Abstract
The aim of this study was to determine the relationships between free testosterone (FT) level and parameters including laboratory data and data from questionnaires and to determine symptoms leading to the detection of late onset hypogonadism (LOH). We retrospectively reviewed medical records of patients in whom serum FT was measured in our hospital. Aging Male Symptoms (AMS) score, self-rating depression scale (SDS) and frequency scale for the symptoms of gastroesophageal reflux disease (FSSG) score were used for questionnaires. A total of 205 patients were included in the analysis (55.2 ± 15.6 years of age, mean ± SD). Among them, 119 patients (58.0%) had an FT level of less than 8.5 pg/mL, which fulfills the diagnostic criterion of LOH syndrome according to the clinical practice manual for LOH in Japan. It was revealed that FSSG score was inversely correlated to serum FT levels (r = –0.3395, p < 0.001), although SDS and AMS scales did not show significant correlations to FT levels. Our study revealed a high prevalence of LOH syndrome among patients in whom the majority complained of general symptoms. Although GERD symptoms are generally not considered to be typical symptoms of LOH, our study indicates that those symptoms might be clues for the detection of LOH.
LATE ONSET HYPOGONADISM (LOH) is a clinical and biochemical state with advancing age that is characterized by particular symptoms and a low level of serum testosterone [1, 2]. LOH syndrome is a condition that can affect the functions of multiple organ systems, and widely recognized clinical signs of LOH syndrome are erectile dysfunction, decrease in muscle strength, obesity, osteoporosis, anemia, depression and deterioration of insulin resistance [3-5]. In Japan, a clinical practice manual for LOH was published by the Japanese Urological Association and the Japanese Society for the Study of Aging Males [6].
In the European guidelines, the algorithm for diagnosis of LOH syndrome is based on total testosterone (TT) level [1]. However, several studies showed that there was no age-dependent decrease of TT level in Japanese men, and it was found in other studies that free testosterone (FT) level shows a linear decrease with aging in the Japanese population [7, 8]. Therefore, the clinical practice manual for LOH recommends the use of FT levels for the diagnosis of LOH in Japan [6].
Although measurement of FT is important for early detection and diagnosis of LOH syndrome, studies on the relationships between FT level and other parameters including laboratory data and questionnaires associated with LOH syndrome have been limited. Moreover, clinicians should know what symptoms will lead to the detection of low FT level and the diagnosis of LOH syndrome. The aim of the present study was to determine those relationships by investigating the symptoms and other parameters in patients in whom serum FT level was measured.
Patients and Methods
Study subjects
We retrospectively reviewed medical records of 273 patients in whom serum FT level was measured between January 2014 and December 2017 in Okayama University Hospital. The following patients were excluded from this study: female patients, patients younger than 20 years, patients with primary or secondary hypopituitarism, and those who were receiving androgen replacement therapy (ART). Patient selection is shown in Fig. 1. During the study period, serum FT was measured in 273 patients; 205 patients were found to be eligible and were included in the analysis. Serum FT was measured once in 154 patients and multiple times in 51 patients during the study period. When FT was measured multiple times in one patient in the study period, the first measurement during the study period was used for analysis. This study was approved by the Ethical Committee of Okayama University Hospital and adhered to the Declaration of Helsinki (No. 1703-026).
Study protocol
Crucial symptoms that were triggers for measurement of serum FT concentrations were reviewed and evaluated, and the major one to three symptoms from each patient were used for analysis. The symptoms were categorized into six groups: general symptoms, psychological symptoms, urological symptoms, musculoskeletal symptoms, gastrointestinal symptoms and others. We also compared FT values in each group. In addition, according to the FT values, patients were categorized into three groups based on cut-off values of 8.5 and 11.8 pg/mL. These cut-off values were determined in a previous study in which the mean value of FT in Japanese young adults in their 20 s was calculated and in which it was shown that the standard value for diagnosis of LOH is FT of less than 8.5 pg/mL, and that for diagnosis of LOH borderline is FT of less than 11.8 pg/mL [7]. In our study, the FT < 8.5 pg/mL group, 8.5 ≤ FT < 11.8 pg/mL group, and 11.8 ≤ FT pg/mL group were defined as the LOH group, Borderline group, and Non-LOH group, respectively.
Analysis of biochemical and radiological markers
Data were obtained from hospital medical records. Body mass index (BMI), blood biochemical data including HbA1c, low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), triglycerides (TG), c-reactive protein (CRP), total testosterone (TT), and dehydroepiandrosterone sulfate (DHEA-S), and radiological data including dual energy X-ray absorptiometry % young adult mean (DEXA %YAM) in the femoral bone and lumbar spine were evaluated. Biochemical data obtained within 2 weeks from the day when FT was measured were used.
Aging Male Symptoms (AMS) score was evaluated when LOH syndrome was suspected. In addition, patients who visited the outpatient department of general internal medicine (GIM) in our hospital were routinely asked to complete some questionnaires including questionnaires on self-rating depression scale (SDS) and frequency scale for the symptoms of gastroesophageal reflux disease (FSSG) score, and those data were also evaluated. FSSG score is commonly used to assess symptoms of GERD [9]. The FSSG questionnaire is a questionnaire comprising 12 questions including 7 questions for assessing acid reflux symptoms and 5 questions for assessing dysmotility-related symptoms, and an FSSG score ≥8 is considered to indicate probable GERD [9, 10]. FT values were measured by a radioimmunoassay using a commercially available kit (Beckman Coulter, Brea, CA, USA).
Statistical analysis
The data were subjected to ANOVA and linear regression analysis to determine differences. If differences were detected by ANOVA, Tukey-Kramer’s post-hoc test was used to determine which means differed. p-values <0.05 were considered statistically significant. All statistical analyses were performed using EZR (Saitama Medical Center, Jichi University, Saitama, Japan), a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria, ver. 3.1.1) [11].
Results
Patients’ characteristics and serum free testosterone (FT) levels
The primary departments and patients’ ages are summarized in Table 1. The mean ages of patients were 55.2 ± 15.6 (range: 20–89) years in all cases and 59.2 ± 15.7 (range: 22–89) years in LOH cases. FT was most frequently measured in the GIM department (61.5% of all cases, 61.3% of LOH cases), followed by the urology department (23.4% of all cases, 24.4% of LOH cases). Table 2 shows the main symptoms of patients when FT was measured in all cases and in LOH cases. Among all patients, frequent symptoms when FT was measured were general fatigue (22.0%), dizziness (9.4%), erectile dysfunction (8.4%), weight loss (7.3%), hot flashes (6.3%) and headache (6.3%). In LOH patients, frequent symptoms were general fatigue (22.1%), weight loss (9.8%), dizziness (9.8%), erectile dysfunction (9.2%), hot flashes (6.7%) and depression (6.1%).
Table 1
| Primary departments and patients’ age |
All cases |
LOH cases |
| Number of patients |
205 patients (%) |
119 patients (%) |
| Primary departments |
|
|
| General internal medicine |
126 (61.5%) |
73 (61.3%) |
| Urology |
48 (23.4%) |
29 (24.4%) |
| Endocrinology |
11 (5.4%) |
6 (5.0%) |
| Other internal medicine |
20 (9.8%) |
11 (9.2%) |
| Age ± SD (range) |
55.2 ± 15.6 (20–89) |
59.2 ± 15.7 (22–89) |
Table 2
|
All cases |
LOH cases |
| Main symptoms |
287 symptoms (%) |
163 symptoms (%) |
| General symptoms |
130 (45.3%) |
76 (46.6%) |
| General fatigue |
63 (22.0%) |
36 (22.1%) |
| Weight loss |
21 (7.3%) |
16 (9.8%) |
| Hot flashes |
18 (6.3%) |
11 (6.7%) |
| Loss of appetite |
11 (3.8%) |
6 (3.7%) |
| Fever |
10 (3.5%) |
4 (2.5%) |
| General pain |
7 (2.4%) |
3 (1.8%) |
| Psychological symptoms |
69 (24.0%) |
40 (24.5%) |
| Dizziness |
27 (9.4%) |
16 (9.8%) |
| Headache |
18 (6.3%) |
8 (4.9%) |
| Depression |
15 (5.2%) |
10 (6.1%) |
| Anxiety |
9 (3.1%) |
6 (3.7%) |
| Urological symptoms |
38 (13.2%) |
24 (14.7%) |
| Erectile dysfunction |
24 (8.4%) |
15 (9.2%) |
| Ejaculation disorder |
5 (1.7%) |
3 (1.8%) |
| Testicular disorder |
4 (1.4%) |
2 (1.2%) |
| Frequent urination |
3 (1.0%) |
2 (1.2%) |
| Decreased libido |
2 (0.7%) |
2 (1.2%) |
| Musculoskeletal symptoms |
19 (6.6%) |
8 (4.9%) |
| Numbness |
12 (4.2%) |
3 (1.8%) |
| Muscle weakness |
7 (2.4%) |
5 (3.1%) |
| Gastointestinal symptoms |
16 (5.6%) |
9 (5.5%) |
| Heartburn |
6 (2.1%) |
2 (1.2%) |
| Abdominal pain |
4 (1.4%) |
3 (1.8%) |
| Abdominal distension |
2 (0.7%) |
1 (0.6%) |
| Dysphagia |
2 (0.7%) |
2 (1.2%) |
| Nausea |
2 (0.7%) |
1 (0.6%) |
| Others |
15 (5.2%) |
6 (3.7%) |
When FT was categorized into three groups based on cut-off values of 8.5 and 11.8 pg/mL, the LOH group (FT < 8.5 pg/mL) had 119 patients (58.0%), the Borderline group (8.5 ≤ FT < 11.8 pg/mL) had 53 patients (25.9%), and the Non-LOH group (11.8 ≤ FT pg/mL) had 33 patients (16.1%; Fig. 2A). The numbers of patients less than 50 years of age were 37 (31.0%) in the LOH group, 22 (41.5%) in the Borderline group, and 24 patients (72.7%) in the Non-LOH group. Correlation between FT and age is shown in Fig. 2B. Linear regression analysis showed that FT decreased with advance of age (r = –0.4191, p < 0.001). Fig. 3 shows a comparison of the main symptoms of patients who visited the GIM department and other departments. Among the patients who visited GIM, frequent symptoms were general symptoms (52%), psychological symptoms (28%), and gastrointestinal symptoms (8%). Frequent symptoms of patients who visited other departments were general symptoms (34%), urological symptoms (32%), and psychological symptoms (17%).
Fig. 4 shows the correlations between FT and each categorized group in all cases and in LOH cases. The differences in FT between categorized groups were not statistically significant. The correlation coefficients between FT and parameters are shown in Table 3. Total testosterone levels were significantly correlated with FT levels (r = 0.5588, p < 0.001). The correlations of FT with SDS, AMS scale and FSSG score and the correlation of TT with FSSG score are shown in Fig. 5. Although SDS and AMS scale did not show significant correlations with FT levels (SDS: r = 0.0717, p = 0.4527; AMS scale: r = 0.1966, p = 0.4061), FSSG score was inversely correlated with FT levels (r = –0.3395, p < 0.001). In addition, FSSG score tended to be high in patients with low TT levels; however, it was not statistically significant (r = –0.4312, p = 0.0653). FSSG scores were higher in patients with low FT levels, and the average FSSG scores in the LOH group, Borderline group and Non-LOH group were 10.69 ± 8.32, 6.00 ± 5.62 and 4.79 ± 4.84, respectively (Fig. 5E). The LOH group had a statistically higher FSSG score than those in the Borderline group (p = 0.03) and Non-LOH group (p < 0.01).
Table 3
| Correlations between the level of free
testosterone and related markers |
| Factors |
Number |
r |
p values |
| BMI |
192 |
0.1145 |
0.1137 |
| HbA1c |
157 |
–0.0845 |
0.2940 |
| LDL-C |
139 |
0.1055 |
0.2183 |
| HDL-C |
127 |
0.0593 |
0.5095 |
| TG |
141 |
0.1515 |
0.0739 |
| CRP |
146 |
–0.0976 |
0.2413 |
| Total testosterone |
77 |
0.5588 |
<0.001** |
| DHEA-S |
12 |
0.3222 |
0.3070 |
| DEXA %YAM (Femoral) |
24 |
0.3637 |
0.0806 |
| DEXA %YAM (Lumbar) |
24 |
0.3695 |
0.0756 |
BMI, body mass index; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; TG, triglycerides; CRP, c-reactive protein; DHEA-S, dehydroepiandrosterone sulfate; DEXA %YAM, dual energy X-ray absorptiometry % young adult mean. **p < 0.01: significant correlations between the indicated factors.
Discussion
In our study, among the patients who were suspected of having LOH syndrome and in whom FT level was measured, 119 patients (58.0%) had an FT level of less than 8.5 pg/mL, which fulfills the diagnostic criterion of LOH syndrome according to the clinical practice manual for LOH in Japan [6]. Among those patients, 37 patients (31.0%) were less than 50 years of age. Considering that the common symptoms of our patients were systemic symptoms such as fatigue and weight loss, these findings suggest a high prevalence of LOH syndrome in middle-age males who visit hospital with general complaints. We analyzed the associations between FT levels and categories of symptoms but did not find any significant association. This finding may imply that a deficiency of testosterone can widely affect the functions of multiple organs including psychological and physical systems [1].
SDS and AMS scale are frequently used to assess the severity of depression and LOH syndrome, respectively [7, 12, 13]. Since typical symptoms of LOH include depression, they are believed to be helpful for LOH screening [5, 6, 14]. However, in our study, no significant association between FT and SDS or AMS scale was found. Since our results together with results of previous studies suggest that SDS or AMS scale has a poor degree of specificity, these scales alone are not ideal tools for screening latent conditions of testosterone deficiency [15-17].
In the present study, it was of note that FSSG score showed a linear decrease with FT level. To the best of our knowledge, the present study is the first study showing a negative correlation between androgen level and GERD symptoms. Although the direct effect of androgen on GERD is still unknown, several studies have been performed to determine the relationships between sex hormone and gastroesophageal diseases. Men develop esophageal adenocarcinoma more frequently than do women [18]. Barrett’s esophagus is a known precursor to the development of esophageal adenocarcinoma, and a positive association between FT level and Barrett’s esophagus has been shown [19, 20]. In women, hormone replacement therapy and pregnancy are associated with GERD symptoms, suggesting that estrogen plays a role in precipitating GERD; however, a direct relationship between female sex hormone levels and GERD has not been shown [21].
Although the etiology of the negative association between FT level and GERD symptoms is unclear, there are several possible explanations. First, psychological stress might be related to both LOH and exacerbation of GERD symptoms. Psychosocial stress and work-related stress are thought to decrease testosterone levels in males [22, 23]. In addition, some studies have suggested that psychosocial stress has a significant relationship with GERD, exacerbating heartburn symptoms by enhancing perceptual response to acid exposure in the esophagus [24, 25]. Furthermore, GERD itself is also associated with depression and anxiety due to poor sleep quality, which in turn may lead to a decrease of androgen levels [26].
Secondly, obesity is related to both conditions of LOH syndrome and GERD. An association between LOH and obesity was previously reported [27, 28]. Obesity is also a risk factor for GERD because surrounding adipose tissue compresses the stomach, leading to an increase in intragastric pressure and subsequent relaxation of the lower esophageal sphincter [29]. Although no significant association between FT and BMI was found in our study, this may explain a direct interrelationship between FT level and FSSG scores. Finally, the potential mechanism of our hypothesis is related to inflammation. Testosterone is thought to play an anti-inflammatory role [30]. Given that the pathogenesis of GERD may be cytokine-mediated inflammation rather than the result of chemical injury [31], these findings suggest that lowered endogenous testosterone in LOH syndrome might be associated with delayed wound healing of the esophagus due to the lack of an anti-inflammatory effect, inducing GERD symptoms.
There were some limitations of the present study. Firstly, the sample size was relatively small. Secondly, the reason for measuring FT level in each patient could not be fully revealed. Thirdly, not all FT levels were measured early in the morning. The situation and exact time when blood samples were drawn were not completely determined. Finally, this was a retrospective study, being conducted in a single, tertiary care center; thus, the results obtained in this study may not be directly interpreted for general hospitals or primary care clinics.
In conclusion, our study revealed a prevalence of LOH syndrome among patients in whom the majority complained of general symptoms. We also found that a high FSSG score was associated with low FT level. Although GERD symptoms are generally not considered to be typical symptoms of LOH, our study indicates that those symptoms might be clues for the detection of LOH syndrome. Further study is needed to reveal the relationship between LOH syndrome and GERD.
Acknowledgements
We are sincerely grateful to all of the clinical staff in the Department of General Medicine who contributed to the present work.
Disclosure
The authors declare no conflicts of interest in association with the present study.
References
- 1 Nieschlag E, Swerdloff R, Behre HM, Gooren LJ, Kaufman JM, et al. (2006) Investigation, treatment, and monitoring of late-onset hypogonadism in males: ISA, ISSAM, and EAU recommendations. J Androl 27: 135–137.
- 2 Wu FC, Tajar A, Beynon JM, Pye SR, Silman AJ, et al. (2010) Identification of late-onset hypogonadism in middle-aged and elderly men. N Engl J Med 363: 123–135.
- 3 Lunenfeld B, Mskhalaya G, Zitzmann M, Arver S, Kalinchenko S, et al. (2015) Recommendations on the diagnosis, treatment and monitoring of hypogonadism in men. Aging Male 18: 5–15.
- 4 Bassil N, Alkaade S, Morley JE (2009) The benefits and risks of testosterone replacement therapy: a review. Ther Clin Risk Manag 5: 427–448.
- 5 Wang C, Nieschlag E, Swerdloff R, Behre HM, Hellstrom WJ, et al. (2009) Investigation, treatment and monitoring of late-onset hypogonadism in males. Int J Androl 32: 1–10.
- 6 Namiki M, Akaza H, Shimazui T, Ito N, Iwamoto T, et al. (2008) Clinical practice manual for late-onset hypogonadism syndrome. Int J Urol 15: 377–388.
- 7 Iwamoto T, Yanase T, Horie H, Namiki M, Okuyama A (2009) Late-onset hypogonadism (LOH) and androgens: validity of the measurement of free testosterone levels in the diagnostic criteria in Japan. Int J Urol 16: 168–174.
- 8 Okamura K, Ando F, Shimokata H (2005) Serum total and free testosterone level of Japanese men: a population-based study. Int J Urol 12: 810–814.
- 9 Kusano M, Shimoyama Y, Sugimoto S, Kawamura O, Maeda M, et al. (2004) Development and evaluation of FSSG: frequency scale for the symptoms of GERD. J Gastroenterol 39: 888–891.
- 10 Nonaka T, Kessoku T, Ogawa Y, Yanagisawa S, Shiba T, et al. (2013) Comparative study of 2 different questionnaires in Japanese patients: the Quality of Life and Utility Evaluation Survey Technology Questionnaire (QUEST) versus the Frequency Scale for the Symptoms of Gastroesophageal Reflux Disease Questionnaire (FSSG). J Neurogastroenterol Motil 19: 54–60.
- 11 Kanda Y (2013) Investigation of the freely available easy-to-use software ‘EZR’ for medical statistics. Bone Marrow Transplant 48: 452–458.
- 12 Daig I, Heinemann LA, Kim S, Leungwattanakij S, Badia X, et al. (2003) The Aging Males’ Symptoms (AMS) scale: review of its methodological characteristics. Health Qual Life Outcomes 1: 77.
- 13 Okada K, Miyake H, Ishida T, Sumii K, Enatsu N, et al. (2016) Improved lower urinary tract symptoms associated with testosterone replacement therapy in Japanese men with late-onset hypogonadism. Am J Mens Health (in press).
- 14 Bhasin S, Cunningham GR, Hayes FJ, Matsumoto AM, Snyder PJ, et al. (2006) Testosterone therapy in adult men with androgen deficiency syndromes: an endocrine society clinical practice guideline. J Clin Endocrinol Metab 91: 1995–2010.
- 15 Hamanoue N, Tanabe M, Tanaka T, Akehi Y, Murakami J, et al. (2017) A higher score on the Aging Males’ Symptoms scale is associated with insulin resistance in middle-aged men. Endocr J 64: 521–530.
- 16 Morley JE, Perry HM 3rd, Kevorkian RT, Patrick P (2006) Comparison of screening questionnaires for the diagnosis of hypogonadism. Maturitas 53: 424–429.
- 17 Blumel JE, Chedraui P, Gili SA, Navarro A, Valenzuela K, et al. (2009) Is the Androgen Deficiency of Aging Men (ADAM) questionnaire useful for the screening of partial androgenic deficiency of aging men? Maturitas 63: 365–368.
- 18 Xie SH, Lagergren J (2016) The male predominance in esophageal adenocarcinoma. Clin Gastroenterol Hepatol 14: 338.e331–347.e331.
- 19 Cook MB, Wood SN, Cash BD, Young P, Acosta RD, et al. (2015) Association between circulating levels of sex steroid hormones and Barrett’s esophagus in men: a case-control analysis. Clin Gastroenterol Hepatol 13: 673–682.
- 20 Ek WE, Lagergren K, Cook M, Wu AH, Abnet CC, et al. (2016) Polymorphisms in genes in the androgen pathway and risk of Barrett’s esophagus and esophageal adenocarcinoma. Int J Cancer 138: 1146–1152.
- 21 Menon S, Prew S, Parkes G, Evans S, Smith L, et al. (2013) Do differences in female sex hormone levels contribute to gastro-oesophageal reflux disease? Eur J Gastroenterol Hepatol 25: 772–777.
- 22 Nilsson PM, Moller L, Solstad K (1995) Adverse effects of psychosocial stress on gonadal function and insulin levels in middle-aged males. J Intern Med 237: 479–486.
- 23 Singer F, Zumoff B (1992) Subnormal serum testosterone levels in male internal medicine residents. Steroids 57: 86–89.
- 24 Levy RL, Olden KW, Naliboff BD, Bradley LA, Francisconi C, et al. (2006) Psychosocial aspects of the functional gastrointestinal disorders. Gastroenterology 130: 1447–1458.
- 25 Song EM, Jung HK, Jung JM (2013) The association between reflux esophagitis and psychosocial stress. Dig Dis Sci 58: 471–477.
- 26 On ZX, Grant J, Shi Z, Taylor AW, Wittert GA, et al. (2017) The association between gastroesophageal reflux disease with sleep quality, depression, and anxiety in a cohort study of Australian men. J Gastroenterol Hepatol 32: 1170–1177.
- 27 Kaplan SA, Meehan AG, Shah A (2006) The age related decrease in testosterone is significantly exacerbated in obese men with the metabolic syndrome. What are the implications for the relatively high incidence of erectile dysfunction observed in these men? J Urol 176: 1524–1527; discussion 1527–1528.
- 28 Corona G, Vignozzi L, Sforza A, Mannucci E, Maggi M, (2015) Obesity and late-onset hypogonadism. Mol Cell Endocrinol 418 Pt 2: 120–133.
- 29 El-Serag H (2008) The association between obesity and GERD: a review of the epidemiological evidence. Dig Dis Sci 53: 2307–2312.
- 30 Schmidt M, Naumann H, Weidler C, Schellenberg M, Anders S, et al. (2006) Inflammation and sex hormone metabolism. Ann N Y Acad Sci 1069: 236–246.
- 31 Dunbar KB, Agoston AT, Odze RD, Huo X, Pham TH, et al. (2016) Association of acute gastroesophageal reflux disease with esophageal histologic changes. JAMA 315: 2104–2112.
