Relationship of body composition with cortisol and dehydroepiandrosterone sulfate levels in Korean men and women

Abstract

Background

Glucocorticoids, primarily cortisol, mediated by the hypothalamic–pituitary–adrenal axis, significantly influence metabolic responses, including insulin resistance, lipolysis, and body fat distribution. Dehydroepiandrosterone sulfate, an adrenal androgen precursor, often counteracts some of cortisol’s metabolic effects. An imbalance between these hormones is hypothesized to impact body composition. This study aimed to investigate the relationship between cortisol and dehydroepiandrosterone sulfate levels, their ratio, and body composition in a Korean population.

Methods

In total, 802 Korean adults participated in this study between 2018 and 2023. Data on sociodemographic and lifestyle factors, body composition, blood pressure, and metabolic variables were assessed. Cortisol and dehydroepiandrosterone sulfate levels were measured and analyzed using partial correlation and multiple logistic regression, adjusting for age, sex, and other relevant confounders, to determine their relationship with body composition.

Results

Cortisol levels were negatively correlated with body weight, waist circumference, and body mass index after adjustment, and this correlation was more significant in women than in men. Dehydroepiandrosterone sulfate levels showed positive correlations with height, body weight, and waist circumference. The cortisol/dehydroepiandrosterone sulfate ratio demonstrated a negative association with height and body weight. Logistic regression analysis revealed that elevated cortisol levels and a high cortisol/dehydroepiandrosterone sulfate ratio were significant positive predictors of central obesity in men.

Conclusions

Elevated cortisol levels are associated with lower adiposity, particularly in women. Dehydroepiandrosterone sulfate levels are positively correlated with height and body mass. The prediction of central obesity is positively associated with cortisol levels and the cortisol/dehydroepiandrosterone sulfate ratio in men and negatively associated with dehydroepiandrosterone sulfate levels.

Introduction

Glucocorticoids directly impair insulin sensitivity in adipocytes ¹ and promote free fatty acid release from mature adipocytes through hormone sensitive lipase-mediated lipolysis. ² Abdominal obesity promotes insulin resistance and is considered a key part of metabolic syndrome.^3,4 Although cortisol has been implicated as a pathophysiological contributor to idiopathic obesity, circulating cortisol levels have not consistently been shown to be elevated in individuals with obesity compared with those in individuals with normal weight.^5,6 A recent study using both cross-sectional and longitudinal analyses to determine the associations between body composition and serum cortisol levels in men showed that weight change drives changes in cortisol levels, and not vice versa. ⁷

Patients with insulin resistance syndrome (IRS) exhibit abnormal levels of dehydroepiandrosterone sulfate (DHEA-S); the DHEA-S level is decreased in men and increased in women with IRS. An association between low dehydroepiandrosterone (DHEA) levels and increased cardiovascular mortality has been reported in men, whereas, in women, high DHEA-S levels are associated with a higher prevalence of cardiovascular disease. ⁸ Among men, hyperinsulinemia favors a reduction in adrenal androgen levels by inhibiting the enzyme 17,20-lyase, which determines androstenedione and DHEA levels without affecting glucocorticoid and mineralocorticoid production. ⁹ Among the various indices used to assess obesity, body mass index (BMI) is the most practical. However, BMI cannot distinguish between fat mass, lean mass, and adiposity types. Individuals with similar BMI may possess distinct body compositions; for instance, individuals with high BMI may be physically fit with a low mortality risk compared to individuals with low BMI but malfunctioning adiposity.^10,11 The type of adiposity seems to play a role in the obesity paradox. Indices such as waist circumference (WC), waist-to-hip ratio, and body fat percentage (BF%) better reflect body fatness. ¹²

The secretion of cortisol and DHEA-S in an appropriate balance is essential for the maintenance of biological function and homeostasis. A high cortisol/DHEA-S ratio is reportedly associated with mortality, ¹³ dementia, ¹⁴ metabolic syndrome, ¹⁵ and reduced immunity following physical stress. ¹⁶ Furthermore, discordant secretion of these two adrenal hormones is related to aging.^17,18 Mechanisms linking hormones such as cortisol and DHEA-S with IRS and body composition are not completely understood; additionally, there are still many controversial aspects. This study aimed to assess the relationship between cortisol, DHEA-S, and the cortisol/DHEA-S ratio with body composition in healthy individuals.

A preprint has been published in medRxiv in March 2025 (https://www.medrxiv.org/content/10.1101/2025.03.23.25324482v1).

Materials and methods

Study population

This retrospective cross-sectional analysis initially enrolled a total of 1,341 Korean adults aged ≥20 years who underwent routine medical check-ups at Ajou University Hospital between January 2018 and March 2023. Participants who were taking steroid medications or lacked body composition data (N = 539) were excluded. The final study population comprised 802 participants, of whom 71.4% were women.

Institutional review board (IRB) statement

The study protocol was reviewed and approved by the IRB of Ajou University Hospital (AJOUIRB-DB-2025-102). As this was a retrospective review of medical records, the requirement for informed consent was waived by the IRB. All data were fully anonymized before analyses. The study was conducted in accordance with the principles of the Declaration of Helsinki (1975, as revised in 2024).

Data collection

Sociodemographic characteristics, smoking status, alcohol consumption, and activity level were assessed using questionnaires. Body composition was measured using bioelectrical impedance analysis (Inbody 3.0, Biospace, Korea, 2021) after an overnight fast, with participants wearing light clothing and no shoes. BMI was calculated by dividing the weight in kg by the squared value of the height (kg/m²). ¹⁹ WC was measured between the lower rib and iliac crest by a trained nurse. Blood pressure was measured using a semiautomated blood pressure monitor (TM-2650A; PMS Instruments, Tokyo, Japan) after the patient had rested for at least 15 min.

Venous blood samples were collected after 12-h overnight fasting and 24 h of abstinence from vigorous physical activity. Cortisol and DHEA-S levels were assessed using radioimmunoassay (RIA) (Diagnostic Product Cooperation, CA, USA). The author has de-identified all patient details. The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. ²⁰

Statistical analyses

Nominal variables were analyzed using chi-square analysis, and Student’s t-test was used to compare the mean values for general characteristics and laboratory test results between men and women. Partial correlations were calculated to adjust for age and sex. Logistic regression analysis was used to evaluate serum levels of cortisol and DHEA-S as well as the cortisol/DHEA-S ratio as predictors of central obesity. Statistical significance was set at p < 0.05. Statistical Package for Social Sciences (SPSS) software (version 29.0) was used for all statistical analyses.

Results

Demographic characteristics

The study population consisted of 802 individuals, including 229 men (29.6%) and 573 women (71.4%) (Table 1). The mean age and BMI were 52.6 ± 11.7 years and 24.2 ± 4.1 kg/m², respectively. The mean cortisol and DHEA-S levels were 12.6 ± 4.5 μg/dL and 104.5 ± 80.7 μg/dL, respectively. Mean (± SD) values of study variables are presented by sex in Table 1. Sex differences were observed in several demographic and clinical variables, including WC, BMI, cortisol and DHEA-S levels, and the cortisol/DHEA-S ratio.

Table 1.

General characteristics of the study participants.

	Men(n = 229)	Women(n = 573)	p value
Age (years)	53.4 ± 12.3	52.3 ± 11.4	0.216
Height (cm)	171.0 ± 5.9	158.5 ± 5.7	<0.001
Body weight (kg)	74.9 ± 14.2	59.3 ± 10.2	<0.001
WC (cm)	89.6 ± 10.2	84.3 ± 9.1	<0.001
BMI (kg/m2)	25.5 ± 4.2	23.6 ± 3.9	<0.001
SBP (mmHg)	126.2 ± 13.2	120.0 ± 13.9	<0.001
DBP (mmHg)	78.5 ± 9.8	72.5 ± 9.5	<0.001
FBS (mg/dL)	102.4 ± 21.4	95.3 ± 14.1	<0.001
T-Chol (mg/dL)	196.6 ± 39.0	201.1 ± 36.6	0.215
TG (mg/dL)	156.1 ± 104.1	110.4 ± 57.2	<0.001
HDL-C (mg/dL)	51.5 ± 13.1	63.3 ± 15.2	<0.001
LDL-C (mg/dL)	116.8 ± 36.1	116.9 ± 32.3	0.949
Cortisol (μg/dL)	13.5 ± 4.3	12.3 ± 4.5	0.001
DHEA-S (μg/dL)	146.4 ± 96.2	87.6 ± 66.6	<0.001
Cortisol/DHEA-S	0.18 ± 0.37	0.28 ± 0.39	0.002
Alcohol consumption (No, %)	50 (21.8)	243 (42.4)
Current smoker (Yes, %)	20 (8.7)	11 (1.9)
The practice of exercise (No, %)	46 (20.1)	131 (22.9)

Values are presented as mean ± SD unless otherwise indicated.

Sex differences in continuous variables were compared using the t-test.

BMI: body mass index; WC: waist circumference; SBP: systolic blood pressure; DBP: diastolic blood pressure; FBS: fasting blood sugar; T-Chol: total cholesterol; TG: triglyceride; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; DHEA-S: dehydroepiandrosterone sulfate.

Correlations between cortisol, DHEA-S, cortisol/DHEA-S ratio, and body composition

Results of the partial correlation analyses evaluating the association between cortisol levels and body composition parameters are shown in Table 2. Cortisol levels adjusted for age and sex were negatively correlated with body weight (BW), WC, and BMI (p = 0.004, p = 0.023, and p = 0.003, respectively). This result was more significant in women than in men.

Table 2.

Partial correlation between cortisol and body composition in Korean individuals.

Factor	r	p value	r	p valuea,b	r	p value c
Height (cm)	0.068	0.046	0.077	0.029	−0.009	0.798
Body weight (kg)	−0.023	0.500	−0.022	0.531	−0.102	0.004
WC (cm)	−0.050	0.160	−0.050	0.161	−0.081	0.023
BMI (kg/m2)	−0.084	0.014	−0.076	0.032	−0.104	0.003

r: Pearson correlation coefficient; all parameters showed statistical significance (p < 0.05).

^aadjusted for age.

^bbody weight and BMI showed significantly negative association in women (p = 0.029 and 0.019, respectively) (p values are reported in the text).

^cadjusted for age and sex.

WC: waist circumference; BMI: body mass index.

DHEA-S levels were significantly associated with height, BW, and WC (p < 0.001, p < 0.001, and p = 0.018, respectively) after adjusting for age (Table 3).

Table 3.

Partial correlation between DHEA-S and body composition in Korean individuals.

Factor	r	p value	r	p value a
Height (cm)	0.314	<0.001	0.288	<0.001
Body weight (kg)	0.247	<0.001	0.197	<0.001
WC (cm)	0.079	0.026	0.064	0.018
BMI (kg/m2)	0.023	0.499	0.069	0.051

r: Pearson correlation coefficient; all parameters showed statistical significance (p < 0.05).

^aadjusted for age.

DHEA-S: dehydroepiandrosterone sulfate; WC: waist circumference; BMI: body mass index.

The cortisol/DHEA-S ratio was negatively associated with height after adjusting for sex (p < 0.001). The cortisol/DHEA-S ratio was also associated with low BW (p = 0.043) (Table 4). No association was observed between the cortisol/DHEA-S ratio and other body compositions parameters.

Table 4.

Partial correlation between cortisol/DHEA-S ratio and body composition in Korean individuals.

Factor	r	p value	r	p value a
Height (cm)	−0.154	<0.001	−0.135	<0.001
Body weight (kg)	−0.104	0.002	−0.072	0.043
WC (cm)	−0.011	0.751	0.016	0.658
BMI (kg/m2)	0.026	0.450	−0.016	0.644

r: Pearson correlation coefficient; all parameters showed statistical significance (p < 0.05).

^aadjusted for sex.

WC: waist circumference; BMI: body mass index.

Logistic regression analysis for identifying predictors of central obesity

In men, higher cortisol levels were associated with an increased risk of central obesity after adjusting for age and BMI (p = 0.009) (Table 5). In contrast, DHEA-S levels were negatively associated with an increased risk of central obesity (p = 0.011) (Table 6). The cortisol/DHEA-S ratio was associated with central obesity (p = 0.02) (Table 7). However, this effect was not statistically significant among women.

Table 5.

Logistic regression for cortisol as a predictor of central obesity.

	Men				Women
	B coefficient (95% CI)	SE	p value	Exp (B)	B coefficient (95% CI)	SE	p value	Exp (B)
Model 1	0.016	0.031	0.611	1.016	−0.018	0.019	0.344	0.982
Model 2	0.018	0.031	0.563	1.018	−0.017	0.019	0.379	0.984
Model 3	0.135	0.052	0.009	1.144	0.028	0.028	0.308	1.028

Central obesity was defined as WC ≥90 cm for men and ≥85 cm for women.

Model 1 in both categories was an unadjusted model, model 2 was adjusted for age, and model 3 was created by adjusting BMI in Model 2.

CI: confidence interval; BMI: body mass index; WC: waist circumference; SE: standard error.

Table 6.

Logistic regression for DHEA-S as a predictor of central obesity.

	Men				Women
	B coefficient (95% CI)	SE	p value	Exp (B)	B coefficient (95% CI)	SE	p value	Exp (B)
Model 1	−0.002	0.002	0.148	0.998	−0.001	0.001	0.340	0.999
Model 2	−0.005	0.002	0.014	0.995	0.001	0.002	0.386	1.001
Model 3	−0.009	0.004	0.011	0.991	0.002	0.002	0.339	1.002

Central obesity was defined as WC ≥90 cm for men and ≥85 cm for women.

Model 1 in both categories was an unadjusted model, model 2 was adjusted for age, and model 3 was created by adjusting BMI in Model 2.

CI: confidence interval; BMI: body mass index; WC: waist circumference; SE: standard error.

Table 7.

Logistic regression for cortisol/DHEA-S ratio as a predictor of central obesity.

	Men				Women
	B coefficient (95% CI)	SE	p value	Exp (B)	B	SE	p value	Exp (B)
Model 1	0.461	0.429	0.282	1.586	0.124	0.215	0.565	1.132
Model 2	0.710	0.519	0.171	2.033	−0.085	0.226	0.707	0.919
Model 3	2.004	0.859	0.020	7.418	−0.280	0.323	0.385	0.755

Central obesity was defined as WC ≥90 cm for men and ≥85 cm for women.

Model 1 in both categories was an unadjusted model, model 2 was adjusted for age, and model 3 was created by adjusting BMI in Model 2.

CI: confidence interval; BMI: body mass index; WC: waist circumference; SE: standard error.

Discussion

This study showed that cortisol levels and the cortisol/DHEA-S ratio were negatively correlated with body composition parameters. In contrast, DHEA-S levels were positively correlated with body composition parameters. In men, higher cortisol levels and cortisol/DHEA-S ratios were associated with an increased risk of central obesity. DHEA-S levels were negatively associated with an increased risk of central obesity. Cortisol levels were associated with neuroendocrine abnormalities such as inhibited gonadal function, low growth hormone levels, and increased sympathetic nervous system activity. However, these findings have not been confirmed in women, ²¹ although a correlation between low plasma cortisol levels and central obesity has been reported. ²² Aldosterone levels were correlated directly with two other adrenal steroids (cortisol and DHEA-S) in women and cortisol alone in men. DHEA-S levels were correlated strongly with visceral adipose tissue in women and, to a lesser extent, in men. ²³ However, there is a lack of direct mechanistic evidence.

The association between low cortisol levels and obesity has been explained by a hypothalamic–pituitary–adrenal (HPA) axis disorder that causes inversion of the normal circadian variation in cortisol with low morning and high afternoon levels. In addition, a high renal cortisol clearance rate has been reported, suggesting that it may have been developed to compensate for increased cortisol secretion and maintain low plasma cortisol levels to prevent its deleterious effects on target organs. There is considerable uncertainty concerning the significance of this association in women. ²² Results of the present study suggest that cortisol levels and cortisol/DHEA-S ratio are negatively correlated with body composition.

There are still contradictory data regarding the relationship between DHEA-S and body fat accumulation, and although some studies have reported no association, ²⁴ others have found an inverse relationship between the hormone and obesity, such as the anti-obesity effect of DHEA-S through the in vitro stimulation of adipose tissue lipolysis.^25,26 Differences in the age of study participants may have represented an important confounding factor in the study of the relationships between obesity, body fat distribution, and endogenous DHEA-S. ²⁷ One study found a statistically significant association of DHEA-S levels with percent body fat, total adipose mass, and percent lean body mass in men. ²⁴

The mechanisms linking obesity, body fat distribution, and serum DHEA or DHEA-S levels have not been fully elucidated, although several possibilities have been proposed. Gordon et al. observed that cell conversion to mature adipocytes is impeded in the presence of DHEA, whereas DHEA-S exhibits a minor inhibitory effect on glucose-6-phosphate dehydrogenase activity. ²⁸ DHEA as a precursor for active androgens and estrogens may also play a role in the relationship between obesity, body fat distribution, and circulating DHEA. Steroids are lipophilic compounds that are highly soluble in adipose tissue.^29,30 Moreover, adipose tissue expresses several steroid-converting enzymes necessary for the local synthesis of active androgens/estrogens from inactive precursors such as DHEA.^29,31,32 The enzymes responsible for the inactivation of androgens/estrogens are also present in the adipose tissue, ³³ thereby modulating the intracellular levels of active steroids. The intracrine conversion of DHEA to androgens or estrogens in a site-specific manner could affect adipocyte physiology and modulate adipose tissue accumulation and mobilization. ³²

Although DHEA and DHEA-S levels decline in both men and women with age, ³⁴ testosterone levels are generally well maintained in men till an older age and are always several-fold higher in men than in women. There was a negative association of cortisol levels with BMI and WC in women; however, no association between cortisol levels and BMI or WC was found in men. These findings are consistent with those of previous studies showing that cortisol levels are not higher in individuals with obesity than in those with normal weight^5,6; these results are also in agreement with a study suggesting diminished stimulability of the HPA axis in individuals with obesity. ³⁵ Cortisol is released from the subcutaneous adipose tissue by 11 beta-hydroxysteroid dehydrogenase type 1 (11β-HSD1) in humans, and increased enzyme expression in obesity is likely to increase local glucocorticoid signaling and contribute to whole-body cortisol regeneration. ³⁶ However, because the previous findings are not entirely consistent, whether enzymatic overactivity is a cause or a result of obesity remains unclear.

Strengths of the present study include the direct examination of the contentious association between cortisol and adiposity as well as the sex-specific analysis of the relationships between body composition and the hormones cortisol and DHEA-S. Additionally, the ability to predict central obesity was stratified by sex, and the cortisol/DHEA-S ratio was evaluated in relation to central obesity in both men and women.

The study population comprised patients who visited a university hospital, which may have limited the generalizability of our findings. Second, because this study employed a cross-sectional design, it was not possible to determine whether cortisol and DHEA-S levels were causally related to changes in body composition parameters or the development of central obesity. Longitudinal studies could assess whether circulating cortisol and/or DHEA-S levels increase changes in body composition parameters and thereby influence the risk of developing central obesity. In addition, variables that could influence body composition, especially in women, such as hormone therapy, were not considered. Further research on the mechanisms underlying these observations may lead to new therapeutic approaches to reduce the influence on body composition parameters and central obesity. This study provides a foundation for future research aimed at understanding the complex interplay between hormones and body composition. By identifying specific hormonal imbalances such as the cortisol/DHEA-S ratio in men, the study may help in developing targeted interventions to address central obesity and related metabolic issues. Moreover, the sex-specific findings suggest that interventions should be tailored to individual patients.

Conclusion

This study demonstrates that plasma cortisol and DHEA-S levels may contribute to variations in body composition parameters in Korean adults and suggests a relationship between HPA activity, as indicated by increased fasting DHEA-S level, and increased central obesity, independent of adiposity in Korean men.

Data availability

The data that support the study are available upon reasonable request to the corresponding author.

Disclosure summary

The author has nothing to disclose.