Howes et al. 2006; Kurzer 2008; Kurzer 2009; Messina and Hughes 2003; Nelson et al. 2006; Williamson-Hughes et al. 2006). Others, among them the German BfR, doubt the effect by pointing to the heterogeneous data in publications.

The assessment of metaanalyses helps in the clarification of seemingly contradictory findings between publications. E.g., Nelson et al. (2006) only came to a verdict on isoflavones when studies were included into the analysis where the origin of the hot flushes was not menopausal, but an adverse effect of a cancer treatment with tamoxifen (Nelson et al. 2006). In the debate of the question of efficacy of soy against menopausal disorders one should, however, expect that only studies in this indication are used for comparisons. The mechanism of action of tamoxifen-induced hot flushes is fundamentally different from that of hot flushes in natural menopause. When studies in women with natural hot flushes were examined, the overall effect was suddenly positive: Nelson et al. (2006) found a statistically significant positive signal in favour of a superiority of isoflavones over placebo, the more so in studies with an adequate design!

The heterogeneity of isoflavones studies against menopausal disorders is undoubtedly existent. The reasons for this heterogeneity become clearer when the details of the single studies are analyzed:

a) Lack of statistical power by insufficient case numbers
An effect on number and severity of hot flushes was observed in almost all studies, but could not always be statistically secured. Some studies only showed a statistically non-significant trend towards an improvement of symptoms (Campagnoli et al. 2005; Khaodhiar et al. 2008; Knight et al. 2001; Murkies et al. 1995; Penotti et al. 2003). The placebo effect was obviously underestimated in these studies. The majority of the studies with insufficient statistical power do not prove the absence of the effect, but are rather unable to demonstrate a significant difference between isoflavones and placebo. Obviously there is leeway for misunderstandings: biological effects with a high natural variability (including hot flushes) always require larger study populations than studies with more focused effects, as otherwise there is the danger of distorting the result by single out-liers. Correspondingly studies performed with insufficient populations and not showing a difference in comparison with placebo cannot be interpreted as having demonstrated the absence of the effect.

A typical example of this phenomenon is the study of Khaodiar et al. (2008): In this study two doses of soy isoflavones (40 and 60 mg per day) were compared with placebo. With 40 mg (48 women) a reduction of symptoms by 52 % was found, with 60 mg (49 women) a reduction by 51 %. With placebo a reduction by 39 % was observed. Due to the high variability of the single values the difference to placebo did in both cases not reach statistical significance (p = 0.07 and 0.09, for 40 and 60 mg, respectively). On pooling of the two soy groups, however, the difference to placebo was suddenly statistically significant for the total group of 97 women (Khaodhiar et al. 2008). Although the original intention of the study could not be reached for formal reasons, the study still demonstrates the existence of the benefit of soy isoflavones in menopausal complaints.

b) Insufficient degree of complaints
In studies of psychosomatic complaints the severity of symptoms at baseline is at least as important as the size of the study population. This is also the case for menopausal complaints. By experience, the higher the severity of symptoms at baseline the higher the overall effect found in controlled trials. This is reflected in the recommendations of the FDA for the design of trials in menopause (FDA 2003). These recommendations call for a minimum of seven hot flushes per day as an inclusion criterion of the study. In some of the soy/isoflavone studies the average number of daily hot flushes was two or even less (Kotsopoulos et al. 2000; St Germain et al. 2001). With such a low grade of symptoms the placebo response must be expected to increase over-proportionally, and the study population would have to be adjusted correspondingly.

c) Type of soy preparation
Very different soy preparations were applied in the studies, ranging from simple dietary recommendations to the supplementation with unfermented soy flour, soy protein and soy drinks, to extracts and even to isolated genistein or equol. Quite obviously the type of preparation has an impact on the study results. Isoflavones are intensively metabolized by the intestinal flora, provided they are adequately liberated from their food matrix. According to the type of preparation large differences in bioavailability of isoflavones and lignans must be expected (Cassidy et al. 2006; Nielsen and Williamson 2007). Such differences should also be reflected in the degree of the potential effect. Differences in bioavailability (next to other factors) may have contributed to the finding of a stronger effect in the control group (wheat flour and isoflavone-depleted soy protein, respectively) than determined in the soy group receiving the isoflavones in the form of unprocessed soy flour respectively soy protein (Burke et al. 2003; Dalais et al. 1998). The impact of the taste of the preparation must not be neglected. Unprocessed soy preparations tend to have a very bad taste, which should have a negative influence on compliance and the attitude of the patient towards the individual treatment.

Among 18 studies performed with soy extracts or soy food adjusted to a defined content of isoflavones 10 yielded a positive study result (Albert et al. 2002; Cheng et al. 2007; Colacurci et al. 2004; Drapier Faure et al. 2002; Han et al. 2002; Kaari et al. 2006; Nahas et al. 2007; Petri Nahas et al. 2004; Scambia et al. 2000; Uesugi et al. 2004). In six studies no difference between groups was found (Burke et al. 2003; Campagnoli et al. 2005; Knight et al. 2001; Penotti et al. 2003; Verhoeven et al. 2007). One further study with soy extract showed a statistically significant effect on the frequency of hot flushes only after six, but not after 12 weeks (Upmalis et al. 2000). The severity of hot flushes was, however, found significantly improved after 12 weeks.

d) Glycosides vs. aglycones
Some studies did not indicate the quantity of isoflavones in the verum arm of the study (Brzezinski et al. 1997; Murkies et al. 1995), in other studies a differentiation between glycosides and aglycones was not possible (Cheng et al. 2007; Colacurci et al. 2004; Khaodhiar et al. 2008; Nahas et al. 2007; Petri Nahas et al. 2004; Scambia et al. 2000; Uesugi et al. 2004; Washburn et al. 1999). It can be expected that in the more recent studies the dosage statements in the studies referred to aglycones. Some positive studies explicitly mention the calculation of the daily dose of isoflavones as aglycones (Albertazzi et al. 1998; Welty et al. 2007). The correct differentiation between the two forms, glycosides and aglycones, is important, as the carbohydrate-free part of the isoflavone molecule accounts for approximately 60 % of the weight of the glycosides. The difference in weight may therefore contribute to a gross overestimation of exposure to isoflavones, and to an underestimation of the effect size.

Individual studies on the effects of soy preparations against menopausal complaints can hardly be compared. The effect was, however, positively demonstrated for soy food, soy extracts and pure genistein and equol. As the robustness of the effect has been confirmed in metaanalyses, the existence of the effect can no longer be doubted. Preparations containing isoflavones, including soy extracts used in supplements, have the potential for the reduction of hot flushes and profuse sweating (Kurzer 2008).

An as yet unpublished trial of the Isoflavone Research Initiative was presented in conferences (Gocan et al. 2007; Imhof et al. 2008). Over a period of 12 weeks, 192 menopausal women were examined for the frequency and severity of hot flushes. The double-blind study part was followed by 12 weeks of open post-observation with all study participants taking the soy extract. Severity of symptoms was assessed by the Greene Climacteric Scale. In addition safety parameters were measured, such as hormonal levels, the impact on vaginal endothelium (test for proliferation-enhancing effects), liver function tests and thyroidal gland parameters. The inclusion criterion was a minimum number of seven hot flushed upon inclusion into the trial.

In comparison with placebo a statistically significant reduction of the frequency and severity of hot flushes was found, next to a significant improvement of 18 out of 21 single symptoms of the Greene Climacteric Scale. According to these results isoflavones have only little effects on somatic menopausal complaints (feeling of heaviness of limbs, paraesthesias and shortness of breath). An effect on these symptoms had, however, not been expected. None of the safety parameters pointed to undesired effects or an impact on hormonal levels and the function of the thyroidal gland.

Overall, the results from the clinical trials demonstrate efficacy and safety of soy preparations for the application in menopausal women.

Download:

Table: Clinical trials with isoflavone-containing preparations for menopausal complaints

References

Isoflavone Research Initiative

Information

Subscribe to Isoflavone News