10 September, 2012:
A recent paper by Kroenke and colleagues discussed the dietary intervention component of the Women's Health Initiative . The intervention was a reduction in fat intake whilst increasing that of fruit, vegetables and whole grains. The outcome was the impact of this, whether accompanied by weight loss or not, on vasomotor symptoms (VMS) in postmenopausal women. A total of 17,500 postmenopausal women aged 50–79 years were recruited into this arm, none of whom were meant to be taking menopausal hormone replacement therapy (HRT). The study found that there was a significant weight loss in the dietary intervention arm that was associated with elimination of symptoms among women who had vasomotor symptoms at baseline (odds ratio (OR) 1.14; 95% confidence interval (CI) 1.01–1.28), who lost more than 10 lbs (OR 1.23; 95% CI 1.05–1.46) or lost 10% or more of their baseline body weight (OR 1.56; 95% CI 1.21–2.02) between baseline and year 1. These groups were significantly more likely to cease having VMS, compared with those who maintained weight. Finally, women who lost substantial weight as a part of the intervention group (OR 1.89; 95% CI 1.39–2.57), but not as part of the control group (OR 1.40; 95% CI 0.92–2.13), were significantly more likely to stop having VMS. Large weight loss (> 22 lbs), but not dietary changes, was related to the elimination of moderate/severe VMS. The conclusion to the study was that weight loss, as part of healthy diet modification, may help eliminate vasomotor symptoms amongst postmenopausal women.
Women were randomized to a diet with 20% of energy derived from fat and five servings of fruit and vegetables daily as well as six of whole grain. They were, in addition, given an intensive behavioral modification program to assist them in achieving this diet. The control group received a copy of a publication 'Dietary guidelines for Americans' but had no direct contact with the nutrition interventionists. Although weight loss was not a primary end to the diet modification, 21% of the intervention group and 7% of the controls lost weight. Weight change was assessed looking at absolute values as well as a percentage change in weight to allow categorization. VMS were reported using a questionnaire assessing hot flush and night sweat occurrence and severity in the previous 4 weeks that was scored from none (0) to severe (3). A score of 1 indicated that there was no interference with the usual activities, 2 that they interfered somewhat, and 3, interfered a lot. Covariates such as demographic factors, years since hysterectomy and depressive symptoms were taken into account.
A total of 6104 women had VMS at baseline and were examined separately; 26% of women reported hot flushes at baseline with only 1% being severe, i.e. 61 women. These women, not surprisingly, were younger and more recently menopausal as one would expect and the VMS were more likely to occur in African-American women, those who were less educated, smoked and with depressive symptoms as well as lower alcohol intake. No comment was made on whether the women had taken HRT in the past or not. The intervention led to an improvement in symptoms, although it is very difficult to take account of natural history since it is known that the women with symptoms were younger and some would be expected to improve anyway. Not all VMS, particularly in older women, respond well to hormone intervention.
The importance of VMS is often underestimated in terms of the impact on quality of life as well as the length of time for which they may last [2-4]. Recent work has suggested that women who are overweight and obese are more likely to have hot flushes  and this may be related to the insulating effect of the fat itself. Interestingly, there has been little study of the impact of weight loss, partly due to the fact that this is so difficult to achieve in a short period of time. Women tend to gain weight with age  and also fat distribution changes [7-9]. They need to be encouraged to watch their diet carefully and therefore symptom relief may add an additional incentive to do this.
In all, nearly 49,000 were recruited into the WHI dietary modification trial. Half were randomly assigned to dietary intervention and half to the control. These women were all postmenopausal, as has been discussed frequently over recent years, and most did not have severe symptoms. Had more women had symptoms, it would inevitably have led to unblinding of the study, and also those with severe symptoms would normally opt to have a treatment for them. Of the 48,835 trial participants, 8050 participated in the hormone trial and 21,335 reported baseline intake of hormone therapy and were therefore excluded. In addition, a further 2271 were excluded because of missing data on VMS or weight at baseline or at year 1; therefore the final sample size was 17,473.
The study noted that women who gained weight were more likely to have flushing at the end of 1 year, compared with those who maintained weight. However, for women with large weight gain, this was not statistically significant, although it was for those with small weight gain. Possibly, this may be simply due to numbers. What is more than a little confusing is that the women in the intervention group who gained 10 1bs or more were also more likely to eliminate symptoms and this was statistically significant. Women in the control group who either lost or gained weight were not more likely than those who maintained weight to eliminate or reduce symptoms. It is impossible to determine the significance of this finding. Although the baseline characteristics, as determined by VMS status and severity are quoted in this publication, analysis of the data is complex, using a proportional odds logistic model in order to assess effect across the range of severity. Covariates were, in addition, taken into account. However, it is not possible for the clinician with average statistical ability to comment on the statistical methods used.
The discussion does not really take into account the natural history of hot flushing and also gives little information on the weight of the women at the start of the study, except that the mean weight was 79 kg. Little further information on body mass index is included apart from the table which shows distribution of weight. I found this way of expressing data to be not necessarily helpful and, even though significance was included in the table that covers the baseline characteristics of the 17,000 women, it was often a little difficult to determine exactly what this meant. Since the benefits were greatest in those with mild symptoms, it is possible that these are the women who would have got better anyway, and no information is given as to whether these particular women were younger or not than those who did not notice an improvement.
This is a large sample size and the women are reasonably well characterized, although the information on hot flushing is somewhat limited and does not include frequency which is generally considered to be a very important factor. It is unfortunate that the study is so complicated, as this tends to undermine the very important message.
Mary Ann Lumsden
Consultant Gynecologist, Head of Reproductive & Maternal Medicine, University of Glasgow, UK
1. Kroenke CH Caan BJ, Stefanick ML, et al. Effects of a dietary intervention and weight change on vasomotor symptoms in the Women's Health Initiative. Menopause 2012 July 9. Epub ahead of publication. http://www.ncbi.nlm.nih.gov/pubmed/22781782
2. Pines A, Sturdee DW, MacLennan AH. Quality of life and the role of menopausal hormone therapy. Climacteric 2012;15:213-16. http://www.ncbi.nlm.nih.gov/pubmed/22612606
3. Politi MC, Schleinitz MD, Col NF. Revisiting the duration of vasomotor symptoms of menopause: a meta-analysis. J Gen Intern Med 2008;23:1507-13. http://www.ncbi.nlm.nih.gov/pubmed/18521690
4. Thurston RC, Joffe H. Vasomotor symptoms and menopause: findings from the Study of Women's Health across the Nation. Obstet Gynecol 2011;38:489-501. http://www.ncbi.nlm.nih.gov/pubmed/21961716
5. Huang AJ, Subak LL, Wing R, et al. An intensive behavioral weight loss intervention and hot flushes in women. Arch Intern Med 2010;170:1161-7. http://www.ncbi.nlm.nih.gov/pubmed/20625026
6. Sowers M, Zheng H, Tomey K, et al. Changes in body composition in women over 6 years at midlife: ovarian and chronological aging. J Clin Endocrinol Metab 2007;92:895-901. http://www.ncbi.nlm.nih.gov/pubmed/17192296
7. Matthews KA, Abrams B, Crawford S, et al. Body mass index in midlife women: relative influence of menopause, hormone use, and ethnicity. Int J Obes Relat Metab Disord 2001;25:863-73. http://www.ncbi.nlm.nih.gov/pubmed/11439301
8. Ho SC, Wu S, Chan SG, Sham A. Menopausal transition and changes of body composition: a prospective study in Chinese perimenopausal women. Int J Obes 2010;34:1265-74. http://www.ncbi.nlm.nih.gov/pubmed/20195288
9. Sutton-Tyrrell K, Zhao X, Santoro N, et al. Reproductive hormones and obesity: 9 years of observation from the Study of Women's Health Across the Nation (SWAN). Am J Epidemiol 2010;171:1203-13. http://www.ncbi.nlm.nih.gov/pubmed/20427327
Content updated 10 September 2012