ABSTRACT
Background
The purpose of this study was to explore the associations of low muscle mass with serum sex hormones and sex hormone‐binding globulin (SHBG) using data from the National Health and Nutrition Examination Survey (NHANES).
Methods
We analysed 4403 adults aged 20–59 years from the 2013–2016 NHANES. Appendicular skeletal muscle index (ASMI) was calculated as appendicular skeletal muscle (kg) divided by height squared (m2). Low muscle mass was defined as ASMI < 7.0 kg/m2 (men) or < 5.5 kg/m2 (women). Logistic regression was performed separately for male and female participants to assess the associations of low muscle mass with serum sex hormones and sex hormone‐binding globulin (SHBG).
Results
The mean age of the participants was 38.6 years (± 11.35 years), and 46.3% of the participants were female. In men, both total testosterone (TT) (Model 3: OR = 1.003, 95% CI: 1.002–1.004, p < 0.001) and free testosterone (Model 3: OR = 1.007, 95% CI: 1.001–1.013, p = 0.022) levels correlated positively with low muscle mass. The highest TT quartile had 4.529 times the odds compared to the lowest quartile (Model 3: OR = 4.529, 95% CI: 1.927–10.645, p = 0.003). Each unit increase in SHBG was associated with higher odds (Model 3: OR = 1.035, 95% CI: 1.024–1.045, p < 0.001). The highest SHBG quartile (> 45.26 nmol/L) showed 6.442‐fold higher odds (Model 3: OR = 6.442, 95% CI: 3.134–13.242, p < 0.001). For women, the highest estradiol quartile (> 116 pg/mL) had 56.4% lower odds of low muscle mass than the lowest quartile (Model 3: OR = 0.436, 95% CI: 0.268–0.709, p = 0.004). Each unit increase in free testosterone (FT) showed an inverse association (Model 3: OR = 0.710, 95% CI: 0.605–0.834, p < 0.001). The highest FT quartile (> 3.70 pg/mL) had 73.7% lower odds (Model 3: OR = 0.263, 95% CI: 0.146–0.473, p < 0.001). Each unit increase in SHBG showed a positive association with low muscle mass (Model 3: OR = 1.009, 95% CI: 1.007–1.012, p < 0.001). The highest SHBG quartile (> 87.21 nmol/L) showed 5.482‐fold higher odds (Model 3: OR = 5.482, 95% CI: 2.854–10.527, p < 0.001).
Conclusions
In women, estradiol and free testosterone levels are negatively associated with low muscle mass. In men, elevated total testosterone was unexpectedly associated with an increased risk of LMM. SHBG elevation is a consistent risk factor across sexes.
Keywords: appendicular skeletal muscle index, free testosterone, low muscle mass, sarcopenia, sex hormone‐binding globulin, sex hormones
1. Introduction
As medical care improves, increasing life expectancy is leading to a rapidly growing elderly population. Sarcopenia—characterized primarily by loss of muscle mass, reduced muscle strength and declined physical performance—is attracting growing research attention. The prevalence of sarcopenia in community‐based populations ranges from 1% to 29% [1]. The disease is most common in older people, but it can also affect middle‐aged and younger people. Sarcopenia may result in decreased physical function, diminished ability to perform daily activities and an increased risk of falling. Several studies have suggested that sarcopenia may be associated with poor prognosis in cardiovascular, neoplastic and trauma patients [2, 3]. Sarcopenia increases the risk of hospitalization and the cost of hospital care for patients, adding to the financial burden on the health systems [4, 5]. Causes of sarcopenia may involve factors such as poor nutrition, chronic diseases, lack of exercise, medication side effects and endocrine disorders. Maintaining a certain amount of muscle mass is important for normal physiological activity in the body. Muscle mass is generally thought to be correlated with testosterone and oestrogen [6, 7]. Total testosterone levels do not fully reflect bioactive fractions. Only 2%–3% of circulating plasma testosterone is in the free state, with the rest bound to sex hormone‐binding globulin and albumin [8]. There are fewer studies on the effects of sex steroid hormones and SHBG on muscle mass. The National Health and Nutrition Examination Survey is an ongoing cross‐sectional study conducted by the National Center for Health Statistics that provides high‐quality, large sample and nationally representative data. We conducted an observational study of the 2013–2016 National Health and Nutrition Examination Survey to assess the association of low muscle mass with serum sex hormones and SHBG.
2. Methods
2.1. Study Population
We utilized NHANES cross‐sectional data from the 2013–2016. Initially, 7746 participants aged 20–59 years were included. After applying exclusion criteria, the analytic cohort consisted of 4403 participants. Specifically, we excluded 2661 participants who lacked reliable data on appendicular skeletal muscle index, height or sex hormones; 495 individuals with malignant tumours (defined by MCQ220), thyroid disease (defined by MCQ170m) or liver disease (defined by MCQ170l); and 187 women with undocumented menstrual status (Figure 1).
FIGURE 1.
Flowchart of participant screen.
2.2. SHBG and Sex Steroid Hormone
Serum SHBG, estradiol and total testosterone were obtained from the TST file in the laboratory data. Serum concentrations of total testosterone and estradiol were determined by isotope dilution liquid chromatography tandem mass spectrometry (ID‐LC–MS/MS). Serum sex hormone‐binding globulin was quantified by a two‐step chemiluminescence immunoassay. Detailed laboratory protocols are available in the NHANES Laboratory Methods document. Free testosterone was calculated using the Vermeulen formula [9, 10].
2.3. Appendicular Skeletal Muscle Index
Appendicular skeletal muscle mass was evaluated using a dual‐energy X‐ray absorptiometry (DXA). Appendicular skeletal muscle index was obtained by appendicular skeletal muscle mass divided by height squared. Low muscle mass was defined using cutoff points of 7.0 kg/m2 for men and 5.5 kg/m2 for women [11].
2.4. Covariate
The covariates included age, race, education level, income‐to‐poverty ratio (PIR), alcohol consumption status, total cholesterol (TC), albumin, diabetes mellitus, serum cotinine, physical activity and menstrual status (female). Physical activity levels were quantified as metabolic equivalent task‐minutes per week (MET‐min/week), calculated using the formula MET‐min/wk = MET value × weekly frequency × duration in minutes based on the NHANES Physical Activity Questionnaire (PAQ). Activity levels were categorized into low (< 600 MET‐min/week), moderate (600–1200 MET‐min/week) and high (> 1200 MET‐min/week) [12].
2.5. Statistical Analysis
Continuous variables are presented as mean ± standard deviation (SD) and were compared using Student's t test. Categorical variables are expressed as percentages and analysed by chi‐square test. Logistic regression models were used to examine the associations between sex hormones, SHBG and low muscle mass. The three logistic regression models were constructed as follows:
-
Model 1:
Unadjusted.
-
Model 2:
Adjusted for demographic factors: age, race, education level and income‐to‐poverty ratio (PIR).
-
Model 3:
Further adjusted for clinical and behavioural covariates: alcohol consumption status, total cholesterol, albumin, diabetes mellitus, serum cotinine, physical activity and menopausal state (female).
3. Results
Table 1 summarizes the baseline characteristics of the participants. Significant differences were observed between males and females in race, education level, alcohol consumption, body mass index, smoking status, physical activity level, low muscle mass prevalence, albumin, cotinine and sex hormones (total testosterone, free testosterone, estradiol and SHBG). There were differences in hormone levels between males and females, so the analysis was stratified by gender.
TABLE 1.
Characteristics of participants.
| Variable | Male (n = 2364) | Female (n = 2039) | p |
|---|---|---|---|
| Age (year) | 38.72 ± 11.44 | 38.42 ± 11.24 | 0.376 |
| Race | 0.005 | ||
| Mexican American, n (%) | 397 (16.8) | 388 (19.0) | |
| Other Hispanic, n (%) | 247 (10.4) | 234 (11.5) | |
| Non‐Hispanic White, n (%) | 824 (34.9) | 669 (32.8) | |
| Non‐Hispanic Black, n (%) | 443 (18.7) | 425 (20.8) | |
| Other Race, n (%) | 453 (19.2) | 323 (15.8) | |
| Education | < 0.001 | ||
| ≤ High school, n (%) | 1062 (44.9) | 778 (38.2) | |
| > High school, n (%) | 1302 (55.1) | 1261 (61.8) | |
| Income to poverty ratio | 0.120 | ||
| 1.3, n (%) | 721 (30.5) | 674 (33.1) | |
| 1.3–3.5, n (%) | 738 (31.2) | 589 (28.9) | |
| 3.5, n (%) | 905 (38.3) | 776 (38.1) | |
| BMI (kg/m2) | 28.33 ± 6.01 | 29.25 ± 7.31 | < 0.001 |
| Drink | < 0.001 | ||
| No, n (%) | 499 (21.1) | 614 (30.1) | |
| Moderate, n (%) | 641 (27.1) | 405 (19.9) | |
| Heavy, n (%) | 1224 (51.8) | 1020 (50.0) | |
| Diabetes | 0.661 | ||
| Yes, n (%) | 156 (6.6) | 127 (6.2) | |
| No, n (%) | 2208 (93.4) | 1912 (93.8) | |
| Smoking | < 0.001 | ||
| Yes, n (%) | 1096 (46.4) | 606 (29.7) | |
| No, n (%) | 1268 (53.6) | 1433 (70.3) | |
| Menopausal state | |||
| Yes, n (%) | N/A | 527 (25.8) | |
| No, n (%) | N/A | 1512 (74.2) | |
| Physical activity | < 0.001 | ||
| Low, n (%) | 603 (25.5) | 776 (38.1) | |
| Medium, n (%) | 263 (11.1) | 268 (13.1) | |
| High, n (%) | 1498 (63.4) | 995 (48.8) | |
| Low muscle mass | |||
| Yes, n (%) | 161 (6.8) | 230 (11.3) | < 0.001 |
| No, n (%) | 2203 (93.2) | 1809 (88.7) | |
| Albumin (g/L) | 44.63 ± 2.98 | 42.31 ± 3.05 | < 0.001 |
| Cotinine (ng/mL) | 75.69 ± 145.32 | 43.12 ± 107.33 | < 0.001 |
| Total testosterone (ng/dL) | 433.10 ± 182.68 | 24.63 ± 21.24 | < 0.001 |
| Free testosterone (pg/mL) | 82.84 ± 32.04 | 3.05 ± 2.66 | < 0.001 |
| Estradiol (pg/mL) | 24.61 ± 9.46 | 81.22 ± 100.46 | < 0.001 |
| SHBG (nmol/L) | 36.97 ± 18.67 | 71.79 ± 50.05 | < 0.001 |
| Total cholesterol (mg/dL) | 190.54 ± 41.07 | 188.14 ± 41.54 | 0.054 |
Abbreviations: BMI, body mass index; SHBG = sex hormone‐binding globulin.
Table 2 shows the association of sex hormones and SHBG with low muscle mass in men. In continuous models, total testosterone (OR = 1.003, 95% CI: 1.002–1.004), free testosterone (OR = 1.007, 95% CI: 1.001–1.013) and SHBG (OR = 1.035, 95% CI: 1.024–1.045) were positively associated with low muscle mass in Model 3 (all p < 0.05). Estradiol showed no significant association with low muscle mass in continuous models. In quartile analysis, the third quartile of estradiol (23.30–28.80 pg/mL) was associated with significantly lower odds of low muscle mass across all models (Model 1: OR = 0.422, 95% CI: 0.203–0.877, p = 0.023; Model 2: OR = 0.436, 95% CI: 0.202–0.939, p = 0.036; Model 3: OR = 0.426, 95% CI: 0.184–0.984, p = 0.046). In the analysis stratified by total testosterone quartiles, men in Q2–Q4 had higher odds versus Q1, though Q3 did not reach significance in Model 3 (p = 0.069). In the unadjusted Model 1, the highest quartile of free testosterone (Q4: > 98.43 pg/mL) showed significantly higher odds versus Q1 (OR = 1.892, 95% CI: 1.033–3.465, p = 0.040), but this attenuated to nonsignificance in Model 3 (OR = 1.602, 95% CI: 0.743–3.452, p = 0.204). Compared with the lowest quartile (Q1), SHBG demonstrated a positive dose–response association across increasing quartiles. In the fully adjusted Model 3, the highest quartile (Q4: > 45.26 nmol/L) had 6.442‐fold higher odds versus Q1 (OR = 6.442, 95% CI: 3.134–13.242, p < 0.001). Figure 2 shows the forest plot of the logistic regression Model 3 for males.
TABLE 2.
Association of low muscle mass with sex steroid hormone and SHBG in male participants analysed by weighted multivariable binary logistic regression.
| Exposures | Model 1 | Model 2 | Model 3 | |||
|---|---|---|---|---|---|---|
| OR [95% CI] | p | OR [95% CI] | p | OR [95% CI] | p | |
| E2 | 0.995 (0.970–1.020) | 0.668 | 0.997 (0.971–1.024) | 0.830 | 0.999 (0.973–1.025) | 0.912 |
| Quartiles of E2 | ||||||
| Q1 (< 18.50) | Ref | |||||
| Q2 (18.50–23.30) | 0.663 (0.357–1.231) | 0.184 | 0.653 (0.340–1.251) | 0.186 | 0.653 (0.328–1.301) | 0.201 |
| Q3 (23.30–28.80) | 0.422 (0.203–0.877) | 0.023 | 0.436 (0.202–0.939) | 0.036 | 0.426 (0.184–0.984) | 0.046 |
| Q4 (> 28.80) | 0.704 (0.412–1.202) | 0.189 | 0.719 (0.409–1.264) | 0.236 | 0.748 (0.419–1.337) | 0.295 |
| Trend p | 0.150 | 0.189 | 0.227 | |||
| TT | 1.003 (1.002–1.004) | < 0.001 | 1.003 (1.002–1.004) | < 0.001 | 1.003 (1.002–1.004) | < 0.001 |
| Quartiles of TT | ||||||
| Q1 (< 307.00) | Ref | |||||
| Q2 (307.0–401.0) | 2.020 (1.006–4.057) | 0.048 | 2.043 (0.996–4.191) | 0.051 | 2.181 (1.023–4.653) | 0.045 |
| Q3 (401.0–525.0) | 2.360 (1.082–5.144) | 0.032 | 2.258 (1.034–4.931) | 0.042 | 2.189 (0.932–5.145) | 0.069 |
| Q4 (> 525.0) | 4.625 (2.237–9.561) | < 0.001 | 4.289 (2.071–8.883) | < 0.001 | 4.529 (1.927–10.645) | 0.003 |
| Trend p | < 0.001 | < 0.001 | 0.002 | |||
| FT | 1.007 (1.003–1.011) | 0.002 | 1.006 (1.000–1.011) | 0.037 | 1.007 (1.001–1.013) | 0.022 |
| Quartiles of FT | ||||||
| Q1 (< 61.78) | Ref | |||||
| Q2 (61.78–78.90) | 1.055 (0.558–1.997) | 0.863 | 1.007 (0.508–1.995) | 0.984 | 1.070 (0.523–2.189) | 0.839 |
| Q3 (78.90–98.43) | 1.392 (0.704–2.754) | 0.329 | 1.221 (0.596–2.501) | 0.567 | 1.251 (0.599–2.612) | 0.517 |
| Q4 (> 98.43) | 1.892 (1.033–3.465) | 0.040 | 1.469 (0.705–3.059) | 0.287 | 1.602 (0.743–3.452) | 0.204 |
| Trend p | 0.025 | 0.223 | 0.162 | |||
| SHBG | 1.025 (1.017–1.032) | < 0.001 | 1.034 (1.025–1.044) | < 0.001 | 1.035 (1.024–1.045) | < 0.001 |
| Quartiles of SHBG | ||||||
| Q1 (< 24.12) | Ref | |||||
| Q2 (24.12–33.20) | 2.002 (1.093–3.665) | 0.026 | 2.200 (1.212–3.993) | 0.012 | 2.223 (1.123–4.401) | 0.026 |
| Q3 (33.20–45.26) | 2.620 (1.271–5.397) | 0.011 | 3.243 (1.567–6.712) | 0.003 | 3.093 (1.395–6.857) | 0.010 |
| Q4 (> 45.26) | 4.579 (2.539–8.257) | < 0.001 | 6.462 (3.529–11.83) | < 0.001 | 6.442 (3.134–13.242) | < 0.001 |
| Trend p | < 0.001 | < 0.001 | < 0.001 | |||
Note: Model 1 was adjusted for none; Model 2 was adjusted for age, race, education and income; Model 3 was adjusted for age, race, education, income, drinking, total cholesterol, albumin, diabetes, cotinine and physical activity.
Abbreviations: E2, estradiol; FT, free testosterone; SHBG, sex hormone‐binding globulin; TT, total testosterone.
FIGURE 2.
Forest plot of binary logistic regression analysis Model 3 for males. Model 3 was adjusted for age, race, education, income, drinking, total cholesterol, albumin, diabetes, cotinine and physical activity. Abbreviations: E2, estradiol; FT, free testosterone; SHBG, sex hormone‐binding globulin; TT, total testosterone.
Table 3 presents the results for women. In continuous models, free testosterone (FT) was inversely associated with low muscle mass (Model 3: OR = 0.710, 95% CI: 0.605–0.834, p < 0.001), and SHBG was positively associated with low muscle mass (Model 3: OR = 1.009, 95% CI: 1.007–1.012, p < 0.001). Estradiol (E2) and total testosterone (TT) showed no significant association with low muscle mass in the continuous model. In quartile analysis, versus the lowest quartile, higher quartiles of E2 (Q2–Q4) were associated with reduced odds of low muscle mass in Model 3 (Q2: OR = 0.313, 95% CI: 0.192–0.512, p < 0.001; Q3: OR = 0.359, 95% CI: 0.199–0.647, p = 0.003; Q4: OR = 0.436, 95% CI: 0.268–0.709, p = 0.004). For FT, the third and fourth quartiles (Q3: 2.50–3.70 pg/mL and Q4: > 3.70 pg/mL) were associated with reduced odds (Model 3: Q3: OR = 0.464, 95% CI: 0.270–0.798, p = 0.010; Q4: OR = 0.263, 95% CI: 0.146–0.473, p < 0.001). For SHBG, the third and fourth quartiles (Q3: 57.39–87.21 nmol/L and Q4: > 87.21 nmol/L) were associated with increased odds (Model 3: Q3: OR = 3.317, 95% CI: 1.425–7.724, p = 0.010; Q4: OR = 5.482, 95% CI: 2.854–10.527, p < 0.001). Significant gradient trends were observed for FT and SHBG (trend p < 0.001). Figure 3 shows the forest plot of the logistic regression Model 3 for females.
TABLE 3.
Association of low muscle mass with sex steroid hormone and SHBG in female participants analysed by weighted multivariable binary logistic regression.
| Exposures | Model 1 | Model 2 | Model 3 | |||
|---|---|---|---|---|---|---|
| OR [95% CI] | p | OR [95% CI] | p | OR [95% CI] | p | |
| E2 | 1.000 (0.998–1.001) | 0.58 | 0.999 (0.997–1.001) | 0.309 | 0.999 (0.996–1.001) | 0.270 |
| Quartiles of E2 | ||||||
| Q1 (< 14.95) | Ref | |||||
| Q2 (14.95–49.30) | 0.494 (0.299–0.817) | 0.008 | 0.360 (0.226–0.571) | < 0.001 | 0.313 (0.192–0.512) | < 0.001 |
| Q3 (49.30–116.00) | 0.618 (0.373–1.025) | 0.061 | 0.450 (0.268–0.754) | 0.004 | 0.359 (0.199–0.647) | 0.003 |
| Q4 (> 116.00) | 0.743 (0.474–1.165) | 0.187 | 0.538 (0.345–0.838) | 0.009 | 0.436 (0.268–0.709) | 0.004 |
| Trend p | 0.632 | 0.267 | 0.182 | |||
| TT | 0.997 (0.990–1.004) | 0.342 | 0.993 (0.982–1.003) | 0.167 | 0.992 (0.980–1.004) | 0.162 |
| Quartiles of TT | ||||||
| Q1 (< 14.90) | Ref | |||||
| Q2 (14.90–21.30) | 0.896 (0.526–1.528) | 0.676 | 0.845 (0.495–1.444) | 0.520 | 0.879 (0.487–1.586) | 0.636 |
| Q3 (21.30–29.50) | 0.814 (0.529–1.252) | 0.335 | 0.694 (0.408–1.182) | 0.168 | 0.676 (0.378–1.207) | 0.163 |
| Q4 (> 29.50) | 0.995 (0.626–1.582) | 0.984 | 0.762 (0.451–1.287) | 0.291 | 0.752 (0.427–1.326) | 0.289 |
| Trend p | 0.918 | 0.299 | 0.268 | |||
| FT | 0.765 (0.665–0.880) | < 0.001 | 0.731 (0.633–0.843) | < 0.001 | 0.710 (0.605–0.834) | < 0.001 |
| Quartiles of FT | ||||||
| Q1 (< 1.70) | Ref | |||||
| Q2 (1.70–2.50) | 0.838 (0.556–1.262) | 0.384 | 0.818 (0.526–1.270) | 0.351 | 0.754 (0.475–1.197) | 0.203 |
| Q3 (2.50–3.70) | 0.519 (0.326–0.825) | 0.007 | 0.516 (0.320–0.833) | 0.009 | 0.464 (0.270–0.798) | 0.010 |
| Q4 (> 3.70) | 0.362 (0.219–0.598) | < 0.001 | 0.292 (0.170–0.502) | < 0.001 | 0.263 (0.146–0.473) | < 0.001 |
| Trend p | < 0.001 | < 0.001 | < 0.001 | |||
| SHBG | 1.008 (1.005–1.010) | < 0.001 | 1.008 (1.005–1.011) | < 0.001 | 1.009 (1.007–1.012) | < 0.001 |
| Quartiles of SHBG | ||||||
| Q1 (< 38.92) | Ref | |||||
| Q2 (38.92–57.39) | 1.704 (0.857–3.389) | 0.123 | 1.840 (0.903–3.749) | 0.089 | 1.762 (0.809–3.837) | 0.137 |
| Q3 (57.39–87.21) | 3.050 (1.475–6.309) | 0.004 | 3.393 (1.575–7.311) | 0.004 | 3.317 (1.425–7.724) | 0.010 |
| Q4 (> 87.21) | 4.664 (2.547–8.540) | < 0.001 | 4.943 (2.758–8.86) | < 0.001 | 5.482 (2.854–10.527) | < 0.001 |
| Trend p | < 0.001 | < 0.001 | < 0.001 | |||
Note: Model 1 was adjusted for none; Model 2 was adjusted for age, race, education and income; Model 3 was adjusted for age, race, education, income, drinking, total cholesterol, albumin, diabetes, cotinine, menopausal state and physical activity.
Abbreviations: E2, estradiol; FT, free testosterone; SHBG, sex hormone‐binding globulin; TT, total testosterone.
FIGURE 3.
Forest plot of binary logistic regression analysis Model 3 for females. Model 3 was adjusted for age, race, education, income, drinking, total cholesterol, albumin, diabetes, cotinine, menopausal state and physical activity. Abbreviations: E2, estradiol; FT, free testosterone; SHBG, sex hormone‐binding globulin; TT, total testosterone.
4. Discussion
Estradiol levels have been associated with skeletal muscle strength performance in previous studies of women [13]. Studies have shown that estradiol inhibits inflammation‐mediated release of a number of proinflammatory cytokines and promotes muscle repair through specific oestrogen receptors [14, 15]. It has also been suggested that lower oestrogen levels can increase apoptosis in skeletal muscle cells, leading to decreased muscle mass [16]. When stratifying women by estradiol quartiles, we observed significantly lower LMM risk in Q2–Q4 versus Q1 in Models 2 and 3. This association was not observed in men.
A study that included 189 female participants suggested that testosterone levels were significantly and positively correlated with appendicular skeletal muscle [17]. In our study, there was no correlation between testosterone levels and the risk of low muscle mass in women, whereas after calculating free testosterone, we found that female patients with lower levels of free testosterone had a higher risk of low muscle mass. A randomized controlled trial showed an increase in muscle mass with exogenous testosterone compared to a placebo group in a population of men with low levels of testosterone [18]. Studies on resistance‐trained men have demonstrated that changes in basal hormone levels within the normal physiologic range exerted minimal influence on muscle hypertrophy outcomes [19, 20]. Lower androgen receptor (AR) expression was detected in skeletal muscle biopsy samples from older adults [21]. In experiments where exogenous testosterone was supplied, high levels of testosterone were required to significantly improve lean body mass (LBM) and appendicular skeletal muscle mass (ASMM) in older men [22]. Exogenous testosterone demonstrated more pronounced effects on muscle maintenance in the elderly, testosterone‐deficient individuals or those with accelerated muscle loss [23]. Exogenous testosterone improved muscle mass and strength in resistance‐trained populations with testosterone deficiency [24]. Compared to previous studies, our participants were younger. This age difference may explain discrepancies in testosterone‐muscle mass associations in our male cohort versus prior research. Basal testosterone levels are much higher in young men than in women, so the effects of changes in physiologic testosterone levels on muscle mass may be masked by diet, lifestyle or physical activity, leading to the observation of opposite results to those observed in female participants [25, 26, 27]. Free testosterone may be a more important determinant of muscle mass than total testosterone levels. A study from the UK Biobank also showed that although total testosterone levels were negatively correlated with total lean body mass in men, free testosterone showed a positive correlation with it [28].
There are few studies on the effect of sex hormone‐binding globulin on skeletal muscle mass. Notably, a randomized controlled trial demonstrated that elderly men with low testosterone receiving testosterone replacement therapy (TRT) achieved significantly increased lean body mass alongside decreased SHBG levels compared with placebo [29]. Similarly, in frail elderly men, TRT was associated with lean mass gain and SHBG reduction [30]. A study of postmenopausal women reported that elevated serum SHBG was negatively correlated with muscle mass, possibly related to intramuscular fat infiltration [31]. These findings aligned with our observations. One mechanistic study found hepatic SHBG mRNA expression decreased with higher body mass index [32]. SHBG is known to regulate free sex steroid levels by binding estradiol and testosterone [33]. Elevated SHBG could reduce bioactive free testosterone delivery to muscle tissue. Recent evidence suggested SHBG actively modulated hormone signalling and liberated bioactive hormones in target tissues [33]. Our study demonstrates that elevated SHBG levels associate with increased low muscle mass risk. This finding contrasts with extensive literature linking low SHBG to heightened incidence of cardiometabolic disorders (hypertension, insulin resistance, type 2 diabetes, coronary heart disease and stroke) and neoplastic conditions [34, 35, 36, 37, 38]. The mechanism remains unclear, but abnormal levels of sex hormone‐binding globulin are associated with disease progression.
5. Limitations
The strengths of this study are the large sample size, which is representative of the general population in the United States, and the ability of DXA to more objectively assess the muscle mass of the participants. There are some limitations to this study. First, muscle strength is also an important indicator for the diagnosis of sarcopenia, but because there were no data related to muscle strength in the 2013–2016 NHANES database, our study focused on the loss of muscle mass. Second, this study was based on cross‐sectional data, so a causal relationship could not be determined.
In conclusion, our study elucidates that SHBG is closely related to low muscle mass and provides new insights into the endocrine mechanisms of low muscle mass.
Ethics Statement
Data for this study were obtained from the National Health and Nutrition Examination Survey, and informed consent was obtained from all participants.
Conflicts of Interest
The authors declare no conflicts of interest.
Qin H., Jiao W., and Liao G., “The Association of Low Muscle Mass With Serum Sex Hormones and Sex Hormone‐Binding Globulin,” Journal of Cachexia, Sarcopenia and Muscle 16, no. 5 (2025): e70056, 10.1002/jcsm.70056.
Funding: The authors received no specific funding for this work.
Data Availability Statement
Data used in this study are all publicly available.
References
- 1. Cruz‐Jentoft A. J., Landi F., Schneider S. M., et al., “Prevalence of and Interventions for Sarcopenia in Ageing Adults: A Systematic Review. Report of the International Sarcopenia Initiative (EWGSOP and IWGS),” Age and Ageing 43, no. 6 (2014): 748–759. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Wong A. and Frishman W., “Sarcopenia and Cardiac Dysfunction,” Cardiology in Review 28, no. 4 (2020): 197–202. [DOI] [PubMed] [Google Scholar]
- 3. Englesbe M. J., Patel S. P., He K., et al., “Sarcopenia and Mortality After Liver Transplantation,” Journal of the American College of Surgeons 211, no. 2 (2010): 271–278. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Sousa A. S., Guerra R. S., Fonseca I., Pichel F., Ferreira S., and Amaral T. F., “Financial Impact of Sarcopenia on Hospitalization Costs,” European Journal of Clinical Nutrition 70, no. 9 (2016): 1046–1051. [DOI] [PubMed] [Google Scholar]
- 5. Cawthon P. M., Lui L.‐Y., Taylor B. C., et al., “Clinical Definitions of Sarcopenia and Risk of Hospitalization in Community‐Dwelling Older Men: The Osteoporotic Fractures in Men Study,” Journals of Gerontology. Series A, Biological Sciences and Medical Sciences 72, no. 10 (2017): 1383–1389. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Anderson L. J., Liu H., and Garcia J. M., “Sex Differences in Muscle Wasting,” in Sex and Gender Factors Affecting Metabolic Homeostasis, Diabetes and Obesity, ed. Mauvais‐Jarvis F. (Springer International Publishing, 2017): 153–197. [Google Scholar]
- 7. Vingren J. L., Kraemer W. J., Ratamess N. A., Anderson J. M., Volek J. S., and Maresh C. M., “Testosterone Physiology in Resistance Exercise and Training: The Up‐Stream Regulatory Elements,” Sports Medicine 40, no. 12 (2010): 1037–1053. [DOI] [PubMed] [Google Scholar]
- 8. Narinx N., David K., Walravens J., et al., “Role of Sex Hormone‐Binding Globulin in the Free Hormone Hypothesis and the Relevance of Free Testosterone in Androgen Physiology,” Cellular and Molecular Life Sciences 79, no. 11 (2022): 543. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Vermeulen A., Verdonck L., and Kaufman J. M., “A Critical Evaluation of Simple Methods for the Estimation of Free Testosterone in Serum,” Journal of Clinical Endocrinology and Metabolism 84, no. 10 (1999): 3666–3672. [DOI] [PubMed] [Google Scholar]
- 10. Chen Z., Shen X., Tian K., et al., “Bioavailable Testosterone Is Associated With Symptoms of Depression in Adult Men,” Journal of International Medical Research 48, no. 8 (2020): 300060520941715. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Cruz‐Jentoft A. J., Bahat G., Bauer J., et al., “Sarcopenia: Revised European Consensus on Definition and Diagnosis,” Age and Ageing 48, no. 1 (2019): 16–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Piercy K. L. and Troiano R. P., “Physical Activity Guidelines for Americans From the US Department of Health and Human Services,” Circulation: Cardiovascular Quality and Outcomes 11, no. 11 (2018): e005263. [DOI] [PubMed] [Google Scholar]
- 13. Pöllänen E., Kangas R., Horttanainen M., et al., “Intramuscular Sex Steroid Hormones Are Associated With Skeletal Muscle Strength and Power in Women With Different Hormonal Status,” Aging Cell 14, no. 2 (2015): 236–248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Velders M., Schleipen B., Fritzemeier K. H., Zierau O., and Diel P., “Selective Estrogen Receptor‐β Activation Stimulates Skeletal Muscle Growth and Regeneration,” FASEB Journal 26, no. 5 (2012): 1909–1920. [DOI] [PubMed] [Google Scholar]
- 15. Geraci A., Calvani R., Ferri E., Marzetti E., Arosio B., and Cesari M., “Sarcopenia and Menopause: The Role of Estradiol,” Frontiers in Endocrinology 12 (2021): 682012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Karvinen S., Juppi H. K., Le G., et al., “Estradiol Deficiency and Skeletal Muscle Apoptosis: Possible Contribution of MicroRNAs,” Experimental Gerontology 147 (2021): 111267. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Kenny A. M., Dawson L., Kleppinger A., Iannuzzi‐Sucich M., and Judge J. O., “Prevalence of Sarcopenia and Predictors of Skeletal Muscle Mass in Nonobese Women Who Are Long‐Term Users of Estrogen‐Replacement Therapy,” Journals of Gerontology. Series A, Biological Sciences and Medical Sciences 58, no. 5 (2003): M436–M440. [DOI] [PubMed] [Google Scholar]
- 18. Storer T. W., Basaria S., Traustadottir T., et al., “Effects of Testosterone Supplementation for 3 Years on Muscle Performance and Physical Function in Older Men,” Journal of Clinical Endocrinology and Metabolism 102, no. 2 (2017): 583–593. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Melville G. W., Siegler J. C., and Marshall P. W. M., “The Effects of D‐Aspartic Acid Supplementation in Resistance‐Trained Men Over a Three Month Training Period: A Randomised Controlled Trial,” PLoS ONE 12, no. 8 (2017): e0182630. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Morton R. W., Oikawa S. Y., Wavell C. G., et al., “Neither Load nor Systemic Hormones Determine Resistance Training‐Mediated Hypertrophy or Strength Gains in Resistance‐Trained Young Men,” Journal of Applied Physiology (1985) 121, no. 1 (2016): 129–138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. di Donato M., Moretti A., Sorrentino C., et al., “Filamin A Cooperates With the Androgen Receptor in Preventing Skeletal Muscle Senescence,” Cell Death Discovery 9, no. 1 (2023): 437. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Sattler F., Bhasin S., He J., et al., “Testosterone Threshold Levels and Lean Tissue Mass Targets Needed to Enhance Skeletal Muscle Strength and Function: The HORMA Trial,” Journals of Gerontology. Series A, Biological Sciences and Medical Sciences 66, no. 1 (2011): 122–129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Isidori A. M., Balercia G., Calogero A. E., et al., “Outcomes of Androgen Replacement Therapy in Adult Male Hypogonadism: Recommendations From the Italian Society of Endocrinology,” Journal of Endocrinological Investigation 38, no. 1 (2015): 103–112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Bhasin S., Storer T. W., Berman N., et al., “The Effects of Supraphysiologic Doses of Testosterone on Muscle Size and Strength in Normal Men,” New England Journal of Medicine 335, no. 1 (1996): 1–7. [DOI] [PubMed] [Google Scholar]
- 25. Morton R. W., Sato K., Gallaugher M. P. B., et al., “Muscle Androgen Receptor Content but Not Systemic Hormones Is Associated With Resistance Training‐Induced Skeletal Muscle Hypertrophy in Healthy, Young Men,” Frontiers in Physiology 9 (2018): 1373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Barsky S. T. and Monks D. A., “Myocytic Androgen Receptor Overexpression Does Not Affect Sex Differences in Adaptation to Chronic Endurance Exercise,” Biology of Sex Differences 13, no. 1 (2022): 59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Fielding R. A., Vellas B., Evans W. J., et al., “Sarcopenia: An Undiagnosed Condition in Older Adults. Current Consensus Definition: Prevalence, Etiology, and Consequences. International Working Group on Sarcopenia,” Journal of the American Medical Directors Association 12, no. 4 (2011): 249–256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Hynes M. C., Watling C. Z., Dunneram Y., Key T. J., and Perez‐Cornago A., “Associations of Body Composition Measures With Circulating Insulin‐Like Growth Factor‐I, Testosterone, and Sex Hormone‐Binding Globulin Concentrations in 16,000 Men,” International Journal of Obesity 48, no. 12 (2024): 1809–1817. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Magnussen L. V., Hvid L. G., Hermann A. P., et al., “Testosterone Therapy Preserves Muscle Strength and Power in Aging Men With Type 2 Diabetes—A Randomized Controlled Trial,” Andrology 5, no. 5 (2017): 946–953. [DOI] [PubMed] [Google Scholar]
- 30. Srinivas‐Shankar U., Roberts S. A., Connolly M. J., et al., “Effects of Testosterone on Muscle Strength, Physical Function, Body Composition, and Quality of Life in Intermediate‐Frail and Frail Elderly Men: A Randomized, Double‐Blind, Placebo‐Controlled Study,” Journal of Clinical Endocrinology and Metabolism 95, no. 2 (2010): 639–650. [DOI] [PubMed] [Google Scholar]
- 31. Pöllänen E., Sipilä S., Alen M., et al., “Differential Influence of Peripheral and Systemic Sex Steroids on Skeletal Muscle Quality in Pre‐ and Postmenopausal Women,” Aging Cell 10, no. 4 (2011): 650–660. [DOI] [PubMed] [Google Scholar]
- 32. Winters S. J., Gogineni J., Karegar M., et al., “Sex Hormone‐Binding Globulin Gene Expression and Insulin Resistance,” Journal of Clinical Endocrinology and Metabolism 99, no. 12 (2014): E2780–E2788. [DOI] [PubMed] [Google Scholar]
- 33. Basualto‐Alarcón C., Llanos P., García‐Rivas G., et al., “Classic and Novel Sex Hormone Binding Globulin Effects on the Cardiovascular System in Men,” International Journal of Endocrinology 2021 (2021): 5527973. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Daka B., Rosen T., Jansson P. A., Larsson C. A., Råstam L., and Lindblad U., “Low Sex Hormone‐Binding Globulin Is Associated With Hypertension: A Cross‐Sectional Study in a Swedish Population,” BMC Cardiovascular Disorders 13 (2013): 30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Le T., Nestler J. E., J. F. Strauss, 3rd , and E. P. Wickham, 3rd , “Sex Hormone‐Binding Globulin and Type 2 Diabetes Mellitus,” Trends in Endocrinology and Metabolism 23, no. 1 (2012): 32–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Reinecke H., Bogdanski J., Woltering A., et al., “Relation of Serum Levels of Sex Hormone Binding Globulin to Coronary Heart Disease in Postmenopausal Women,” American Journal of Cardiology 90, no. 4 (2002): 364–368. [DOI] [PubMed] [Google Scholar]
- 37. Madsen T. E., Luo X., Huang M., et al., “Circulating SHBG (Sex Hormone‐Binding Globulin) and Risk of Ischemic Stroke: Findings From the WHI,” Stroke 51, no. 4 (2020): 1257–1264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Zhu J. L., Chen Z., Feng W. J., Long S. L., and Mo Z. C., “Sex Hormone‐Binding Globulin and Polycystic Ovary Syndrome,” Clinica Chimica Acta 499 (2019): 142–148. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data used in this study are all publicly available.