Body composition and reproductive function exert unique influences on indices of bone health in exercising women

Highlights

• The influence of reproductive function, metabolic status, and body composition on components of bone strength were explored in exercising women.

• Circulating estrogen concentrations and age of menarche were the strongest predictors of lumbar spine bone mineral density.

• Lean mass was the strongest predictor of bone density and geometry at the total hip and femoral neck.

• Body composition and reproductive function play key roles in bone health at weight-bearing and primarily trabecular sites, respectively.

• The osteogenic effects of reproductive function, metabolic hormones, and body composition exert their influence at distinct sites.

Abstract

Reproductive function, metabolic hormones, and lean mass have been observed to influence bone metabolism and bone mass. It is unclear, however, if reproductive, metabolic and body composition factors play unique roles in the clinical measures of areal bone mineral density (aBMD) and bone geometry in exercising women. This study compares lumbar spine bone mineral apparent density (BMAD) and estimates of femoral neck cross-sectional moment of inertia (CSMI) and cross-sectional area (CSA) between exercising ovulatory (Ov) and amenorrheic (Amen) women. It also explores the respective roles of reproductive function, metabolic status, and body composition on aBMD, lumbar spine BMAD and femoral neck CSMI and CSA, which are surrogate measures of bone strength. Among exercising women aged 18–30 years, body composition, aBMD, and estimates of femoral neck CSMI and CSA were assessed by dual-energy x-ray absorptiometry. Lumbar spine BMAD was calculated from bone mineral content and area. Estrone-1-glucuronide (E1G) and pregnanediol glucuronide were measured in daily urine samples collected for one cycle or monitoring period. Fasting blood samples were collected for measurement of leptin and total triiodothyronine. Ov (n = 37) and Amen (n = 45) women aged 22.3 ± 0.5 years did not differ in body mass, body mass index, and lean mass; however, Ov women had significantly higher percent body fat than Amen women. Lumbar spine aBMD and BMAD were significantly lower in Amen women compared to Ov women (p < 0.001); however, femoral neck CSA and CSMI were not different between groups. E1G cycle mean and age of menarche were the strongest predictors of lumbar spine aBMD and BMAD, together explaining 25.5% and 22.7% of the variance, respectively. Lean mass was the strongest predictor of total hip and femoral neck aBMD as well as femoral neck CSMI and CSA, explaining 8.5–34.8% of the variance. Upon consideration of several potential osteogenic stimuli, reproductive function appears to play a key role in bone mass at a site composed of primarily trabecular bone. However, lean mass is one of the most influential predictors of bone mass and bone geometry at weight-bearing sites, such as the hip.

Introduction

Amenorrheic exercising women typically present with low bone mineral density (BMD) when compared to their ovulating counterparts [1], [2], [3]. The poor bone health observed among amenorrheic exercising women is due to the uncoupling of bone formation and resorption that occurs in an environment of low energy availability and suppressed estrogen activity [4], [5]. Estrogen has known osteogenic benefits, serving as a key inhibitor of osteoclast action [6]. Amenorrheic women present with suppressed concentrations of estrogen which may contribute to elevated resorption and, ultimately, poor bone health as evidenced by low BMD and impaired bone geometry and microarchitecture [4], [5], [7], [8], [9]. Likewise, the metabolic environment characteristic of exercising women with functional hypothalamic amenorrhea (FHA), i.e. suppressed insulin-like growth factor-1 (IGF-1), leptin, and total triiodothyronine (TT3), may also contribute to compromised bone health due to its potentially detrimental impact on bone formation [4], [5].

In addition to suppressed reproductive hormone concentrations and altered metabolic profile, another characteristic that may differentiate amenorrheic and ovulatory women is fat mass. Exercising women with FHA typically have a lower fat mass or percent body fat than ovulatory exercising women, serving as another possible contributing factor to poor bone health due to the potentially osteogenic effects of the adipocyte-derived hormone leptin on BMD [10], [11], [12]. Furthermore, an energy deficit can cause a decrease in circulating concentrations of leptin prior to changes in fat mass, indicating that a harmful environment for bone health may be present prior to changes in body weight and may be exacerbated by subsequent changes in body composition that occur as energy deficiency is prolonged [13]. Another large component of body composition, lean mass, has been repeatedly demonstrated to exert strong osteogenic effects on bone mass [14], [15]; however, less is known about the influence of body fat on bone health. Therefore, although reproductive function, metabolic hormones and body composition, particularly lean mass, have been observed to be important for bone health when each component is viewed individually, it is currently not clear whether reproductive function, the metabolic milieu, or body composition, to include both lean body mass and fat mass, is a stronger predictor of bone health among exercising women.

When assessing bone health, BMD is obtained from dual-energy x-ray absorptiometry (DXA), a technique that is only capable of measuring areal BMD (aBMD) rather than true volumetric BMD [16]. Areal BMD tends to underestimate true BMD in small, thin individuals and overestimate true BMD in taller, larger individuals. Therefore, an algorithm that corrects for bone size has been developed to estimate volumetric BMD (bone mineral apparent density or BMAD) at the lumbar spine [16], a site that is prone to low BMD among amenorrheic women and osteoporotic fractures in aged women [2], [3]. Investigators have previously reported low lumbar spine BMAD among amenorrheic adolescent athletes compared to eumenorrheic adolescent athletes and non-athletic girls [3], [8]. Likewise, retired elite gymnasts with a history of amenorrhea were also reported to have lower lumbar spine BMAD compared to retired gymnasts without a history of amenorrhea [9]. Among both adolescent girls and adult women with anorexia nervosa, a severe model of energy deficiency, Karlsson et al. [17] and Misra et al. [18] reported lower lumbar spine BMAD compared to age-matched controls. Therefore, it appears that both an estrogen and energy deficiency contribute to lower lumbar spine BMAD among adolescent girls and women; however, lumbar spine BMAD has not been assessed, to date, in exercising women with a current presentation of FHA.

In addition, DXA is unable to measure bone geometry, an important component of bone strength. However, three-dimensional techniques such as quantitative computed tomography (QCT) and peripheral QCT (pQCT) which can assess bone geometry involve a higher radiation dose than DXA and may not be as readily available as DXA. Therefore, due to the widespread clinical use of DXA and the importance of bone geometry in determining bone strength and fracture risk, a method of estimating geometric properties of the femoral neck using DXA, termed hip strength analysis (HSA) has been developed [19], [20]. HSA provides an estimate of femoral neck strength via measurements of cross-sectional area (CSA) and cross-sectional moment of inertia (CSMI) [19], [21], and, in essence, has enhanced traditional DXA measurements by allowing for an estimate of not only bone mass but also bone geometry, two key components of bone strength. Investigations that go beyond DXA-derived aBMD may not only provide a better estimate of bone strength and fracture risk in amenorrheic women but may also help to identify the determinant factors that affect skeletal fragility in this population.

The roles of body composition, metabolic status, and reproductive function in the parameters of bone health derived from DXA such as aBMD, lumbar spine BMAD, and femoral neck CSA and CSMI are currently not well-understood. Therefore, the purpose of this study is twofold. This study seeks 1) to determine if amenorrheic and ovulatory exercising women differ with regard to DXA-derived estimates of volumetric density of the lumbar spine (lumbar spine BMAD) and femoral neck strength (femoral neck CSMI and CSA) and 2) to explore the respective roles of reproductive function, metabolic status, and body composition, i.e. fat mass and lean mass, in aBMD, lumbar spine BMAD and femoral neck CSMI and CSA. It is hypothesized that 1) amenorrheic exercising women will demonstrate lower lumbar spine BMAD, and lower femoral neck CSMI and CSA compared to ovulatory exercising women, and 2) estrogen and progesterone exposure, age of menarche, leptin and TT3, and lean mass and fat mass will be significant predictors of bone health parameters (aBMD, lumbar spine BMAD, femoral neck CSA and CSMI) among exercising women but will exert differing influences depending on the bone site.