Circulating growth hormone, cortisol and testosterone in relation to vitamin D status: influence of lower and upper body wingate anaerobic test in elite artistic gymnasts

Abstract

Background and study aim

It is well known that professional physical training may be one of the factors modifying s circulating serum level of growth hormone, testosterone and cortisol. However, the effect of high-intensity upper and lower body Wingate Anaerobic Test (WAnT) on the serum hormone levels in association to vitamin D status still remains unspecified. The aim of the current study was to verify hypotheses that a longstanding background in elite gymnastics training induces adaptive changes in hormonal homeostasis during upper- and lower-body WAnT, and that these changes are modulated by muscle group engagement and vitamin D status.

Materials and methods

Fifteen elite male artistic gymnasts (21.3 ± 3.4 years-old) and 14 physically active men (the control group, 20.2 ± 1.1) voluntarily participated in this study. Blood was collected using venipuncture procedures (antecubital vein) in tree timepoints: before, immediately and 60 min after WAnT. Hormone measurements consisted of levels of free human growth hormone (hGH), testosterone and cortisol in blood serum. Measurement was made using chemiluminescence method. Vitamin D active metabolites, 25-hydroxyvitamin D₂ [25(OH)D₂] and 25-hydroxyvitamin D₃ [25(OH)D₃], as a proportion of the total serum concentration of 25-hydroxyvitamin D [25(OH)D], were analysed using the commercially available Total 25OH Vitamin D ELISA kits.

Results

Significantly higher performance during upper-body WAnT were observed in professional gymnasts’ groups, for mean power normalized to body mass. Furthermore, gymnasts showed higher serum concentration for hGH, and testosterone immediately after upper-body WAnT. An inverse relationship was observed in cortisol, whose concentration changes were greater in the control group. Additionally, in control group, baseline vitamin D positively correlated with cortisol changes post lower-body WAnT but negatively with testosterone changes immediately after lower-body WAnT.

Conclusions

Gymnastic training affects anaerobic performance and hormonal status by altering the serum concentrations of hGH, cortisol, and testosterone in response to anaerobic exercise. Moreover, hormonal status is associated with vitamin D concentration, and shows its significant regulating properties in post exercises response.

Introduction

Hormone secretion during exercise is primarily associated with body system changes during prolonged exercises or as a post-exercises body response [1]. It is thought that the changes in some hormone secretions may be related to energic demands during different types of exercise [2–5] or the nature of specific exercises (ex. emotionally dependent cortisol release, aerobic exercise depends on testosterone secretion, and many others). As a result, both acute and chronic changes in hormonal homeostasis play a critical role in regulate energy metabolism and supporting high-performance physical activity [6].

One of the key anabolic hormones involved in exercise adaptation is testosterone [7]. This primary male hormone produced predominantly by the Leydig cells in the testes, plays a crucial role in muscle development and glycogen synthesis [8]. High-intensity but short-lasting exercise appears to increase testosterone secretion [9, 10], upregulate resting testosterone levels. Many factors may directly or indirectly affect testosterone content. One of these factors is vitamin D levels.

Many authors have suggested that vitamin D may play an essential role in physical activity by modulating energy demands, influencing inflammation status, and affecting power generation [11–14]. In addition, vitamin D supplementation may increase testosterone levels [15, 16]. It has been proven in an animal model that vitamin D receptor-knockout mice show hypergonadotropic hypogonadism [17]. Moreover, it has been proven that appropriate vitamin D status in men is directly associated with serum testosterone level [18]. However, some authors have indicated no direct correlation between this specific hormone and vitamin D levels [19]. Considering testosterone secretion differences and their association with vitamin D status, it has been proven that there is another sport-related hormone that may affect testosterone content, which plays a key role in muscle activity, in which the concentration may be modulated by vitamin D status to cortisol. In most cases, testosterone is a major anabolic hormone, whereas cortisol is generally described as a catabolic hormone [7]. This hormone plays a major role in immune function and metabolism through glucocorticoid activity and energy substrate mobilization (e.g., carbohydrates, fats, and proteins) [20]. Cortisol affects several tissues (ex. skeletal muscles), and plays a crucial role in the regulation of physiological processes involved in muscle recovery and energy supply (ex. protein breakdown, liberation of free amino acids, and triglyceride degradation into free fatty acids) [21]. Cortisol plays an essential role in the recovery of the body during stressful situations (ex. physical activity). However, excessive cortisol levels have direct effects on cardiac activity and are directly associated with hypertension [22]. Moreover, cortisol release has been shown to be stimulated by activation of the hypothalamic-pituitary-adrenal axis [23], and its typical changes are associated with time of day, body functioning, and ex. physical activity. Excessive cortisol secretion has been found to be associated with a number of negative health consequences [24] and affects the homeostasis of other body hormones. Considering the previous association between vitamin D status and testosterone level, it is worth noting that some authors have indicated that there may be a negative correlation between cortisol level and vitamin D status. Vitamin D supplementation has been suggested to significantly reduce cortisol levels [25]. However, many authors did not observe significant differences in hormone levels (testosterone and cortisol) resulting from the implementation of dietetics (ex. vitamin D) factors [26–28].

During the detailed analysis of testosterone and cortisol properties, it is possible to determine the testosterone/cortisol ratio [29, 30]. It is considered a good indicator of metabolic status [21] that’s shows physiological responses during physical activity and may be proportional to the intensity of physical effort [31]. On the other hand, similar to most of the secreted factors during activity, its regulation may be directly associated with the body’s adaptation to the type of exercise and the training status of the analyzed individuals [32]. Moreover, it has not been thoroughly determined whether these two hormones are directly related to the secretion of human growth hormone (hGH). It is one of the most widely analyzed and well-established human peptide hormones. hGH also called somatotropin, regulates many aspects of the body development, from the initial body growth to metabolism, reproduction and many other processes [33–36]. This specific hormone has anabolic properties and is mainly released from the pituitary gland in a pulsatory and sex-dependent manner [36]. Serum hGH levels may be affected by many factors; additionally, its contents correlate with physical activity [37], especially with training intensity, and it has a pleiotropic effect on carbohydrate, lipid, and protein metabolism, affecting post-exercises regeneration [38, 39].

Anaerobic sports such as artistic gymnastics are characterized by repeated bouts of short-duration, high-intensity efforts that predominantly rely on the phosphagen system and anaerobic glycolysis. Gymnastic routines involve brief explosive actions such as vaulting, tumbling, and dismounting, typically lasting less than 10 s, and are interspersed with short recovery periods. These movements require rapid generation of force, placing high metabolic demands on intramuscular energy reserves [40, 41]. Initially, energy is supplied by the breakdown of phosphocreatine and ATP, but as the effort continues, especially during longer sequences or repeated attempts, anaerobic glycolysis becomes increasingly engaged. The combined stress from rapid ATP turnover, metabolite accumulation, and neuromuscular recruitment leads to significant perturbations in physiological homeostasis [42].

One of the most widely used protocols for evaluating high-intensity, short-term anaerobic performance is the Wingate Anaerobic Test (WAnT), which has also been applied in studies assessing long-term neuromuscular and metabolic adaptations in trained athletes, including gymnasts [43–45]. The WAnT is particularly suited for such investigations due to its high reproducibility and its relevance to sports involving maximal efforts of brief duration [46, 47].

Importantly, in addition to phosphagen system activation, the WAnT elicits a substantial contribution from anaerobic glycolysis—especially during the middle and later phases of the 30-second effort. This pathway supports ATP resynthesis through rapid glycogenolysis and lactate production, resulting in elevated intramuscular acidosis. The glycolytic component is essential for sustaining high power output beyond the initial seconds dominated by phosphocreatine breakdown [48]. Consequently, the WAnT imposes a physiologically robust challenge involving both phosphagen and glycolytic energy systems. In elite artistic gymnasts, whose sport-specific demands include rapid and forceful movements (e.g., vaults, tumbling, apparatus dismounts), the WAnT closely replicates the neuromuscular and metabolic demands of competition. Thus, assessing endocrine responses to WAnT enables the investigation of acute hormonal adaptations potentially reflective of chronic training-induced modulation of anabolic and stress-related signaling pathways. While WAnT is typically performed using the lower body, studies have shown that, for gymnasts, the upper-body variation may be more suitable for evaluating the post-exercise inflammatory response, as the overwhelming majority of their exercise and training routines involve the upper body [45]. It is therefore advisable to take into account both the lower- and upper-body WAnT protocols when analysing the hormonal response of gymnasts to exercise.

Given the complexity of hormonal regulation and the influence of exercise modality, it is essential to examine endocrine dynamics in athletes exposed to prolonged training loads. Furthermore, the role of vitamin D status as a potential modulator of post-exercise hormonal responses remains insufficiently defined. Therefore, the aim of this study was to determine the effects of lower- and upper-body WAnT on serum concentrations of cortisol, human growth hormone (hGH), and testosterone in elite artistic gymnasts, while considering the potential modulatory role of vitamin D status.

Materials and methods

Experimental overview

A group of elite senior-level Polish artistic gymnasts and age-equal physically active men participated in the study. Both groups performed maximal anaerobic effort in the form of the WAnT with adaptations on the upper and lower body. Blood samples were collected during each type of WAnT: before the test, immediately after, and 60 min after ending the exercise. The samples were analyzed in the context of hormonal homeostasis (circulating growth hormone, cortisol, and testosterone) in relation to vitamin D status. All testing sessions were conducted at the Exercise Physiology Laboratory of the Gdańsk University of Physical Education and Sport. To minimize circadian variability, both testing and blood collection were performed between 9:00 and 12:00 a.m. under standardized laboratory conditions (20–23 °C, controlled lighting, and no external disturbances). Participants refrained from physical activity for 24 h and followed an overnight fast prior to testing.

Subjects

Fifteen elite senior-level Polish gymnasts (21.3 ± 3.4 years-old; training 6 times per week, 5–6 h per session) and 14 healthy, male undergraduate physical education students (20.2 ± 1.1 years-old) whom self-reported more than 180 min per week of moderate physical activity participated in the study (control group). Inclusion criteria for the gymnast group included at least 5 years of continuous training in competitive artistic gymnastics, regular participation in national-level competitions, and absence of injury in the 6 weeks prior to testing. For the control group, inclusion required engagement in general physical activity at least twice per week but no structured athletic training. Exclusion criteria for both groups included chronic illness, current medication affecting hormonal or metabolic status, vitamin D supplementation in the past three months, or any acute illness at the time of testing. The group in this study was recruited from the same overarching research project described in our previous publication [45]. All the participants provided informed consent by signing a written statement confirming they had been fully informed and voluntarily agreed to take part in the study. The study protocol was approved by the Bioethics Committee of the University of Physical Education and Sport in Gdańsk (Resolution No. 2, dated December 16, 2024), and complied with the principles of the Declaration of Helsinki. All participants were aware of the study goals and test procedures and were informed of the possibility of withdrawing consent at any time and for any reason. Six months before the study, all participants were healthy and had a negative medical history regarding systemic and orthopedic diseases, injuries, mental problems, and did not suffer from no other factors that might have affected the results. The descriptive physical characteristics and basal (resting) levels of vitamin D are presented in Table 1.

Table 1.

Physical characteristics and resting values of serum vitamin D

Variable	Control (n = 14)	Gymnasts (n = 15)	p value Control/Gymnasts
	Mean ± SD	Mean ± SD
Body Height (cm)	176 ± 5	171 ± 4	< 0.01*
Body Mass (kg)	72.24 ± 8.80	69.25 ± 7.03	0.31
BMI (kg/m2)	23.34 ± 3.36	23.57 ± 1.59	0.81
Body Fat (%)	10.88 ± 4.81	6.71 ± 3.01	< 0.01*
25(OH)D (ng/ml)	22.94 ± 3.07	21.20 ± 4.16	0.06

Note: BMI, body mass index; 25(OH)D, total 25-hydroxyvitamin D

* Significant difference from the control group at P ˂0.01

Lower-body and upper-body WAnT

A WAnT was conducted for both lower- and upper-body assessments using a cycle ergometer. The protocol followed the original procedure described by Bar-Or [48], with adaptations specific to upper-body testing as outlined by [44]. The WAnT has been widely validated and is known to demonstrate high test–retest reliability for both peak and mean power output in trained and untrained populations [48]. Two days before the start of the experiment, all participants attended a familiarization session to ensure that they were familiar with the study protocol and testing equipment, and all procedures and basic anthropometric characteristics were measured. During the week before the experiment, all participants were asked to refrain from extensive exercise, stay hydrated, and maintain normal dietary habits.

The study protocol started with lower-body WAnT, and after a week break, the same groups performed upper-body WAnT. All experiments were performed in the early morning hours.

Lower-body WAnT was performed on a Monark 894E cycle ergometer (Peak Bike, Sweden, all testing saddle heights were adjusted individually, and all participants started from standardized warm-up (5 min at 60 rpm, 1 W/kg on the cycle ergometer). After the warm-up, a proper test was performed. Lower-body WAnT lasted for 30 s, where participants were required to pedal with maximum effort against a fixed resistive load of 75 g/kg of total body mass.

The upper-body WAnT was performed using a hand-cycle ergometer (Monark 891E; Peak Bike, Sweden) [49]. Similar to the case of the lower-body WAnT, seat height and back rest were adjusted individually, and all participants completed a standardized warm-up (5 min at 60 rpm, 1 W/kg on the hand cycle ergometer).

During testing, each participant was required to arm-crank for 30 s with maximum effort against a standard resistive load equivalent to 50 g/kg of total body mass. In both WAnTs, the procedure was started without prior spinning of the flywheel.

The measured physiological and biomechanical parameters were recorded using a personal computer with MCE 5.1 software [50]. The measured WAnT variables include the highest single point of power output (peak power [W] and relative peak power[W/kg]) and the calculated average power output during the 30 s (mean power [W] and relative mean power [W/kg]).

Blood sample collection and serum analysis

Blood was collected using venipuncture procedures (antecubital vein) by a professional medical diagnostic using 5 mL BD Vacutainer Clot Activator Tubes (Becton Dickinson and Company, NJ, USA). Afterwards, the serum was separated by centrifugation (4000 × g 10 min) and aliquoted (500 µL portions). The samples were frozen and stored (no longer than six months) at − 80 °C until further analysis. During testing, all samples were collected at three time points: before, immediately after (no more than 5 min after the test), and 60 min after the test. Hormone measurements were analyzed based on the levels of free hGH, testosterone, and cortisol in the blood serum. All hormone levels were measured using chemiluminescent immunoassays performed on fully automated platforms routinely used in clinical laboratory procedures, manufactured by Abbott GmbH & Co, Germany and Beckman Coulter Inc., USA.

Vitamin D active metabolites, 25-hydroxyvitamin D₂ [25(OH)D₂] and 25-hydroxyvitamin D₃ [25(OH)D₃], as a proportion of the total serum concentration of 25-hydroxyvitamin D [25(OH)D], were analysed using the commercially available Total 25OH Vitamin D ELISA kits (DRG Instruments GmbH, Germany) according to the manufacturer’s protocol.

All assays were performed in duplicates. Only samples that were not hemolyzed were analyzed.

Statistical analysis

The results are expressed as the mean ± standard deviation (± SD) for all measured variables. A two-way analysis of variance (ANOVA) with repeated measures test was performed to evaluate performance differences between the upper- and lower-body WAnT (exercise factor) in the gymnasts and the control group (group factor). To compare free testosterone, total cortisol, and hGH levels and their ratios in blood between gymnasts and the control group (group factor) at three points (time factor: before the WAnT, immediately after, and 60 min after the WAnT), during the upper- and lower-body WAnT, a two-way ANOVA with repeated measures was used. Tukey’s post hoc test was performed when there was a significant main effect or interaction. The normality of the distribution was checked using the Shapiro–Wilk test, and Levene’s test was used to check the homogeneity of variance.

Additionally, Pearson correlation coefficient and its 95% confidence intervals (calculated using Fisher’s z-transformation) was computed to determine the relationship between the baseline vitamin D concentration and the changes in the analyzed hormone levels (immediately after vs. baseline; 60 min vs. baseline).

The effect size of the biomarkers was determined using eta-squared statistics (ƞ²). In the analysis, ƞ² values equal to or greater than 0.01 (r = 0.1), 0.06 (r = 0.3), and 0.14 (r = 0.5) were the threshold values for a small, moderate, and large effect size, respectively [51].

The appropriate sample size for analyzing the interactions between the effects was determined through a power analysis using GPower [52] ver. 3.1.9.2. Based on a medium effect size and test power of 0.80, a minimum of 28 participants was required. All calculations and graphics were generated using GraphPad Prism 6.0 (GraphPad Software, MA, USA), and all calculations were performed using Statistica 12 (StatSoft, OK, USA). Statistical significance was set at a p-value of ≤ 0.05.

Results

Significant differences were found in body height and body fat percentage between gymnasts and the control group (Table 1). Body mass, BMI and basal vitamin D concentration did not differ significantly between the groups.

The performance of the lower- and upper-body WAnTs for all participants are presented in Table 2. Regardless of the group factor, ANOVA revealed a 28% higher both the absolute mean (F_2,54 = 58.92; p < 0.01; η² = 0.50) and absolute peak power (F_2,54 = 51.02; p < 0.01; η² = 0.46) during the lower-body compared to the upper-body WAnT. Similarly, ANOVA indicated that relative mean (F_2,54 = 145.78; p < 0.01; η² = 0.71) and peak (F_2,54 = 152.05; p < 0.01; η² = 0.62) power were 28% higher during the lower-body WAnT compared to the upper-body WAnT. A significant group × exercise interaction was observed exclusively for relative mean power (F_2,54 = 7.26; p < 0.01; η² = 0.11). Post hoc analysis indicated that gymnasts demonstrated a 15% (p = 0.05) higher relative mean power during the upper-body WAnT compared to the control group (Table 2).

Table 2.

Lower- and upper-body wingate anaerobic test characteristics of elite artistic gymnasts and physically active men (control)

Variable	Lower-body Wingate Anaerobic Test		Upper-body Wingate Anaerobic Test
	Control (n = 14)	Gymnasts (n = 15)	Control (n = 14)	Gymnasts (n = 15)
	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD
Absolute PP (W)	801.13 ± 120.68	775.61 ± 135.24	542.2 ± 126.8	606.10 ± 121.22
Absolute MP (W)	619.91 ± 105.9	595.78 ± 89.90	421.69 ± 82.42	463.62 ± 86.31
Relative PP (W/kg)	11.14 ± 1.17	11.22 ± 1.18	7.46 ± 1.13	8.86 ± 1.4
Relative MP (W/kg)	8.32 ± 0.77	8.63 ± 0.68	5.83 ± 0.71	6.80 ± 0.91*

Note: PP, peak power; MP, mean power

* Significant difference from the control group at p ˂0.01

Figure 1 presents the changes in cortisol, hGH, testosterone, and the testosterone/cortisol ratio concentrations induced by the lower-body WAnT in both gymnasts and control participants.

Fig. 1.

Concentrations of cortisol (A), human growth hormone - hGH (B), testosterone (C) and testosterone/cortisol ratio (D) before (I), immediately after (II), and 60 min after (III) the lower-body Wingate Anaerobic Test (WAnT) in gymnasts and control participants (mean and standard deviation). Note: Blue color: gymnasts, black color: control group. *Significant difference between groups at p < 0.01. # significant difference vs. before WAnT at p < 0.01. ## significant difference vs. before and immediately after the WAnT at p < 0.01

ANOVA with repeated measurement showed a significant group effect (p < 0.01). Regardless of the time point of blood sampling, the gymnasts had a 49% (p < 0.01) higher concentration in hGH than the control group. ANOVA with repeated measurement showed also a significant effect of time in hGH (p < 0.01), cortisol (p < 0.01), and testosterone (p < 0.01) levels during lower-body WAnT (Table 3). Post-hoc analysis showed, that regardless of the group, concentration of hGH increased immediately (6200%, p < 0.01) and 60 min (2600%, p < 0.01) after lower-body WAnT. In case of the cortisol levels, it was shown that it increased by 24% (p < 0.01) and 131% (p < 0.01) immediately and 60 min after lower-body WAnT, respectively. Furthermore, post-hoc analysis showed that testosterone levels were highest immediately after lower-body WAnT, which was 59% (p < 0.01) and 41% (p < 0.01) higher in compare to before and 60 min levels after WAnT, respectively. The testosterone/cortisol ratio showed similar outcome to testosterone analysis, yet ratio at 60 min after lower-body WAnT was also 20% (p < 0.01) lower than before values.

Table 3.

Two-way repeated measures ANOVA for hormonal responses induced by lower-body wingate anaerobic test (WAnT) in gymnasts and control group

Variable	Effect	F	Df	p	Effect Size (η2)	Post-hoc Outcome
hGH	GR RM GR × RM	21.31 578.10 33.05	1, 27 2, 54 2, 54	0.01* 0.01* 0.01*	0.44 0.95 0.55	G > C I < III < II GII > CII
Cortisol	GR RM GR × RM	0.18 300.97 10.25	1, 27 2, 54 2, 54	0.67 0.01* 0.01*	0.01 0.91 0.27	I < II < III
Testosterone	GR RM GR × RM	1.99 149.42 7.09	1, 27 2, 54 2, 54	0.16 0.01* 0.01*	0.07 0.84 0.21	II > I,III
Testosterone/cortisol Ratio	GR RM GR × RM	0.83 75.28 2.45	1, 27 2, 54 2, 54	0.36 0.01* 0.09	0.03 0.73 0.08	II > I > III

hGH - human growth hormone, GR – group, RM – repeated measure, G – gymnasts, C – control group, I – before WAnT, II – immediately after WAnT, III – 60 min after WAnT, * significant differences at p < 0.01

A significant interaction between group and time factor was also observed for cortisol (p < 0.01, Fig. 1A), hGH (p < 0.01, Fig. 1B) and testosterone (p < 0.01, Fig. 1C).

Figure 2 illustrates the changes in cortisol, hGH, testosterone, and the testosterone/cortisol ratio concentrations before, immediately after and 60 min after the upper-body WanT in gymnasts and control participants.

Fig. 2.

Concentrations of cortisol (A), human growth hormone - hGH (B), testosterone (C) and testosterone/cortisol ratio (D) before (I), immediately after (II), and 60 min after (III) following the upper-body Wingate Anaerobic Test (WAnT) in gymnasts and control participants (mean and standard deviation). Note: Blue color: gymnasts; black color: control group. *Indicates significant between-group difference at p < 0.01. # Indicates significant difference vs. before WAnT at p < 0.01. ## Indicates significant difference vs. before and immediately after WAnT at p < 0.01

A significant group effect showed that, regardless time points related to upper-body WAnT, hGH and testosterone levels were 170% (p < 0.01) and 15% (p < 0.01) higher, respectively, in gymnasts compared to the control group. On the other hand, cortisol levels were 19% (p < 0.01) lower in gymnasts compared to the control group.

ANOVA with repeated measures showed a significant time effect for hGH (p < 0.01), cortisol (p < 0.01), and testosterone (p < 0.01) during upper-body WAnT (Table 4). Post-hoc analysis showed, that regardless of the group, concentration of hGH and testosterone increased immediately (4300%, p < 0.01 and 67%, p < 0.01, respectively) and 60 min (2700%, p < 0.01 and 35%, p < 0.01, respectively) after upper-body WAnT. In case of the cortisol levels, it was shown that only 60 min after upper-body WAnT the significant increase (262%, p < 0.01) was observed. The testosterone/cortisol ratio showed similar outcome presented by cortisol changes, where only ratio at 60 min after upper-body WAnT showed significant decrease (76%, p < 0.01). The significant interaction of time × group factor was observed for cortisol (p < 0.01, Fig. 2A), hGH (p < 0.01, Fig. 2B), and testosterone (p < 0.01, Fig. 2C).

Table 4.

Two-way repeated measures ANOVA for hormonal responses induced by upper-body wingate anaerobic test (WAnT) in gymnasts and control group

Variable	Effect	F	Df	p	Effect Size (η2)	Post-hoc Outcome
hGH	GR RM GR × RM	67.22 275.19 56.62	1, 27 2, 54 2, 54	0.01** 0.01** 0.01**	0.71 0.91 0.67	G > C I < III < II GII > CII
Cortisol	GR RM GR × RM	17.29 228.32 17.93	1, 27 2, 54 2, 54	0.01** 0.01** 0.01**	0.39 0.89 0.39	G < C I, II < III GIII < CIII
Testosterone	GR RM GR × RM	11.88 125.34 4.73	1, 27 2, 54 2, 54	0.01** 0.01** 0.01*	0.30 0.82 0.15	G > C I < III < II GII > CII
Testosterone/cortisol Ratio	GR	1.81	1, 27	0.18	0.06
	RM	24.17	2, 54	0.01**	0.47	III < I, II
	GR × RM	6.40	2, 54	0.01**	0.20	GII > CII

hGH - human growth hormone, GR – group, RM – repeated measure, G – gymnasts, C – control group, I – before WAnT, II – immediately after WAnT, III – 60 min after WAnT, significant differences at *p ≤ 0.05, **p < 0.01

The correlation coefficients between baseline vitamin D levels and exercise-induced changes in hGH, cortisol, and testosterone concentrations are presented separately for the lower- and upper-body WAnTs in Tables 5 and 6, respectively.

Table 5.

Correlation coefficient between changes in hormonal response induced by lower-body wingate anaerobic test and baseline level of vitamin D

Variable	Change	Control			Gymnasts			All
		Pearson r	p	95% CI (LL; UL)	Pearson r	p	95% CI (LL; UL)	Pearson r	p	95% CI (LL; UL)
hGH	II-I	-0.42	0.13	(-0.78; 0.14)	0.09	0.76	(-0.46; 0.59)	-0.26	0.17	(-0.58; 0.13)
	III-I	-0.38	0.18	(-0.78; 0.18)	-0.10	0.73	(-0.60; 0.45)	-0.25	0.19	(-0.57; 0.14)
Cortisol	II-I	0.75	0.01**	(0.36; 0.92)	-0.20	0.46	(-0.66; 0.37)	0.37	0.04*	(-0.00; 0.65)
	III-I	0.60	0.02*	(0.10; 0.86)	-0.06	0.84	(-0.57; 0.49)	0.33	0.07	(-0.05; 0.63)
Testosterone	II-I	-0.72	0.01**	(-0.90; -0.31)	0.38	0.16	(-0.19; 0.76)	0.18	0.32	(-0.21; 0.52)
	III-I	-0.26	0.36	(-0.69; 0.31)	0.27	0.33	(-0.30; 0.70)	0.06	0.73	(-0.31; 0.42)
Testosterone/ cortisol ratio	II-I	-0.79	0.01**	(-0.93; -0.45)	0.39	0.15	(-0.18; 0.76)	-0.30	0.11	(-0.61; 0.08)
	III-I	-0.63	0.02*	(-0.87; -0.15)	0.25	0.44	(-0.32; 0.69)	-0.45	0.02*	(-0.70; -0.09)

CI – confidence intervals; LL- lower limit; UL – upper limit; hGH – human growth hormone; I – before WAnT; II – immediately after WAnT; III – 60 min after WAnT

Significant correlation at *p ≤ 0.05, **p < 0.01

Table 6.

Correlation coefficient between changes in hormonal response induced by upper-body wingate anaerobic test and baseline level of vitamin D

Variable	Change	Control			Gymnasts			All
		Pearson r	p	95% CI (LL; UL)	Pearson r	p	95% CI (LL; UL)	Pearson r	p	95% CI (LL; UL)
hGH	II-I	-0.41	0.14	(-0.77; 0.15)	-0.06	0.83	(-0.57; 0.49)	-0.24	0.16	(-0.56; 0.15)
	III-I	-0.37	0.17	(-0.75; 0.20)	-0.05	0.85	(-0.56; 0.50)	-0.25	0.18	(-0.57; 0.14)
Cortisol	II-I	-0.40	0.16	(-0.77; 0.17)	0.09	0.74	(-0.46; 0.59)	0.11	0.59	(-0.27; 0.46)
	III-I	0.36	0.21	(-0.21; 0.75)	0.06	0.83	(-0.49; 0.57)	0.28	0.15	(-0.10; 0.59)
Testosterone	II-I	-0.45	0.11	(-0.79; 0.11)	-0.49	0.06	(-0.81; 0.05)	-0.50	0.01*	(-0.74; -0.16)
	III-I	0.22	0.45	(-0.35; 0.67)	-0.22	0.42	(-0.67; 0.35)	-0.11	0.55	(-0.46; 0.27)
Testosterone/ cortisol ratio	II-I	0.03	0.90	(-0.51; 0.55)	-0.22	0.42	(-0.67; 0.35)	-0.23	0.22	(-0.56; 0.16)
	III-I	-0.15	0.59	(-0.63; 0.41)	-0.34	0.21	(-0.74; 0.23)	-0.27	0.15	(-0.58; 0.11)

CI – confidence intervals; LL- lower limit; UL – upper limit; hGH – human growth hormone; I – before WAnT; II – immediately after WAnT; III – 60 min after WAnT

* Significant correlation at p ≤ 0.05

In physically active men, a significantly positive correlation was observed between the baseline vitamin D concentration and the change in cortisol concentration immediately (p < 0.01) and 60 min (p = 0.02) after the lower-body WAnT. Considering changes immediately after lower-body WAnT in all participants, this correlation was also significant (p = 0.04). Conversely, a significant negative correlation was found between baseline vitamin D levels and the change in testosterone concentration immediately after WAnT (p < 0.01). Additionally, a significantly negative correlation was also observed for the testosterone/cortisol ratio at both immediately (p < 0.01) and 60 min (p = 0.02) after the lower-body WAnT. Considering changes 60 min after upper-body WAnT in all participants, this correlation was also significant (p = 0.01). Similarly, a significantly negative correlation between baseline vitamin D concentration and the change in testosterone concentration was noted immediately after the upper-body WAnT (p = 0.01), regardless of group division, among young adult men.

Discussion

The main purpose of this study was to determine the effects of upper- and lower-body anaerobic exercise (WAnT) on cortisol, hGH, and testosterone levels in elite artistic gymnasts compared to physically active controls. The primary findings revealed significant differences between groups in their acute hormonal responses to WAnT. Specifically, gymnasts demonstrated significantly higher relative mean power during the upper-body WAnT, accompanied by greater immediate post-exercise increases in hGH and testosterone compared to controls. Conversely, physically active controls exhibited more pronounced cortisol concentrations, particularly during the recovery period. Although differences in lower-body WAnT performance were less pronounced, gymnasts still showed distinct hormonal response (higher increase in hGH) patterns compared to controls, reinforcing the influence of specific training adaptations. The obtained results for upper-body anaerobic performance were consistent with previous findings reported by Jemni et al. [43] and Franchini et al. [53], indicating that the engagement of specific muscle groups during sports training and competition directly influences their anaerobic capacity. Consequently, the application of the WAnT was justified in this investigation, as it provides a validated measure of short-term maximal anaerobic power that closely reflects the explosive strength demands associated with vaulting, tumbling, and apparatus dismounts typical of elite gymnastics. Additionally, baseline vitamin D status correlated positively with cortisol and negatively with testosterone responses, particularly in physically active controls following the lower-body WAnT, suggesting its potential role as a modulator of hormonal dynamics in response to anaerobic exercise involving non-dominant muscle groups.

Regular participation in high-intensity sports training, such as gymnastics, is well known to significantly impact endocrine homeostasis, influencing both acute and chronic secretion of anabolic and catabolic hormones [54]. The acute hormonal fluctuations observed in this study can be explained by the physiological mechanisms regulating hGH, testosterone, and cortisol secretion. hGH secretion is stimulated by physical stress, hypoglycemia, and increased metabolic demands characteristic of intense anaerobic exercise. Specifically, short-duration, high-intensity activities enhance sympathetic nervous system activity, stimulating hypothalamic release of growth hormone-releasing hormone and subsequently increasing pituitary hGH secretion. This response facilitates metabolic adaptation through increased lipolysis, protein synthesis, and carbohydrate metabolism. Testosterone secretion, regulated via the hypothalamic-pituitary-gonadal axis, typically increases acutely after high-intensity exercise due to luteinizing hormone-mediated stimulation of testicular steroidogenesis [55, 56]. However, the magnitude of testosterone increase may vary depending on total exercise volume and muscle mass involved. In contrast, cortisol, secreted by the adrenal cortex in response to adrenocorticotropic hormone and regulated by the hypothalamic-pituitary-adrenal axis, increases following exercise primarily due to metabolic stress, elevated energy demand, and psychological arousal. Higher cortisol concentrations observed in controls, particularly during recovery, likely reflect lower adaptation to anaerobic stress compared to elite gymnasts. Maximal anaerobic exercise involving large muscle groups triggers stress-associated physiological responses that significantly affect hormonal status.

It is well visible, especially in the case of groups that are not adapted to maximal exercise [57]. Despite this fact, when cortisol should reach higher concentrations, testosterone should not change dramatically during and within a short period after physical activity. This conclusion is similar to the results shown in the previously presented data [58]. However, some researchers have observed a short-term negative physiological relationship between serum cortisol and total testosterone levels. It remains difficult to determine why such a response occurs. According to Cumming et al. [59], cortisol disrupt the testicular steroidogenic process in Leydig cells due to enzymatic inhibition. It was especially visible 60 min after upper-body WAnT in both groups. Such observations were already supported by the research of Bambino and Hsueh [60], which has proven in animal models that there is a direct inhibitory effect of infused pharmacological dosages of cortisol on the luteinizing hormone receptor activity and content of the testes, and the inhibition of the testicular steroidogenic is dependent from the volume of the used dosage of the cortisol. Brownlee et al. [61] suggested there is need of a critical level of cortisol which should be reached in order to substantially influence circulating testosterone level.

During physical performance examination, it is essential to analyze not only testosterone or cortisol by itself, but more interestingly, the testosterone/cortisol ratio [30]. Measurement of the ratio between these two hormones may provide an answer about endocrine homeostasis during acute phase, which is why its results can be useful in the assessment of overtraining [62] or intensive emotional response during physical activity, which may directly affect physical performance [63]. The testosterone/cortisol ratio seems to be more sensitive to training stress than either testosterone or cortisol measured alone.

Analysis of testosterone/cortisol ratio showed significantly higher values only in the upper-body WAnT, especially immediately after the test (79%, p < 0.01).

It should also be noted that the observed differences in endocrine responses may be partially attributed to the distinct histological and functional profiles of upper and lower limb musculature. The lower limbs are predominantly composed of type I fibers, whereas the upper limbs—especially in gymnasts—often exhibit a higher proportion of type II fibers due to sport-specific adaptations related to muscular strength [64, 65]. These histological differences may influence the degree of metabolic stress and, consequently, the hormonal response. Another possible explanation is that upper-body exercises may elicit a stronger emotional or psychological response in gymnasts, given the dominant role of upper-body musculature in their sport-specific routines. This heightened arousal could further contribute to the endocrine patterns observed. However, this interpretation remains speculative and warrants further investigation.

Vitamin D concentration in testosterone/cortisol hormonal homeostasis

The relationship between 25(OH)D and serum testosterone has been widely studied. This is mostly because vitamin D receptors are located in the male reproductive tract tissues, including Leydig cells [66]. Furthermore, the development of hypogonadism in the absence of vitamin D receptors activity suggests that vitamin D plays a role in testosterone production and secretion [67]. However, results demonstrating a potential association between these parameters in sports are still contradictory. Some authors show that there is no association between vitamin D and testosterone changes - like in work of [27] or [68] and many other. By contrast, Lombardi et al. [69] and Budak et al. [70] showed positive correlations between vitamin D and free testosterone which may suggests adequate vitamin D levels may by associated with upregulation of steroidogenic enzymes, facilitating testosterone synthesis, whereas deficiency may impair this pathway.

In the current study, a significant association between baseline vitamin D levels and hormonal responses was observed exclusively in the control group following the lower-body WAnT. Specifically, in physically active controls, baseline 25(OH)D levels correlated positively with acute increases in cortisol and negatively with both testosterone levels and the testosterone/cortisol ratio immediately after exercise. No significant correlations between baseline vitamin D status and hormonal changes were found in gymnasts, neither in the lower- nor upper-body WAnT protocols. Vitamin D may modulate the stress response by suppressing activation of the hypothalamic–pituitary–adrenal axis, potentially reducing the release of corticotropin-releasing hormone and adrenocorticotropic hormone through its immunomodulatory and anti-inflammatory effects. Some studies suggest that sufficient vitamin D levels are associated with attenuated cortisol responses under stress conditions, possibly explaining the differential cortisol kinetics observed between the groups in this study.

Growth hormone concentration in testosterone/cortisol hormonal homeostasis

Many previous studies have shown that hGH levels in individuals fluctuate depending on various factors, including sex, age, stress, glucose state, food, hormonal balance, and exercises [71, 72]. Physical exercise can elicit marked elevations in hGH levels, similar to the modulation of testosterone and cortisol secretion.

The hGH activity is mainly linked to its main mediator, insulin-like growth factor 1 (IGF-1) [73]. IGF-1 circulates in the bloodstream, binds to insulin-like growth factor binding protein-3, and then performs its systemic functions by binding to specific receptors [74]. Several metabolic effects of growth hormone release (ex. changes in lipid and glucose metabolism, decrease in insulin concentration, increase in lipolysis and lipid oxidation, and stimulation of glucose uptake in the muscles and liver [75].

According to the results, there were no statistically significant differences between the groups before and after the upper- and lower- WAnTs. It may result from the fact that both groups did not differ significantly from each other (only observed difference’s where in few centimeters of high and percent body fat). Observing post-exercises response of hGH concentration clearly showed that regardless of the blood sample collection, the gymnasts had significantly higher concentrations of hGH than the control group, both in the upper and lower-body WAnTs. It may result from the changes of insulin release and glucose demands in gymnastic populations which muscles metabolism is well adapted to the increased energy demands [76]. The role of glucose status in the hGH concentration has been well established and proven [76]. However, it is likely that not only one factor affects the differences in post-exercises hGH levels. The hGH regulation during exercise at the neuroendocrine level is clearly multifactorial, but it seems that in such short time intervals, it is not affected by changes in cortisol and testosterone levels. Moreover, similar to many other studies, the absence of a relationship between 25(OH)D status and hGH change was confirmed [77]. In summary, our study did not show any short time-dependent relationships between hGH, cortisol, testosterone, and vitamin D concentrations during upper- and lower-body anaerobic exercise.

Strengths and limitations

One of the key strengths of this study is its novel design combining upper- and lower-body WAnT in a population of elite artistic gymnasts, allowing for a direct comparison of hormonal responses in relation to specific muscle group engagement [78, 79]. The integration of endocrine profiling with vitamin D status evaluation provides new insights into the hormonal adaptation mechanisms associated with high-intensity anaerobic exercise [13, 80]. Moreover, the inclusion of an age-matched, physically active control group enhances the internal validity and interpretability of the group-related effects. All exercise and biochemical assessment protocols were standardized and based on validated methodologies, contributing to the methodological robustness of the study.

However, the study had some limitations. Although the acute hormonal response induced by upper- and lower-body WAnT, was evaluated, the analysis was limited to three hormonal parameters that affect hormonal status and may influence the post-exercises response. This approach may not fully reflect the complexity of the adaptation process induced by many years of training, especially when considering the hormonal aspects of the process. The energy status during physical activity in any active population is reflected by insulin and glucagon secretion. These two hormones may regulate each other and affect testosterone, hGH, and cortisol secretion, liberating the glucose needed for anaerobic exercise [81].

Additionally, the blood samples were not adjusted for potential hemoconcentration effects, which could influence the interpretation of absolute hormone concentrations [57]. Therefore, future studies should consider plasma volume changes to more accurately reflect circulating hormone levels. The analysis was also limited to free testosterone, cortisol, and hGH, without assessing their precursors or metabolites. As cortisol and testosterone share structural similarities and are derived from a common precursor, their binding to carrier proteins may be altered by exercise-induced shifts in pH and temperature, potentially affecting the ratio of free to bound hormone and confounding interpretation [81].

Moreover, only serum 25-hydroxyvitamin D [25(OH)D] was assessed in this study, which, although commonly used to estimate vitamin D status, is not the most biologically active form [82]. Furthermore, the use of enzyme-linked immunosorbent assay (ELISA), while practical and widely adopted, offers lower specificity than the gold-standard method of liquid chromatography–mass spectrometry (LC-MS/MS) [83]. Future studies should consider using LC-MS/MS for more accurate vitamin D quantification.

Finally, the relatively small sample size, while justified by the elite nature of the gymnast population, may limit the generalizability of our findings. Future research should consider broader populations with diverse body types, training levels, age groups, and both sexes to confirm and expand on these observations.

Conclusions

In conclusion, this study demonstrated distinct acute hormonal responses to upper- and lower-body WAnT between elite artistic gymnasts and physically active controls, specifically for cortisol, testosterone, and hGH. Elite gymnasts demonstrated superior upper-body anaerobic performance, evidenced by significantly higher relative mean power values compared to controls. This enhanced capacity may have contributed to the greater post-exercise elevations in testosterone and hGH concentrations observed in this group. Conversely, the control group exhibited more pronounced cortisol responses during the recovery phase, particularly following upper-body WAnT. These findings underscore the role of training-specific adaptations and muscle group engagement in shaping endocrine responses to short-term, high-intensity anaerobic exercise.

Moreover, a significant inverse relationship between cortisol and testosterone concentrations during recovery was observed, indicating a potential antagonistic interaction between catabolic and anabolic hormonal responses following high-intensity exercise. In non-athletic participants, baseline vitamin D levels were significantly associated with the magnitude of cortisol and testosterone responses following upper-body WAnT, suggesting a modulatory role of vitamin D in regulating hormonal dynamics under conditions of acute physiological stress. These associations were absent in gymnasts, potentially reflecting the stabilizing influence of chronic training on hormonal homeostasis.

Future research should further elucidate these interactions using detailed endocrine profiling, including both free and total hormone concentrations, and investigate the mechanistic role of vitamin D in modulating the hypothalamic–pituitary–adrenal and hypothalamic–pituitary–gonadal axes in response to high-intensity anaerobic exercise.

Author contributions

Conceptualization: J.M., A.K. B.N., J.A., Data curation: J.M., P.B., M.K., B.N., Formal Analysis: P.B, B.N., R.G., A.K. Funding acquisition: J.M., A.K., P.S. Investigation (Data Collection): J.M., M.K., R.G., A.K., P.S. Methodology: J.M., A.K. B.N., J.A., Project Administration: J.M., P.B., M.K., A.K., R.G., Resources: J.M., P.B., M.K., B.N., A.K. Software: P.B., M.K., R.G., Supervision: J.M., B.N., J.A., A.K. Validation J.M., A.K., P.S. Visualization P.B., A.K. Writing – Original Draft: J.M. A.K., B.N. Writing – Review & Editing: J.M., P.B., M.K., B.N., J.A., A.K.

Funding

No external funding or sponsorship was received for the conduct of this study or the preparation of this article.

Data availability

The data will be made publicly available once the scientific project to which they pertain has been completed. Interested researchers may contact the corresponding authors to discuss data access in compliance with applicable ethical and institutional guidelines.

Declarations

Ethics approval and consent to participate

The study protocol was approved by the Bioethics Committee of the University of Physical Education and Sport in Gdańsk (Resolution No. 2, dated December 16, 2024) and complied with the principles of the Declaration of Helsinki. All participants were informed about the aims and procedures of the study and provided written informed consent before participation.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.