Abstract
Testosterone production is important in males, and various physical and psychological abnormalities occur in individuals with low testosterone levels. In the present study, we aimed to examine the effects of longitudinal changes in total testosterone levels in the same cohort. We included 178 male subjects who visited our hospital multiple times between 2018 and 2023 for medical checkups for at least 3 years. The median baseline age and total testosterone level (TT) of the cohort were 61 years and 4.74 ng/mL, respectively. The patients were divided into four groups based on the difference in TT (ΔTT) between baseline and last visit (Q1, n = 45; Q2, n = 45; Q3, n = 44; Q4, n = 44). ΔTT values ranged from –3.07 to –0.78 ng/mL in Q1, from –0.75 to –0.05 ng/mL in Q2, from –0.03 to 0.73 ng/mL in Q3, and from 0.75 ng/mL to 3.4 ng/mL in Q4. The median ΔTT were –1.22 for Q1, –0.35 for Q2, +0.19 for Q3, and +1.43 for Q4. Decreased TT tended to increase body weight, body mass index, waist circumference, and visceral fat (p for trend 0.0136, 0.0272, 0.0354, and 0.0032, respectively), and decrease adiponectin level (p for trend 0.0219). Herein, we found that decreased TT increases visceral fat and decreases adiponectin levels.
Introduction
Testosterone affects several aspects of physical and mental health [1]. Testosterone levels are frequently associated with the sex drive and play an important role in sperm production [2]. Testosterone is a steroid hormone, produced primarily in the testes of males, ovaries of females, and adrenal glands of both males and females. Moreover, testosterone plays a pivotal role in maintaining muscle mass, bone density, and red blood cell production [3]. Low testosterone levels have been associated with several symptoms, including decreased libido, reduced energy, weight gain, depression, moodiness, low self-esteem, and decreased body hair [4]. Testosterone levels decline with age [3], and considering that recent medical advancements have extended life spans [5], maintaining testosterone levels is crucial for sustaining a healthy life expectancy. In addition, testosterone affects fat metabolism and distribution. Low testosterone levels have been associated with an increased risk of central or visceral adiposity, characterized by fat deposition in the abdominal cavity, particularly around the visceral organs [6].
Late-onset hypogonadism (LOH) syndrome, in which testosterone levels decline with age, has garnered interest in recent years [7]. Symptoms of LOH syndrome vary, with obesity revealing that body fat percentage increases with age, especially visceral fat (VF) [8]. Increased VF is detrimental to health and increases the risk of metabolic syndromes [8], cardiovascular diseases [9], liver damage, and cancer [10]. Understanding the relationship between testosterone and VF is critical to elucidate the underlying physiological mechanisms and to develop targeted interventions to reduce the health risks associated with visceral obesity.
The negative correlation between testosterone and VF was reported by Seidell in 1990 [11], following which several cross-sectional studies have shown that elevated of VF can be observed in individuals with low testosterone levels [12]. Testosterone levels naturally fluctuate with age and are influenced by several external factors, including lifestyle, diet, and physical activity [7]. Similarly, VF increases and decreases over time and is influenced by aging, hormonal changes, and environmental factors [8]. Understanding the causal relationship between testosterone and VF requires the longitudinal tracking of these changes.
Although several studies have examined the clinical effects of testosterone replacement therapy (TRT) in patients with low testosterone levels [13], few reports have examined testosterone levels and clinical changes in healthy cohorts [14]. Furthermore, to date, there have been no reports examining the relationship between natural longitudinal changes in testosterone levels and clinical parameters, such as blood pressure, hemoglobin A1c (HbA1c), and cholesterol levels. In the present study, we aimed to examine the relationship between testosterone levels and VF and clinical data related to blood pressure, diabetes, and hyperlipidemia, which may be affected by testosterone changes.
Material and Methods
Study Population (Fig. 1)
Fig. 1
Inclusion criteria
Between April 2018 and April 2023, we retrospectively analyzed data from 382 males who had visited our clinic multiple times for medical checkups and had their total testosterone levels measured. Notably, our hospital incorporates testosterone measurements as part of a late-onset hypogonadism screening protocol, providing additional insights into testosterone profiles beyond standard medical checkups. We excluded patients with total testosterone levels <1.0 ng/mL to exclude those with extremely low testosterone levels (n = 6). In addition, we excluded patients who were receiving TRT (n = 16) and also excluded patients with <3 years between the first and last measurements to observe changes in total testosterone and clinical findings (n = 178).
Laboratory and clinical data
Height, body weight (BW), waist circumference (WC), and blood pressure were measured after an overnight fast. WC was measured at the level of the umbilicus based on the recommendation of the Japan Society for the Study of Obesity [15].
Blood samples were collected between 8 AM and 11 AM to measure fasting plasma glucose (FPG), fasting serum immunoreactive insulin (F-IRI), HbA1c, serum lipid concentrations, and total testosterone. FPG levels were measured using auto-analyzers equipped with a glucose oxidase-immobilized membrane and a hydrogen peroxide-sensing electrode (GA-1160 and GA-1171; Arkray, Kyoto, Japan). F-IRI levels were measured with auto-analyzers using a chemiluminescent enzyme immunoassay (Lumipulse and Lumipulse Presto; Fujirebio, Tokyo, Japan). HbA1c (National Glycohemoglobin Standardization Program) levels were measured using high-performance liquid chromatography (ADAMS; Arkray). The homeostasis model assessment of insulin resistance was calculated using the following formula: F-IRI (μU/mL) × FPG (mg/dL)/405. Serum adiponectin concentrations were measured using a latex particle-enhanced turbidimetric immunoassay (Human Adiponectin Latex Kit; Otsuka Pharmaceuticals, Tokyo, Japan). Serum triglyceride concentrations were measured using an enzymatic method (Determiner-L; Kyowa Medex, Tokyo, Japan). The assay systems used to measure serum concentrations of low-density lipoprotein cholesterol and high-density lipoprotein cholesterol were Determiner and Metabolead, respectively (Kyowa Medex). Serum activities of aspartate aminotransferase and alanine aminotransferase were measured with an Iatro LQ auto-analyzer (LSI Medience Corporation, Tokyo, Japan) using a kinetic method. Throughout the study period, the biochemical analyzers were rigorously calibrated using reference materials certified by a national agency to be traceable to international primary reference materials. The same batches of calibrators for plasma glucose and IRI were used throughout the study period, and batch-to-batch consistency of the HbA1c calibrator was confirmed by a national agency. Total testosterone (TT) levels were calculated using an electrochemiluminescence immunoassay (SRL, Inc., Tokyo, Japan).
The questionnaire asked whether the patients were taking diabetes, hyperlipidemia, or antihypertensive medications, and about their exercise habits. They were asked whether if they exercised at least three times a week for 20 minutes or more.
Imaging study
The VF area was computed and manually or automatically measured using commercial software for computed tomography scans taken at the umbilical level in the supine position, based on the Japanese guidelines for obesity treatment (Japan Society for the Study of Obesity, in Japanese) [15].
Statistical analyses
All data were analyzed using JMP® 17 (SAS Institute Inc., Cary, NC, USA). We analyzed within-subject changes from baseline to the last follow-up period using a paired t-test. P-values <0.05 were considered statistically significant. We analyzed four groups of patients according to the differences in changes in testosterone levels. We used analysis of variance to determine whether there were significant differences in the initial values of each parameter in the four groups. Fisher’s exact test was used to determine whether there were significant differences in the initial medications and exercise habits in the four groups. Covariates that were significant in this test were employed to analyze the effect of the difference in change in testosterone levels on the difference in change in each parameter using analysis of covariance. P for trend <0.05 was considered statistically significant.
Ethics statement
The study protocol was approved by the Institutional Review Board of Sumitomo Hospital (Reg. No. 2023-12).
Results
Participant characteristics (Tables 1, 2)
Table 1 Patient characteristics
|
Baseline median (IQRa) |
Last visit median (IQRa) |
p-value |
| Age (year) |
61 (55–69) |
65 (59–73) |
|
| Body weight (kg) |
70.2 (65.3–77.7) |
70.6 (64.8–77.4) |
0.0577 |
| BMI (kg/m2) |
24.7 (22.8–26.7) |
24.8 (22.8–27.1) |
0.8517 |
| Waist circumference (cm) |
89.4 (84–95.1) |
89.5 (84.3–95) |
0.4318 |
| Total testosterone (ng/mL) |
4.74 (3.86–5.89) |
4.70 (3.67–6.1) |
0.7914 |
| AST (U/L) |
22 (19–26) |
22 (19–26) |
0.8501 |
| ALT (U/L) |
20 (15–26) |
21 (15–26) |
0.2569 |
| TG (mg/dL) |
102 (73–143) |
93 (68–135) |
0.0501 |
| HDL cholesterol (mg/dL) |
60 (50–69) |
61 (53–72) |
0.0001 |
| LDL cholesterol (mg/dL) |
119 (103–136) |
118 (100–135) |
0.0402 |
| Adiponectin (μg/mL) |
7.2 (5.1–9.8) |
7.7 (6–11.2) |
0.0001 |
| HbA1c (%) |
5.7 (5.5–6.0) |
5.8 (5.6–6.2) |
0.0014 |
| Visceral fat (cm2) |
148 (102–189) |
137 (106–182) |
0.0568 |
| sBP (mmHg) |
125 (115–133) |
124 (116–134) |
0.3813 |
| dBP (mmHg) |
76 (69–81) |
73 (66–80) |
0.0038 |
| Fasting blood glucose (mg/dL) |
100 (95–110) |
104 (97–114) |
0.0616 |
| Insulin (μU/mL) |
6.6 (4.3–11.1) |
6.7 (4.6–10.2) |
0.4551 |
| HOMA-β |
61 (43–98) |
60 (38–89) |
0.1069 |
| HOMA-IR |
1.66 (1.17–2.84) |
1.76 (1.16–2.81) |
0.2513 |
Values are presented as medians (interquartile ranges).
IQR, interquartile range; BMI, body mass index; AST, aspartate aminotransferase; ALT, alanine aminotransferase; TG, triglyceride; HDL, high-density lipoprotein; LDL, low-density lipoprotein; HbA1c, hemoglobin A1c; sBP, systolic blood pressure; dBP, diastolic blood pressure; HOMA-β, homeostatic model assessment for β-cell function; HOMA-IR, homeostatic model assessment for insulin resistance.
Table 2 Patient characteristics
| Medication |
Baseline median |
Last visit median |
| Anti-hypertensive drug (%) |
38.2 (68/178) |
42.7 (76/178) |
| Anti-lipidemic drug (%) |
33.7 (60/178) |
35.4 (63/178) |
| Anti-diabetic drug (%) |
7.3 (13/178) |
10.1 (18/178) |
| Exercise habit (%) |
26.4 (47/178) |
28.6 (51/178) |
Exercise habits were defined as exercise of at least 20 minutes at least three times a week.
Tables 1 and 2 summarize the characteristics of the patients at baseline and last visit. At baseline, the median age was 61 years, and the median weight, BMI, WC, and VF were 70.2 kg, 24.7 kg/m2, 89.4 cm, and 148 cm2, respectively. The median baseline TT level was 4.74 ng/mL, and only six patients met the diagnostic criteria for LOH, <2.5 ng/mL, according to the Japanese LOH guidelines [16]. The median TT level at the last visit was 4.70 ng/mL. Only seven patients met the criteria for LOH. The comparison of the values at baseline and last visit revealed that HDL cholesterol (60 mg/dL vs. 61 mg/dL, p = 0.0001), HbA1c (5.7% vs. 5.8%, p = 0.0014), and adiponectin (7.2 μg/mL vs. 7.7 μg/mL, p = 0.0001) showed a significant increase and LDL cholesterol (119 mg/dL vs. 118 mg/dL, p = 0.0001) and diastolic blood pressure (76 mmHg vs. 73 mmHg, p = 0.0038) showed a significant decrease.
Relationship between changes in total testosterone (ΔTT) level and initial parameter values for each group (Tables 3, 4)
Table 3 Relationship between differences in testosterone levels (ΔTT) and initial parameter values for each
| Variables |
Q1 |
Q2 |
Q3 |
Q4 |
ANOVA |
| n = 45 |
n = 45 |
n = 44 |
n = 44 |
p-value |
| ΔTT (ng/mL) |
–1.22 |
–0.35 |
0.19 |
1.43 |
|
| Duration (month) |
47 |
47 |
46.5 |
47.5 |
0.1622 |
| iTT (ng/mL) |
5.79 |
4.5 |
4.28 |
4.68 |
0.0006 |
| Age (year) |
65 |
63 |
58 |
59 |
0.0325 |
| Body weight (kg) |
69 |
71.5 |
70.8 |
70.4 |
0.5998 |
| BMI (kg/m2) |
24.3 |
24.9 |
24.4 |
25.5 |
0.6307 |
| Waist circumference (cm) |
90 |
88 |
89.5 |
89.5 |
0.7742 |
| AST (U/L) |
21 |
23 |
22 |
21.5 |
0.7499 |
| ALT (U/L) |
19 |
22 |
18 |
21 |
0.8435 |
| TG (mg/dL) |
104 |
95 |
103 |
100 |
0.3176 |
| HDL cholesterol (mg/dL) |
57 |
61 |
60 |
63 |
0.4598 |
| LDL cholesterol (mg/dL) |
119 |
119 |
119 |
119 |
0.8371 |
| Adiponectin (μg/mL) |
7.2 |
7.5 |
7.2 |
7.0 |
0.9531 |
| HbA1c (%) |
5.6 |
5.9 |
5.6 |
5.7 |
0.4087 |
| Visceral fat (cm2) |
140 |
138 |
168 |
166 |
0.4395 |
| sBP (mmHg) |
125 |
124 |
125 |
122 |
0.4560 |
| dBP (mmHg) |
74 |
76 |
76 |
76 |
0.6224 |
| Fasting blood glucose (mg/dL) |
99 |
99 |
100 |
104 |
0.0381 |
| Insulin (μU/mL) |
6.4 |
5.6 |
5.6 |
8.2 |
0.0736 |
| HOMA-β |
60 |
61 |
54 |
71 |
0.7819 |
| HOMA-IR |
1.68 |
1.66 |
1.66 |
1.67 |
0.9025 |
Values are presented as medians.
ANOVA, analysis of variance; ΔTT, difference in total testosterone; iTT, initial total testosterone; BMI, body mass index; AST, aspartate aminotransferase; ALT, alanine aminotransferase; TG, triglyceride; HDL, high-density lipoprotein; LDL, low-density lipoprotein; HbA1c, hemoglobin A1c; sBP, systolic blood pressure; dBP, diastolic blood pressure; HOMA-β, homeostatic model assessment for β-cell function; HOMA-IR, homeostatic model assessment for insulin resistance.
Table 4 Relationship between differences in testosterone levels (ΔTT) and initial patient’s background for each
| Variables |
Q1 |
Q2 |
Q3 |
Q4 |
Fisher’s exact test |
| n = 45 |
n = 45 |
n = 44 |
n = 44 |
p-value |
| ΔTT (ng/mL) |
–1.22 |
–0.35 |
0.19 |
1.43 |
|
| Anti-hypertensive drug (%) |
11.8 |
9.6 |
9.6 |
7.3 |
0.4289 |
| Anti-lipidemic drug (%) |
8.4 |
6.7 |
10.1 |
8.4 |
0.5752 |
| Anti-diabetic drug (%) |
1.1 |
1.7 |
2.2 |
2.2 |
0.7742 |
| Exercise habit (%) |
7.9 |
6.2 |
5.1 |
7.3 |
0.6646 |
Values are presented as medians and percentage of patients.
ΔTT, difference in total testosterone; Exercise habits were defined as exercise of at least 20 minutes at least three times a week.
We divided the cohort into quartile groups (Q1 to Q4) according to the differences in TT levels (ΔTT). We found that baseline age, TT levels and fasting blood glucose differed significantly (baseline age: Q1 vs. Q2 vs. Q3 vs. Q4: 65 vs. 63 vs. 58 vs. 59, p = 0.0325; baseline TT level: Q1 vs. Q2 vs. Q3 vs. Q4: 5.79 ng/mL vs. 4.5 ng/mL vs. 4.28 ng/mL vs. 4.68 ng/mL, p = 0.0006; baseline fasting glucose level: Q1 vs. Q2 vs. Q3 vs. Q4: 99 mg/dL vs. 99 mg/dL vs. 100 mg/dL vs. 104 mg/dL, p = 0.0381). Medications and exercise habits did not affect testosterone changes.
Relationship between ΔTT and changes in each parameter (Table 5, Fig. 2)
Table 5 Relationship between changes in testosterone levels and changes in each parameter
| Variables |
Quartile of ΔTT |
p for trend |
Adjusted p for trend* |
| median (range), ng/mL |
| Q1 |
Q2 |
Q3 |
Q4 |
| –1.22 |
–0.35 |
0.19 |
1.43 |
| (–3.07 to –0.78) |
(–0.75 to –0.05) |
(–0.03 to 0.73) |
(0.75 to 3.4) |
| n = 45 |
n = 45 |
n = 44 |
n = 44 |
| ΔBody weight (kg) |
–0.2 |
0 |
–1 |
–0.55 |
0.0391 |
0.0136 |
| ΔBMI (kg/m2) |
0.2 |
0 |
–0.25 |
0.1 |
0.0596 |
0.0272 |
| ΔWaist circumference (cm) |
–0.5 |
2 |
–0.5 |
–1.5 |
0.0667 |
0.0354 |
| ΔAST (U/L) |
–1 |
–1 |
3 |
–0.5 |
0.4411 |
0.3930 |
| ΔALT (U/L) |
–1 |
–1 |
3 |
–1 |
0.9929 |
0.7437 |
| ΔTG (mg/dL) |
–6 |
–8 |
–14 |
–7.5 |
0.0896 |
0.1223 |
| ΔHDL cholesterol (mg/dL) |
1 |
2 |
5 |
2 |
0.3721 |
0.4109 |
| ΔLDL cholesterol (mg/dL) |
–5 |
–1 |
1.5 |
–3 |
0.3943 |
0.6345 |
| ΔAdiponectin (μg/mL) |
0.6 |
0.6 |
1.1 |
0.6 |
0.1000 |
0.0219 |
| ΔHbA1c (%) |
0.1 |
0.1 |
0.1 |
0.1 |
0.8099 |
0.9609 |
| ΔVisceral fat (cm2) |
11.5 |
0 |
–11.5 |
–13 |
0.0004 |
0.0032 |
| ΔsBP (mmHg) |
2 |
–1 |
2 |
1.5 |
0.4020 |
0.3938 |
| ΔdBP (mmHg) |
–1 |
–3 |
–2 |
–2.5 |
0.5409 |
0.3653 |
| ΔFasting blood glucose (mg/dL) |
7 |
1 |
4 |
2 |
0.0527 |
0.5353 |
| ΔInsulin (μU/mL) |
0.2 |
0.1 |
0.3 |
–0.65 |
0.4023 |
0.9651 |
| ΔHOMA-β |
–4.4 |
–5.4 |
–3.2 |
–5.1 |
0.5413 |
0.6071 |
| ΔHOMA-IR |
0.06 |
0.17 |
0.15 |
–0.01 |
0.6735 |
0.4659 |
Values are presented as medians.
*, adjusted for baseline age and testosterone level
Δ, difference; BMI, body mass index; AST, aspartate aminotransferase; ALT, alanine aminotransferase; TG, triglyceride; HDL, high-density lipoprotein; LDL, low-density lipoprotein; HbA1c, hemoglobin A1c; sBP, systolic blood pressure; dBP, diastolic blood pressure; HOMA-β, homeostatic model assessment for β-cell function; HOMA-IR, homeostatic model assessment for insulin resistance.
Fig. 2
Relationship between changes in testosterone levels and changes in each parameter. Visceral fat increase as testosterone levels decrease. Box-and-whisker plot showing the changes in each parameter. Body weight, body mass index (BMI), waist circumference, and visceral fat increased as testosterone levels decreased (p for trend body weight: 0.0136, BMI: 0.0272, waist circumference: 0.0354, visceral fat: 0.0032). Adiponectin levels showed decreased as testosterone levels decreased (p for trend 0.0219).
As baseline age and TT level affected the ΔTT levels, these three variables were used as covariates to examine the relationship between the ΔTT and the difference in change in each parameter. When TT level tended to increase, BW (difference in BW [ΔBW]: Q1 vs. Q2 vs. Q3 vs. Q4: –0.2 kg vs. 0 kg vs. –1 kg vs. –0.55 kg, p = 0.0136), BMI (difference in BMI [ΔBMI]: Q1 vs. Q2 vs. Q3 vs. Q4: 0.2 kg/m2 vs. 0 kg/m2 vs. –0.25 kg/m2 vs. 0.1 kg/m2, p = 0.0272), WC (difference in WC [ΔWC]: Q1 vs. Q2 vs. Q3 vs. Q4: –0.5 cm vs. 2 cm vs. –0.5 cm vs. –1.5 cm, p = 0.0354), and VF vales decreased (difference in VF [ΔVF]: Q1 vs. Q2 vs. Q3 vs. Q4: 11.5 cm2 vs. 0 cm2 vs. –11.5 cm2 vs. –13 cm2, p = 0.0032), whereas the adiponectin level tended to increase (difference in adiponectin [Δadiponectin]: Q1 vs. Q2 vs. Q3 vs. Q4: 0.6 μg/mL vs. 0.6 μg/mL vs. 1 μg/mL vs. 0.6 μg/mL, p = 0.0219).
Discussion
In the current study, we found that ΔTT negatively correlated with ΔVF, ΔBMI, and ΔWC (Graphical abstract). Several cross-sectional examinations have explored the relationship between testosterone levels, VF, and BW [12, 17, 18], revealing that low testosterone levels are associated with VF accumulation and increased insulin resistance [17]. Moreover, low testosterone levels decrease mitochondrial function and lead to the onset of metabolic syndrome [19].
Graphical Abstract
We also observed that lower testosterone levels were associated with lower adiponectin levels. Adiponectin, a protein hormone, and adipokine produced primarily in the adipose tissue, is involved in regulating glucose levels and fatty acid breakdown [20]. Low adiponectin levels are associated with obesity, insulin resistance, cardiovascular diseases, and dyslipidemia [21]. Testosterone and adiponectin are positively correlated in patients with diabetes [22] and those with obesity [23]. However, few studies have been conducted in healthy populations, and, to the best of our knowledge this is the first study to report a relationship between long-term changes in testosterone and adiponectin levels. Adiponectin prevents atherosclerosis, and it can be speculated that adiponectin production can be maintained by maintaining testosterone levels.
Although few studies have examined long-term changes in testosterone and their effects among healthy participants, some studies estimated the changes before and after TRT in patients with low testosterone levels [2, 12, 13, 24, 25]. The benefits of TRT in patients with type 2 diabetes and obesity accompanied by low testosterone have recently become apparent [24, 25]. In a study by Fui et al., patients with obesity and low testosterone levels were divided into testosterone-treated and non-treated groups with dietary restrictions, and their effects were examined [24]. Both groups experienced weight loss; however, VF was significantly lower in the testosterone-treated group than in the non-treated group. This finding that the artificial maintenance of testosterone may prevent VF gain. In the current study, we found that naturally decreased testosterone levels were associated with VF gain. If an increase in VF could be prevented by maintaining testosterone levels, this may prevent the development of related diseases.
Regarding adiponectin changes before and after TRT, some reports have indicated that TRT increases adiponectin levels [26], whereas others have suggested that TRT decreases adiponectin levels [27]. Currently, there have been no reports of adiponectin changes in individuals with normal testosterone levels, as demonstrated in the present study, and further studies are required.
Maintaining testosterone levels is crucial, and TRT improves erectile function, quality of life [13], and sarcopenia [28] in patients with low testosterone levels. In addition, the possibility of an increased incidence of cardiovascular events was previously suggested as a disadvantage of TRT; however, this has been disrupted in recent years in randomized trials, and TRT is considered a highly valuable treatment [29].
Moreover, few studies have explored natural changes in testosterone levels and their subsequent effects. For example, one study examined how sex hormones and visceral and subcutaneous fat are altered over 10 years in 190 Japanese Americans [14], while another explored the relationship between 5-year changes in testosterone levels and lifestyle habits in 1,382 participants in Australia [30]. Dinh et al. have shown that reduced testosterone levels were associated with elevated VF [14], which is consistent with the results of our study. Shi et al. have reported that a lower BMI was associated with higher testosterone levels and that being unmarried and obese were associated with lower testosterone levels [30]. In the current study, we found that a decrease in testosterone levels was associated with an increase in VF, BW, WC, and BMI, and an increase in testosterone levels was associated with an increase in adiponectin levels.
This study revealed that a longitudinal decrease in testosterone levels was associated with increased VF in addition to BMI, WC, and BW. However, these changes did not affect markers of glucose or lipid metabolism. This may be partly due to the short follow-up period and the fact that some of the patients were on diabetic and antilipidemic medications. Nevertheless, an increased in visceral fat due to decreased testosterone, even in markers of glucose and lipid metabolism were controlled, is considered a new finding. Furthermore, we find it very interesting that even when testosterone is within the normal range, its increase or decrease is closely related to the increase or decrease in visceral fat. In addition, we demonstrated that an increase in testosterone levels over time was associated with a decrease in adiponectin levels. This suggests that maintaining testosterone levels could reduce the increase in VF. Prevention of VF gain may prevent the development of metabolic syndromes, such as diabetes and hypertension, and potentially extend healthy life expectancy. We suggest that if VF can be reduced, it would prevent the decline in testosterone levels and various symptoms caused by low testosterone levels. Further long-term studies are required to confirm these findings.
This study has limitation. Regarding testosterone measurement, we only measured total testosterone. Testosterone exhibits androgenic effects primarily when free testosterone is converted to dihydrotestosterone. Although there are reports that free testosterone is associated with muscle strength [31], there is considerable debate regarding the measurement methods of free testosterone [32]. Therefore, only total testosterone was measured in this study.
Acknowledgments
Author Contribution
Conceptualization: SK, SF. Data curation: SK, HK. Formal analysis: SK, HK. Investigation: SK, AT, JS, TA, GT, YM, TF. Methodology: SK, HK, GT, TI, NU, KT. Project administration: SK, SF, YM. Supervision: SF, NN. Writing—original draft: SK. Writing—review & editing: SF, YM, NN.
Funding
None.
Data Sharing Statement
The data required to reproduce these findings cannot be shared at this time due to legal and ethical reasons.
Disclosure
Conflict of interest: None of the authors have any potential conflicts of interest associated with this research.
References
- 1 Zitzmann M (2020) Testosterone, mood, behaviour and quality of life. Andrology 8: 1598–1605.
- 2 Rizk PJ, Kohn TP, Pastuszak AW, Khera M (2017) Testosterone therapy improves erectile function and libido in hypogonadal men. Curr Opin Urol 27: 511–515.
- 3 Finkelstein JS, Lee H, Burnett-Bowie SA, Pallais JC, Yu EW, et al. (2013) Gonadal steroids and body composition, strength, and sexual function in men. N Engl J Med 369: 1011–1022.
- 4 Morales A, Bebb RA, Manjoo P, Assimakopoulos P, Axler J, et al. (2015) Diagnosis and management of testosterone deficiency syndrome in men: clinical practice guideline. Can Med Assoc J 187: 1369–1377.
- 5 Lutz W, Sanderson W, Scherbov S (2008) The coming acceleration of global population ageing. Nature 451: 716–719.
- 6 Tchernof A, Brochu D, Maltais-Payette I, Mansour MF, Marchand GB, et al. (2018) Androgens and the regulation of adiposity and body fat distribution in humans. Compr Physiol 8: 1253–1290.
- 7 Tsujimura A (2013) The relationship between testosterone deficiency and men’s health. World J Mens Health 31: 126–135.
- 8 Tchernof A, Després JP (2013) Pathophysiology of human visceral obesity: an update. Physiol Rev 93: 359–404.
- 9 Karlsson T, Rask-Andersen M, Pan G, Höglund J, Wadelius C, et al. (2019) Contribution of genetics to visceral adiposity and its relation to cardiovascular and metabolic disease. Nat Med 25: 1390–1395.
- 10 Katzmarzyk PT, Brown JC, Yang S, Mire EF, Wu XC, et al. (2022) Association of abdominal visceral adiposity and total fat mass with cancer incidence and mortality in white and black adults. Cancer Epidemiol Biomarkers Prev 31: 1532–1538.
- 11 Seidell JC, Björntorp P, Sjöström L, Kvist H, Sannerstedt R (1990) Visceral fat accumulation in men is positively associated with insulin, glucose, and C-peptide levels, but negatively with testosterone levels. Metabolism 39: 897–901.
- 12 Bianchi VE, Locatelli V (2018) Testosterone a key factor in gender related metabolic syndrome. Obes Rev 19: 557–575.
- 13 Corona G, Sforza A, Maggi M (2017) Testosterone replacement therapy: long-term safety and efficacy. World J Mens Health 35: 65–76.
- 14 Dinh KT, Amory JK, Matsumoto AM, Marck BT, Fujimoto WY, et al. (2020) Longitudinal changes in plasma sex hormone concentrations correlate with changes in CT-measured regional adiposity among Japanese American men over 10 years. Clin Endocrinol (Oxf) 93: 555–563.
- 15 Examination Committee of Criteria for ‘Obesity Disease’ in Japan; Japan Society for the Study of Obesity (2002) New criteria for ‘Obesity Disease’ in Japan. Circ J 66: 987–992.
- 16 Ide H, Akehi Y, Fukuhara S, Ohira S, Ogawa S, et al. (2023) Summary of the clinical practice manual for late-onset hypogonadism. Int J Urol 30: 422–430.
- 17 Su M, Wei H, Chen L, Guan Y, Dong W, et al. (2023) The impact of visceral adiposity on testosterone levels in American adult men: a cross-sectional analysis. Med Sci Monit 29: e941394.
- 18 Sebo ZL, Rodeheffer MS (2021) Testosterone metabolites differentially regulate obesogenesis and fat distribution. Mol Metab 44: 101141.
- 19 Pitteloud N, Mootha VK, Dwyer AA, Hardin M, Lee H, et al. (2005) Relationship between testosterone levels, insulin sensitivity, and mitochondrial function in men. Diabetes Care 28: 1636–1642.
- 20 Khoramipour K, Chamari K, Hekmatikar AA, Ziyaiyan A, Taherkhani S, et al. (2021) Adiponectin: structure, physiological functions, role in diseases, and effects of nutrition. Nutrients 13: 1180.
- 21 Zhao S, Kusminski CM, Scherer PE (2021) Adiponectin, leptin and cardiovascular disorders. Circ Res 128: 136–149.
- 22 Elsaied MA, Masallat D, Abdel-Hamid IA (2019) Correlation of adiponectin with testosterone in patients with and without type 2 diabetes and erectile dysfunction. Am J Mens Health 13: 155798831880704.
- 23 Muhammed AA, Eid RMHM, Mohammed WS, Abdel-Fadeil MR (2022) An association between adropin hormone and total testosterone in obese men: a case-control study. BMC Endocr Disord 22: 192.
- 24 Ng Tang Fui M, Prendergast LA, Dupuis P, Raval M, Strauss BJ, et al. (2016) Effects of testosterone treatment on body fat and lean mass in obese men on a hypocaloric diet: a randomised controlled trial. BMC Med 14: 153.
- 25 Hackett G, Cole N, Mulay A, Strange RC, Ramachandran S (2020) Long-term testosterone therapy in type 2 diabetes is associated with decreasing waist circumference and improving erectile function. World J Mens Health 38: 68–77.
- 26 Dudek P, Kozakowski J, Zgliczyński W (2020) The effects of testosterone replacement therapy in men with age-dependent hypogonadism on body composition, and serum levels of leptin, adiponectin, and C-reactive protein. Endokrynol Pol 71: 382–387.
- 27 Lanfranco F, Zitzmann M, Simoni M, Nieschlag E (2004) Serum adiponectin levels in hypogonadal males: influence of testosterone replacement therapy. Clin Endocrinol (Oxf) 60: 500–507.
- 28 Bhasin S, Ellenberg SS, Storer TW, Basaria S, Pahor M, et al. (2018) Effect of testosterone replacement on measures of mobility in older men with mobility limitation and low testosterone concentrations: secondary analyses of the Testosterone Trials. Lancet Diabetes Endocrinol 6: 879–890.
- 29 Lincoff AM, Bhasin S, Flevaris P, Mitchell LM, Basaria S, et al. (2023) Cardiovascular safety of testosterone-replacement therapy. N Engl J Med 389: 107–117.
- 30 Shi Z, Araujo AB, Martin S, O’Loughlin P, Wittert GA (2013) Longitudinal changes in testosterone over five years in community-dwelling men. J Clin Endocrinol Metab 98: 3289–3297.
- 31 Roy TA, Blackman MR, Harman SM, Tobin JD, Schrager M, et al. (2002) Interrelationships of serum testosterone and free testosterone index with FFM and strength in aging men. Am J Physiol Endocrinol Metab 283: E284–E294.
- 32 Bikle DD (2021) The free hormone hypothesis: when, why, and how to measure the free hormone levels to assess vitamin d, thyroid, sex hormone, and cortisol status. JBMR Plus 5: e10418.
https://orcid.org/0000-0002-0564-6404
