Relationships between lumbar spine bone mineral density, energy status biomarkers and sex steroid hormones in male endurance athletes.

Published: 2019/12/04 Number of words: 3969


Aims: Low bone mineral density (BMD) has been identified in male endurance athletes with some reports of negative associations with training volume. However, the underlying pathophysiological mechanisms remain unclear. Recent research indicates that energy status and oestrogen (E2) can influence skeletal health. This study investigated relationships between biomarkers of energy availability, sex steroid concentrations and lumbar spine BMD in male endurance athletes.
Methods: Thirty fivemale endurance athletes, mean age 33.2 ± 8.0 years, participated in the study. Dual energy X-ray absorptiometry (Lunar iDXA) was utilised to measure body composition and BMD at the total body, lumbar spine (L2-L4) and total hip. Biochemical markers included serum testosterone (T), 17β-estradiol (E2), and energy status biomarkers (thyroid function, insulin-like growth factor-1 and cortisol).
Results: Low lumbar spine BMD (Z-score < -1.0) was evident in 6 participants (17%). A further 14 (40%) athletes had a lumbar spine Z-score below 0. Lumbar spine BMD Z-score was significantly correlated with E2 and lean tissue mass (r = 0.348; p=0.040 and r = 338; p=0.048, respectively). No significant associations were evident between lumbar spine BMD and other biochemical or body composition measures. E2 and T were significantly correlated (r =0.53; p=0.01). Stepwise multiple regression (E2, T and total body lean tissue as predictor variables) showed E2 and lean tissue to be the significant predictors of lumbar spine Zscore (combined adjusted r2= 17.6%; p=0.017).
Conclusions: Lower lumbar spine BMD in male endurance athletes may be causally related to lower lean tissue mass and E2 concentration. Further research, including markers of bone turnover, on the influence of sex steroid hormones on BMD in male endurance athletes is warranted.

Male osteoporosis is becoming increasingly prevalent, with one in five men over the age of fifty suffering an osteoporotic fracture [1]. Regular weight-bearing exercise can contribute to the maximisation of peak bone mass and is a prophylaxis against osteoporosis in later life [2]. Conversely, the beneficial effect of exercise on bone health has been diminished in hypoestrogenic female athletes with menstrual irregularities and low body mass [3]. Further, Bilanin et al. (1989) first identified low bone mineral density (BMD) in male endurance runners [4], and this has since been confirmed in male cyclists, triathletes and runners [5-7]. This is of relevance because low BMD for age is a risk factor for the development of osteoporosis and fracture [8].
Over the last three decades, evidence has established a role for the sex steroids and energy biomarkers in the regulation of bone metabolism in female athletes [9-11]. To date, research has revealed inverse associations between weekly running volume and lumbar spine BMD [5, 6]. However, the pathophysiology underlying low BMD in male athletes is less well understood. There is a need for research at the biochemical level, which may provide further insights into the physiology of bone health in male athletes [7, 12].
It has been hypothesised that chronic endurance training can cause sub- clinical alterations to the male reproductive endocrine system, including reduced total and free testosterone (T) levels [5, 7, 13, 14]. The link between lowered T and low lumbar spine BMD in male athletes is equivocal [14], and evidence suggests that oestradiol (E2) is the dominant sex steroid influencing bone metabolism in men [15-17]. Further, the actions of T on the male skeleton may be mediated in part by aromatisation to E2 [18]. Consideration of hormonal and other factors associated with BMD in men is required. The purpose of the present study was to evaluate associations between biomarkers of sex steroids and energy availability, and lumbar spine BMD, in male endurance athletes.



Thirty five Caucasian male endurance athletes, aged between 20 to 50 years, were selected for this report based on availability of complete BMD and biochemical data, from a group of male and female athletes (n=78) who participated in a larger study. The recruitment criteria with respect to age allowed for peak bone mass accrual and avoid bone loss due to age related factors. All participants reported over 3 years experience of competitive athletic training, and had a weekly training mileage of at least 50km, during the 3months prior to their assessment. Lifestyle factors and medications that influence bone mass were considered. Any athletes who self reported supplementation of amino acids, creatine and recreational drugs in the last six months; or had ever previously supplemented anabolic steroids, were excluded from the study. Written informed consent was obtained from each participant. Ethical approval was obtained through the National Research Ethics Service, Leeds (Central) Research Ethics Committee. All experimental testing procedures complied with the British Association of Sport and Exercise Science guidelines.
Participants attended one visit to the Carnegie Research Institute (CRI), where all measurements were undertaken. Participants were requested to refrain from vigorous exercise 24hours prior to their visit. During the visit each participant completed a validated questionnaire which assessed medical, injury and training history [6].

A 25ml sample of venous blood was collected from the antecubital vein of each participant, dispensed into silicon-coated vacutainer tubes (Becton-Dickinson Ltd .Oxford, UK) and centrifuged at 20oc, 2000rpm for 10 minutes. Serum was distributed by 1ml graduated pipettes (Pastette, UK) into aliquots sufficient for analysis of each hormone and frozen at -800c until analysis. Insulin-like growth factor one (IGF-1) analyses were undertaken by IMMULITE 2000 Systems Analysers (Siemens Healthcare Diagnostics Inc, UK).
All other hormonal analyses were undertaken by electrochemiluminescense immunoassay (ECLIA) using an Elecsys and Cobase immunoassay analyser (Roche Diagnostics GmbH, D-66298 Mannheim, Indianappolis, USA). The total testosterone and oestrogen assays are based on competitive test principles using monoclonal antibodies specifically directly against testosterone and 17β-estradiol respectively. Endogenous testosterone released from the sample by ANS (8-anilino 1- naphthalene sulfonic acid) and norgestrel competes with the added testosterone derivative labelled with ruthenium complex for the binding sites on the biotinylated antibody. Endogenous estradiol released from the sample by mesterolone competes with the added estradiol derivative labelled with a ruthenium complex for the binding sites on the biotinylated antibody. Cortisol, triiodothyronine (T3), thyroxine (T4), and thyrotopin (TSH) assays employ antibodies which are specifically directed against; polyclonal cortisol, polyclonal T3, T4, and monoclonal TSH respectively. Free thyroxine (FT4) determination is made with the aid of a specific anti-T4 antibody labelled with a ruthenium complex. All biochemical analyses were all undertaken in the Biochemistry Laboratory, Manchester Royal Infirmary.

Anthropometry and DXA assessment
Anthropometry was performed using standardised methods. Body mass was measured on digital scales (Seca alpha, Germany) and height was measured to the nearest millimetre using a stadiometer (Seca, Germany). Body mass index (BMI) was calculated from body mass divided by height 2 (kg.m-2).
Total body (TB), anteroposterior lumbar spine (L2-L4), and total hip BMD ( were measured by dual energy X-ray absorptiometry (DXA) (Lunar iDXA, GE Medical systems, UK). The same DXA technician performed the acquisition and analysis of all bone scans, machine calibration occurred daily against a spinal phantom to ensure validity and its performance was followed with a quality control protocol, which indicated no significant machine drift throughout the study. Precision error for the lumbar spine and hip using this DXA machine is 0.41% and 0.58% respectively. In accordance with the manufacturers (Lunar iDXA, GE Medical systems, UK) reference databases of age and sex matched healthy controls, BMD values were converted into Z-scores. Total body measurements acquired by DXA identified regional measurements of BMD, fat free and fat mass for every participant. Precision for this DXA unit is 0.8% for body fat percentage assessment.

SPSS version 17.0 (LEAD technologies Inc) was utilised for statistical analysis. The Kolmogorov-Smirnoff test confirmed the data was normally distributed. Based on L2-L4 Z-scores (aged and sex matched) the sample was split into 3 groups; according to whether lumbar spine Z-scores: were above 0.0, below 0.0, or represented low BMD (<-1.0). A one way analysis of variance (ANOVA) with Bonfferoni post hoc tests compared energy availability biomarkers and sex steroid levels between the three groups. Univariate associations were assessed by Pearson’s Product-moment correlation coefficient, to determine predictor variables for L2-L4 Z-scores. The significant univariate predictors of site-specific BMD were then entered into stepwise multiple linear regression models. A level of P<0.05 was considered as statistically significant.

Descriptive characteristics of the sample are presented in Table 1. Among the 35 participants, 6 athletes were classified as having low lumbar spine BMD (17%) and a further 14 (40%) had Z-scores below the age and sex adjusted mean (Z-score <0.0).
No significant differences were evident between the Z-score categories for age, physical (BMI, body mass, lean tissue & body fat) or biochemical markers of energy availability (IGF-1, cortisol, T3 & leptin) (Table 2). There was a significant difference across the three groups for total E2 (p= 0.048) but not total T (p=0.398) concentrations. Weekly running volume was not significantly different between groups. Bonfferoni post hoc test indicated a significant difference between groups 1 and 3 for total body BMD (p=0.008) and total hip BMD (p=0.007).

Within all participants lumbar spine (L2-L4) Z-scores correlated significantly with E2 (r =0.348, p=0.040) and lean tissue (r =0.338, p=0.047). Correlations were also observed between biochemical markers T and E2 (r =0.531, p=0.001). Total hip and total body Z-scores also significantly correlated with lean tissue (r =0.472, p=0.004 and r =0.670, p=0.000 respectively). No other variables correlated with BMD Z-scores, including body mass and BMI.
The results from the stepwise multiple regression analysis are presented in Table 3. Serum total oestrogen concentration independently explained 9.5% (p<0.05) of the variance in L2-L4 BMD Z-score. Oestrogen and lean tissue mass together explained 27.1% (p<0.05) of the variance in lumbar spine BMD Z-score.

The results of this study provide further evidence of compromised bone health in some young male endurance athletes, with 17% of participants presenting lumbar spine Z-scores below -1.0. A further 40% of the endurance athletes sampled had lumbar spine BMD values below that of the age and sex matched reference mean.
Elsewhere, low lumbar spine BMD has also previously been reported among male endurance athletes [6, 7]. However, no causal mechanisms for the potential reduction in bone density in male endurance athletes have yet been established. We have demonstrated a significant association between reduced total serum E2 concentrations and reduced lumbar spine BMD, in male endurance athletes. Further, E2 and lean tissue were found to be the main predictors of lumbar spine BMD in this group. Interestingly, these associations reached significance in a relatively small sample of athletic males (n=35). We found no correlations between BMD and energy availability biomarkers (such as IGF-1 and leptin). This suggests that the reduced BMD found within the present group is unlikely to be related to biomarkers of energy deprivation.
Endurance training has been associated with a dysfunction within the hypothalamic-pituitary-testicular axis [13]. Hetland et al. (1993) reported negative correlations between free testosterone index and weekly running volume in male athletes, and Hind et al. (2006), between lumbar spine BMD and weekly running volume in male endurance runners. However, contrary to the above reports [5, 6], lumbar spine BMD in the present study was not related to training volume. Although MacKelvie et al, (2000) reported no difference in lumbar spine BMD or testosterone between male runners and non-athletic controls, there was a significant negative correlation between total testosterone and free testosterone and running volume in runners [19].
In the present study, total testosterone concentrations were not significantly different between endurance athletes with low and normal BMD. However, concentrations of oestrogen were lower in athletes with low lumbar spine BMD and correlated significantly with lumbar spine BMD in the entire group. Oestrogen specifically affects the activity of osteoblasts and osteoclasts, and much of the action of testosterone to reduce bone resorption is indirect, via the aromatisation to oestrogen [20].

We founda significant correlation between oestrogen and testosterone concentrations in our sample of male athletes. Evidence elsewhere suggests that oestrogens and androgens are inter-related [18, 21], with the peripheral conversion of androgen to oestrogen accounting for approximately 80-85% of the total production of oestrogen [16]. A previous study in elderly men has shown that the ratio of oestradiol to testosterone is lower in men with osteoporosis, suggesting a role for an impairment of aromatase activity in bone loss [22]. The findings of the present study, although relying on total hormone concentrations, are consistent with a predominant effect of E2 on bone.
The set point of the bone mechanostat, and thus the relation between mechanical loading and bone strength; has been considered to be modulated by oestrogens [23]. We found a significant correlation between lean tissue and BMD, and oestrogen and lean tissue were the main predictors of lumbar spine BMD Z-scores in our group of male athletes. Rector et al. (2009) evaluated the effects of long-term endurance training in males on whole-body and regional BMD [24]. Lean tissue was a significant predictor of BMD in resistance trained and cycling males. Travison et al. (2008) examined body composition and bone mineral content (BMC) in a racially diverse sample of 1,209 males and identified that while body weight, BMI, waist circumference and fat mass were associated with BMC to certain thresholds, lean tissue independently exhibited strong constant associations with BMC [25].

Consistent with some [6, 26] but not others [27], we identified that male athletes with low lumbar BMD had significantly lower total body and hip BMD compared to athletes with normal results. The high cortical composition of the hip bone may protect this site from premature bone loss. Furthermore, it has been previously reported that middle distance runners with low lumbar spine BMD have high BMD in the legs, reflecting a protective mechanism of bone tissue against mechanical stress [28]. These results suggest that the mechanical impact from endurance exercise may not be sufficient to protect against bone loss at weight-bearing sites.

In the absence of the analysis of bone turnover, we cannot infer the pathophysiological explanation for how reduced oestrogen levels may decrease bone mass within this group. In male cyclists, Rector et al. (2008) reported E2 was negatively correlated with bone formation markers and total testosterone was positively associated with resorption markers. Furthermore, it was not possible to control for all variables associated with bone health. Factors such as calcium intake, psychological stress [29] and resistance exercise [2] can potentially influence BMD. Finally, although DXA is an important clinical tool, its utility is somewhat limited by the fact that it cannot clearly separate trabecular from cortical bone or provide information on bone size or geometry. Further research into the volumetric bone density and structure in male athletes would be valuable, in order to inform on overall bone strength in this population.

The cross-sectional nature of these data prevents inference of any causal relationships. However, this investigation provides further evidence for low BMD in male endurance runners. Unlike female endurance athletes [30], no characteristics of energy deficiency were noted in males with lower BMD. Instead, our findings emphasise the role for sex steroids in bone health, and highlights the potential role of E2 in preserving skeletal health in male athletes. Further research, incorporating markers of bone turnover in a larger sample of male endurance athletes is required.

Table 1. Descriptive results for participants (n=35)

(kg.m-2 )
Fat (%)Weekly
training volume(km)

Table 2. Results (mean SD values) for male endurance athletes (n=35) with L2-L4 Z-scores below -1.0 (Group 1), below 0 (Group2) or above 0 (Group 3)


Group 1
Z-score <-1.0
(n = 6)

Group 2
Z-score >-1.0 – <0.0
(n = 14)

Group 3
Z-score >0.0

P – Value

Age (years)
Training volume (km)
BMI (kg.m-2)
Lean Tissue (kg)
Body Fat (%)
Total Body BMD
Femoral Neck BMD
Total body Z-score
L2-L4 Z-score
Femoral Neck Z-score

33.33 ± 5.2
81.75 ± 37.55
20.87 ± 1.15
51.85 ± 4.11
16.95 ± 3.66
1.15 ± 0.06
-1.03 ± 0.02
0.98 ± 0.70
-0.47 ± 0.59
-1.73 ± 0.23
-0.45 ± 0.70

32.71 ± 8.5
91.73 ± 33.01
21.93 ± 3.10
56.36 ± 5.25
18.52 ± 6.58
1.23 ± 0.07
-1.18 ± 0.04
1.07 ± 0.76
0.30 ± 0.66
-0.44 ± 0.28
0.09 ± 0.76

33.53 ± 8.7
82.95 ± 33.23
22.13 ± 1.11
56.27 ± 5.06
16.05 ± 4.75
1.27 ± 0.08
1.32 ± 0.09
1.13 ± 0.85
0.81 ± 0.86
0.87 ± 0.53
0.57 ± 0.85

0.000 *

E2 (pmol/L)
Testosterone (nmol/L)
Cortisol (nmol/L)
IGF-1 (ng/ml)
T3 (nmol/L)
TSH (nmol/L)
FT4 (pmol/L)

85.86 ± 30.22
19.20 ± 5.80
381.12 ±165.0
162.17 ±52.95
1.72 ± 0.28
1.93 ± 1.04
17.09 ± 1.41

88.39 ± 30.77
17.77 ± 4.71
435.4 ± 141.56
186.29 ± 42.0
1.87 ± 0.3
1.59 ± 0.51
16.62 ± 1.76

116.81± 36.12
20.36 ± 5.09
442.37 ± 130.7
188.0 ± 39.65
1.83 ± 0.25
2.38 ± 1.08
16.02 ± 2.02


Values are presented as mean ± SD. P-values = significant differences between group means.

Table 3. Summary of stepwise regression model; Lumbar spine (L2-L4) BMD z-score as dependent variable, oestrogen and lean tissue as predictor variables.

L2-L4 BMD (Z-Score) Adjusted R2 Beta T Sig (n=35)(P-value)
Oestrogen 9.5 0.348 2.1 0.040Lean Tissue 17.6 0.333 2.1 0.017


1. Khosla, S., Update in Male Osteoporosis. J Clin Endocrinol Metab. 95(1): p. 3-10.
2. Kohrt, W.M., et al., Physical Activity and Bone Health. Medicine & Science in Sports & Exercise, 2004. 36(11): p. 1985-1996.
3. Zanker, C.L. and I.L. Swaine, The relationship between serum oestradiol concentration and energy balance in young women distance runners. International Journal of Sports Medicine, 1998. 19(2): p. 104-108.
4. Bilanin, J.E., M.S. Blanchard, and E. Russek-Cohen, Lower vertebral bone density in male long distance runners. Medicine & Science in Sports & Exercise, 1989. 21(1): p. 66-70.
5. Hetland, M., J. Haarbo, and C. Christiansen, Low bone mass and high bone turnover in male long distance runners. J Clin Endocrinol Metab, 1993. 77(3): p. 770-775.
6. Hind, K., J.G. Truscott, and J.A. Evans, Low lumbar spine bone mineral density in both male and female endurance runners. Bone, 2006. 39(4): p. 880-885.
7. Smathers, A.M., M.G. Bemben, and D.A. Bemben, Bone Density Comparisons in Male Competitive Road Cyclists and Untrained Controls. Medicine & Science in Sports & Exercise, 2009. 41(2): p. 290-296.
8. Seeman, E., Invited Review: Pathogenesis of osteoporosis. J Appl Physiol, 2003. 95(5): p. 2142-2151.
9. Burrows, M., et al., Physiological factors associated with low bone mineral density in female endurance runners. British Journal of Sports Medicine, 2003. 37(1): p. 67-71.
10. Jurimae, J. and T. Jurimae, BONE METABOLISM IN YOUNG FEMALE ATHLETES: A REVIEW. / 666 – METABOLIZAM KOSTI KOD MLADIH SPORTAÅ ICA. Kinesiology, 2008. 40(1): p. 39-49.
11. Zanker, C.L. and I.L. Swaine, Responses of bone turnover markers to repeated endurance running in humans under conditions of energy balance or energy restriction. European Journal of Applied Physiology, 2000. 83(4/5).
12. Bilanin, J.E., M.S. Blanchard, and E. Russek-Cohen, Lower vertebral bone density in male long distance runners. / La densite osseuse du rachis lombaire chez des coureurs de fond. Medicine & Science in Sports & Exercise, 1989. 21(1): p. 66-70.
13. Hackney, A.C., W.E. Sinning, and B.C. Bruot, Reproductive hormonal profiles of endurance-trained and untrained males. Medicine & Science in Sports & Exercise, 1988. 20(1): p. 60-65.
14. Maimoun, L., et al., Testosterone Is Significantly Reduced in Endurance Athletes without Impact on Bone Mineral Density. Hormone Research in Paediatrics, 2003. 59(6): p. 285-292.
15. Khosla, S., L.J. Melton, III, and B.L. Riggs, Estrogen and the Male Skeleton. J Clin Endocrinol Metab, 2002. 87(4): p. 1443-1450.
16. Grumbach, M.M., Mutations in the Synthesis and Action of Estrogen: The Critical Role in the Male of Estrogen on Pubertal Growth, Skeletal Maturation, and Bone Mass. Annals of the New York Academy of Sciences, 2004. 1038(Understanding and Optimizing Human Development: From Cells to Patients to Populations): p. 7-13.
17. Lapauw, B.M., et al., Serum Estradiol Is Associated With Volumetric BMD and Modulates the Impact of Physical Activity on Bone Size at the Age of Peak Bone Mass: A Study in Healthy Male Siblings. Journal of Bone and Mineral Research, 2009. 24(6): p. 1075-1085.
18. Tuck, S.P., et al., Sex steroids and bone turnover markers in men with symptomatic vertebral fractures. Bone, 2008. 43(6): p. 999-1005.
19. MacKelvie, K.J., et al., Bone mineral density and serum testosterone in chronically trained, high mileage 40–55 year old male runners. British Journal of Sports Medicine, 2000. 34(4): p. 273-278.
20. Riggs, B.L., S. Khosla, and L.J. Melton, III, Sex Steroids and the Construction and Conservation of the Adult Skeleton. Endocr Rev, 2002. 23(3): p. 279-302.
21. Rochira, V., et al., Skeletal effects of long-term estrogen and testosterone replacement treatment in a man with congenital aromatase deficiency: Evidences of a priming effect of estrogen for sex steroids action on bone. Bone, 2007. 40(6): p. 1662-1668.
22. Fink, H.A., et al., Association of Testosterone and Estradiol Deficiency with Osteoporosis and Rapid Bone Loss in Older Men. J Clin Endocrinol Metab, 2006. 91(10): p. 3908-3915.
23. Frost, H.M., Muscle, bone, and the Utah paradigm: a 1999 overview. Medicine & Science in Sports & Exercise, 2000. 32(5): p. 911-917.
24. Rector, R.S., et al., Lean Body Mass and Weight-Bearing Activity in the Prediction of Bone Mineral Density in Physically Active Men. The Journal of Strength & Conditioning Research, 2009. 23(2): p. 427-435 10.1519/JSC.0b013e31819420e1.
25. Travison, T., et al., The relationship between body composition and bone mineral content: threshold effects in a racially and ethnically diverse group of men. Osteoporosis International, 2008. 19(1): p. 29-38.
26. Smathers, A.M., M.G. Bemben, and D.A. Bemben, Bone Density Comparisons in Male Competitive Road Cyclists and Untrained Controls. Medicine & Science in Sports & Exercise, 2009. 41(2): p. 290-296 10.1249/MSS.0b013e318185493e.
27. Maïmoun, L., et al., Effects of physical activities that induce moderate external loading on bone metabolism in male athletes. Journal of Sports Sciences, 2004. 22(9): p. 875 – 883.
28. MacDougall, J.D., et al., Relationship among running mileage, bone density, and serum testosterone in male runners. J Appl Physiol, 1992. 73(3): p. 1165-1170.
29. Bennell, K.L., P.D. Brukner, and S.A. Malcolm, Effect of altered reproductive function and lowered testosterone levels on bone density in male endurance athletes. British Journal of Sports Medicine, 1996. 30(3): p. 205-208.
30. Zanker CL, and Hind K. The effect of energy balance on endocrine function and bone health in youth. Med Sport Sci 2007 51;81-101.

Cite this page

Choose cite format:
Online Chat Messenger Email
+44 800 520 0055