OBES SURG

DOI 10.1007/s11695-014-1233-y

ORIGINAL CONTRIBUTIONS

Effects of Bariatric Surgery on Male Obesity-Associated

Secondary Hypogonadism: Comparison of Laparoscopic Gastric
Bypass with Restrictive Procedures
Berniza Calderón & Alba Galdón & Alfonso Calañas & Roberto Peromingo & Julio Galindo &
Francisca García-Moreno & Gloria Rodriguez-Velasco & Antonia Martín-Hidalgo &
Clotilde Vazquez & Héctor F. Escobar-Morreale & José I. Botella-Carretero

# Springer Science+Business Media New York 2014

Abstract Bariatric surgery results in the complete resolution concentrations. Fasting glucose and insulin levels served for
of male obesity-associated secondary hypogonadism (MOSH) homeostatic model assessment of insulin resistance
in many patients. However, the effects of different bariatric (HOMAIR). MOSH was present in 25 and 16 of the 35
surgical procedures on male sexual hormone profiles and patients when considering TT and FT concentrations respec-
sexual dysfunction have not been compared to date. We tively, resolving after surgery in all but one of them. When
compared the pre- and post-operative (at least 6 months after considering all obese men as a whole, patients submitted to
initial surgery) sex hormone profiles of 20 severely obese men LGB or restrictive procedures did not differ in terms of excess
submitted to laparoscopic gastric bypass (LGB) with 15 sim- weight loss, in the decrease of fasting glucose and insulin,
ilar patients submitted to restrictive techniques (sleeve gas- HOMAIR and waist circumference, or in the increase of
trectomy in 10 and adjustable gastric banding in 5). We serum 25-hydroxyvitamin D, TT and FT levels. The improve-
calculated free testosterone (FT) levels from total testosterone ment in TT correlated with the decrease in fasting glucose (r=
(TT) and sex hormone binding globulin (SHBG) −0.390, P = 0.021), insulin (r = −0.425, P = 0.015) and
HOMAIR (r=−0.380, P=0.029), and with the increase in
SHBG (r=0.692, P<0.001). The increase in FT correlated
B. Calderón : A. Galdón : A. Calañas : C. Vazquez :
H. F. Escobar-Morreale : J. I. Botella-Carretero (*)
with the decrease in fasting glucose (r=−0.360, P=0.034).
Department of Endocrinology and Nutrition, Hospital Universitario LGB and restrictive techniques are equally effective in pro-
Ramón y Cajal & Universidad de Alcalá & Instituto Ramón y Cajal ducing a remission of MOSH.
de Investigación Sanitaria (IRYCIS), Carretera de Colmenar Km. 9.1,
28034 Madrid, Spain
e-mail: [Link]@[Link]
Keywords Laparoscopic gastric bypass . Sleeve
R. Peromingo : J. Galindo : F. García-Moreno : gastrectomy . Band lap . Obesity . Surgery . Hypogonadism .
G. Rodriguez-Velasco Androgens . Insulin resistance
Department of Digestive and General Surgery, Hospital Universitario
Ramón y Cajal & Universidad de Alcalá & Instituto Ramón y Cajal
de Investigación Sanitaria (IRYCIS), 28034 Madrid, Spain

A. Martín-Hidalgo Introduction
Department of Biochemistry-Research, Hospital Universitario
Ramón y Cajal & Universidad de Alcalá & Instituto Ramón y Cajal Obesity is a major public health problem with increasing
de Investigación Sanitaria (IRYCIS), 28034 Madrid, Spain
prevalence [1] and is associated with an increased all cause
A. Martín-Hidalgo : C. Vazquez : J. I. Botella-Carretero mortality and medical co-morbidity [2]. The use of bariatric
Centro de Investigación Biomédica en Red-Fisiopatología de surgery has increased because it offers more successful weight
Obesidad y Nutrición (CIBERobn), Madrid, Spain loss and long-term weight maintenance than life style modifi-
cation [3]. Even though modern bariatric surgical techniques
H. F. Escobar-Morreale
Centro de Investigación Biomédica en Red Diabetes y Enfermedades are not entirely free of long-term nutritional and metabolic
Metabólicas Asociadas (CIBERDEM), Madrid, Spain issues [4–7], the success and safety of current bariatric
 OBES SURG

procedures clearly compensate the very important health risks at least 10 % of initial weight was achieved. We compared the
associated with severe obesity. pre-and post-operative hormone profiles of the 20 patients
Long-term outcomes after bariatric surgery have shown submitted to LGB with the 15 patients submitted to restrictive
resolution of many complications associated with obesity. techniques (considering those submitted to SG and those
Overall, bariatric surgery results in the resolution of diabetes treated with AGB as a whole).
mellitus, hypertension, dyslipidemia, and sleep apnea that Ten healthy controls were matched with patients in terms of
occur in more than 70, 60, 60, and 85 % of severely obese age for the purpose of establishing reference ranges for some
patients, respectively [8, 9]. Obesity-associated sexual dys- hormonal determinations. Written informed consent was ob-
function, including polycystic ovary syndrome in women and tained from every participant, and the study was approved by
male obesity-associated secondary hypogonadism (MOSH) in the institutional review board of our institution.
men, may also achieve complete remission after bariatric Anthropometric parameters were recorded, and body mass
surgery [10, 11]. index (BMI) was calculated both before operation and during
Regarding MOSH, a recent meta-analysis concluded that follow-up, and the percentage of BMI lost was also deter-
weight loss attained after bariatric surgery is associated with mined. The same applied to waist circumference, which was
an increase in both total testosterone (TT) and free testoster- measured as the smallest perimeter between the costal border
one (FT) and the normalization of serum sex hormone-binding and the anterior suprailiac spines. Excess weight loss (EWL)
globulin (SHBG), resulting in the complete resolution of was calculated in our study as the percentage of the ratio
MOSH in most patients [12]. Since the different bariatric between weight loss attained at revaluation from the time
procedures may result in different rates of remission of immediately before operation and ideal weight. The latter
obesity-related complications such as diabetes mellitus [13, was calculated as the weight corresponding for a BMI of 25,
14], the same could happen with MOSH. However, to our best given previous lack of consensus for the precise definition of
knowledge, a direct comparison of the different bariatric tech- EWL [16, 17]. Excess body weight (EBW) at baseline was
niques on the beneficial effects on MOSH has not been calculated as the difference between baseline body weight and
conducted to date. Therefore, we aimed to compare laparo- ideal weight.
scopic gastric bypass (LGB) with restrictive bariatric surgical
procedures in terms of improvement of serum androgen con-
centrations and resolution of MOSH. Analytical Procedures and Reference Ranges

Serum creatinine, alanine amino-transferase, aspartic amino-

Patients and Methods transferase, and serum glucose levels were measured by stan-
dard colorimetric methods, using the Architect ci8200 analyz-
Patients er (Abbot Diagnostics, Berkshire, UK). Levels of HDL cho-
lesterol were measured in supernatant after plasma precipita-
We studied thirty-five patients who underwent bariatric sur- tion with phosphotungstic acid and Mg2+ (Boehringer Mann-
gery at the Metabolic Surgery Unit of the Department of heim GmbH, Mannheim, Germany). Levels of total choles-
Endocrinology and Nutrition of the Hospital Universitario terol and triglycerides were measured by enzymatic methods
Ramón y Cajal. Indication for surgery was based on the (Menarini Diagnostica, Florence, Italy). The LDL cholesterol
recommendations of the National Institutes of Health [15]. level was calculated by using Friedewald’s formula.
Inclusion criteria needed attending the scheduled follow-up Fasting insulin, TT, sex hormone-binding globulin
visits at the unit and not presenting heart disease, kidney, or (SHBG), luteinizing hormone (LH), follicle-stimulating hor-
liver failure. Previous diagnoses of hypogonadism or treat- mone (FSH), and estradiol were also assayed. Briefly, TT was
ment for sexual dysfunction, thyroid disease, measured by radioimmunoassay (Spectria, Orion Diagnostica,
hyperprolactinemia or any other condition, or drug that could Espoo, Finland) with a coefficient of variation (CV) <10 %.
interfere with normal gonadal function were considered ex- The FT concentration was calculated from total testosterone
clusion criteria. and SHBG concentrations [18]. Serum insulin was measured
Twenty patients were treated by LGB, 10 by sleeve gas- by immunochemiluminescence (Immulite 2000, Siemens
trectomy (SG), and 5 by adjustable gastric banding (AGB) Healthcare Diagnostics Inc., Gwynedd, UK), with a CV
according to the hospital’s established protocols and after the <10 %. Insulin resistance in the fasting state was estimated
evaluation and choice of the surgeon. For this reason, patients by the homeostasis model assessment method (HOMAIR). A
were not randomized to each of the surgical techniques. Pa- commercial enzyme-linked immunosorbent assay (ELISA)
tients were re-evaluated after surgery at different time intervals was employed for the measurement of 25-hydroxyvitamin D
always after the initial 6 post-operative months (mean±SD, concentrations (IDS Ltd., Boldon, UK). The specificity of this
12.3±5.6 months) and only when a significant weight loss of assay is 100 % for 25-hydroxyvitamin D3 and 75 % for 25-
OBES SURG

hydroxyvitamin D2, with negligible cross-reactivities with (35 %) with hypogonadism in the LGB group and nine
vitamin D3 and vitamin D2 (<0.01 and <0.30 %, respectively). (60 %) in the restrictive procedures group (Fisher’s test, P=
Reference ranges were those provided by the Central Lab- 0.182). Before surgery, TT inversely correlated with BMI,
oratory of Hospital Universitario Ramón y Cajal, except for systolic blood pressure, diastolic blood pressure, waist cir-
free testosterone concentrations and HOMAIR which were cumference, triglycerides, fasting glucose, insulin and
obtained by calculating the 95 % confidence intervals of the HOMAIR, and directly with HDL and SHBG; FT showed
control group’s means (since our Central Laboratory had no similar associations (Table 2).
standard reference range for these parameters). Normal ranges When patients were revaluated after weight loss induced by
were 300–900 ng/dL for TT, 117–639 μg/dL for SHBG, 6.5– bariatric surgery, all patient normalized their serum concen-
18.3 ng/dL for FT, and 6.0–27.0 μU/mL for insulin. trations of TT and FT, with the exception of one patient
submitted to SG who did not normalize TT concentrations
Statistics (his baseline TT of 200 ng/dL increased to 253 ng/dL when
revaluated 6 months after surgery). TT and FT concentrations
A priori power analysis was performed using Ene 3.0 software increased after bariatric surgery irrespective of the bariatric
([Link] To detect a difference equal or technique applied (LGB or restrictive techniques, Table 3).
above 1.4 ng/dL in serum FT concentrations in patients before The near-significant difference (P=0.053) between surgical
and after surgery, with a SD of 1.4 ng/dL, power=0.8 and α= techniques observed in the increase in TT was probably due to
0.05, at least 10 patients were needed for a two-tail estimate. the significant change in SHBG, because no differences were
To detect such a difference between groups with a one-tail observed between techniques in the increase in FT (Table 3).
estimate favoring LGB, at least 14 patients in each group were This highlights the importance of calculating FT in these
needed. patients, as results in TT may be spuriously low due to the
Results are expressed as means±SD unless otherwise stat- lowering effects of obesity on SHBG levels.
ed. The Kolmogorov–Smirnov statistic was applied to contin- Fasting glucose, insulin, HOMAIR, and waist circumfer-
uous variables. Logarithmic transformation was applied as ence showed a similar decrease after surgery between groups
needed to ensure normal distribution of the variables. Un- (Table 3). Serum 25-hydroxyvitamin D levels also showed a
paired t test or Mann–Whitney U test was used to compare similar increase after surgery between groups (Table 3), and
the central tendencies of the different groups as needed. To EWL did not reach a significant difference between groups
evaluate the association between discontinuous variables, we either (67±26 % and 54±26 % for LGB and restrictive
used the χ2 test and Fisher’s exact test as appropriate. Com- procedures, respectively, t=1.373, P=0.179). On the other
parisons of continuous variables before and after bariatric hand, total weight loss, the decrease in BMI and the increase
surgery were performed using repeated-measures GLM anal- in SHBG after surgery were higher with LGB than with
ysis. Changes in selected variables after surgery were restrictive techniques (Table 3). When restricting the analysis
expressed as percentage of the difference between final and to patients presenting MOSH at baseline (defined by low FT
baseline values divided by the baseline values. Bivariate cor- concentrations,) only SHBG showed a higher increase with
relation was employed to study lineal association between two LGB compared with restrictive surgery (Fig 1).
quantitative variables using Pearson’s or Spearman’s tests as The increase in TT correlated with the decrease in fasting
appropriate. Finally, a backward stepwise multiple linear re- glucose (r=−0.390, P=0.021), insulin (r=−0.425, P=0.015),
gression model was applied to evaluate the effects of several and HOMAIR (r=−0.380, P=0.029), and with the increase in
dependent variables on the changes of serum FT concentration. SHBG (r=0.692, P<0.001). The increase in FT correlated with
Analyses were performed using SPSS 15 (SPSS Inc, Chicago, the decrease in fasting glucose (r=−0.360, P=0.034). Finally,
Illinois). P<0.05 was considered statistically significant. multivariate linear regression was employed to evaluate the
effects of several dependent variables (EWL, waist circumfer-
ence, fasting glucose, insulin, and estradiol) on the changes of
Results serum FT concentration. The model retained only the decrease
in fasting glucose (β=−0.995, P=0.017) as predictor of the
Baseline characteristics of the included 35 patients who increase in serum FT (R2 =0.467, F=4.945, P=0.046).
underwent bariatric surgery are shown in Table 1. Patients
submitted to LGB presented with higher baseline weight,
BMI, EBW, and serum estradiol (Table 1). When considering Discussion
TT concentrations, there were 12 patients (60 %) with
hypogonadism in the LGB group and 11 (73.3 %) in the Our present results indicate that LGB and restrictive tech-
restrictive procedures group (Fisher’s test, P=0.489). When niques such as SG and AGB are equally effective in inducing
considering FT concentrations, there were seven patients the remission of MOSH in severely obese men. In fact, all
 OBES SURG

Table 1 Clinical and analytical

characteristics of the severely Laparoscopic gastric Restrictive P value
obese men submitted to bariatric bypass (n=20) surgery (n=15)
surgery
Age (year) 38±9 41±10 0.468
Weight (kg) 156±26 132±14 0.002
Body mass index (kg/m2) 50.4±8.7 42.9±2.7 0.001
Excess body weight (kg) 79±26 55±8 0.001
Waist circumference (cm) 142±18 132±10 0.078
Waist to hip ratio 1.0±0.1 1.1±0.1 0.345
Systolic blood pressure (mmHg) 147±12 142±19 0.361
Diastolic blood pressure (mmHg) 87±9 87±10 0.925
Serum creatinine (mg/dL) 1.0±0.2 0.9±0.2 0.166
Aspartate amino-transferase (U/L) 30±18 26±8 0.379
Alanine amino-transferase (U/L) 58±49 47±25 0.452
Total cholesterol (mg/dL) 189±28 189±29 0.952
HDL cholesterol (mg/dL) 40±5 40±9 0.989
LDL cholesterol (mg/dL) 120±28 123±29 0.717
Triglycerides (mg/dL) 157±93 132±44 0.347
Total testosterone (ng/dL) 302±102 262±79 0.224
Sex hormone-binding globulin (μg/dL) 182±67 181±85 0.982
Free testosterone (ng/dL) 7.7±2.6 6.6±1.8 0.194
Prolactin (ng/mL) 8.8±2.6 11.7±1.0 0.184
Luteinizing hormone (mU/mL) 3.4±1.2 2.9±1.0 0.354
Follicle-stimulating hormone (mU/mL) 4.1±2.2 3.3±1.3 0.380
Estradiol (pg/mL) 40.6±13.0 30.8±3.9 0.027
Fasting glucose (mg/dL) 120±61 114±22 0.720
Fasting insulin (μU/mL) 34.2±22.2 39.8±40.8 0.608
HOMAIR 9.8±6.1 11.9±14.3 0.591
Data are means±SD Serum PTH (pg/mL) 52.4±16.1 56.9±17.9 0.466
HOMAIR homeostasis model as- 25-hydroxyvitamin D (ng/mL) 18.9±9.4 18.4±8.0 0.866
sessment of insulin resistance

men presenting with MOSH before surgery normalized their

Table 2 Correlation of total testosterone and free testosterone with other
serum FT concentrations after all bariatric surgical procedures.
clinical and analytical variables before surgery when considering the 35
patients submitted to bariatric surgery as a whole The normalization of FT occurred in parallel to weight loss, to
the increase in serum vitamin D, and to the decrease in waist
Total testosterone Free testosterone circumference, glycemia, insulinemia, and insulin resistance.
Moreover, the decrease in glycemia after surgery was the main
r P r P
factor associated with the increase in FT.
Body mass index (kg/m2) −0.477 0.001 −0.445 0.002
Decreased testosterone concentrations in men have been
Waist circumference (cm) −0.584 <0.001 −0.540 <0.001
found not only in obesity but also in patients with type 2
Systolic blood pressure (mmHg) −0.382 0.011 −0.319 0.037
diabetes mellitus and in association with the metabolic syn-
Diastolic blood pressure (mmHg) −0.471 0.001 −0.432 0.004
drome [19–21]. Moreover, low testosterone concentrations
HDL cholesterol (mg/dL) 0.651 <0.001 0.551 <0.001
have been associated with insulin resistance in non-diabetic
Triglycerides (mg/dL) −0.443 0.002 −0.436 0.003
men [22], and weight loss induced by dieting has been shown
SHBG (μg/dL) 0.633 <0.001 0.313 0.036
to increase testosterone levels in parallel to the improvement
Fasting glucose (mg/dL) −0.426 0.003 −0.412 0.005
in insulin sensitivity [23].
Fasting insulin (μU/mL) −0.311 0.037 −0.296 0.048
Although the precise pathophysiological mechanisms
HOMAIR −0.340 0.022 −0.334 0.025
linking reduced serum testosterone concentrations with insu-
HDL high-density lipoprotein, HOMAIR homeostasis model assessment lin resistance in men are not clearly understood, it has been
of insulin resistance, SHBG sex hormone-binding globulin proposed that the increased aromatase activity characteristic of
OBES SURG

Table 3 Changes in selected variables after bariatric surgery

Laparoscopic gastric Restrictive P value for within P value for between P value for
bypass (n=20) surgery (n=15) subjects effect subjects effect interaction

Weight (kg) −51.0±20.7 −31.1±17.6 <0.001 0.031 0.005

Body mass index (kg/m2) −16.4±6.8 −9.9±5.0 <0.001 0.042 0.004
Waist circumference (cm) −26.0±19.5 −20.2±15.2 <0.001 0.455 0.410
Total testosterone (ng/dL) 251.2±168.7 148.3±119.5 <0.001 0.002 0.053
Sex hormone-binding globulin (μg/dL) 243±207 54±72 <0.001 0.010 0.002
Free testosterone (ng/dL) 2.2±2.9 3.0±2.8 <0.001 0.421 0.398
Luteinizing hormone (mU/mL) 1.0±1.5 0.5±1.6 0.078 0.146 0.549
Follicle-stimulating hormone (mU/mL) 1.0±1.8 0.9±1.4 0.051 0.265 0.831
Estradiol (pg/mL) −0.2±11.4 −5.8±5.2 0.288 0.087 0.320
Fasting glucose (mg/dL) −30.8±58.0 −21.1±20.8 0.002 0.901 0.542
Fasting insulin (μU/mL) −27.0±23.8 −30.9±41.3 <0.001 0.477 0.742
HOMAIR −8.0±6.6 −10.1±14.7 <0.001 0.378 0.587
Serum parathyroid hormone (pg/mL) 8.9±30.7 −8.3±14.6 0.946 0.459 0.065
25-hydroxyvitamin D (ng/mL) 9.7±20.7 7.9±10.2 0.009 0.879 0.776

Data are means±SD

HOMAIR homeostasis model assessment of insulin resistance

obesity, by increasing estrogen levels, may inhibit gonadotro- insulin resistance [25]. Other pathophysiologic pathways that
pin secretion thereby reducing androgen concentrations [24]. link hypogonadism with obesity and insulin resistance may
These changes may also downregulate GLUT4 by an aug- include adypokines such as leptin and adiponectin [26], as
mentation of the estrogen receptor beta expression, favoring testosterone replacement therapy decreases leptin and in-
creases adiponectin levels in type 2 diabetic men [27].
Aside from the decrease in glycemia, another factor which
might have influenced the increase in testosterone after bar-
iatric surgery was the increase observed in vitamin D concen-
trations, as it has been positively associated with TT and FT
[28, 29]. Moreover, vitamin D is positively associated with
semen quality apart from its association with androgen levels
[30, 31]. On the other hand, to date, no trial has shown that
vitamin D administration may correct androgen deficiency, so
reduced vitamin D may be only a surrogate marker of ill-
health in these patients and the association of increased an-
drogen and vitamin D concentrations after bariatric surgery
may only reflect the improvement in their metabolic function.
Massive weight loss after bariatric surgery is a constant
finding associated with the remission of MOSH. A recent
study conducted in 33 men has shown that age and the
percentage of weight loss were the only independent factors
associated with the increase in TT after bariatric surgery [32].
Also, a recent meta-analysis has shown that body weight loss
Fig 1 Selected anthropometric and analytical variables in obese patients induced by bariatric surgery was the main factor associated
presenting with MOSH at baseline, showing their changes after bariatric
surgery as percentage of pre-surgery values. Data are means±SEM. Black
with a relevant increase in gonadotropins and in TT and FT,
bars correspond to restrictive techniques (considering sleeve gastrectomy and with a decline in estrogen levels [12]. Although not
and adjustable gastric banding as a whole) and gray bars correspond to thoroughly analyzed in this study, we may speculate with
laparoscopic gastric bypass (LGB). BMI body mass index, EWL excess the possibility that weight reduction is associated with in-
weight loss, FT free testosterone, SHBG sex hormone-binding globulin,
TT total testosterone, WC waist circumference. Only SHBG showed a
creased insulin sensitivity, and in turn with an activation of
significant difference between the effects of LGB and those of restrictive the hypothalamic-pituitary-gonadal axis, as insulin exerts a
techniques (*P=0.016 by Mann–Whitney U test) permissive role on GnRH neuron activity [33]. However, in
 OBES SURG

our study, we did not find an association of excess weight loss 6 months post-operatively, suggesting that a longer period is
and the increase in FT in multivariate analysis. This may have needed for MOSH to resolve. Future studies in which patients
been due to the fact that all patients were revaluated at least are reevaluated after bariatric surgery at fixed intervals will
6 months after surgery, when extensive weight loss has al- help in elucidating if resolution of MOSH is only a matter of
ready taken place. We might hypothesize that once a threshold weight loss or is also a time-dependent effect of bariatric
for weight loss is reached, gonadal function is restored, the surgery.
relationship between excess weight loss and remission of In conclusion, weight loss after bariatric surgery results in
MOSH not being lineal. the complete remission of MOSH, with LGB and restrictive
In conceptual agreement with the beneficial effects of techniques being equally beneficial for male gonadal function.
weight loss on the gonadal function of severely obese men
observed in our study, both the Endocrine Society Guidelines
Acknowledgments We thank the nurse staff of the Department of
and the Third International Consultation on Sexual Medicine Endocrinology and Nutrition for their help with the anthropometric and
recommend that lifestyle modification must be strongly en- blood sampling of the patients.
couraged in hypogonadal men with obesity, type 2 diabetes,
and the metabolic syndrome [34, 35]. However, the increase Conflict of Interest The authors declare no conflict of interest.
in testosterone induced by lifestyle interventions was clearly
inferior to that induced by bariatric surgery (9.8 % with diet vs
32 % with surgery).
Therefore, we suggest that bariatric surgery should be References
considered for the treatment of MOSH [12]. However, it is
unclear which bariatric technique is the most appropriate for 1. Finucane MM, Stevens GA, Cowan MJ, et al. National, regional, and
inducing the remission of MOSH since a direct comparison of global trends in body-mass index since 1980: systematic analysis of
health examination surveys and epidemiological studies with 960
these bariatric surgical procedures has not been conducted to country-years and 9.1 million participants. Lancet. 2011;377:557–67.
date. There have been controversial results when comparing 2. Berrington de Gonzalez A, Hartge P, Cerhan JR, et al. Body-mass
the effects of weight loss after bariatric surgery with different index and mortality among 1.46 million white adults. N Engl J Med
techniques on TT and FT. Although Bastounis et al. [36] 2010;363:2211–9.
3. Buchwald H, Oien DM. Metabolic/bariatric surgery worldwide 2011.
found years ago an improvement in SHBG and TT after Obes Surg. 2013;23:427–36.
vertical gastroplasty without any change in FT, other authors 4. Lynch RJ, Eisenberg D, Bell RL. Metabolic consequences of bariatric
recently found an increase in FT after gastric bypass [37] or surgery. J Clin Gastroenterol. 2006;40:659–68.
banded gastroplasty [38]. Even the aforementioned meta- 5. Balsa JA, Botella-Carretero JI, Peromingo R, et al. Role of calcium
malabsorption in the development of secondary hyperparathyroidism
analysis [12] gives no clue on this aspect and, as far as we after biliopancreatic diversion. J Endocrinol Invest. 2008;31:845–50.
know, our study is the first one to report a direct comparison of 6. Balsa JA, Botella-Carretero JI, Peromingo R, et al. Chronic increase
the effects of LGB and restrictive techniques on sexual dys- of bone turnover markers after biliopancreatic diversion is related to
function and the first to include an evaluation of the changes in secondary hyperparathyroidism and weight loss. Relation with bone
mineral density. Obes Surg. 2010;20:468–73.
gonadal steroids after SG. In fact, our data indicated that 7. Balsa JA, Botella-Carretero JI, Gomez-Martin JM, et al. Copper and
changes in BMI and SHBG were the only significant ones zinc serum levels after derivative bariatric surgery: differences be-
between the different surgical techniques. This might be of tween Roux-en-Y gastric bypass and biliopancreatic diversion. Obes
interest, since SHBG produced by the liver has been shown to Surg. 2011;21:744–50.
8. Mechanick JI, Youdim A, Jones DB, et al. Clinical practice guide-
have important effects on insulin sensitivity and the regulation lines for the perioperative nutritional, metabolic, and nonsurgical
of body weight [39]. support of the bariatric surgery patient–2013 update: cosponsored
Our study, however, was not free of limitations. First, by American Association of Clinical Endocrinologists, The Obesity
because this study was not a randomized trial, patients in the Society, and American Society for Metabolic & Bariatric Surgery.
Obesity (Silver Spring). 2013;21 Suppl 1:S1–S27.
LGB group were more obese at baseline than those submitted 9. Yip S, Plank LD, Murphy R. Gastric bypass and sleeve gastrectomy
to restrictive procedures. This difference in weight, however, for type 2 diabetes: a systematic review and meta-analysis of out-
does not invalidate the results because both groups were comes. Obes Surg. 2013;23:1994–2003.
comparable in terms of total testosterone, SHBG and free 10. Escobar-Morreale HF, Botella-Carretero JI, Alvarez-Blasco F, et al.
The polycystic ovary syndrome associated with morbid obesity may
testosterone levels, and age. Second, our relatively small resolve after weight loss induced by bariatric surgery. J Clin
sample size did not permit a separate comparison of LGB, Endocrinol Metab. 2005;90:6364–9.
SG, and AGB, explaining why we considered patients sub- 11. Botella-Carretero JI, Balsa JA, Gomez-Martin JM, et al. Circulating
mitted to SG and AGB as a whole. Third, not all patients were free testosterone in obese men after bariatric surgery increases in
parallel with insulin sensitivity. J Endocrinol Invest. 2013;36:227–32.
revaluated at the same time intervals after bariatric surgery 12. Corona G, Rastrelli G, Monami M, et al. Body weight loss reverts
and, in fact, the only MOSH patient who did not normalize obesity-associated hypogonadotropic hypogonadism: a systematic
entirely his sex hormone profile was revaluated after only review and meta-analysis. Eur J Endocrinol. 2013;168:829–43.
OBES SURG

13. Mingrone G, Panunzi S, De Gaetano A, et al. Bariatric surgery versus 26. Fischer-Posovszky P, Wabitsch M, Hochberg Z. Endocrinology of
conventional medical therapy for type 2 diabetes. N Engl J Med. adipose tissue—an update. Horm Metab Res. 2007;39:314–21.
2012;366:1577–85. 27. Kapoor D, Clarke S, Stanworth R, et al. The effect of testosterone
14. Schauer PR, Kashyap SR, Wolski K, et al. Bariatric surgery versus replacement therapy on adipocytokines and C-reactive protein in
intensive medical therapy in obese patients with diabetes. N Engl J hypogonadal men with type 2 diabetes. Eur J Endocrinol.
Med. 2012;366:1567–76. 2007;156:595–602.
15. NIH conference. Gastrointestinal surgery for severe obesity. 28. Lee DM, Tajar A, Pye SR, et al. Association of hypogonadism with
Consensus Development Conference Panel. Ann Intern Med vitamin D status: the European Male Ageing Study. Eur J Endocrinol.
1991;115:956–61. 2012;166:77–85.
16. Shah B, Sucher K, Hollenbeck CB. Comparison of ideal body weight 29. Nimptsch K, Platz EA, Willett WC, et al. Association between
equations and published height-weight tables with body mass index plasma 25-OH vitamin D and testosterone levels in men. Clin
tables for healthy adults in the United States. Nutr Clin Pract. Endocrinol (Oxf). 2012;77:106–12.
2006;21:312–9. 30. Lerchbaum E, Obermayer-Pietsch B. Vitamin D and fertility: a sys-
17. Montero PN, Stefanidis D, Norton HJ, et al. Reported excess weight tematic review. Eur J Endocrinol. 2012;166:765–78.
loss after bariatric surgery could vary significantly depending on 31. Hammoud AO, Meikle AW, Peterson CM, et al. Association of 25-
calculation method: a plea for standardization. Surg Obes Relat Dis. hydroxy-vitamin D levels with semen and hormonal parameters.
2011;7:531–4. Asian J Androl. 2012;14:855–9.
18. Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of 32. Pellitero S, Olaizola I, Alastrue A, et al. Hypogonadotropic
simple methods for the estimation of free testosterone in serum. J Clin hypogonadism in morbidly obese males is reversed after bariatric
Endocrinol Metab. 1999;84:3666–72. surgery. Obes Surg. 2012;22:1835–42.
19. Dandona P, Dhindsa S, Chaudhuri A, et al. Hypogonadotrophic 33. Burcelin R, Thorens B, Glauser M, et al. Gonadotropin-releasing
hypogonadism in type 2 diabetes, obesity and the metabolic syn- hormone secretion from hypothalamic neurons: stimulation by insu-
drome. Curr Mol Med. 2008;8:816–28. lin and potentiation by leptin. Endocrinology. 2003;144:4484–91.
20. Guay AT. The emerging link between hypogonadism and metabolic 34. Bhasin S, Cunningham GR, Hayes FJ, et al. Testosterone therapy in
syndrome. J Androl. 2009;30:370–6. men with androgen deficiency syndromes: an Endocrine Society
21. Katabami T, Kato H, Asahina T, et al. Serum free testosterone and clinical practice guideline. J Clin Endocrinol Metab. 2010;95:2536–
metabolic syndrome in Japanese men. Endocr J. 2010;57:533–9. 59.
22. Yeap BB, Chubb SA, Hyde Z, et al. Lower serum testosterone is 35. Buvat J, Maggi M, Gooren L, et al. Endocrine aspects of male sexual
independently associated with insulin resistance in non-diabetic older dysfunctions. J Sex Med. 2010;7:1627–56.
men: the Health In Men Study. Eur J Endocrinol. 2009;161:591–8. 36. Bastounis EA, Karayiannakis AJ, Syrigos K, et al. Sex hormone
23. Vikan T, Schirmer H, Njolstad I, et al. Low testosterone and sex changes in morbidly obese patients after vertical banded gastroplasty.
hormone-binding globulin levels and high estradiol levels are inde- Eur Surg Res. 1998;30:43–7.
pendent predictors of type 2 diabetes in men. Eur J Endocrinol. 37. Hammoud A, Gibson M, Hunt SC, et al. Effect of Roux-en-Y gastric
2010;162:747–54. bypass surgery on the sex steroids and quality of life in obese men. J
24. Saboor Aftab SA, Kumar S, Barber TM. The role of obesity and type Clin Endocrinol Metab. 2009;94:1329–32.
2 diabetes mellitus in the development of male obesity-associated 38. Globerman H, Shen-Orr Z, Karnieli E, et al. Inhibin B in men with
secondary hypogonadism. Clin Endocrinol. 2013;78:330–7. severe obesity and after weight reduction following gastroplasty.
25. Cohen PG. Obesity in men: the hypogonadal-estrogen receptor rela- Endocr Res. 2005;31:17–26.
tionship and its effect on glucose homeostasis. Med Hypotheses. 39. Lancha A, Frübeck G, Gómez-Ambrosi J. Peripheral signalling in-
2008;70:358–60. volved in energy homeostasis control. Nutr Res Rev. 2012;25:223–48.