Association of testosterone and sex hormone-binding globulin with fat distribution in men: a quantitative water-fat MRI study

Introduction

Obesity is a major risk factor for male hypogonadism (1). Adipose tissue plays a central role in the metabolism of sex hormones. In obesity, increased visceral adipose tissue (VAT) is thought to enhance aromatase activity, which may promote the conversion of androgens to estrogens. This, in turn, could contribute to the suppression of hypothalamic-pituitary-gonadal axis activity and lower testosterone synthesis (2). Testosterone has a significant impact on both reproductive health and the preservation of muscle and lean body mass. Its deficiency not only impairs reproductive health but also contributes to a reduction in muscle mass and promotes further accumulation of visceral fat, thereby establishing a vicious cycle (3). Therefore, a comprehensive investigation into the complex relationship between obesity and sex hormone metabolism is essential for gaining critical scientific and clinical insights, which are vital for developing effective prevention and treatment strategies to interrupt this vicious cycle.

Tsai’s research highlights a significant finding: in middle-aged American males, both abdominal visceral fat area and subcutaneous fat area, measured at the L4–5 lumbar level using computed tomography (CT), exhibited a strong negative correlation with levels of total testosterone (TT), free testosterone (FT), and bioavailable testosterone (BT) (4). Similarly, in a cohort of young Danish males, the volume of VAT, assessed by magnetic resonance imaging (MRI), shows a negative correlation with TT, BT, and FT levels, while sex hormone-binding globulin (SHBG) concentration inversely correlates with the volume of subcutaneous adipose tissue (SAT) (5). Furthermore, an in-depth study of a Chinese adult community cohort reveals a negative association between SHBG concentration and liver fat content, determined by ultrasonography (6). However, it is important to note that previous research has predominantly focused on the relationship between abdominal or liver fat accumulation and TT, FT, BT, and SHBG levels. In contrast, there remains a significant gap in the systematic exploration of potential correlations between testosterone (including TT, FT, and BT) and SHBG with the volume of epicardial adipose tissue (EAT), as well as fat content in the pancreas, lumbar region, and muscles.

MRI is a non-invasive medical imaging method extensively used in quantitative research on body fat distribution. The multi-echo quantitative Dixon (Q-Dixon) technique employed in this study enables detailed evaluation of tissue-specific fat fraction—quantified as proton density fat fraction (PDFF)—alongside fat volume. It is designed to provide robust quantitative measurements, although PDFF quantification can be affected by artifacts at low fat fractions. Compared to ultrasound and CT, MRI provides a more detailed view for exploring the link between obesity and sex hormone levels (7).

This study examines body fat distribution and testosterone and SHBG levels in obese versus normal-weight individuals to elucidate obesity-hormone relationships. It primarily focuses on two key associations: testosterone with VAT and EAT volumes, and SHBG with liver PDFF and VAT volume. Adipose tissue depots (visceral, subcutaneous, and epicardial) were assessed by volume, whereas ectopic fat infiltration in the liver, pancreas, vertebra, and skeletal muscle was quantified using the PDFF. This methodological distinction is grounded in the physiology of mature white adipose tissue, which exhibits a consistently high intrinsic PDFF, making volume a more sensitive measure of depot size. Correlation and multiple linear regression analyses were employed to identify independent relationships. The aim is to provide evidence for obesity-related disease prevention/treatment and seek biomarkers for obesity-related androgen deficiency. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-2013/rc).

Methods

Study population

Twenty-six male obese patients who visited the Endocrinology Outpatient Department of Our Hospital between March 2023 and September 2024 were included, along with ten male subjects with a normal body mass index (BMI) from the Physical Examination Department, serving as the control group.

Inclusion criteria:

• Obesity group: male individuals aged 18 to 50 years; BMI ≥28 kg/m²; weight change <5% in the past 3 months.
• Normal control group: male individuals aged 18 to 50 years; BMI 18.5 to 23.9 kg/m²; weight change <5% in the past 3 months.

Exclusion criteria:

• Hemoglobin A1c ≥6.5%;
• Severe liver dysfunction (alanine aminotransferase ≥3 times the upper limit of normal), severe renal insufficiency (estimated glomerular filtration rate ≤60 mL/min/1.73 m²), or heart failure (New York Heart Association Class III to IV);
• Secondary obesity due to endocrine disorders;
• History of using weight-loss medications in the past 3 months;
• Malignant tumors;
• Presence of MRI contraindications or inability to undergo MRI examination (e.g., patients with pacemakers, artificial metal heart valves, ferromagnetic aneurysm clips, or intraocular ferromagnetic foreign bodies).

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments, and approved by the Ethics Committee of The First Affiliated Hospital of Zhengzhou University (No. 2020-KY-0235-006). Informed consent was obtained from all participants for the collection of clinical data.

Anthropometric indices assessment

Trained medical professionals measured the height (cm) and weight (kg) of all participants. Participants stood barefoot, wearing lightweight clothing, with their backs against the stand of an ultrasonic body composition analyzer (SK-X80, SONKA, Shenzhen, China). The readings displayed on the LCD screen were recorded to obtain the height and weight. BMI was calculated using the formula: BMI = weight (kg) / height² (m²).

Measurement of testosterone and SHBG

Blood samples were collected from participants in the morning after a fasting period of more than 10 hours. TT was measured using chemiluminescent microparticle immunoassay (ARCHITECT i2000SR, Abbott, Chicago, USA). SHBG was quantified using a chemiluminescent method (Versa Cell X3, SIEMENS, Berlin, Germany). FT and BT were calculated using software (http://www.issam.ch/freetesto.ht), based on the levels of TT and serum SHBG (8).

MRI acquisition

MRI scans were performed using the 3.0T MRI system (MAGNETOM Prisma, Siemens Healthcare, Erlangen, Germany), equipped with a dedicated 18-channel phased-array body coil. Participants abstained from solid food and water for 4–6 hours prior to the MRI scan.

The axial Q-Dixon scan (repetition time/echo time = 10.50 ms/1.16 ms, 2.66 ms, 4.15 ms, 5.66 ms, 7.16 ms, 8.66 ms; slice thickness: 5.0 mm; slice gap: 1.0 mm; field of view: 450 mm × 337 mm; voxel size: 2.0 mm × 2.0 mm × 5.0 mm; flip angle: 3.0°; number of slices =26) covered the entire abdomen, from the diaphragm to the pelvic inlet at the level of the pubic symphysis. The sagittal Q-Dixon scan (repetition time/echo time: 9.00 ms/1.12 ms, 2.46 ms, 3.69 ms, 4.92 ms, 6.15 ms, 7.38 ms; slice thickness: 3.5 mm; slice gap: 1.0 mm; field of view: 350 mm × 306 mm; voxel size: 1.1 mm × 1.1 mm × 3.5 mm; flip angle: 4.0°; number of slices =14) covered the lumbar vertebral region. The heart was examined using gated ECG breath-holds with a multi-shot turbo spin-echo sequence (flip angle: 125°; slice thickness: 8 mm; matrix: 143´256; field of view = 400 mm × 400 mm; number of slices =18) in a four-chamber view orientation. Water suppression was performed using the spectral inversion recovery method.

Image reconstruction and fat quantification

After data acquisition, the multi-echo source images were processed on the scanner using the manufacturer’s integrated reconstruction software (Siemens Healthcare). The reconstruction pipeline applied automated corrections for B0 field inhomogeneity and phase errors to ensure robust water‑fat separation. Quantitative PDFF maps were then generated using a complex‑based, multi‑peak fitting algorithm. This algorithm models the known multi‑peak fat spectrum (with chemical shifts of approximately −3.4, −3.3, −2.6, −1.9, −0.5, and 0.5 ppm relative to the water peak) and includes the effective transverse relaxation time (T₂^*) decay correction simultaneously, thereby providing a confounder‑corrected PDFF estimate for each voxel.

Image analysis

All MRI imaging data were independently reviewed by two senior board-certified radiologists with extensive experience in hybrid imaging. The PDFF of visceral organs, lumbar spine and skeletal muscle was analyzed using the MRI system. Fat volume quantifications (VAT, SAT, EAT) were performed via three-dimensional (3D) segmentation across the entire volumes of interest. In contrast, PDFF measurements for all organs and tissues [liver, pancreas, third lumbar vertebra (L3), erector spinae muscles] were acquired using 2D regions of interest (ROIs) placed on a single representative slice per structure. Fat segmentation was performed using a combination of semi-automated and manual approaches in the open-source software 3D Slicer (version 5.0.2). This strategy was adopted following pilot tests, which indicated that fully automated algorithms frequently led to substantial over- or under-segmentation errors that could not be reliably corrected through parameter adjustment alone.

For VAT and SAT adipose tissue, segmentation was based on the quantitative fat fraction (FF) maps. First, a PDFF threshold of 70–80% was applied to the fat fraction maps to generate an initial mask. This mask was then subjected to meticulous, slice-by-slice review and targeted manual refinement by two independent radiologists (with 8 and 12 years of experience in abdominal imaging), who were blinded to all clinical and biochemical data. Refinement was primarily focused on regions known to challenge threshold-based algorithms, such as hyperintense intestinal contents and ambiguous boundaries between fat and adjacent non-fatty tissues (e.g., fibrous septa, small vessels). Final VAT and SAT volumes were calculated from the diaphragmatic dome to the pelvic inlet at the level of the pubic symphysis (Figure 1).

Figure 1 Representative MRI segmentation of regional adipose tissue depots in a 32-year-old male with obesity. (A-C,F) 3D volumetric renderings of abdominal adipose tissue, displayed in axial (F), sagittal (B), and coronal (C) planes, with corresponding 3D surface reconstruction (A). VAT is colored in red, and SAT in yellow. (D,E,G,H) Manual segmentation of EAT shown on cardiac short-axis (H), outflow tract (D), and two-chamber (E) views, with a corresponding 3D surface reconstruction (G). EAT is highlighted in blue. The quantified volumes were: VAT volume =3,132.55 cm³, SAT volume =2,449.30 cm³, EAT volume =92.40 cm³. 3D, three-dimensional; EAT, epicardial adipose tissue; MRI, magnetic resonance imaging; SAT, subcutaneous adipose tissue; VAT, visceral adipose tissue.

For EAT, a fully manual contouring approach was necessary due to its complex anatomical distribution and poor boundary definition against the myocardium on the cardiac sequences. Using 3D Slicer’s contouring tool, the same two radiologists, who were blinded to all clinical and biochemical data, manually delineated EAT on each slice according to predefined anatomical guidelines, covering adipose tissue from the aortic root to the cardiac apex while carefully excluding non-adipose structures (Figure 1).

To obtain a robust estimate of whole-liver steatosis, PDFF was measured using three large ROIs placed in different lobes—a method widely adopted in quantitative MRI for its reproducibility and correlation with overall fat content. Specifically, ROIs were placed in the left lobe, right anterior lobe, and right posterior lobe, at the level near the hepatic hilum. They were drawn as large as possible within areas of homogeneous signal, with careful exclusion of large blood vessels, bile ducts, organ boundaries, focal hepatic lesions, and artifacts. The final whole-liver PDFF was calculated as the mean of these three ROIs. The pancreatic PDFF was assessed by drawing a single, large ROI at the gland’s maximum horizontal extension to sample representative parenchyma, thereby mitigating the effects of local inhomogeneity. Care was taken to exclude extra-pancreatic fat, ducts, vessels and artifacts. The bone marrow PDFF was assessed using an ROI drawn at the inner edge of the L3 on the PDFF maps of Q-Dixon, avoiding vertebral nourishing blood vessels and artifacts. Fat infiltration of the paraspinal muscles was assessed by measuring the PDFF of the erector spinae muscle group. This was performed on a single axial Q-Dixon slice at the level of the L3/4 intervertebral disc. Manual ROIs were placed within the fascia of the left and right erector spinae muscles, excluding visible intermuscular fat and vasculature. The average PDFF of these two ROIs was recorded (Figure 2). In addition to PDFF, T₂* was also calculated from the multi-echo Q-Dixon source data for the liver, pancreas, L3, and erector spinae muscles using the scanner’s proprietary reconstruction algorithm and the same ROI as those defined for PDFF measurement (Figure 2).

Figure 2 Representative quantitative parametric maps of PDFF and T₂* acquired using the Q‑Dixon sequence in a 32‑year‑old male with obesity. ROIs are delineated in white. (A,B) PDFF map and corresponding T₂* map of the liver, with PDFF of 15.56% and T₂* of 104.31 ms. (C,D) PDFF map and T₂* map of the pancreas, with PDFF of 2.23% and T₂* of 236.11 ms. (E,F) PDFF map and T₂* map of the L3 vertebra, with PDFF of 40.32% and T₂* of 55.26 ms. (G,H) PDFF map and T₂* map of the erector spinae muscles at the L3/4 level, with PDFF of 3.85% and T₂* of 291.14 ms. L3, third lumbar; PDFF, proton density fat fraction; ROIs, regions of interest; T₂*, effective transverse relaxation time.

Statistical analysis

A blinded approach was adopted for all data collection. Statistical analysis utilized SPSS 25.0. Normally distributed data were expressed as mean ± standard deviation (SD) and compared via independent-samples t-tests. Non-normally distributed data were presented as median (interquartile range) and compared using non-parametric tests. Normality tests preceded Pearson (for normally distributed data) or Spearman (for non-normally distributed data) correlation analyses. Inter-reader reproducibility of VAT, SAT, and EAT volumes, as well as liver, pancreas, L3, and erector spinae muscles PDFF was evaluated using Bland-Altman analysis and intraclass correlation coefficients (ICCs). Multiple linear regression models, adjusted for age and height, assessed the independent associations of regional fat metrics (adipose tissue volumes: VAT, SAT, EAT; tissue PDFF: liver, pancreas, L3, erector spinae muscles) with serum levels of TT, FT, BT, and SHBG. A two-tailed P value <0.05 denoted statistical significance.

Measurement reproducibility analysis

Inter-reader reproducibility was assessed for all MRI-derived quantitative metrics across the entire cohort. All measures-including adipose tissue volumes (VAT, SAT, EAT), proton-PDFF of the liver, pancreas, L3, and erector spinae muscles, and T₂* values of the corresponding tissues-showed excellent agreement, with ICCs consistently exceeding 0.85. Detailed ICCs and 95% confidence intervals are provided in Table 1. Based on this high consistency, measurements from the first reader were used for all subsequent analyses.

Results

Study population

No statistically significant differences were observed in age or height between the obese group and the control group (P>0.05, Table 2). BMI was significantly higher in the obese group compared to the control group (P<0.05, Table 2).

Comparison of hormone levels, body fat distribution, and T2* between the obese (OB) and control (CON) groups

The levels of TT, FT, BT, and SHBG were significantly lower in the obese group compared to the control group (P<0.05, Table 2). The obese group exhibited higher volumes of VAT, SAT, EAT, as well as higher PDFF for the liver, pancreas, and erector spinae muscles compared to the control group (P<0.05). However, no significant difference in the L3 PDFF was observed between the OB and control groups (P=0.739, Table 2). T₂* were measured for the liver, pancreas, L3, and erector spinae muscles. No statistically significant differences were observed between the OB and control groups for any of these tissues (all P>0.05, Table 2).

Correlation and multiple linear regression analysis of testosterone, SHBG and body fat content

Correlation analyses were performed using data from the entire cohort (N=36). The correlation analysis revealed significant negative correlations of both TT and SHBG with VAT, SAT, and EAT volumes, as well as with liver and erector spinae muscles PDFF (all P<0.05). FT, BT exhibited notable negative correlations with the volumes of VAT, SAT, and EAT, as well as erector spinae muscles PDFF (P<0.01). The overall correlational pattern was integrated into a heatmap (Figure 3), and the corresponding correlation coefficients (r) and exact P values are detailed in Table 3.

Figure 3 Heatmap depicting correlations between testosterone and SHBG levels and regional fat metrics. Heatmap displaying correlation coefficients (r) across all participants. The color gradient represents Pearson (or Spearman) correlation coefficient (r), with blue tones indicating positive correlations (or closer to zero) and yellow/green tones indicating negative correlations. ^#, P<0.05. BT, bioavailable testosterone; EAT, epicardial adipose tissue; FT, free testosterone; L3, third lumbar; Muscle, erector spinae muscle at the L3/4 level; PDFF, proton density fat fraction; SAT, subcutaneous adipose tissue; SHBG, sex hormone‑binding globulin; TT, total testosterone; VAT, visceral adipose tissue.

In line with our primary focus, the multiple linear regression analysis showed that TT, FT, and BT were negatively associated with the volumes of VAT and EAT (P<0.05). Furthermore, SHBG was negatively correlated with the volume of VAT and liver PDFF (P<0.05). The independent associations were visualized in a heatmap of standardized beta (β) coefficients (Figure 4), with complete numerical results in Table 4.

Figure 4 Heatmap depicting the standardized β coefficients from multiple linear regression analyses, illustrating the associations of testosterone measures (TT, FT, and BT) and SHBG with fat-related parameters: the VAT, SAT, and EAT, and PDFF of the liver and muscle. The color gradient corresponds to the magnitude of β: darker blue indicates a less negative (or closer to zero) coefficient, while bright yellow indicates a more strongly negative coefficient. White represents missing values. ^#, P<0.05. BT, bioavailable testosterone; EAT, epicardial adipose tissue; FT, free testosterone; PDFF, proton density fat fraction; SAT, subcutaneous adipose tissue; SHBG, sex hormone-binding globulin; TT, total testosterone; VAT, visceral adipose tissue.

Discussion

This study found that male patients with obesity had significantly lower levels of testosterone (TT, FT, BT) and SHBG, alongside a substantially higher body fat content compared to control group. To further explore these associations, we performed multivariate linear regression analysis. The results reveal a significant negative correlation between TT, FT, and BT with the volumes of VAT and EAT. Additionally, SHBG levels show a negative correlation with liver PDFF and VAT volume. These findings offer a deeper understanding of the intricate relationship among testosterone, SHBG and body fat distribution in male patients with obesity.

Obesity is a global epidemic, with reports indicating that over half of adult individuals in China are either overweight or obese (9,10). Recent epidemiological surveys have shown that the prevalence of obesity among Chinese men is 17.6%, nearly double that of women (11). As BMI increases and fat accumulation surpasses the body’s metabolic capacity, fat begins to accumulate ectopically in organs such as the heart, liver, pancreas, and skeletal muscles (12). Correlations have been observed between BMI and the areas of subcutaneous fat, visceral fat, epicardial fat thickness, and fat content in the liver, muscle, and pancreas (13-15). Bredella et al. used magnetic resonance spectroscopy to show that bone marrow fat is unrelated to BMI in healthy individuals (16).

In this study, we found elevated VAT, SAT, and EAT volumes, along with increased liver, pancreatic, and erector spinae muscles PDFF in male patients with obesity, while L3 PDFF showed no significant change. These findings are consistent with previous studies but also present important distinctions. By applying a contemporary quantitative water‑fat MRI (Q‑Dixon) protocol to a homogeneous cohort of Chinese men with obesity, this study extends previous research. The protocol provides accurate, radiation‑free fat measurements and incorporates T₂* correction to mitigate confounding from factors such as iron deposition (17). An exploratory analysis of T₂* relaxation times found no significant differences between groups, further suggesting that iron deposition was not a major confounding factor in this cohort. Building on previous MRI-based fat research (15,18), this study extends the evidence by specifically focusing on male fat distribution and incorporating both volumetric assessment of adipose tissue depots (VAT, SAT, EAT) and multi-organ PDFF measurements.

Testosterone, the primary male sex hormone, exerts a wide range of physiological effects in the male body. SHBG specifically binds to and transports androgens, thereby regulating the concentration of active testosterone in the bloodstream. Obese individuals often exhibit lower levels of testosterone (TT, FT, BT) as well as SHBG (19). In line with previous studies (20), this study also found that males in the obese group had significantly lower testosterone levels compared to those in the normal-weight group.

In studies conducted in Hunan, China, TT was negatively correlated with subcutaneous fat, as measured by dual-energy X-ray absorptiometry. However, no correlation was found between TT and visceral fat (21). In middle-aged American men, negative correlations were observed between visceral fat area and subcutaneous fat area at the L4–5 level, as measured by CT, with TT, FT, and BT (4). Among overweight American men with impaired glucose tolerance, TT was negatively correlated with both abdominal visceral fat area and subcutaneous fat area (measured at the L2–3 and L3–4 levels by CT), while SHBG showed no correlation with either fat area (22). A Danish cohort study found that in young men, TT was negatively correlated with both the volumes of VAT and SAT, as measured by MRI. BT and FT were negatively correlated with the volume of VAT but not SAT, while SHBG was negatively correlated with the volume of SAT but not VAT (5). Additionally, a study involving a multi-ethnic American population found that, after adjusting for age, race, and BMI, CT-measured hepatic fat content in men was negatively correlated with SHBG (23).

In this study, TT, FT, and BT levels showed a negative correlation with VAT volume, but not with SAT volume. This specific association with VAT aligns with the established pathophysiology in which visceral adiposity promotes aromatase activity and inflammation, thereby suppressing gonadal function (2,24). Notably, the absence of a correlation with SAT in our Chinese cohort differs from some reports in Western populations (4,21,22). This discrepancy may be explained, in part, by ethnic differences in fat distribution: Asian men, including the present cohort, are known to accumulate a proportionally higher amount of visceral fat at any given BMI compared to Western individuals (25). This characteristic fat patterning could accentuate the metabolic and hormonal impact of VAT, potentially making its relationship with testosterone more pronounced while attenuating the relative contribution of SAT.

The present study identified a negative correlation between both FT and BT with the volume of EAT, this finding has been infrequently reported in the existing literature. Additionally, a negative correlation was found between TT and EAT, which contradicts prior research (26). In that study, no correlation was observed between echocardiographically measured epicardial fat thickness and TT in patients with Klinefelter syndrome. However, it is important to consider that the previous study focused on a specific patient group and employed echocardiography, a method highly dependent on operator skill, positioning, and angle. This can limit its ability to comprehensively represent overall epicardial fat content. In contrast, the current study utilized MRI imaging, which provided a more accurate assessment of the overall volume of epicardial fat. Furthermore, existing research has consistently shown a close association between epicardial fat and obesity, which is in turn linked to an increased risk of cardiovascular events (27). Low testosterone levels in males are also associated with obesity and elevated cardiovascular risk (28,29), which indirectly supports the findings of the present study.

SHBG plays a critical role in the regulation of androgens. In the present study, we observed a significant negative correlation between SHBG levels and both VAT volume and liver PDFF in Chinese men. Notably, the association between SHBG and VAT volume contrasts with certain previous reports (5). Multivariable regression confirmed that both VAT volume and liver PDFF were independently associated with SHBG, suggesting distinct yet complementary pathological contributions. This dual association aligns with established yet separate pathophysiological pathways: hepatic steatosis directly suppresses SHBG synthesis within the liver, largely through insulin resistance and local inflammation triggered by excess intrahepatic fat (30,31). Visceral adiposity, in turn, likely lowers SHBG indirectly by driving systemic insulin resistance and chronic inflammation, which further inhibits hepatic SHBG production (32). Additionally, low SHBG levels may contribute to further insulin resistance by modulating the PI3K/AKT signaling pathway, leading to the phosphorylation of downstream mTOR and resulting in changes in fat distribution (33). Thus, VAT volume and liver PDFF appear to contribute to lower SHBG levels through parallel mechanisms, rather than one being the sole driver of the other.

Limitations

First, it is a single-center clinical study with a relatively small sample size, which may constrain the generalizability of the findings and the statistical power for subgroup analyses. Second, being a cross-sectional design, it cannot establish causal relationships or explore the underlying mechanisms linking fat distribution, male testosterone, and SHBG levels; therefore, longitudinal or interventional studies are required to elucidate their directionality and underlying mechanisms. Third, while this study examines the relationship between obesity, testosterone and SHBG levels, and body fat content, it does not account for other factors that may influence testosterone and SHBG levels, such as dietary habits, physical activity, and genetic factors. Fourth, the study focused on sex hormones and did not include other metabolic parameters [e.g., lipid profiles, glycated hemoglobin (HbA1c), estradiol], which could provide additional mechanistic insights and should be considered in future research. Furthermore, the quantification of ectopic fat in organs like the pancreas relied on a single-slice ROI, which may not capture their full volumetric fat heterogeneity. Future studies employing automated 3D segmentation tools for organs could allow for volumetric analysis of fat distribution, potentially providing deeper insights beyond single-slice ROI measurements.

Conclusions

In conclusion, this cross-sectional Q‑Dixon MRI study in Chinese men with obesity demonstrates that specific patterns of fat distribution are differentially associated with testosterone and SHBG levels. Lower testosterone levels are independently linked to greater volumes of visceral and epicardial fat, whereas lower SHBG levels are independently associated with both increased visceral fat volume and hepatic fat content. These results clarify the phenotypic links between specific fat depots and androgen regulation. Given the cross‑sectional design, causality cannot be inferred; however, these findings provide an imaging‑biology framework for future research into obesity-related hypogonadism.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-2013/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-2013/dss

Funding: This study was supported by the Medical Science, Technology Project of Henan Province, China (No. SBGJ202102111 to J.L.), The First Affiliated Hospital of Zhengzhou University (No. YNQN2017158 to M.H.), the National Natural Science Foundation (No. 82470893 to H.Z.), Joint Construction Project of Henan Medical Science, Technology Project (No. SBGJ202102135 to H.Z.), New Technology and New Project of The First Affiliated Hospital of Zhengzhou University (No. 2023-C19-XJS to H.Z. and No. 2024-C286-XXM to M.H.), Key Scientific Research Project of Colleges and Universities in Henan Province (No. 24A320079 to H.Z.), and Henan Provincial Natural Science Foundation (No. 242300421270 to H.Z.).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-2013/coif). M.H. reports the funding from The First Affiliated Hospital of Zhengzhou University (No. YNQN2017158) and the New Technology and New Project of The First Affiliated Hospital of Zhengzhou University (No. 2024-C286-XXM). H.Z. reports the funding from the National Natural Science Foundation (No. 82470893), Joint Construction Project of Henan Medical Science, Technology Project (No. SBGJ202102135), New Technology and New Project of The First Affiliated Hospital of Zhengzhou University (No. 2023-C19-XJS), Key Scientific Research Project of Colleges and Universities in Henan Province (No. 24A320079), and Henan Provincial Natural Science Foundation (No. 242300421270). J.L. reports the funding from the Medical Science, Technology Project of Henan Province, China (No. SBGJ202102111). The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments, and approved by the Ethics Committee of The First Affiliated Hospital of Zhengzhou University (No. 2020-KY-0235-006). Written informed consent was obtained from all participants prior to data collection.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Lenart-Lipińska M, Łuniewski M, Szydełko J, Matyjaszek-Matuszek B. Clinical and Therapeutic Implications of Male Obesity. J Clin Med 2023;12:5354. [Crossref] [PubMed]
2. Eng PC, Phylactou M, Qayum A, Woods C, Lee H, Aziz S, Moore B, Miras AD, Comninos AN, Tan T, Franks S, Dhillo WS, Abbara A. Obesity-Related Hypogonadism in Women. Endocr Rev 2024;45:171-89. [Crossref] [PubMed]
3. Escobar-Morreale HF, Santacruz E, Luque-Ramírez M, Botella Carretero JI. Prevalence of 'obesity-associated gonadal dysfunction' in severely obese men and women and its resolution after bariatric surgery: a systematic review and meta-analysis. Hum Reprod Update 2017;23:390-408. [Crossref] [PubMed]
4. Tsai EC, Matsumoto AM, Fujimoto WY, Boyko EJ. Association of bioavailable, free, and total testosterone with insulin resistance: influence of sex hormone-binding globulin and body fat. Diabetes Care 2004;27:861-8. [Crossref] [PubMed]
5. Nielsen TL, Hagen C, Wraae K, Brixen K, Petersen PH, Haug E, Larsen R, Andersen M. Visceral and subcutaneous adipose tissue assessed by magnetic resonance imaging in relation to circulating androgens, sex hormone-binding globulin, and luteinizing hormone in young men. J Clin Endocrinol Metab 2007;92:2696-705. [Crossref] [PubMed]
6. Wang X, Xie J, Pang J, Zhang H, Chen X, Lin J, Li Q, Chen Q, Ma J, Xu X, Yang Y, Ling W, Chen Y. Serum SHBG Is Associated With the Development and Regression of Nonalcoholic Fatty Liver Disease: A Prospective Study. J Clin Endocrinol Metab 2020;105:dgz244. [Crossref] [PubMed]
7. Zhang W, Fu C, Yan D, Yuan Y, Zhang W, Gu D, Wu Y, Zhang D, Wang L, Cheng X. Quantification of volumetric thigh and paravertebral muscle fat content: comparison of quantitative Dixon (Q-Dixon) magnetic resonance imaging (MRI) with high-speed T(2)-corrected multiecho MR spectroscopy. Quant Imaging Med Surg 2024;14:4490-505. [Crossref] [PubMed]
8. Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab 1999;84:3666-72. [Crossref] [PubMed]
9. Preda A, Carbone F, Tirandi A, Montecucco F, Liberale L. Obesity phenotypes and cardiovascular risk: From pathophysiology to clinical management. Rev Endocr Metab Disord 2023;24:901-19. [Crossref] [PubMed]
10. Pan XF, Wang L, Pan A. Epidemiology and determinants of obesity in China. Lancet Diabetes Endocrinol 2021;9:373-92. [Crossref] [PubMed]
11. Chen K, Shen Z, Gu W, Lyu Z, Qi X, Mu Y, Ning Y. Prevalence of obesity and associated complications in China: A cross-sectional, real-world study in 15.8 million adults. Diabetes Obes Metab 2023;25:3390-9. [Crossref] [PubMed]
12. Nabasenja C, Barry K, Nelson T, Chandler A, Hewis J. Imaging individuals with obesity. J Med Imaging Radiat Sci 2022;53:291-304. [Crossref] [PubMed]
13. Willar B, Tran KV, Fitzgibbons TP. Epicardial adipocytes in the pathogenesis of atrial fibrillation: An update on basic and translational studies. Front Endocrinol (Lausanne) 2023;14:1154824. [Crossref] [PubMed]
14. Liang KW, Tsai IC, Lee WJ, Lee IT, Lee WL, Lin SY, Wan CJ, Fu CP, Ting CT, Sheu WH. MRI measured epicardial adipose tissue thickness at the right AV groove differentiates inflammatory status in obese men with metabolic syndrome. Obesity (Silver Spring) 2012;20:525-32. [Crossref] [PubMed]
15. Al-Mrabeh A, Hollingsworth KG, Shaw JAM, McConnachie A, Sattar N, Lean MEJ, Taylor R. 2-year remission of type 2 diabetes and pancreas morphology: a post-hoc analysis of the DiRECT open-label, cluster-randomised trial. Lancet Diabetes Endocrinol 2020;8:939-48. [Crossref] [PubMed]
16. Bredella MA, Gill CM, Gerweck AV, Landa MG, Kumar V, Daley SM, Torriani M, Miller KK. Ectopic and serum lipid levels are positively associated with bone marrow fat in obesity. Radiology 2013;269:534-41. [Crossref] [PubMed]
17. Hu F, Yang R, Huang Z, Wang M, Yuan F, Xia C, Wei Y, Song B. 3D Multi-Echo Dixon technique for simultaneous assessment of liver steatosis and iron overload in patients with chronic liver diseases: a feasibility study. Quant Imaging Med Surg 2019;9:1014-24. [Crossref] [PubMed]
18. Peng G, Yu B, Luo T, Tao L, Zhang Z. Multi-echo three-dimensional (3D) Dixon sequence combined with disodium gadolinium for risk stratification of advanced fibrosis in metabolic dysfunction-associated steatotic liver disease. Quant Imaging Med Surg 2024;14:6660-8. [Crossref] [PubMed]
19. Choi S, Chon J, Yoo MC, Shim GY, Kim M, Soh Y, Won CW. The Association of Free Testosterone with Sarcopenic Obesity in Community-Dwelling Older Men: A Cross-Sectional Study. Medicina (Kaunas) 2024;60:754. [Crossref] [PubMed]
20. Barud W, Piotrowska-Swirszcz A, Ostrowski S, Palusiński R, Makaruk B. Association of obesity and insulin resistance with serum testosterone, sex hormone binding globulin and estradiol in older males. Pol Merkur Lekarski 2005;19:634-7.
21. Zhou YH, Guo Y, Wang F, Zhou CL, Tang CY, Tang HN, Yan DW, Zhou HD. Association of Sex Hormones and Fat Distribution in Men with Different Obese and Metabolic Statuses. Int J Gen Med 2022;15:1225-38. [Crossref] [PubMed]
22. Kim C, Dabelea D, Kalyani RR, Christophi CA, Bray GA, Pi-Sunyer X, Darwin CH, Yalamanchi S, Barrett-Connor E, Golden SH, Boyko EJ. Changes in Visceral Adiposity, Subcutaneous Adiposity, and Sex Hormones in the Diabetes Prevention Program. J Clin Endocrinol Metab 2017;102:3381-9. [Crossref] [PubMed]
23. Lazo M, Zeb I, Nasir K, Tracy RP, Budoff MJ, Ouyang P, Vaidya D. Association Between Endogenous Sex Hormones and Liver Fat in a Multiethnic Study of Atherosclerosis. Clin Gastroenterol Hepatol 2015;13:1686-93.e2. [Crossref] [PubMed]
24. Yuxin L, Chen L, Xiaoxia L, Yue L, Junjie L, Youzhu L, Huiliang Z, Qicai L. Research Progress on the Relationship between Obesity-Inflammation-Aromatase Axis and Male Infertility. Oxid Med Cell Longev 2021;2021:6612796. [Crossref] [PubMed]
25. Gallagher D, Kuznia P, Heshka S, Albu J, Heymsfield SB, Goodpaster B, Visser M, Harris TB. Adipose tissue in muscle: a novel depot similar in size to visceral adipose tissue. Am J Clin Nutr 2005;81:903-10. [Crossref] [PubMed]
26. Granato S, Barbaro G, Di Giorgio MR, Rossi FM, Marzano C, Impronta F, Spaziani M, Anzuini A, Lenzi A, Radicioni AF. Epicardial fat: the role of testosterone and lipid metabolism in a cohort of patients with Klinefelter syndrome. Metabolism 2019;95:21-6. [Crossref] [PubMed]
27. Iacobellis G. Epicardial fat links obesity to cardiovascular diseases. Prog Cardiovasc Dis 2023;78:27-33. [Crossref] [PubMed]
28. Araujo AB, Dixon JM, Suarez EA, Murad MH, Guey LT, Wittert GA. Clinical review: Endogenous testosterone and mortality in men: a systematic review and meta-analysis. J Clin Endocrinol Metab 2011;96:3007-19. [Crossref] [PubMed]
29. Blinc A, Schernthaner GH, Poredoš P, Anagnostis P, Jensterle M, Studen KB, Antignani PL, Mikhailidis DP, Šabović M. Testosterone and Peripheral Arterial Disease. Curr Vasc Pharmacol 2023;21:297-303. [Crossref] [PubMed]
30. Song MJ, Choi JY. Androgen dysfunction in non-alcoholic fatty liver disease: Role of sex hormone binding globulin. Front Endocrinol (Lausanne) 2022;13:1053709. [Crossref] [PubMed]
31. Sarkar M, VanWagner LB, Terry JG, Carr JJ, Rinella M, Schreiner PJ, Lewis CE, Terrault NCoronary Artery Risk Development in Young Adults (CARDIA) Cohort. Sex Hormone-Binding Globulin Levels in Young Men Are Associated With Nonalcoholic Fatty Liver Disease in Midlife. Am J Gastroenterol 2019;114:758-63. [Crossref] [PubMed]
32. Kahn SE, Hull RL, Utzschneider KM. Mechanisms linking obesity to insulin resistance and type 2 diabetes. Nature 2006;444:840-6. [Crossref] [PubMed]
33. Zhu JL, Chen Z, Feng WJ, Long SL, Mo ZC. Sex hormone-binding globulin and polycystic ovary syndrome. Clin Chim Acta 2019;499:142-8. [Crossref] [PubMed]