9 February, 2015 

Vasomotor symptoms (VMS) are common, but whether VMS are associated with fracture incidence or bone mineral density (BMD) levels is unknown. The following data are derived from the Women's Health Initiative Clinical Trial participants (n= 23,573) aged 50–79 years not using menopausal hormone therapy, and 4867 participants of the BMD sub-study [1]. This was a prospective observational study with mean (standard deviation) follow-up of 8.2 (1.7) years (1993–2005). Baseline VMS, incident adjudicated fractures, and BMD (baseline, annual visits 1, 3, 6, and 9) were measured. The mean age of those with moderate/severe VMS was 60 years, and mean body mass index (BMI) was 30.7 kg/m². After adjustment for baseline age, BMI, race/ethnicity, smoking, and education, the hazard ratio for hip fracture among women with baseline moderate/severe VMS (vs. no VMS) was 1.78 (95% confidence interval, CI 1.20–2.64;p= 0.01). There was no association between VMS and vertebral fracture. VMS severity was inversely associated with BMD during follow-up (p= 0.004 for femoral neck,p= 0.045 for lumbar spine). In repeated measures models, compared with women who reported no VMS, women with moderate/severe VMS had 0.015 g/cm²lower femoral neck BMD (95% CI -0.025 to -0.005) and 0.016 g/cm²lower lumbar spine BMD (95% CI -0.032 to -0.004).

Comment

Although this seems quite strange, but as Crandall and colleagues state and a PubMed search verifies, the associations between VMS and fracture risk have not been frequently investigated. Previous studies examined the potential relationship of VMS and BMD, and indeed most found that women with more severe VMS had lower BMD values. Gast and colleagues analyzed data from a population-based sample of 5,600 women, aged 46–57 years and free from bone diseases, who participated in the first cross-sectional part of the Eindhoven Perimenopausal Osteoporosis Study between 1994 and 1995 [2]. They rated the degree of menopausal symptoms and correlated it with spine BMD results, and demonstrated that, after multivariate adjustments for age, BMI, menopause status, smoking, education, exercise, and hormone use, women with the highest frequency of symptoms had a 0.022 g/cm (95% CI -0.03 to -0.01) lower BMD compared with asymptomatic women. Women who reported having the highest frequency of night sweats had a 0.011 g/cm (95% CI -0.02 to -0.001) lower BMD compared with women with no symptoms of night sweats. Interestingly, the same investigators were able to establish similar associations with coronary artery disease: women with menopausal night sweats had a significantly moderately increased risk which could not be totally explained by the levels of cardiovascular risk factors: multivariable-adjusted hazard ratio 1.33 (95% CI 1.05–1.69), attenuated but not eliminated after correction for BMI, blood pressure, and total cholesterol (hazard ratio 1.25; 95% CI 0.99–1.58) [3]. However, symptoms of flushing were not associated with risk of coronary artery disease.

One of the problems in doing such correlations is the need for very large cohorts and long-term follow-up. If one picks an older population, such as the WHI one, higher fracture rates are expected on the one hand but, on the other hand, the chance of having enough women with ongoing menopausal symptoms obviously becomes smaller. The current WHI cohort started with about 28,000 potential participants but, after implementing all the exclusion criteria, only 549 with moderate/severe VMS remained for evaluation, among whom only 29 suffered hip fracture during follow-up [1]. Also to note that 280 out of the 549 participants were 60–79 years old. Are these older women who still have VMS different from perimenopausal women with VMS? This parameter may be important in view of previous data from the WHI study related to timing of VMS and the risk for cardiovascular disease [4]. In women with late onset of VMS compared with women having no VMS, there was an increased risk of major coronary events, stroke, total cardiovascular events, and all-cause mortality in the minimally adjusted model. Contrarily, women experiencing only early VMS compared with no VMS had a modest, yet significant decreased risk of stroke, total cardiovascular disease events, and all-cause mortality. Finally, in those with persistent VMS (early onset and continuing for many years – 25% of the included symptomatic women), there was no significant increase or decrease in risk of major coronary events, stroke, total cardiovascular disease, or all-cause mortality compared with women with no VMS. This raised the question whether VMS at older age are the result of pathophysiological mechanisms which may not be similar to those involved in early menopause. If this is the case, VMS at older age may mark certain, not fully understood metabolic alteration that may have implications on bone turnover as well. To follow the same way of thinking, elucidation of the biological mechanisms underlying the VMS/fracture associations is problematic. Fracture risk is multifactorial, involving many extra osseous factors, such as gait, sight, cardiovascular integrity and function, etc. Nevertheless, it is tempting to explain the associations by addressing the sex hormone profile. VMS and bone loss are both tied to estrogen deficiency states, whereas menopausal estrogen replacement therapy is very effective in alleviating VMS and reducing fracture risk. Unfortunately, information on estrogen levels was not available for the present WHI study. The authors state that it is unlikely that hormone therapy use confounded their results because the cohort included only women who reported that they had never used hormones at baseline, and the overwhelming majority of these women did not initiate hormone therapy during follow-up.

Crandall and colleagues [1] concluded their manuscript by saying that 'Women with VMS may benefit from greater attention to healthy lifestyle habits to maintain bone health.' I would add that women with moderate/severe VMS, being at risk for future hip fracture, may certainly benefit from hormone therapy, simultaneously achieving two targets: an improvement in quality of life and a reduced fracture risk.

Amos Pines
Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel

References 

1. Crandall CJ, Aragaki A, Cauley JA, et al. Associations of menopausal vasomotor symptoms with fracture incidence. J Clin Endocrinol Metab 2014 Dec 18. Epub ahead of print
http://www.ncbi.nlm.nih.gov/pubmed/25522264
2. Gast GC, Grobbee DE, Pop VJ, et al. Vasomotor symptoms are associated with a lower bone mineral density. Menopause 2009;16:231-8
http://www.ncbi.nlm.nih.gov/pubmed/19034053
3. Gast GC, Pop VJ, Samsioe GN, et al. Vasomotor menopausal symptoms are associated with increased risk of coronary heart disease. Menopause 2011;18:146-51
http://www.ncbi.nlm.nih.gov/pubmed/21127438
4. Szmuilowicz ED, Manson JE, Rossouw JE, et al. Vasomotor symptoms and cardiovascular events in postmenopausal women. Menopause 2011;18:603-10
http://www.ncbi.nlm.nih.gov/pubmed/21358352