Hypogonadotropic hypogonadism among a population of obese men: Prevalence, risk factors and reversibility after weight loss induced by bariatric surgery

Hypogonadotropic hypogonadism among a population of obese men: Prevalence, risk factors and reversibility after weight loss induced by bariatric surgery

e-SPEN Journal 8 (2013) e37ee43 Contents lists available at SciVerse ScienceDirect e-SPEN Journal journal homepage: http://www.elsevier.com/locate/c...

226KB Sizes 0 Downloads 7 Views

e-SPEN Journal 8 (2013) e37ee43

Contents lists available at SciVerse ScienceDirect

e-SPEN Journal journal homepage: http://www.elsevier.com/locate/clnu

Original article

Hypogonadotropic hypogonadism among a population of obese men: Prevalence, risk factors and reversibility after weight loss induced by bariatric surgery Vanessa Ippersiel a, Ariane Lepot b, Damien Gruson b, Jacques Jamart c, Orsalia Alexopoulou b, Dominique Maiter b, Jean-Paul Thissen b, * a

Department of Internal Medicine, Grand Hôpital de Charleroi, Gilly 6060, Belgium Department of Endocrinology and Nutrition, Saint-Luc Academic Hospital, Institut de Recherches Expérimentales et Cliniques (IREC), Pôle d’Endocrinologie, Diabétologie et Nutrition (EDIN), Université Catholique de Louvain, Avenue Hippocrate, B1.55.06, B-1200 Brussels, Belgium c Scientific Support Unit, Centre Hospitalier Universitaire de Mont-Godinne, Université Catholique de Louvain, Yvoir 5530, Belgium b

a r t i c l e i n f o

s u m m a r y

Article history: Received 16 July 2012 Accepted 30 December 2012

Background and aims: Obesity in men is frequently associated with low levels of testosterone, loss of libido and/or erectile dysfunction. Our goal was to estimate the prevalence of hypogonadism among obese men and to determine its risk factors and reversibility after bariatric surgery. Methods: Seventy-five obese men were studied at baseline. Metabolic and hormonal parameters were measured, body composition was assessed by bioelectrical impedance and hypogonadism was evaluated by the ADAM (Androgen Deficiency in Aging Males) questionnaire. Twenty-one patients were reevaluated after bariatric surgery. Results: At baseline, 39% of obese men had hypotestosteronemia, while symptoms of androgen deficiency were present in 93%. Total Testosterone (TT) was inversely related to body mass index (BMI) (P < 0.05), waist circumference (P ¼ 0.012) and body fat mass (P ¼ 0.022). Bariatric surgery was associated with an increase in TT (P ¼ 0.001) and decreases in estradiol (E2) (P ¼ 0.008) and in the E2/TT ratio (P ¼ 0.001). Conclusions: Low testosterone levels are frequently observed among morbidly obese men and are correlated with the degree of abdominal adiposity, but not strongly with the presence of sexual dysfunction. Bariatric surgery leads to normalized TT and to decreased E2 and E2/TT ratio, suggesting a role of excessive aromatization in the hypotestosteronemia associated with obesity. Ó 2013 European Society for Clinical Nutrition and Metabolism. Published by Elsevier Ltd. All rights reserved.

Keywords: Hypogonadism Obesity Bariatric surgery

1. Introduction Male hypogonadism results from the failure of the testis to produce physiologic levels of testosterone and/or a normal number of spermatozoa due to disruption at one or more levels of the hypothalamo-pituitary-gonadal axis. The diagnosis of hypogonadism is based on the presence of symptoms and/or clinical signs of androgen deficiency, combined with low circulating levels of total testosterone.1 In the general population, the prevalence of hypogonadism is about 5%,2 but this condition is particularly frequent among obese male subjects. Some reports have shown prevalence as high as 58% of hypotestosteronemia among obese men,3 while others have found a more conservative value. The relationship between obesity and hypogonadism seems to be a reflective one and the direction of causality remains unclear. On * Corresponding author. Tel.: þ32 2 764 54 69; fax: þ32 2 764 54 79. E-mail address: [email protected] (J.-P. Thissen).

one hand, total testosterone levels are negatively correlated with body mass index (BMI),3,4 even after adjustment for comorbidities,5 suggesting that obesity itself causes testosterone to decline. Furthermore, not only obesity but also its comorbidities such as metabolic syndrome and type 2 diabetes predict low testosterone, independently of BMI.6 On the other hand, low testosterone levels predict a higher risk of excessive gain weight and the development of the metabolic syndrome and diabetes.7,8 In addition, in some randomized control trials, testosterone replacement therapy is associated with improvement in metabolic syndrome.6 This suggests that low testosterone might exacerbate the risk of obesity and its complications,9 creating a vicious circle. Since the relationship between obesity and hypogonadism can be bidirectional, we undertook the study to delineate the reversibility of hypogonadism in response to the dramatic weight loss caused by bariatric surgery. Pathogenic mechanisms linking low testosterone levels with obesity appear to be complex. Insulin resistance, decreased sex hormone binding-globulin (SHBG), decreased luteinizing hormone (LH)

2212-8263/$36.00 Ó 2013 European Society for Clinical Nutrition and Metabolism. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clnme.2012.12.003


V. Ippersiel et al. / e-SPEN Journal 8 (2013) e37ee43

secretion and excessive aromatization have all been proposed to contribute to the low testosterone levels in obese subjects.10 Similarly, the risk factors for obesity-related hypogonadism are not well defined.11 The goal of our study was to analyze in the same population the prevalence and risk factors of obesity-related hypogonadism and its reversibility by weight loss obtained by bariatric surgery. 2. Subjects and methods 2.1. Study subjects All obese men aged between 18 and 65 years and seeking for bariatric surgery in the Obesity clinic of our institution between February 2007 and March 2010 were invited to participate in this study (n ¼ 93). An oral consent was obtained from all the participants to the study. Patients with a known hypogonadism of other etiology than obesity (n ¼ 3), with a known endocrinopathy other than diabetes (n ¼ 1), using drugs suspected to interfere with gonadal function (n ¼ 1), drug-addicted (n ¼ 1) or with incomplete data (n ¼ 2) were excluded from the study. Ten patients did not undergo initial biological assessment. Finally, 75 patients were included and evaluated at baseline (transversal study). Among them, 21 subjects underwent bariatric surgery (a laparoscopic Roux-en-Y Gastric bypass in 19 and a sleeve gastrectomy in 2), mainly for morbid obesity (19/21), and were re-evaluated 3 and 12 months after surgery (longitudinal study). 2.2. Methods Measurements of height, weight, waist circumference, arterial blood pressure and body composition were performed in all 75 patients at baseline. The BMI was calculated as the ratio of weight (kg) divided by the square of height (m2). Obesity was defined accordingly to the WHO criteria as a BMI 30 kg/m2 and morbid obesity as a BMI 40 kg/m2. Hypertension was defined as a systolic arterial pressure more than 140 mmHg and/or a diastolic pressure more than 90 mmHg. Hypogonadism was defined by the presence of symptoms and/or clinical signs of androgen deficiency, combined with low circulating levels of total testosterone, as proposed by the Endocrine Society.1 Body composition was assessed by bioelectrical impedance (BIA) (Tanita BC-418MA). Fasting blood samples were obtained for measurements of usual biochemistry, lipid profile, ferritin, glucose, glycosylated hemoglobin (HbA1c), insulin and C peptide, thyroid stimulating hormone (TSH) and free T4, SHBG, prolactin, LH and follicle stimulating hormone (FSH), androstenedione, dehydroepiandosterone sulfate (DHEA-S), estrone (E1) and estradiol (E2), Insulin-like Growth Factor-I (IGF-I), total testosterone (TT) and free testosterone (FT). A 75-g oral glucose tolerance test (OGTT) and a Homeostatic Model Assessment (HOMA) test were also performed. HOMA modeling was used to calculate insulin sensitivity (HOMA-S), b-cell function (HOMA-b) and hyperbolic product (b-cell function  insulin sensitivity [B  S]).12 The Androgen Deficiency in Aging Males (ADAM) questionnaire was used to screen for potential androgen deficiency. This questionnaire includes 10 questions to which the subject may answer “yes” or “no”. A questionnaire was considered as positive in case of an affirmative answer to questions 1 (“Do you have a decrease in libido?”) or 7 (“Are your erections less strong?”) or to any three other questions.13 When medical history was suggestive of sleep apnea syndrome (SAS), the patients also underwent a nocturnal oxymetry and/or a polysomnography (n ¼ 28). For the 21 patients who underwent bariatric surgery, clinical evaluation and biological tests were performed at baseline, 3 and 12 months after surgery. The ADAM questionnaire was recorded at baseline and 12 months after surgery.

2.3. Biochemical assessment Blood was collected from antecubital vein into serum tubes. After centrifugation within 1 h, serum was carefully separated and stored at 80  C until assayed. All laboratory measurements were performed in the Laboratory of the Cliniques Saint-Luc (Brussels, Belgium). Glucose, total cholesterol, low density lipoprotein-cholesterol (LDL-C), high density lipoprotein-cholesterol (HDL-C), triglycerides, HbA1c, were measured using standard methods. Determinations of SHBG and E2 were performed on ArchitectÒ i4000SR (Abbott Diagnostics, Chicago, IL, USA, reference range, respectively, 13e 71 nmol/L and 13e40 pg/ml). TT was measured by competitive immunoassay on UnicelÒ DxI 800 (Beckman Coulter, Fullerton, CA, USA, reference range 10e35 nmol/L). The free testosterone assay was an equilibrium dialysis assay based on the ultrafiltrable fraction of added [3H] testosterone. The functional sensitivity of the assay is 0.03 nM. The within-run coefficients of variation using serum pools with concentrations of 0.100 nM and 0.450 nM were 9.8 and 8.3%, respectively. The between run coefficients of variation for the same pools were 11.2 and 10.2%, respectively. The reference range of free testosterone is 0.170e0.700 nM. The value of free testosterone, as measured in our study by equilibrium analysis, has been demonstrated correlated not only to mass spectrometry value but also to free testosterone value obtained by calculation from T and SHBG.14 Enzyme-linked Immuno Sorbent Assay (ELISA) method was used to measure E1 (DRGÒ, Mountainside, NJ, USA, reference range 25e 85 pg/ml) and androstenedione (LDNÒ, Nordhorn,Germany, reference range 0.3e3.5 ng/ml). The E1 assay was a solid phase enzymelinked immunosorbent assay based on the principle of competitive binding competitive immunoassay principle and was developed by DRG. The working range for this assay extends from 13 to 1000 pg/ mL. The documented cross-reactivities are less than 1.0% for progesterone and testosterone, 2.5% for estradiol and 2.1% for estriol. The within-run coefficients of variation within serum pools with concentrations of 23 and 34 pg/mL were 6.5 and 9.4%, respectively. The between run coefficients of variation for the same pools were 13.6 and 12.8%, respectively. DHEA-S was measured by an automated immunoassay (DxI 800, Beckman Coulter, Fullerton, CA, USA, reference range: 2.5e12 mM). IGF-1 and insulin were measured by immunometric assay using chemiluminescence-based method (LiaisonÒ Diasorin, Saluggia, Italy) and normal values for insulin were <10 mUI/ml. 2.4. Statistical analysis All numerical data are presented as means  standard deviation (SD) unless otherwise specified and were compared between groups by the Wilcoxon rank sum test. When indicated, non-parametric tests were performed and results presented as medians [P25eP75]. In the longitudinal study, numerical variables were compared by Friedman test, followed in case of significant heterogeneity by Wilcoxon signed rank tests for 2 by 2 comparisons. Correlations between parameters were assessed by Spearman rank correlation analysis. Frequencies were compared by chi square and Mc Nemar tests for unpaired or paired data, respectively. A P value less than 0.05 was considered as statistically significant. All statistical tests were two-tailed and were performed by SPSS 15.0 software (SPSS Inc., Chicago, Ill.) 3. Results 3.1. Clinical and biological characteristics of the study population Seventy-five patients were enrolled in the study, with a median age of 42 yrs (35e43) (Table 1). The mean body weight was

V. Ippersiel et al. / e-SPEN Journal 8 (2013) e37ee43 Table 1 Clinical characteristics of the study population. Age (years) BMI (kg/m2) Waist circumference (cm) Body fat mass (%)

43 (35e53) 41.5  6.9 130.9  13.4 36.1  5.7

75 75 65 57

ADAM questionnaire Number of positive questionnaires Number of patients with decreased sex drive Number of patients with erections less strong Number of patients answering "yes" at any 3 other questions Blood pressure (BP) Systolic BP (mmHg) Diastolic BP (mmHg)

58 54 26 31 53

140 (120e160) 90 (80e100)

(93%) (45%) (53%) (92%)

75 75


range, 39% of the patients had TT levels lower than 10.0 nmol/L and FT levels lower than 0.170 nmol/L. According to the ADAM questionnaire obtained in 52 patients, 93% of our obese patients presented clinical symptoms suggestive of possible androgen deficiency. Among these patients, 45% complained of decreased libido and 53% of less strong erections (Table 1). Forty-two percent of the patients combined low TT and signs of androgen deficiency as assessed by the ADAM questionnaire and were considered therefore as hypogonadal males. Patients who underwent bariatric surgery (n ¼ 21) were characterized by lower TT levels (P ¼ 0.006) and higher frequency of hypogonadism (P ¼ 0.049) by comparison with the rest of the group. 3.3. Determinants of hypotestosteronemia and hypogonadism

Glycemic status Type 2 diabetes Impaired glucose tolerance (IGT) Impaired fasting glucose (IFT) IGT þ IFT

20 4 14 3

Obstructive sleep apnea syndrome

35 (47%)

(27%) (5%) (19%) (4%)

Values are given as mean  SD or median (P25eP75).

128.9  23.9 kg and the mean BMI was 41.5  6.9 kg/m2. Most of the patients (61%) were morbidly obese (BMI  40 kg/m2). About half of the patients (51%) were hypertensive. SAS was present (n ¼ 13) or had been diagnosed during the evaluation (n ¼ 22) in almost half of the population (47%) and 13 patients were treated by a Continuous Positive Airway Pressure (CPAP). More than a quarter of the population (27%) had type 2 diabetes. In addition, 14 patients had an impaired fasting glucose (IFG), 4 an impaired glucose tolerance (IGT) and 3 had both IFG and IGT. Median HbA1c was 39 mmol/mol (37e45) (Table 2). According to the HOMA test (Table 2), insulin resistance was frequent (median HOMA-S: 33 % (24e47); 82% with HOMA-S <50%) and often compensated by increased insulin secretion (median HOMA-b: 152% (103e213); 51% with HOMA-b > 150%). Patients who underwent bariatric surgery (n ¼ 21) were more obese, based on BMI (P < 0.02) and fat mass (P ¼ 0.05) and had a more abdominal fat distribution based on waist circumference (P ¼ 0.001). Comorbidities such as diabetes (P ¼ 0.048) and hypertension (P ¼ 0.003) were also more common in this subgroup.

TT level was negatively correlated with BMI (r ¼ 0.251; P ¼ 0.038), body fat mass (r ¼ 0.320; P ¼ 0.022) and waist circumference (r ¼ 0.322; P ¼ 0.012) while FT level was only negatively correlated with body fat mass (r ¼ 0.296; P ¼ 0.035)(Table 3). In contrast, no significant correlation was found between TT and FT levels on one hand and HbA1c, HOMA-S and diabetes on the other hand. There was no significant difference in TT and FT levels in patients with or without sleep apnea syndrome. Neither androstenedione, nor E1 or E2 levels were significantly correlated with BMI or body fat mass. Among clinical symptoms of hypogonadism, erectile dysfunction and decreased libido were studied in relationship with hypotestosteronemia (Table 4). Total and free testosterone levels were lower in patients with erectile dysfunction (P ¼ 0.044 and P ¼ 0.036, respectively), but not in patients with decreased libido. Age, BMI, adiposity and insulin resistance were not determinants of erectile dysfunction or of decreased libido. Conditions like sleep obstructive apnea or hypertension were not more frequent in patients with erectile dysfunction or decreased libido. Although diabetes was more frequent in patients with erectile dysfunction, this difference did not reach significance (P ¼ 0.054). A multiple regression analysis was performed in order to evaluate the role of different confounders such as BMI, age, body fat, insulin sensitivity, waist circumference, and the presence of diabetes on TT values. None of the above mentioned parameters were independently associated with TT values. 3.4. Improvement of metabolic syndrome after bariatric surgery

3.2. Prevalence of hypotestosteronemia and hypogonadism The main hormonal characteristics of the population are shown in Table 2. Although mean TT and FT levels were still in the normal

Table 2 Metabolic and hormonal characteristics of the study population.

HOMA B (%) S (%) BxS (%) HbA1c (mmol/mol) Total testosterone (nmol/L) Free testosterone (nmol/L) LH (mUI/ml) FSH (mUI/ml) SHBG (nmol/L) Estradiol (pg/ml) Estrone (pg/ml) Androstenedione (ng/ml) DHEA-S (mmol/L)

152 (103e213) 33 (24e47) 55 (39e74) 39 (37e45) 10.9  2.4 0.195  0.052 2.8 (2.2e3.5) 4.6 (3.1e6.6) 30.4 (24.8e37.7) 30 (23e35) 32 (28e44) 1.4 (1.1e2.1) 5.0 (2.6e7.1)

Values are given as mean  SD or median (P25eP75).


Normal range

73 73 73 74 69 69 75 75 69 75 69 69 75

20e42 10.0e35.0 0.170e0.700 1.7e12.1 1.4e9.9 13e71 13e40 25e85 0.3e3.5 2.5e12

Of the 75 male patients, 21 underwent bariatric surgery. The median age of these patients was 40 (33e53) years. Baseline median body weight was 138.6 (129.2e151.3) kg before surgery and was reduced to 98.9 (84.9e111.9) kg 12 months after surgery (P < 0.001). In parallel, initial BMI was 45.3  5.6 kg/m2 and decreased to 31.0  4.2 kg/m2 12 months after surgery (P < 0.001). Body fat mass decreased from 36.1  5.7% to 24.9  8.7% (P < 0.001), systolic blood pressure from 150 (140e160) mmHg to 125 (120e 130) mmHg (P ¼ 0.001) and diastolic blood pressure from 95 (90e 105) mmHg to 80 (70e85) mmHg (P < 0.001). Fasting plasma glucose levels did not change significantly, while HbA1c improved

Table 3 Relationships between testosterone and anthropometric parameters.

BMI (kg/m2) Body fat mass (%) Waist circumference (cm)

Total testosterone

Free testosterone


P value


P value

0.251 0.320 0.322

0.038 0.022 0.012

0.146 0.296 0.226

0.230 0.035 0.082

Data are Pearson’s r correlation coefficients.


V. Ippersiel et al. / e-SPEN Journal 8 (2013) e37ee43

Table 4 Determinants of clinical symptoms of androgen deficiency. Erectile dysfunction

Total testosterone (nmol/L) Free testosterone (nmol/L) Age (years) BMI (kg/m2) Body fat mass (%) HOMA-S (%) Waist circumference (cm) Sleep obstructive apnea (n) Diabetes (n) Hypertension (n)

Decreased libido



P value



P value

10.2  2.8 0.182  0.053 43 (38e55) 42.6  7.1 36.1  5.7 32 (22e49) 134.3  12.2 16 (55.2%) 10 (76.9%) 16 (55.2%)

11.2  2.0 0.205  0.042 40 (32e48) 41.0  6.6 36.2  5.0 31 (25e47) 129.8  13.0 13 (44.8%) 3 (23.1%) 13 (44.8%)

0.044 0.036 NS (0.069) NS NS NS NS NS NS (0.054) NS

10.4  3.0 0.189  0.056 43 (39e54) 41.8  5.4 36.2  5.8 35 (26e50) 134.1  12.4 13 (44.8%) 8 (61.5%) 12 (41.4%)

10.9  2.0 0.195  0.045 40 (32e53) 41.9  7.9 36.1  5.1 29 (24e46) 130.4  12.9 16 (55.2%) 5 (38.5%) 17 (58.6%)


Values are given as mean  SD or as numbers or as median (P25eP75). NS, not significant.

from 44  9 mmol/mol to 37  5 mmol/mol (P ¼ 0.001). A remission of diabetes was observed in 7 out of 9 patients, with antidiabetic drugs being completely stopped in 6. The evolution of the HOMA profile (Table 5) was characterized by an increase in sensitivity from 36  22% to 114  59% (P < 0.001) and a decrease in compensatory b-cell function from 166  64% to 97  31% (P ¼ 0.002). Bariatric surgery also improved all parameters of lipid profile: total cholesterol decreased from 166  38 mg/dl to 146  28 mg/dl (P < 0.001), LDL-C from 103  37 mg/dl to 84  18 mg/dl (P ¼ 0.028) and triglycerides from 125  45 mg/dl to 77  34 mg/dl (P ¼ 0.001) while HDL-C increased from 38  8 mg/dl to 46  12 mg/dl (P ¼ 0.001). 3.5. Reversibility of hypogonadism after bariatric surgery TT level increased significantly by 36% during the 12 months following surgery (P ¼ 0.001) (Table 5). However, the main increase of TT level took place during the 3 months after surgery, while no further significant change occurred between 3 and 12 months (Fig. 1A). At baseline, 10 out of 17 patients had TT level below the lower limit of 10 nmol/L. Three months after surgery, only 6 patients still had hypotestosteronemia and after 12 months, all patients had TT level above 10 nmol/L. In contrast, the increase of FT levels was not significant and no change was observed in the concentrations of LH, FSH, SHBG, androstenedione, DHEA-S and estrone. However, estradiol levels decreased significantly from baseline to 12 months after surgery (P ¼ 0.008) as well as the ratio E2/TT ratio (P ¼ 0.001) (Fig. 1B). Clinical symptoms of hypogonadism improved after bariatric surgery since the number of positive ADAM questionnaires decreased from 18 at baseline to 10 after 12 months (Fig 2). Likewise, the number of patients with erectile

dysfunction decreased from 13 to 5 (P ¼ 0.021) and the number of patients with reduced libido decreased from 11 to 6, although this change was not significant. 4. Discussion Our study confirms a high prevalence of hypogonadism in obese middle-aged male subjects, as defined by the recent guidelines of the American Endocrine Society.1 In fact, while suggestive symptoms such as erectile dysfunction and decreased libido are present in nearly all obese men, hypotestosteronemia is present in about 40% of them only. Our data also show that the greater the obesity, and more specifically abdominal adiposity, the lower the testosterone levels are. On the other hand, diabetes, insulinoresistance and sleep obstructive apnea were not predictive of hypotestosteronemia in our study. Finally, the weight loss induced by bariatric surgery completely normalized the hypotestosteronemia in all patients and markedly improved signs of androgen deficiency. The prevalence of hypotestosteronemia in our population of obese male subjects was slightly lower than that (50%) observed in the HIM study15 and in the study of Hostra et al.3 This difference could perhaps be explained by a higher age of the subjects in the HIM study, by different population characteristics and by different sizes of the study populations (2165 patients in the HIM study and 149 patients in the study of Hostra et al.). Nevertheless, male hypogonadism is frequent and should be considered nowadays as a classical endocrine complication in obese men. The responses to the ADAM questionnaire show that symptoms suggestive of androgen deficiency were present in the vast majority of our obese patients (93%). Surprisingly enough, total and free testosterone concentrations were significantly lower in patients

Table 5 Metabolic and hormonal response to bariatric surgery.

HOMA B (%) S (%) BxS (%) HbA1c (mmol/mol) Total testosterone (nmol/L) Free testosterone (nmol/L) LH (mUI/ml) FSH (mUI/ml) SHBG (nmol/L) Estradiol (pg/ml) Estrone (pg/ml) Androstenedione (ng/ml) DHEA-S (mmol/L)



3 months

12 months

P value

18 18 18 20 17 17 21 21 17 21 17 16 21

152 (101e226) 31 (19e41) 51 (29e74) 40 (36e51) 9.6  2.1 0.182  0.045 2.7 (2.5e4.1) 4.5 (2.9e5.6) 28.7 (25.6e36.3) 30 (22.5e35.5) 30.7 (27.8e39.7) 1.3 (0.9e2.2) 5.7 (2.5e6.9)

122 (91e148) 69 (48e95) 99 (66e102) 37 (32e39) 12.1  3.0 0.187  0.049 2.8 (2.2e5.0) 4.4 (2.9e7.1) 34.1 (26.6e46.3) 30 (20.5e39.5) 49 (36.5e117.4) 1.4 (1.0e2.0) 4.7 (2.4e6.8)

94 (73e116) 97 (58e168) 87 (68e134) 36 (34e40) 13.1  2.6 0.194  0.049 2.8 (2.2e3.6) 4.5 (3.4e7.4) 34.6 (27.6e43.4) 25 (15.5e29.5) 94 (39.9e109.6) 1.5 (1.1e2.6) 4.6 (2.4e7.7)

0.002 <0.001 <0.001 0.001 0.001 NS NS NS NS 0.008 0.056 0.079 NS

P values are for comparisons between groups at baseline and 12 months after bariatric surgery. Values are given as mean  SD or median (P25eP75). NS, not significant.

V. Ippersiel et al. / e-SPEN Journal 8 (2013) e37ee43




Total Testosterone (nmol/L)

14 12

*** **

10 8 6 4 2 0







Ratio Estradiol/ Total testosterone (pmol/pmol x 1000)

14 12 10 8



*** 4 2 0




Months Fig. 1. Circulating concentrations of sex hormones before and after bariatric surgery. Panel A: total testosterone levels (n ¼ 17) at baseline, 3 months and 12 months after bariatric surgery. Panel B: ratio estradiol/total testosterone (n ¼ 17) at baseline, 3 months and 12 months after bariatric surgery. The data are shown here as mean  SEM. *P  0.05; **P  0.01; ***P  0.001 vs. baseline values.

with erectile dysfunction but not in patients with decreased libido. This suggests that reduction of testosterone would be more tightly associated with erectile disorders than decreased libido. This hypothesis is supported by the parallel and significant improvement in TT levels and erectile dysfunction that we observed after bariatric surgery. However, in diabetic or older subjects, testosterone levels have been shown to be poorly correlated with symptoms of androgen deficiency, in particular when assessed by the ADAM questionnaire. The ADAM questionnaire has indeed poor sensitivity and specificity to detect androgen deficiency, being more influenced by age and illness than testosterone levels, as illustrated by several recent studies.16e19 Furthermore, it should be kept in mind that erectile dysfunction was assessed by the patient himself and not by objective methods. Obese patients typically exhibit hypogonadotropic hypogonadism with low normal gonadotropins levels, particularly LH, in association with hypotestosteronemia. The fact that testosterone levels were negatively correlated with obesity and visceral adiposity is in agreement with previous observations, showing an inverse relationship between testosterone level and waist circumference, BMI and body fat mass.20,21,16 Multivariate analysis was

however unable to reveal a prominent factor out of these variables which are clearly interrelated. In this study, the level of testosterone was not correlated with the presence of diabetes or with the level of glycemic control. This is in agreement with some studies which have not found a relationship between testosterone level and the level of hyperglycaemia or the presence of diabetes.22 However, it is fair to say that most studies have observed such associations16,23 in contrast to our findings. Obesity could be a confounding factor in our study as we selected primarily severely obese and not necessarily diabetic patients. Likewise, the role of insulin resistance was not confirmed in our study as testosterone concentrations were not correlated with HOMA-S, but this association might have been masked by the narrow spectrum of insulin sensitivity in our insulin-resistant obese patients. However, despite the fact that weight loss induces marked increase of both testosteronemia and insulin sensitivity, there was no significant correlation between these changes. Hypotestosteronemia seemed not to be explained by a decrease of SHBG level in our population. In fact, our study did not confirm the reduction of SHBG level which is usually observed in obese patients. Mean SHBG concentrations were in the normal range in


V. Ippersiel et al. / e-SPEN Journal 8 (2013) e37ee43

Fig. 2. Signs of androgen deficiency before and after bariatric surgery. Clinical signs of androgen deficiency as assessed by the ADAM questionnaire at baseline (n ¼ 20) and 12 months (n ¼ 19) after bariatric surgery. As illustrated, bariatric surgery was associated with a significant decrease of erectile dysfunction and an increase, albeit non-significant, of libido. *P < 0.05, significant differences.

both cross-sectional and longitudinal studies and did not change significantly after bariatric surgery. Moreover, SHBG was not correlated with insulin sensitivity, in contrast with previous observations.24,25 The cause of this discrepancy remains unclear. Sleep obstructive apnea might also be implicated in the development of hypogonadotropic hypogonadism, as LH secretion is impaired in response to sleep fragmentation.26 However, levels of LH, total and free testosterone were similar in patients with and without sleep obstructive apnea. This absence of difference cannot be due to the CPAP treatment as only one third of our SAS patients were treated with CPAP at the time of inclusion into the study. Nevertheless, SAS screening was not systematically performed in all patients. For this reason, the diagnosis cannot be completely excluded in patients presenting without any symptoms. Although mean serum concentrations of estradiol and estrone remained within the normal range in our study population, it appears likely that increased aromatization of testosterone to estrogens by the adipose tissue played an important role in the hypogonadism observed in obese men, as previously suggested by several studies.4,27 Indeed, while total testosterone increased, estradiol levels decreased significantly after bariatric surgery, thus indicating a reduction of aromatase activity. We could also observe a net decrease in the estradiol/testosterone ratio, which is also in favor of a reduction of this enzymatic activity. The role of aromatization in the obesity-related decline of testosterone has been previously demonstrated by the normalization of testosterone levels obtained by administration of aromatase inhibitors.28 Because men represents only 20% of the patients who undertake bariatric surgery,29 only few studies have focused on the reversibility of male hypogonadism after weight loss, especially following bariatric surgery. Reis et al.29 have studied the effect of weight loss obtained by a gastric bypass on erectile dysfunction and hormone levels. They showed a significant improvement of total testosterone levels and of erectile dysfunction as evaluated by the International Index of Erectile Function (IEFF)-5 questionnaire. They also found an increase of free testosterone and FSH but these changes were not

significant. Bastounis et al.27 reported that after weight loss induced by vertical banded gastroplasty, SHBG, total testosterone and FSH increased significantly while estradiol decreased. Again, there was no significant change of free testosterone. Finally, Hammoud et al.4 showed significant increases of SHBG and total and free testosterone together with an estradiol decrease two years after gastric bypass. In our study, an increase of total testosterone was already observed 3 months after surgery with limited increase thereafter. As in previous studies,4 we could not demonstrate any increase of the free testosterone levels. Some of these studies have also focused on the reversibility after weight loss of symptoms related with androgen deficiency. In several studies,4,29 the quality of sexual life and erectile function markedly improved after weight loss induced by bariatric surgery. Likewise, in our study, the ADAM questionnaire, positive at baseline in 86% of the patients elected for surgery, significantly improved 12 months later, since it was positive in only 48% of them. The number of patients complaining from erectile dysfunction but not of patients suffering from decreased libido also decreased significantly. Our study has several strengths, including a long follow up period (1 yr) and a homogenous population. Weaknesses of this study include the small number of participants, some missing data and the use of immunoassays to assess circulating estrogens levels. In conclusion, the prevalence of hypogonadotropic hypogonadism is high among severely obese male patients. Several mechanisms are likely responsible for this condition. Our study indicates that excessive aromatization of testosterone to estrogens may play a predominant role in the decline of testosterone caused by morbid obesity. In contrast, diabetes and sleep apnea syndrome did not emerge as specific risk factors of hypogonadism in this specific population. More importantly from a therapeutic perspective, the major weight loss induced by bariatric surgery is able to fully reverse hypogonadism since total testosterone levels normalize and clinical symptoms of androgen deficiency improve significantly within the first year following surgery. Our observation supports therefore the implementation of lifestyle measures

V. Ippersiel et al. / e-SPEN Journal 8 (2013) e37ee43

to raise testosterone and reverse signs of androgen deficiency in obese men. Statement of authorship Dr Thissen had full access to all the data in the study and takes the responsibility for the integrity of the data. Study concept and design: VI, DM and JPT. Acquisition of the data: VI, AL, DG and JPT. Analysis and interpretation of the data: VI, AL, DG, DM, OA and JPT. Drafting of the manuscript: VI, DM and JPT. Statistical analysis: JJ, OA. Critical revision of the manuscript: VI, DM, OA and JPT. Study supervision: DM and JPT. Conflict of interest statement The authors declared no competing interests. Acknowledgments The authors thank R Detry and B Navez, who performed the bariatric procedures and P Lause for technical assistance. References 1. Bhasin S, Cunningham GR, Hayes FJ, Matsumoto AM, Snyder PJ, Swerdloff RS, et al. Testosterone therapy in men with androgen deficiency syndromes: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2010;95: 2536e59. 2. Rao SR, Kini S, Tamler R. Sex Hormones and Bariatric Surgery in Men. Gend Med 2011;8:300e11. 3. Hofstra J, Loves S, van WB, Ruinemans-Koerts J, Jansen I, de BH. High prevalence of hypogonadotropic hypogonadism in men referred for obesity treatment. Neth J Med 2008;66:103e9. 4. Hammoud A, Gibson M, Hunt SC, Adams TD, Carrell DT, Kolotkin RL, et al. Effect of Roux-en-Y gastric bypass surgery on the sex steroids and quality of life in obese men. J Clin Endocrinol Metab 2009;94:1329e32. 5. Corona G, Mannucci E, Fisher AD, Lotti F, Petrone L, Balercia G, et al. Low levels of androgens in men with erectile dysfunction and obesity. J Sex Med 2008;5: 2454e63. 6. Corona G, Monami M, Rastrelli G, Aversa A, Tishova Y, Saad F, et al. Testosterone and metabolic syndrome: a meta-analysis study. J Sex Med 2011;8:272e83. 7. Corona G, Monami M, Rastrelli G, Aversa A, Sforza A, Lenzi A, et al. Type 2 diabetes mellitus and testosterone: a meta-analysis study. Int J Androl 2011;34:528e40. 8. Grossmann M. Low testosterone in men with type 2 diabetes: significance and treatment. J Clin Endocrinol Metab 2011;96:2341e53. 9. Tajar A, Huhtaniemi IT, O’Neill TW, Finn JD, Pye SR, Lee DM, et al. Characteristics of androgen deficiency in late-onset hypogonadism: results from the European Male Aging Study (EMAS). J Clin Endocrinol Metab 2012;97:1508e16. 10. Diaz-Arjonilla M, Schwarcz M, Swerdloff RS, Wang C. Obesity, low testosterone levels and erectile dysfunction. Int J Impot Res 2009;21:89e98.


11. Mah PM, Wittert GA. Obesity and testicular function. Mol Cell Endocrinol 2010;316:180e6. 12. Hermans MP, Levy JC, Morris RJ, Turner RC. Comparison of tests of beta-cell function across a range of glucose tolerance from normal to diabetes. Diabetes 1999;48:1779e86. 13. Morley JE, Charlton E, Patrick P, Kaiser FE, Cadeau P, McCready D, et al. Validation of a screening questionnaire for androgen deficiency in aging males. Metabolism 2000;49:1239e42. 14. Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab 1999;84:3666e72. 15. Mulligan T, Frick MF, Zuraw QC, Stemhagen A, McWhirter C. Prevalence of hypogonadism in males aged at least 45 years: the HIM study. Int J Clin Pract 2006;60:762e9. 16. Kapoor D, Aldred H, Clark S, Channer KS, Jones TH. Clinical and biochemical assessment of hypogonadism in men with type 2 diabetes: correlations with bioavailable testosterone and visceral adiposity. Diabetes Care 2007;30:911e7. 17. Biswas M, Hampton D, Newcombe RG, Rees DA. Total and free testosterone concentrations are strongly influenced by age and central obesity in men with type 1 and type 2 diabetes but correlate weakly with symptoms of androgen deficiency and diabetes-related quality of life. Clin Endocrinol (Oxf) 2011. 18. Emmelot-Vonk MH, Verhaar HJ, Nakhai-Pour HR, Grobbee DE, van der Schouw YT. Low testosterone concentrations and the symptoms of testosterone deficiency according to the Androgen Deficiency in Ageing Males (ADAM) and Ageing Males’ Symptoms rating scale (AMS) questionnaires. Clin Endocrinol (Oxf) 2011;74:488e94. 19. Chueh KS, Huang SP, Lee YC, Wang CJ, Yeh HC, Li WM, et al. The comparison of the aging male symptoms (AMS) scale and androgen deficiency in the aging Male (ADAM) questionnaire to detect androgen deficiency in middle-aged men. J Androl 2012;33:817e23. 20. Biswas M, Hampton D, Turkes A, Newcombe RG, Aled RD. Reduced total testosterone concentrations in young healthy South Asian men are partly explained by increased insulin resistance but not by altered adiposity. Clin Endocrinol (Oxf) 2010;73:457e62. 21. Svartberg J, von MD, Sundsfjord J, Jorde R. Waist circumference and testosterone levels in community dwelling men. The Tromso study. Eur J Epidemiol 2004;19:657e63. 22. Dhindsa S, Miller MG, McWhirter CL, Mager DE, Ghanim H, Chaudhuri A, et al. Testosterone concentrations in diabetic and nondiabetic obese men. Diabetes Care 2010;33:1186e92. 23. Tripathy D, Dhindsa S, Garg R, Khaishagi A, Syed T, Dandona P. Hypogonadotropic hypogonadism in erectile dysfunction associated with type 2 diabetes mellitus: a common defect? Metab Syndr Relat Disord 2003;1:75e80. 24. Birkeland KI, Hanssen KF, Torjesen PA, Vaaler S. Level of sex hormone-binding globulin is positively correlated with insulin sensitivity in men with type 2 diabetes. J Clin Endocrinol Metab 1993;76:275e8. 25. Falkner B, Sherif K, Sumner A, Kushner H. Hyperinsulinism and sex hormones in young adult African Americans. Metabolism 1999;48:107e12. 26. Luboshitzky R, Aviv A, Hefetz A, Herer P, Shen-Orr Z, Lavie L, et al. Decreased pituitary-gonadal secretion in men with obstructive sleep apnea. J Clin Endocrinol Metab 2002;87:3394e8. 27. Bastounis EA, Karayiannakis AJ, Syrigos K, Zbar A, Makri GG, Alexiou D. Sex hormone changes in morbidly obese patients after vertical banded gastroplasty. Eur Surg Res 1998;30:43e7. 28. Loves S, Ruinemans-Koerts J, de BH. Letrozole once a week normalizes serum testosterone in obesity-related male hypogonadism. Eur J Endocrinol 2008;158: 741e7. 29. Reis LO, Favaro WJ, Barreiro GC, de Oliveira LC, Chaim EA, Fregonesi A, et al. Erectile dysfunction and hormonal imbalance in morbidly obese male is reversed after gastric bypass surgery: a prospective randomized controlled trial. Int J Androl 2010;33:736e44.