Menopause signals the end of child-bearing capacity, and is also associated with changes in susceptibility to various chronic diseases, including breast cancer, heart disease, and osteoporosis.66 Differences in age at menopause between vegetarian and omnivorous women, should they exist, could be associated with differences in chronic disease patterns between these groups. Furthermore, some women experience unpleasant symptoms during menopause (vasomotor symptoms such as night sweats and hot flushes, mood swings, insomnia, weight gain, headaches, and fatigue),67 and these symptoms have been observed to differ among women in different cultures.67,68 Whether dietary variables contribute to these differences in symptom experiences has not been clearly established, but there is speculation that they could.68-70 Some of these dietary differences may also exist between vegetarian and omnivorous women. Accordingly, after defining and describing the menopausal transition, available research on variables associated with age at menopause will be reviewed, and the question of whether age at menopause differs between vegetarians and omnivores will be examined. The effect of plant components on meno-pausal symptoms will also be discussed.
Research on the menopausal transition has been complicated by a lack of consistency in definitions. "Menopause" is not an event that occurs at a single point in time, but instead, is a complex endocrinological transition that occurs over a period of years.71 The World Health Organization72 defines the perimenopause as "the period immediately before the menopause (when the endocrinological, biological, and clinical features of approaching menopause commence) and the first year after menopause." "Menopause" is defined here as the final menstrual period, which can only be retrospectively assessed after 12 months without flow.72 Because of the problems associated with this retrospective identification, and because hormonal changes continue during the year after the last period occurs, menopause has also been defined as beginning when a year without flow has elapsed.71
The entire transition may occur over a period of up to 6 years, and has been described as occurring in five phases.71 In the first and second phases, menstrual cycles are regular but may have ovulatory disturbances, may be shorter than normal for a given woman, and may occur with either normal or increased flow. Premenstrual symptoms increase, with increased breast symptoms, cramping, and headaches. At this time, vasomotor symptoms first appear, usually as early morning night sweats at mid-cycle or before the onset of flow. FSH levels are normal or intermittently elevated, and estradiol levels may be moderately elevated. During the third and fourth phases, cycles become irregular, first with alternating short and long cycles, and then progressing to more frequent periods of oligomenorrhea. Flow may be either very high or reduced, and this often alternates. Premenstrual symptoms decrease, but vasomotor symptoms intensify and occur more often during the daytime. FSH levels are persistently elevated, and estradiol levels alternate between normal and high. Finally, the last phase represents the 12 months after the last menstrual flow. Vasomotor symptoms often increase in frequency and intensity, but may disappear in some women who experienced them earlier. FSH remains high and estradiol levels decrease, but may be intermittently high.
Epidemiological studies have assessed the impact of various factors on age at natural menopause, and it is important to review this literature briefly, especially since the potential exists for some of these variables to confound the assessment of differences between vegetarians and omnivores.
Based on differences between populations in mean age at menarche and menopause, Frisch73 postulated that an early menarche was associated with a later menopause, particularly among well-nourished populations. As discussed earlier, the age at menarche in well-nourished vegetarians appears similar to that of well-nourished omnivores, so differences in age at menopause between these groups are unlikely to be confounded by differences in age at menarche. Furthermore, data obtained from individual women, rather than populations, do not support Frisch's hypothesis. For example, no association between age at menarche and menopause was observed in a prospective study of 3756 Dutch women, categorized according to whether their menopause occurred early or late.74 The mean age at menarche was 13.88 years in 1267 women who experienced an early menopause (mean = 45.0 years), very similar to the mean of 13.94 years in the 1144 women who experienced a late menopause (mean = 54.6 years). Similarly, age at menarche was not associated with age at menopause in Scottish and British women studied cross-sectionally.75 Thus, an individual's age at menarche does not appear to influence her age at menopause.
b. Frequency of Ovulation
The oocyte depletion hypothesis for menopause suggests that differences in the rate of depletion of oocytes lead to differences in ovarian depletion and thus age at menopause.76 Accordingly, factors that prevent ovulation, including oral contraceptive use, pregnancy, and lactation, would lead to a later age at menopause. Supportive evidence for this hypothesis is provided by observations of a later age at menopause in women who have used oral contraceptives,66,74,77 and an earlier age in women who are nulliparous.66,74,75,78 Also consistent with this hypothesis is the finding that women who reported a history of irregular menstrual cycles between the ages of 20 and 35 had a later age at menopause.79
2. Lifestyle Variables a. Tobacco, Alcohol and Caffeine Use
Smoking is consistently associated with an earlier age at menopause, in both retrospective and prospective studies.74,75,78-81 Data on other life-style practices are less consistent. For example, alcohol use was associated with a later age at menopause in a cross-sectional study of almost 1500 45-49-year-old Scottish and British women conducted by Torgerson et al.75 These results were confirmed in a subsequent 2-year prospective follow-up of the 1227 women who were not menopausal during the original study.81 Another 11-year prospective study of Danish women, however, found no association with alcohol use.80 Similarly, consuming coffee was associated with a later age at menopause in a cross-sectional study of 4186 Japanese women,78 but no associations were observed in the prospective Danish study.80
b. Dietary Practices
Few studies have related age at menopause with dietary differences. The cross-sectional study of Torgerson et al.75 found that the Odds Ratio for being postmenopausal was 0.25 (95% CI 0.10-0.65) among women who ate meat or poultry once a day, relative to those who reported never eating meat or poultry (P < 0.01). However, in the follow-up study conducted 2 years later with the women who were initially premenopausal, no association was observed with meat intake. Compared with those who never ate meat, the Odds Ratio for being postmenopausal among those eating meat more than once a day was 1.07 (95% CI 0.94-1.22, P = 0.28).81 In Nagata's cross-sectional study of Japanese women, intakes of soy products were significantly higher among postmenopausal than premenopausal women after controlling for age (P<0.001),78 but this association has not been demonstrated in a prospective study.
Body size does not appear to be a signifi cant determinant of age at menopause. No associations with BMI were detected in the cross-sectional study of Scottish and British women,75 which also assessed self-reported body weight at ages 20, 30, and 40 years. Similarly, BMI did not differ between women who became menopausal or remained premenopausal in the 11-year prospective Danish study80 or in the 7-9-year prospective study of 3756 Dutch women,74 despite considerable variability in BMI among the samples. Although the cross-sectional Japanese study of Nagata et al. reported that mean BMI was significantly lower among postmenopausal women, the absolute difference was small (22.2 vs. 22.5 kg/m2).78 In summary, although some associations with several different types of variables and age at menopause have been observed, in many cases these associations are inconsistent or weak. For example, the large prospective study of Dutch women assessed age at menarche, number of siblings, age at first childbirth, parity, BMI, smoking, oral contraceptive use and SES. For the total sample, the multiple regression model (including all variables) predicted only 2.3% of the variance in age at menopause.74
C. Studies of Age at Menopause in Vegetarians and Omnivores
Surprisingly few data are available comparing age at menopause between vegetarians and omnivores. One study, reported only in abstract form,82 retrospectively compared age at menopause in 80 Seventh-Day Adventist lacto-ovo-vegetarians and 280 omnivores who were not Adventists. Median age at menopause was 48 years among vegetarians and 50 years among omnivores, a significant difference. The difference was not explained by smoking, since no Adventists smoked, by differences in body weight (mean BMI was higher among vegetarians), or by differences in other variables for which data were available. However, the abstract did not state whether oral contraceptive use had differed between groups. The data cited earlier on dietary practices (e.g., cross-sectional studies reporting that the use of meat was associated with later menopause and the use of soy products with earlier menopause) appear to provide weak support for this study's observation, but additional research is clearly warranted.
D. Can Plant Components Replace Hormone Therapy?
At the time of menopause, some women choose to take hormone therapy to reduce the severity of menopausal symptoms such as vaginal dryness, hot flushes, and mood swings.83 Others, however, prefer not to use hormones, and seek dietary or herbal treatments for their symptoms. The use of hormones is also suggested to modulate the risk of several chronic diseases.66 That topic, however, is beyond the scope of this review, which will focus on the effect of plant components on symptoms associated with the menopausal transition.
It has been noted that menopausal symptoms appear to occur less frequently and less intensely in women from countries such as Japan and China.68-70 Because phytoestrogen-rich soy products are dietary staples in these areas, it has been speculated that the estrogenic actions of these compounds may explain the differences in menopausal symptomatology.
This hypothesis has been evaluated in several open trials with varying results. Brzezinski et al.84 randomized symptomatic, early menopausal women to a phytoestrogen-rich diet that provided about 25% of daily calories as tofu, soy drink, miso, and flax seed (n = 78), or to maintain their usual omnivorous diets and avoid soy products and flax seed (n = 36). Measurements conducted at baseline and after 12 weeks on the diet included serum hormone and phytoestrogen levels and an assessment of the severity of menopausal symptoms (vaginal dryness, hot flushes, night sweats, palpitations, headache, depression, urinary discomfort, decreased libido, and insomnia). Serum phytoestrogen concentrations increased dramatically in the treatment group, confirming their adherence to the diet. In this open trial, menopausal symptom scores decreased substantially and similarly in both groups, from 10.65 ± 0.60 to 5.31 ± 0.45 in the treatment group, and from 9.23 ± 0.87 to 4.79 ± 0.71 in the control group. When symptoms were analyzed separately, however, the treatment group had greater decreases in the hot flushes and vaginal dryness scores.
In another open trial, Baird et al.85 compared the responses of women at least 2 years postmenopause to a control (usual) diet (n = 25), and to a soy diet in which approximately 33% of energy was provided as soy products (n = 66). Urinary phytoestrogens were monitored, and the vaginal maturation index was assessed as an index of estrogenicity. After 4 weeks on the diet, urinary phytoestrogen excretion had increased substantially (an average of more than 100-fold) in the soy group, but there was no significant difference in the maturation index between groups (P = 0.40). The authors suggest that more time might have been needed to detect differences.85
A randomized double-blind trial was conducted by Murkies et al.,86 with outcome measurements that included the number of hot flushes, a subjective assessment of menopausal symptoms, and the vaginal maturation index. Postmenopausal women experiencing at least 14 hot flushes per week were randomized to diets supplemented with 45 g of either soy flour (n = 28) or wheat flour (n = 30) per day for a period of 12 weeks. Urinary phytoestrogen measurements substantially increased in women in the soy group. The number of hot flushes and the menopause symptom score improved significantly in both groups, and although the improvements were slightly greater in the soy group, there were no differences between treatment groups (P = 0.82 and P = 0.90 for hot fl ushes and symptom scores, respectively). The vaginal maturation index was unchanged in both groups.
A randomized, double-blind trial was also conducted by Albertazzi et al.87 Postmenopausal women, experiencing at least seven moderate to severe hot flushes per day during the prestudy period, consumed 60 g isolated soy protein (n = 40) or 60 g casein placebo (n = 39) daily for 12 weeks. In addition to recording the number of hot flushes, they completed an index that scored the severity of a variety of menopausal complaints (hot flushes, abnormal sensations such as numbness, prickling or tingling, insomnia, nervousness, depression, balance disturbances, weakness, joint pain, headaches, and palpitations). The number of hot flushes decreased significantly in both groups: at 12 weeks, there was a 45% decrease in the soy group, compared with a 30% reduction in the placebo group. The reduction in hot flushes was significantly greater in the soy group, but no improvements or differences between groups were noted in the meno-pausal index scores.87 Moreover, the decrease in hot flushes did not correlate with the increase in serum or urine phytoestrogen levels.88
A number of trials that may provide more definitive results are currently under way in this area. Presently, however, it appears that there is a substantial placebo response to interventions. Although modest differential effects of soy phytoestrogens may exist for hot flushes, other menopausal symptoms do not appear to be affected.
In addition to phytoestrogens, various herbal preparations have been used by women to relieve menopausal symptoms.89 Although little systematic research has been conducted in North America, in other parts of the world, herbal remedies are more widely recognized. The best available data on safety and efficacy is probably produced by Germany's Commission E, which was established in 1978 to review more than 1400 herbal drugs. Commission E has recognized eight herbal remedies as being effective for various menopausal symptoms: balm, black cohosh, chasteberry, gingko, ginseng, passion flower, St. John's wort and valerian.89 In some cases, however, the Commission placed caveats on the use of these remedies, such as limiting ginseng use to no more than 3 months or black cohosh to no more than 6 months. Other herbal remedies recommended in the lay press for the treatment of menopausal symptoms may be dangerous: for example, scullcap and life root have been found to be hepatotoxic, and dong quai to contain a carcinogen.89 Once again, additional placebo-controlled research is warranted.
Was this article helpful?