Few randomized clinical studies have evaluated the impact of diet and physical activity on testosterone levels in obese men with conflicting results. Conversely, studies on bariatric surgery in men generally have shown an increase in testosterone levels. The aim of this study is to perform a systematic review and meta-analysis of available trials on the effect of body weight loss on sex hormones levels.
An extensive Medline search was performed including the following words: ‘testosterone’, ‘diet’, ‘weight loss’, ‘bariatric surgery’, and ‘males’. The search was restricted to data from January 1, 1969 up to August 31, 2012.
Out of 266 retrieved articles, 24 were included in the study. Of the latter, 22 evaluated the effect of diet or bariatric surgery, whereas two compared diet and bariatric surgery. Overall, both a low-calorie diet and bariatric surgery are associated with a significant (P<0.0001) increase in plasma sex hormone-binding globulin-bound and -unbound testosterone levels (total testosterone (TT)), with bariatric surgery being more effective in comparison with the low-calorie diet (TT increase: 8.73 (6.51–10.95) vs 2.87 (1.68–4.07) for bariatric surgery and the low-calorie diet, respectively; both P<0.0001 vs baseline). Androgen rise is greater in those patients who lose more weight as well as in younger, non-diabetic subjects with a greater degree of obesity. Body weight loss is also associated with a decrease in estradiol and an increase in gonadotropins levels. Multiple regression analysis shows that the degree of body weight loss is the best determinant of TT rise (B=2.50±0.98, P=0.029).
These data show that weight loss is associated with an increase in both bound and unbound testosterone levels. The normalization of sex hormones induced by body weight loss is a possible mechanism contributing to the beneficial effects of surgery in morbid obesity.
Obesity is a complex condition with serious social and psychological dimensions, which affects virtually all age and socioeconomic groups and threatens to overwhelm both developed and developing countries. Recently, Finucane et al. (1) estimated the trends of BMI for adults aged 20 years and older in 199 countries and territories. Between 1980 and 2008, age-standardized mean BMI for men increased in every region apart from Central Africa and South Asia, with USA and UK being highest. The global change worldwide was 0.4 kg/m2 per decade for men. Accordingly, the worldwide, age-standardized prevalence of obesity was 9.8% in men in 2008, which was nearly twice the 1980 prevalence of 4.8% (1). Contrary to conventional wisdom, the obesity epidemic is not restricted to developed societies. Accordingly, a recent large population-based study involving more than 3400 subjects from eight European centers (EMAS study) showed that about half of subjects were overweight or obese with an even higher prevalence in Eastern transitional countries (2). Excess body weight is a crucial risk factor for mortality and morbidity not only for cardiovascular diseases (CVD) but also for type 2 diabetes mellitus (T2DM), cancer, and musculoskeletal disorders; these complications cause nearly 3 million deaths every year worldwide (3, 4, 5, 6, 7, 8). Despite this evidence, obesity is one of today's most blatantly visible – yet most neglected – public health problems.
Several studies have demonstrated that, in individuals at risk, intensive lifestyle interventions, which include nutritional counseling and physical activity, are able to reduce body weight and insulin resistance, preventing the progression to overt T2DM and CVD (9, 10, 11, 12). However, many observational studies have shown that most of the weight lost with diet and exercise is regained in the long term, at least in the majority of patients (13). Bariatric surgery is another possibility to rapidly reduce body weight in individuals at risk, with beneficial effects in obesity-related metabolic disturbances. A meta-analysis of observational studies of bariatric surgery involving 3188 patients with T2DM has reported that 78% had normalization of blood glucose in the absence of medications (14). However, available randomized trials are still very few and of limited size and duration (15, 16). Furthermore, some recent observational studies have suggested caution about the benefits of bariatric surgery, showing no improvement in mortality when compared with standard care (17) and a high rate of long-term surgical complications especially for laparoscopic gastric banding procedures (18).
It is known that obesity in men is associated with low testosterone and reduced sex hormone-binding globulin (SHBG) levels (19, 20, 21, 22, 23). An obesity-associated decline in SHBG might partially explain the observed fall in testosterone levels (19, 20, 21, 22, 23). However, it is important to note that an increased BMI was associated with a low measured, or calculated, free and bioavailable testosterone (24). Specific pathogenetic mechanisms involved in this phenomenon are complex and not completely understood. Evidence indicates that testosterone deficiency induces increased adiposity while increased adiposity induces hypogonadism (22, 25). Few randomized clinical studies have specifically evaluated the impact of diet and physical activity on testosterone levels in obese men. The results of these studies are essentially conflicting: some of them showed an increase of testosterone (26, 27, 28, 29, 30, 31, 32, 33, 34), others showed no change (35, 36, 37, 38, 39), and one small study has shown even a decrease in testosterone levels (40). Similarly, in the last 10 years, several trials have evaluated the impact of bariatric surgery on testosterone levels in men generally, showing an increase in testosterone levels (41, 42, 43, 44, 45, 46, 47, 48).
The aim of this study was to perform a systematic review and meta-analysis of available trials on the effect of body weight loss on sex hormones levels.
Materials and methods
A meta-analysis was performed including all studies, comparing testosterone levels before and after the diet or bariatric surgery. An extensive Medline, Embase, and Cochrane search was performed including the following words: ‘testosterone’, ‘diet’, ‘weight loss’, and ‘bariatric surgery’. The search, which accrued data from January 1, 1969 up to August 31, 2012, was restricted to English language articles and studies of human participants. Only studies with an observational period of at least 2 weeks were considered.
The identification of relevant abstracts, the selection of studies based on the criteria described above, and the subsequent data extraction were performed independently by two of the authors (G Corona and G Rastrelli), and conflicts resolved by the third investigator (M Monami). The quality of studies was assessed using the Cochrane criteria (49).
Completed but still unpublished trials were identified through a search of the www.clinicaltrials.gov website and the results, when not already available on the website, were obtained through a formal request to the authors.
Heterogeneity studies were assessed using the I2 statistics for total testosterone (TT). Considering that heterogeneity could not be excluded (I2=51.236%), mean differences in TT, SHBG, estradiol (E2), calculated free testosterone (cFT), LH, and FSH between subjects before and after the diet or bariatric surgery were calculated using both a random- and fixed-effect model. For a more conservative approach, results of random-effect models are presented; fixed-effect model is presented only when it differed from the random-effect model. In addition, in some trials, significance of between-group comparisons (P) was not reported; these studies could not be included in the analysis described above. As a consequence, a sensitivity analysis was performed, evaluating endpoint values of each parameter in different treatment groups, in a non-paired fashion (non-paired analysis). Given that most of the studies that did not describe P values reported a non-significant difference across groups, the mean (paired) analysis, which excludes these data, is likely to overestimate the effect of treatments. On the other hand, the non-paired analysis is a very conservative approach, which could underestimate the treatment effect. Meta-regression analysis was performed to test the effect of age, BMI, diabetes, and delta of weight before or after any body-weight-loss interventions on TT and E2 levels. In addition, different linear age-adjusted regression analysis models, weighting each study for the number of subjects enrolled, were performed to verify the independent effect of baseline BMI and DM as well as percent reduction in BMI (Δ-BMI) and E2 (Δ-E2) on TT after weight loss. A further multivariate model was performed, considering Δ-TT as the dependent variable and baseline TT, baseline BMI, and Δ-BMI as covariates.
All analyses were performed using Comprehensive Meta-analysis Version 2, Biostat (Englewood, NJ, USA). Multivariate analyses were performed on Statistical Package for the Social Sciences (SPSS) for Windows 17.1.
Out of 266 retrieved articles, 24 were included in the study. Of the included studies (see the trial flow summary in Fig. 1), 22 evaluated the effect of diet or bariatric surgery, whereas two compared diet and bariatric surgery. The characteristics of the trials included in the meta-analysis are summarized in Table 1.
Moderators and outcome variables in individual studies included in the meta-analysis. All data are reported as mean±s.d.
|Study||Location||Int.||Type of study||FU (weeks)||Age (years)||BMI (kg/m2)||DM (%)||TT (nmol/l)||LH (U/l)||FSH (U/l)||SHBG (nmol/l)||cFT (pmol/l)||E2 (pg/ml)|
|Stanik et al. (26)||Sepulveda, CA, USA||LCD||BA||8||–||–||4.2||13.9±3.4||–||–||–||–||36.0±14.7|
|Hoffer et al. (27)||Boston, MA, USA||LCD||BA||4||34.0±7.0||33.1||0||13.0±1.0||–||–||–||–||–|
|Pasquali et al. (28)||Bologna, Italy||LCD||BA||10||35.2±5.6||43.4±6.3||22.2||11.9||8.4||8.8||9.0||–||45|
|Strain et al. (29)||New York, NY, USA||LCD||BA||68||34.0±11.0||–||0||8.3±4.0||10.3||6.5±4.7||9.2±3.2||–||54.0±26.0|
|Leenen et al. (36)||Utrecht, The Netherlands||LCD||BA||13||40.0±6.0||30.7±2.2||0||12.7±3.2||–||–||17.0±6.0||–||27.7±7.9|
|Pritchard et al. (30)||Laval, Canada||LCD||BA||13||21.0±0.8||26.2±5.5||0||12.3±4.1||–||–||–||–||–|
|Kraemer et al.a(37)||Pennsylvania, USA||LCD||RCT||12||39.7±6.6||33.1||0||15.9±7.7||–||–||–||–||–|
|Kraemer et al.b(37)||Pennsylvania, USA||LCD||RCT||12||38.0±8.2||31.3||0||17.4±4.2||–||–||–||–||–|
|Kraemer et al.c(37)||Pennsylvania, USA||LCD||RCT||12||39.6±6.6||29.2||0||18.1±5.2||–||–||–||–||–|
|Volek et al. (38)||Storrs, CT, USA||LCD||BA||6||36.0±11.9||–||0||21.5±7.6||–||–||42.8±32.0||–||–|
|Mingrone et al. (70)||Rome, Italy||LCD||RCT||54||–||47.8±8.8||0||–||–||–||19.7±11.0||–||–|
|Kaukua et al. (31)||Helsinki, Finland||LCD||RCT||10||45.9±9.0||39.3±3.3||5.3||11.1±3.4||3.6±2.2||4.9±2.8||29.0±14.0||201.3±58.9||–|
|Niskanen et al. (32)||Kuopio, Finland||LCD||BA||9||46.3±7.5||36.1±3.8||24.0||11.2±3.9||–||–||27.6±11.9||185.0±66.0||12.1±1.1|
|Hufelder et al. (33)||Berlin, Germany||LCD||RCT||52||55.9±6.0||32.5±2.4||100||10.4±0.8||–||–||39.7±8.0||–||–|
|Reis et al. (35)||São Paulo, Brazil||LCD||RCT||104||42.2±10.5||54.0±6.7||0||11.7±5.1||4.7±1.5||3.4±1.6||–||–||37.2±11.2|
|Khoo et al. (34)||Adelaide, Australia||LCD||CBA||8||49.6±11.0||34.4±4.0||38.8||20.2±6.9||–||–||22.2±8.7||507.0±300.6||–|
|Khoo et al. (39)||Adelaide, SA, Australia||LCD||RCT||52||62.3±5.9||35.6±4.8||100||13.9±3.3||–||–||30.4±13.9||296.0±47.0||–|
|Khoo et al.d(39)||Adelaide, SA, Australia||LCD||RCT||52||58.1±11.4||35.1±4.3||100||11.7±3.6||–||–||22.5±9.3||285.0±87.0||–|
|Bastounis et al. (41)||Athens, Greece||BSe||BA||52||34.7±7.7||57.1±7.4||0||12.3±6.2||2.4±1.6||2.9±2.0||20.3±16.8||–||64.4±24.9|
|Mingrone et al. (70)||Rome, Italy||BSf||RCT||54||–||48.0±5.4||0||–||–||–||13.2±6.5||–||–|
|Globerman et al. (42)||Tel-Aviv, Israel||BSe||BA||46.4||38.2±10.1||44.3±6.8||35.3||13.4±7.2||4.9±3.2||8.0±5.6||–||–||11.1±3.4|
|Alagna et al. (43)||Sassari, Italy||BSf||BA||52||–||47.3||75||9.6±3.7||2.4±1.6||2.9±1.9||–||–||44.0±29.0|
|Hammoud et al. (44)||Salt Lake City, UT, USA||BSf||CBA||104||48.9±9.6||46.2±7.2||–||11.8±5.9||–||–||–||244.3±9.9||39.0±26.7|
|Omana et al. (45)||St Louis, MO, USA||BSg||BA||52||48||48.3||43.2||10.6±||5.1||13.4||35.9||199.0||34.4|
|Reis et al. (35)||São Paulo, Brazil||BSf||RCT||104||36.7±11.5||55.7±7.8||0||11.7±4.5||4.8±1.7||4.0±3.4||–||–||47.8±15.5|
|Botella et al.h(46)||Madrid, Spain||BSf||BA||24||40.0±10.3||47.1±6.0||35||10.1±3.5||3.2±2.2||4.2±2.2||18.3±8.3||258.0±100.4||37.6±13.4|
|Facchiano et al. (47)||Florence, Italy||BSg||BA||24||39.2||45.7±6.3||25||8.6±3.1||2.8±1.9||3.7±2.5||19.6±7.6||219.0±62.6||43.8±13.4|
|Pellitero et al. (48)||Barcelona, Spain||BSg||BA||52||40.5±9.9||50.3±6.1||15.2||8.6±3.2||–||–||18.3±11.8||230.2±82.6||42.3±9.9|
|Woodard et al. (50)||Stanford, CA, USA||BSf||BA||52||48.1±10.4||48.2±12.0||42.0||9.0±4.0||–||–||–||–||–|
Int., intervention; LCD, low-calorie diet; BS, bariatric surgery; FU, follow-up (weeks); DM, diabetes mellitus; TT, total testosterone; E2, estradiol; SHBG, sex hormone-binding globulin; cFT, calculated free testosterone; BA, controlled cohort before-and-after comparisons in the same group of patients; RCT, randomized controlled trials; CBA, controlled before-and-after study between two or more groups of participants receiving different interventions.
Low-calorie diet-only group.
Low-calorie diet and aerobic exercise three-times-a-week group.
Low-calorie diet and aerobic and strength training three-times-a-week group.
Low-energy diet group.
High-protein, low-fat diet group.
Effect of Diet or bariatric surgery on hormonal modifications
Studies comparing the effect of diet or bariatric surgery on hormonal parameters enrolled 479 patients with a mean follow-up of 38 weeks. In particular, 13 studies evaluated the effect of low-calorie diet, whereas 11 investigated the role of bariatric surgery (two gastroplasty, six biliopancreatic diversion, and three mixed interventions respectively). Information on TT, SHBG, and cFT, according to Vermeulen's formula, was available in 23, 15, and nine studies respectively. In addition, data on E2, FSH, and LH were available in 14, 10, and 10 studies respectively. The mean percent body weight loss at endpoint was 9.8±4.5% in the low-calorie diet and 32.0±5.4% in bariatric surgery.
The Begg adjusted rank correlation test (Kendall's τ=0.02, P=0.88), calculated on the basis of TT, suggested no major publication bias.
The results of fixed-effect models were not different from those of random-effect models; therefore, only the latter will be reported. When considering data on TT, both low-calorie diet and bariatric surgery were associated with a significant (P<0.0001) increase in plasma TT, with bariatric surgery being more significantly effective in comparison with the low-calorie diet (Fig. 2A). In particular, malabsorptive surgery (including studies which performed biliopancreatic diversion, see references (35, 43, 44, 46, 50)) determined a more consistent increase in TT when compared with non-malabsorptive surgery (including studies which performed gastroplasty approach, see references (41, 42)) (11.58 (7.12–16.03) vs 6.08 (2.38–9.78) nmol/l, respectively; both P<0.0001). However, the latter was not confirmed after adjusting for percent weight loss of BMI (Δ-BMI).
Meta-regression analysis showed that differences in TT, after any approach for body weight loss, were significantly higher in patients with higher TT at baseline as well as in younger and more obese subjects (Fig. 3A, B and C). Furthermore, a higher Δ-BMI was associated with a higher testosterone increase (Fig. 3D). The prevalence of diabetes at baseline was, conversely, inversely related to the TT increase after body weight loss (Fig. 3E). However, meta-regression analysis did not show any association between the Δ-E2 and TT modifications after weight loss (data not shown).
In order to evaluate the major determinants of the percent increase in TT after weight loss (Δ-TT), alternative, age-adjusted, stepwise, logistic regression analyses were explored using Δ-TT as the dependent variable and baseline TT, BMI, and DM as well as percent reduction in BMI (Δ-BMI) and E2 (Δ-E2) after body weight loss as putative predictors. Baseline TT, BMI, and Δ-BMI as well as the percent increase in SHBG (Δ-SHBG) were all significantly associated with Δ-TT (B=5.16±2.13, P=0.022; B=2.51±0.84, P=0.007; B=2.76±0.71, P=0.001; and B=0.70±0.24, P=0.013 respectively), whereas no association between Δ-TT and Δ-E2 or DM at baseline was observed after adjusting for age (data not shown). When a further model was performed considering Δ-TT as the dependent variable and baseline TT, BMI, and Δ-BMI as covariates, only Δ-BMI retained a significant association with Δ-TT (B=2.50±0.98, P=0.029).
Similar to what was observed for TT, SHBG (P<0.0001) and cFT levels significantly increased after body weight loss (P<0.0001; see also Fig. 2B and C). No subgroup analyses for the body-weight-loss reduction approach were performed for SHBG and cFT, due to insufficient available data. All these data were confirmed when non-paired analyses were applied (Fig. 2D, E and F).
When considering data on E2, body weight loss was associated with a significant reduction in plasma E2 (mean changes −8.87 (−15.22;−2.52) pmol/l, P<0.0001; Fig. 4A). The data were confirmed when non-paired analysis was applied (Fig. 4B). No subgroup analyses for the body-weight-loss reduction approach and no meta-regression analyses were performed for E2, due to insufficient data.
Finally, body weight loss, whatever obtained, was significantly associated with a significant increase in both LH and FSH levels (1.31 (0.80–1.82) and 1.79 (1.28–2.30) U/l for LH and FSH, respectively; both P<0.0001, see also Fig. 5A and B). No subgroup analyses for the body-weight-loss reduction approach were performed for LH and FSH due to insufficient available data. All these data were confirmed when non-paired analyses were applied (Fig. 5C and D).
This meta-analysis shows that body weight loss is associated with a relevant increase in gonadotropins and in bound and unbound testosterone, with a decline in the estrogen level. Androgen rise is greater in those patients who lose more weight. The beneficial effects of body weight loss are more apparent in younger, non-diabetic subjects with a greater degree of obesity; however, this could be due to a greater body weight loss.
It is well known that in morbidly obese men, LH levels and pulse amplitude were attenuated, when compared with normal-weight controls (51, 52). Conversely, no modification of LH pulse frequency has been reported in obese men (51, 52). Based on these and other studies (53, 54), obesity-related male hypogonadism is now considered as a form of hypogonadotropic hypogonadism (see for a review, references (55, 56)). Obesity is characterized by a relative abundance of estrogens since P450 aromatase is highly expressed by fat tissue. The increased amount of estrogen levels might, in turn, play a negative feedback on both the hypothalamus and the pituitary, leading to decreased LH secretion (22, 55). Accordingly, it has been reported that a weekly low dose (2.5 mg) of the aromatase inhibitor letrozole can restore testosterone levels and increase LH levels in severely obese hypogonadal men (57). In line with this view, we now show that body weight loss, obtained either by a lifestyle or bariatric intervention, is associated with a decline in estrogen levels and a rise in gonadotropins and testosterone. However, no effect of Δ-E2 on TT after weight loss was observed in the meta-regression analysis. In addition, when Δ-E2 was introduced in a multiple regression model along with other covariates, Δ-BMI, and not Δ-E2, resulted as the most important determinant of the testosterone increase. Hence, other fat-associated factors, besides estrogens, should mediate a weight reduction-induced improvement in testosterone levels. It should be recognized, however, that local aromatization at the hypothalamus–pituitary unit could explain the effect of aromatase inhibitors on LH and testosterone, rather than circulating E2 itself and that, therefore, the best predictor of local E2 concentrations is systemic testosterone concentrations.
A series of adipokines has been proposed as a link between obesity and reproductive axis disorders and among them, the most extensively studied are leptin, ghrelin, and adiponectin (58). However, information on adipokine variation levels after a lifestyle or bariatric intervention is not available.
An alternative possibility is that weight reduction is associated with increased insulin sensitivity, with insulin having a permissive role on GnRH neuron activity (59). Accordingly, patients with T2DM and insulin resistance show a higher prevalence of hypogonadotropic hypogonadism (54, 55, 60), and in a mouse model of central depletion of the insulin receptor, the phenotype recapitulates both the metabolic syndrome (MetS) and hypogonadotropic hypogonadism (61). Interestingly, we now report that the diabetic condition has a negative effect on testosterone rise after the intervention; however, this difference does not retain statistical significance after adjusting for Δ-BMI, suggesting that it could be due to a greater body weight loss in non-diabetic patients.
In line with this data, the Endocrine Society Guidelines (62) and the Third International Consultation on Sexual Medicine (63) emphasized that lifestyle modifications should be strongly encouraged in hypogonadal subjects with obesity, T2DM, and the MetS. This meta-analysis demonstrates that, in fact, body weight loss significantly increases testosterone levels. However, the testosterone rise induced by lifestyle interventions was only modest, probably reflecting the relatively modest results of the targeted diet and physical activity on body weight loss (9.8% with diet vs 32% with surgery in the studies included in this meta-analysis).
Our knowledge of the long-term effects of bariatric surgery is based only on observational, non-randomized studies, often of poor methodological quality. These studies showed a relevant beneficial effect of surgery on body weight loss, survival, and obesity-related comorbidities (64, 65). These effects are evident only in patients with severe forms of morbid obesity (64). This meta-analysis shows that bariatric surgery is associated with a greater increase in testosterone levels than lifestyle interventions, particularly in those with a higher baseline BMI and who also show a greater body weight loss. It is possible that the consistent testosterone rise in patients with higher degrees of obesity might mediate some of the aforementioned positive effects of surgical interventions on health status such as the lower incidence of diabetes, the MetS, CVD, and mortality (66, 67, 68).
Male hypogonadism can be considered as one of the many adverse consequences of overweight and obesity. Body weight loss and lifestyle interventions should be the first approach offered to obese hypogonadal men. However, their effect on testosterone levels is modest. An interesting alternative is the combination of lifestyle interventions and testosterone supplementation. In a small randomized controlled trial (RCT) involving 16 subjects with newly diagnosed T2DM and the MetS, Heufelder et al. (33) demonstrated that the combination of testosterone-replacement therapy (TRT) and lifestyle interventions leads to greater therapeutic improvements in glycemic control and reverses the MetS condition after 52 weeks of treatment in comparison with lifestyle interventions only. Similarly, an 18-week, placebo-controlled RCT in obese patients with obstructive sleep apnea demonstrated that the combination of TRT and lifestyle modifications improved insulin sensitivity, reduced liver fat, and increased muscle mass in comparison with placebo and lifestyle modifications alone (69). Hence, although the combination of TRT and the diet would be expected to further improve weight loss-associated testosterone rise, larger studies are advisable to confirm this hypothesis.
Some limitations should be recognized. First, hormonal assays and reference values were not often comparable. In addition, local aromatization at the hypothalamus–pituitary unit could explain the effect of aromatase inhibitors on LH and testosterone, rather than circulating E2 itself and, therefore, the best predictor of local E2 concentrations should be systemic testosterone concentrations. Accordingly, our data suggest that circulating E2 does not represent the major determinant of testosterone increase after weight loss.
In conclusion, body weight loss induces a consistent increase in TT levels, which is greater in those who lose more body weight. Furthermore, a reduction in body weight is associated with an increase in SHBG, cFT, LH, and FSH and a reduction in E2. The normalization of sex hormones induced by body weight loss is a possible mechanism contributing to the beneficial effects of surgery in morbid obesity.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
This work was supported by the Italian Minister of University, Research and Instruction (grant nos 2009WLNXNT_002 and RBFR10VJ56_002).
FinucaneMMStevensGACowanMJDanaeiGLinJKPaciorekCJSinghGMGutierrezHRLuYBahalimAN. National, regional, and global trends in body-mass index since 1980: systematic analysis of health examination surveys and epidemiological studies with 960 country-years and 9.1 million participants. Lancet2011377557–567. (doi:10.1016/S0140-6736(10)62037-5).
CoronaGLeeDMFortiGO'ConnorDBMaggiMO'NeillTWPendletonNBartfaiGBoonenSCasanuevaFF. Age-related changes in general and sexual health in middle-aged and older men: results from the European Male Ageing Study (EMAS). Journal of Sexual Medicine201011362–1380. (doi:10.1111/j.1743-6109.2009.01601.x).
World Health Organization. Global health risks: mortality and burden of disease attributable to selected major risks. Geneva: World Health Organization 2009
TuomilehtoJLindströmJErikssonJGValleTTHämäläinenHIlanne-ParikkaPKeinänen-KiukaanniemiSLaaksoMLouherantaARastasM. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. New England Journal of Medicine20013441343–1350. (doi:10.1056/NEJM200105033441801).
KnowlerWCBarrett-ConnorEFowlerSEHammanRFLachinJMWalkerEANathanDMDiabetes Prevention Program Research Group. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. New England Journal of Medicine2002346393–403. (doi:10.1056/NEJMoa012512).
GilliesCLAbramsKRLambertPCCooperNJSuttonAJHsuRTKhuntiK. Pharmacological and lifestyle interventions to prevent or delay type 2 diabetes in people with impaired glucose tolerance: systematic review and meta-analysis. BMJ200733429912–29915. (doi:10.1136/bmj.39063.689375.55).
DixonJBSchachterLMO'BrienPEJonesKGrimaMLambertGBrownWBaileyMNaughtonMT. Surgical vs conventional therapy for weight loss treatment of obstructive sleep apnea: a randomized controlled trial. Journal of the American Medical Association20123081142–1149. (doi:10.1001/2012.jama.11580).
CoronaGMannucciEFortiGMaggiM. Following the common association between testosterone deficiency and diabetes mellitus, can testosterone be regarded as a new therapy for diabetes?International Journal of Andrology200932431–441. (doi:10.1111/j.1365-2605.2009.00965.x).
WuFCTajarAPyeSRSilmanAJFinnJDO'NeillTWBartfaiGCasanuevaFFortiGGiwercmanA. Hypothalamic–pituitary–testicular axis disruptions in older men are differentially linked to age and modifiable risk factors: the European Male Aging Study. Journal of Clinical Endocrinology and Metabolism2008932737–2745. (doi:10.1210/jc.2007-1972).
NiskanenLLaaksonenDEPunnonenKMustajokiPKaukuaJRissanenA. Changes in sex hormone-binding globulin and testosterone during weight loss and weight maintenance in abdominally obese men with the metabolic syndrome. Diabetes Obesity & Metabolism20046208–215. (doi:10.1111/j.1462-8902.2004.00335.x).
HeufelderAESaadFBunckMCGoorenL. Fifty-two-week treatment with diet and exercise plus transdermal testosterone reverses the metabolic syndrome and improves glycemic control in men with newly diagnosed type 2 diabetes and subnormal plasma testosterone. Journal of Andrology200930726–733. (doi:10.2164/jandrol.108.007005).
ReisLOFavaroWJBarreiroGCde OliveiraLCChaimEAFregonesiAFerreiraU. Erectile dysfunction and hormonal imbalance in morbidly obese male is reversed after gastric bypass surgery: a prospective randomized controlled trial. International Journal of Andrology201033736–744. (doi:10.1111/j.1365-2605.2009.01017.x).
KraemerWJVolekJSClarkKLGordonSEPuhlSMKozirisLPMcBrideJMTriplett-McBrideNTPutukianMNewtonRU. Influence of exercise training on physiological and performance changes with weight loss in men. Medicine and Science in Sports and Exercise1999311320–1329. (doi:10.1097/00005768-199909000-00014).
KhooJPiantadosiCDuncanRWorthleySGJenkinsANoakesMWorthleyMILangeKWittertGA. Comparing effects of a low-energy diet and a high-protein low-fat diet on sexual and endothelial function, urinary tract symptoms, and inflammation in obese diabetic men. Journal of Sexual Medicine2011102868–2875. (doi:10.1111/j.1743-6109.2011.02417.x).
Botella-CarreteroJIBalsaJAGómez-MartinJMPeromingoRHuertaLCarrascoMArrietaFZamarronIMartin-HidalgoAVazquezC. Circulating free testosterone in obese men after bariatric surgery increases in parallel with insulin sensitivity. Journal of Endocrinological Investigation2013In press.
FacchianoEScaringiSVeltriMSamavatJMaggiMFortiGLuconiMLuccheseM. Age as a predictive factor of testosterone improvement in male patients after bariatric surgery: preliminary results of a monocentric prospective study. Obesity Surgery201223167–172. (doi:10.1007/s11695-012-0753-6).
Eds Higgins JPT & Green S. In Cochrane Handbook for Systematic Reviews of Interventions version 5.0.1 (updated September 2008). The Cochrane Collaboration 2008. (Available from www.cochrane-handbook.org)
VermeulenAKaufmanJMDeslypereJPThomasG. Attenuated luteinizing hormone (LH) pulse amplitude but normal LH pulse frequency, and its relation to plasma androgens in hypogonadism of obese men. Journal of Clinical Endocrinology and Metabolism1993761140–1146. (doi:10.1210/jc.76.5.1140).
CoronaGMannucciMPetroneLRiccaVBalerciaGMansaniRChiariniVGiommiRFortiGMaggiM. Association of hypogonadism and type 2 diabetes in men attending an outpatient erectile dysfunction clinic. International Journal of Impotence Research200618190–197. (doi:10.1038/sj.ijir.3901391).
BhasinSCunninghamGRHayesFJMatsumotoAMSnyderPJSwerdloffRSMontoriVMTask ForceEndocrine Society. Testosterone therapy in men with androgen deficiency syndromes: an Endocrine Society Clinical Practice Guideline. Journal of Clinical Endocrinology and Metabolism2010952536–2559. (doi:10.1210/jc.2009-2354).
HoyosCMYeeBJPhillipsCLMachanEAGrunsteinRRLiuPY. Body compositional and cardiometabolic effects of testosterone therapy in obese men with severe obstructive sleep apnea: a randomised placebo-controlled trial. European Journal of Endocrinology2012167531–541. (doi:10.1530/EJE-12-0525).
MingroneGGrecoAVGiancateriniAScarfoneACastagnetoMPugeatM. Sex hormone-binding globulin levels and cardiovascular risk factors in morbidly obese subjects before and after weight reduction induced by diet or malabsorptive surgery. Atherosclerosis2002161455–462. (doi:10.1016/S0021-9150(01)00667-0).