| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Endocrinology (F.C.W.W.), Manchester Royal Infirmary, University of Manchester, Manchester M13 9WL, United Kingdom; ARC Epidemiology Unit (A.T., S.R.P., A.J.S., J.D.F., T.W.O.), The University of Manchester, Manchester M13 9PT, United Kingdom; Department of Obstetrics, Gynaecology, and Andrology (G.B.), Albert Svent-Gyorgy Medical University, Szeged 6725, Hungary; Department of Medicine (F.C.), Santiago de Compostela University, 15705 Santiago de Compostela, Spain; Andrology Unit (G.F.), Department of Clinical Physiopathology, University of Florence, 50121 Florence, Italy; Scanian Andrology Centre (A.G.), Department of Urology, Malmö University Hospital, University of Lund, SE20502 Lund, Sweden; Department of Reproductive Biology (I.T.H.), Imperial College London, Hammersmith Campus, London SW7 2AZ, United Kingdom; Department of Andrology and Reproductive Endocrinology (K.K.), Medical University of Lodz, 90-419 Lodz, Poland; Andrology Unit (M.P.), United Laboratories of Tartu University Clinics, 50406 Tartu, Estonia; and Department of Andrology and Endocrinology (S.B., D.V.), Catholic University of Leuven, B-3000 Leuven, Belgium
Address all correspondence and requests for reprints to: Professor Frederick C. W. Wu, Department of Endocrinology, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, United Kingdom. E-mail: frederick.wu{at}manchester.ac.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of the study was to investigate the relationships between lifestyle and health with reproductive hormones in aging men.
Design: This was a baseline cross-sectional survey on 3200 community-dwelling men aged 40–79 yr from a prospective cohort study in eight European countries.
Results: Four predictors were associated with distinct modes of altered function: 1) age: lower free T (FT; –3.12 pmol/liter·yr, P < 0.001) with raised LH, suggesting impaired testicular function; 2) obesity: lower total T (TT; –2.32 nmol/liter) and FT (–17.60 pmol/liter) for body mass index (BMI; 
25 to < 30 kg/m2) and lower TT (–5.09 nmol/liter) and FT (–53.72 pmol/liter) for BMI 30 kg/m2 or greater (P < 0.001–0.01, referent: BMI < 25 kg/m2) with unchanged/decreased LH, indicating hypothalamus/pituitary dysfunction; 3) comorbidity: lower TT (–0.80 nmol/liter, P < 0.01) with unchanged LH in younger men but higher LH in older men; and 4) smoking: higher SHBG (5.96 nmol/liter, P < 0.001) and LH (0.77 U/liter, P < 0.01) with increased TT (1.31 nmol/liter, P < 0.001) but not FT, compatible with a resetting of T-LH-negative feedback due to elevated SHBG.
Conclusions: Complex multiple alterations in the hypothalamic-pituitary-testicular axis function exist in aging men against a background of progressive age-related testicular impairment. These changes are differentially linked to specific risk factors. Some risk factors operate independently of but others interact with age, in contributing to the T decline. These potentially modifiable risk factors suggest possible preventative measures to maintain T during aging in men.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Declining testosterone (T) and other anabolic hormones in men aged from the mid-30s onward (sometimes referred to as the andropause but perhaps male midlife transition is more appropriate) may influence the aging-related deteriorations in body function [e.g. frailty, obesity, osteopenia, cognitive decline, and erectile failure (4, 5, 6, 7)]. However, the associations between endocrine dysregulation and adverse health outcomes or health-related quality of life are inconsistent and poorly characterized. This has given rise to much uncertainty regarding hormonal interventions for symptomatic older men (8).
New paradigms for the study of the natural history and mechanisms of the midlife transition in men are required to improve our understanding of the relationships between symptoms and signs with a broad range of endocrine changes that develop with increasing age. This requires better definition of ensemble endocrine regulatory cascades rather than individual hormones. Moreover, these models should also include lifestyle, comorbid, socioeconomic, and genetic factors that may interact with increasing chronological age. This broader-based approach, reflecting the contributions of complex exposures, effect modifiers, and confounders, is more likely to provide clearer answers to critical research questions in endocrine aging.
We hypothesize that: 1) modifications in hypothalamic-pituitary-testicular (HPT) axis are extant in aging men; 2) altered HPT functions during aging reflect interactions between age and exposures to extrinsic risk factors, and 3) age and other risk predictors influence HPT axis function via distinct control pathways.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Study design
The EMAS is a multinational, community-based, prospective cohort study comprising two phases: 1) a cross-sectional baseline survey of a random population sample of ambulatory men aged 40–79 yr recruited in 2003–2005 and 2) a follow-up survey 4 yr later. The eight participating centers were Manchester (UK), Leuven (Belgium), Malmö (Sweden), Tartu (Estonia), Lodz (Poland), Szeged (Hungary), Florence (Italy), and Santiago de Compostela (Spain). Ethics approval for the study was obtained in accordance with local institutional requirements in each center.
Study population sampling
Each center aimed to recruit 400 community-dwelling men from the general population via general practice (three centers), municipal (three centers), and electoral (two centers) registers. Stratified random sampling by computer allocation was used to recruit equal numbers of men into each of four age bands: 40–49, 50–59, 60–69, and 70–79 yr. There were no specific exclusion criteria as long as the subject could respond to the postal invitation and provide written informed consent.
Subjects
A total of 8416 men were invited by letter to attend, of whom 3369 eventually participated. The number recruited per center varied from 396 to 445. The mean adjusted response rate across the eight centers was 45% (range 24–62%). The mean age of subjects recruited, 59.7 yr [95% confidence interval (CI) 59.3–60.1], was similar across all eight centers.
Hormone measurements
A single fasting morning (before 1000 h) venous blood sample was obtained from all subjects. Serum was separated immediately after phlebotomy and stored at –80 C until assay at the end of the baseline study. All samples were transported in frozen state to a single laboratory (General Laboratory, Azienda Ospedaliero-Universitaria Careggi, Florence, Italy) where they were assayed for T, LH, and SHBG by the Modular E170 platform electrochemiluminescence immunoassays (Roche Diagnostics, Mannheim, Germany). Free T levels were derived from total T, SHBG, and albumin concentrations (10). Assay performance was assessed in a serum pool processed 20 times for within-assay imprecision and for 20 separate assays at 1-monthly intervals for between-assay imprecision. Within- and between-assay coefficients of variation for T were 1.05 and 3.72%, LH 1.88 and 3.01%, and SHBG 1.70 and 3.18%, respectively. Detection limits of the respective assays were 0.07 nmol/liter, 0.10 U/liter, and 0.35 nmol/liter. Stability of samples during long-term storage was evaluated by the reproducibility of hormone values in a pool of sera stored under the same condition as the EMAS serum samples (–80 C). The pool was analyzed monthly from January 2003 to July 2006 and a coefficient of variation of 4.06, 5.80, and 2.73% for T, LH, and SHBG, respectively, was obtained with no evidence of any systematic drift with time.
Other measurements and data collection
Self-completed and interviewer-assisted standardized validated questionnaires (in local languages) described in detailed elsewhere (9) gathered information on sociodemographic, general health status, medical conditions, medications, smoking, and alcohol consumption. Height, weight and waist circumference were measured in the standing position. Body mass index (BMI) was calculated as body weight (kilograms) divided by the square of height (meters). Comorbid conditions included self-reported heart condition, high blood pressure, bronchitis, asthma, peptic ulcer, epilepsy, diabetes, cancer, liver conditions, kidney conditions, prostate diseases, and thyroid disorders.
Statistical analyses
From a total of 3369 participants, 149 were excluded because of prevalent pituitary or testicular diseases or current use of medications that could affect pituitary/testicular functions (testosterone, dehydroepiandrosterone, antiandrogens, GnRH agonists, glucocorticoids, and psycholeptic agents) or sex steroid clearance (e.g. anticonvulsants), with 3220 men in the analysis sample.
Because hormones in the pituitary-testicular axis (total T, free T, and LH) are functionally linked by dynamic feed-forward and feed-back interrelationships (11) as well as being influenced by the binding effects of circulating SHBG (12), changes in any one of the four parameters under consideration do not occur in isolation but reflect the altered equilibrium established after compensatory adjustments in each of the other linked components. Our approach to data analyses and interpretation therefore aimed to reflect the collective changes between hormones of the pituitary-testicular axis and SHBG to explore their ensemble relationships to age, anthropometric, lifestyle, and comorbid factors.
Data analyses were conducted in three stages. First, initial exploratory analyses were carried out to assess the distribution of the four outcomes (total T, free T, LH, and SHBG), their age trends, and any between-center differences. Second, the center effect was assessed by multilevel regression models (random intercept) (13), which incorporated the hierarchical structure of the analysis sample (individuals nested within centers). Third, having detected no major difference in the mean hormone levels between the centers (see Cohort characteristics in Results), we then used the entire cohort to investigate the relationships between the four hormones and the covariates of interest. Multivariate multiple regression analysis (also known as multiple outcomes regression) was used to jointly regressed the four outcome variables simultaneously against all the independent predictors: age (as continuous variables), age2 (as continuous variables to describe curvilinear relationships), BMI, smoking, alcohol consumption, and comorbidity (as categorical variables). The regression coefficients were obtained, and then test statistics were considered, the null hypothesis being tested is that the four β-coefficients for each covariate from the regression model are all simultaneously equal to zero, i.e. for an independent variable x, the null hypothesis of the multiple test is Ho: TTx = FTx = LHx = SHBGx = 0, where TTx, FTx, LHx, and SHBGx are, respectively, the regression coefficients of x for total T, free T, LH, and SHBG. By using multiple outcomes regression, the coefficients across equations could be tested. Age and age2 were both centered at the mean age of 60 yr to allow comparison of the regression coefficients.
We considered three regression models: model 1 included only main effects; model 2 added to model 1 any two-way interaction between age or age2 and the categorical covariates; and model 3 added to model 2 any two-way interaction between the categorical covariates. Because model 1 was nested within model 2, and model 2 nested within model 3, the log likelihood ratio test was appropriate to compare model 1 with model 2 and model 2 with model 3. The hypothesis was that the set of parameters added to the simpler model were all nonsignificant. For models 1 and 2 the –2 log (likelihood) value was 87156.6, and 87077.7, respectively (the P value from the log likelihood ratio test comparing model 2 vs. model 1 was P < 0.00, rejecting the null hypothesis of nonsignificance), The –2 log (likelihood) value for model 3 was 87045.0 (the P value from the log likelihood ratio test comparing model 3 vs. model 2 was P < 0.63), indicating that model 2 was the most parsimonious. We therefore used model 2 as the final model for regression analyses. All parameters (significant and nonsignificant) were reported in the final result table (see Table 2
). Potential outliers were excluded from the sample used in the final model. The relative importance of each predictor was estimated by comparing changes in variance (R2) after excluding one of the predictor variables from the model that included only main effects (model 1). Results were considered statistically significant if the null hypothesis could be rejected at the 0.05 (two tailed) level. All data analyses were performed using STATA version 9.2 (Stata Corp., College Station, TX) and SAS proc mixed (SAS version 9; SAS Institute, Cary, NC).
|
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Characteristics of the 3220 EMAS men are shown in Table 1. A quarter (24.4%, BMI 30 kg/m2) to a third (35.1%, waist circumference 102 cm) of the cohort was obese. Current smokers comprised 21.6%, whereas one or more comorbid conditions were reported in 48.0%, this being more prevalent in older men. There was no statistically significant effect of center on hormone levels as judged by the variance and its SE except for free T. The proportion of variance in hormone levels explained by center (after adjustment for age and BMI) was 2% for total T, 4% for free T, 0.5% for LH, and 2% for SHBG, respectively. Having shown that center effect was relatively minor, men from all eight centers were combined for the subsequent analyses.
|
There were significant cross-sectional age trends in total T, free T, LH, and SHBG (Fig. 1): unadjusted annual age trends (β) for total T was –0.04 nmol/liter·yr (P < 0.001); free T, –3.40 pmol/liter·yr (P < 0.001); LH, 0.12 U/liter·yr; and SHBG, 0.65 nmol/liter·yr (P < 0.001). The magnitude of the age trend in total T was modest, compared with that of free T. Total T decreased by only less than 1.5 nmol/liter or 8.6% across the 4 decades, equivalent to a rate of –0.4% per annum (log hormone-age), whereas free T decreased by 116 pmol/liter (33.1%), equivalent to –1.3% per annum.
|
). The adjusted age trends were linear for free T and SHBG but nonlinear for LH. The age2 effect for LH showed progressively higher levels of these parameters in older men. Hormones and BMI
BMI was stratified into three categories: less than 25 kg/m2 (nonobese), 25 or greater to less than 30 kg/m2 (overweight), and 30 kg/m2 (obese). Mean total T was significantly lower in the overweight (–2.32 nmol/liter, P < 0.001) and obese (–5.09 nmol/liter, P < 0.001), compared with the nonobese reference group across all ages (Table 2 and Fig. 2). Mean free T was significantly lower in the overweight (–17.60 pmol/liter, P < 0.01) and obese (–53.72 pmol/liter, P < 0.001), compared with the nonobese reference group. There was no significant interaction between age and BMI for both total T and free T (Table 2).
|
and Table 2).
Mean SHBG was significantly lower in the overweight (–6.90 nmol/liter, P < 0.001) and obese (–12.01 nmol/liter, P < 0.001) groups, respectively, compared with the nonobese (Fig. 2 and Table 2). The significant positive age trends in SHBG in all three groups were preserved and similar.
Hormones and waist circumference
The overall hormone patterns in men categorized by waist circumference were very similar to that using BMI (data available as supplemental information online, published as supplemental data on The Endocrine Society’s Journals Online Web site at http://jcem.endojournals.org).
Hormones and comorbidity
The cohort was initially stratified into three categories: those with none, one, and two or more comorbid conditions. Because there was no difference between one and two or more comorbidity (confirmed by regression models, data not shown), we used none vs. one or more comorbid conditions with each given a similar weighting.
Comorbidity was associated with lower total T (–0.80 nmol/liter, P < 0.01) and SHBG (–2.21 nmol/liter, P < 0.05) (Table 2 and Fig. 3), independently of age. There was no difference in free T between the groups. Although mean LH was not different with respect to the comorbidity status at the mean age of 60 yr, there was a highly significant positive age2 and comorbidity interaction effect with a marked curvilinear trend (P < 0.001), so that mean LH was substantially higher in older men with comorbidity (Table 2 and Fig. 3).
|
The cohort was first stratified into three categories: never-smoked, ex-smoker, and current smoker. There was no significant difference between ex-smoker and never-smoked (data not shown). We therefore combined ex-smoker and never-smoked categories as the reference category in the regression model (Table 2). The mean total T (1.31 nmol/liter, P < 0.001), LH (0.77 U/liter, P < 0.01), and SHBG (5.96 nmol/liter, P < 0.001) were higher in current smokers (Table 2 and Fig. 4). There was no significant effect of smoking status on free T (Table 2 Fig. 4). There was also no significant interaction effect of smoking and age in the four outcomes.
|
The cohort was stratified into frequent (five or more times per week) and infrequent (less than five times per week) drinkers or teetotalers. Individual as well as multiple tests showed no significant association between the three hormones or SHBG with alcohol intake (Table 2).
Relative importance of predictors
The total variance (R2) accounted for in the full regression model including main effects of age, BMI, smoking, alcohol, and comorbidity was 13% for total T, 21% for free T, 14% for LH, and 23% for SHBG (Table 2). For total T, BMI was by far the most important determinant (R2 changed from 13 to 4% if BMI was excluded from the full model) followed far behind by smoking. In contrast, for free T age (R2 changed from 21 to 10%), BMI (R2 changed from 21 to 17%), and comorbidity, and for SHBG age (R2 changed from 23 to 8%), BMI (R2 changed from 23 to 16%), and smoking were important in descending order. Age was by far the most important determinant for LH (R2 changed from 14 to 3%) with only small contributions from obesity, smoking, and comorbidity equally (full table is available as supplemental information online).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The apparent lack of major differences in hormone levels of the HPT axis between centers with different socioeconomic and health conditions is noteworthy. This is the first European study to show such uniformity. Recent data from the United States and Australia also showed that T levels are not significantly different between men from different ethnic groups (14) and geographic regions (15). A more detailed exploration of differences between centers in EMAS will be the subject of a separate report. The main focus of the current analysis was to identify patterns of predictor-outcome relationships that may reveal possible underlying mechanisms, taking the cohort as a whole.
The present study confirms the age-related cross-sectional trends in circulating T in community-dwelling men (6). The magnitude of fall in total T is modest (–0.4% per annum); this is similar to some (16, 17) but less than that of other cross-sectional studies typically reporting a 0.4–2% annual decline (18, 19, 20). Furthermore, the age trend in total T in the present study became statistically insignificant after adjusting for other covariates, suggesting that these risk factors may confound any apparent effects of age. Consistent with previous cross-sectional studies, the age trend in free T was more substantial (–1.3% per annum).
The core hormonal pattern with increasing age is suggestive of incipient primary testicular dysfunction with maintained total T and progressively blunted free T associated with higher LH. This interpretation is supported by the age-related attrition of the Leydig cell population (21), reduced steroidogenic enzymic activity (22), and attenuation of the testicular response to LH. The increasingly higher LH in older men in the face of stable total or free T age trends suggests that testicular function may be further impaired or gonadal feedback action may be attenuated at the hypothalamic level (11). These two possible explanations are not mutually exclusive and other factors (e.g. comorbidity) may contribute (see later text).
Obesity was associated with progressively lower total and free T independent of the simultaneous decrease in SHBG. This is consistent with many previous reports (23, 24 and reviewed in Ref. 25). However, our data highlight the fact that LH was unchanged or even lower in older men in the face of lower T in obesity, suggesting that there may be a failure at the hypothalamic-pituitary level. The effect of increasing BMI (and waist size) on circulating T was more substantial, compared with that of the age (Figs. 1 and 2; see supplementary figure on waist circumference, published as supplemental data on The Endocrine Society’s Journals Online Web site at http://jcem.endojournals.org). Thus, a change in BMI from nonobese to obese may be equivalent to a 15 yr fall in T. Similar findings have recently been reported in longitudinal U.S. population studies (26). It is noteworthy that the effects of obesity on the HPT axis can be dissociated from that of age: at different strata of BMI, the age trends (or the lack of it in the case of total T) of free T and LH were clearly preserved. This pattern supports the hypothesis that different underlying mechanisms influence the functions of the HPT axis: age predominantly affects testicular function, whereas obesity impairs hypothalamic/pituitary function. Thus, whereas the effects of aging on testicular function can be moderated by increased LH compensation for many decades, obesity impairs hypothalamic/pituitary function independent of age, arguably an adaptive response for which there should be no compensatory mechanism.
Although SHBG was negatively associated with increasing strata of obesity, the positive relationship with age was well preserved in obese as well as in the nonobese men, clearly demonstrating the concurrent but opposite (and separate) effects of obesity and age on SHBG (27). Obesity is associated with insulin resistance (28), and the increased circulating insulin inhibits hepatic SHBG synthesis (29). On the other hand, the SHBG increase with age may be related to relative IGF-I deficiency (27), although this has not been directly proven.
Obesity is associated with peripheral and central insulin resistance (30) and proinflammatory cytokine production (TNF
and IL-6) from adipocytes (31) and central nervous system endocannibinoid release (32), all of which are potential candidates for abrogating hypothalamic endocrine and downstream reproductive axis functions.
The relationship between obesity and T can be bidirectional: low T may be the cause rather than consequence of obesity (33, 34). Men with acquired hypogonadism develop redistributed and increased fat mass (35), whereas T replacement in hypogonadal men reduced fat mass (36). However, effects of T treatment on fat mass are modest in magnitude, amounting to no more than 2–3 kg decrease or less than 6% loss of fat mass on whole-body dual-energy x-ray absorptiometry, generally without any discernible shifts in BMI or waist circumference (6).
Despite the significantly lower total T in men aged 40–69 yr with comorbidity, LH was similar between those with and without comorbid conditions. This suggests a failure of the hypothalamus/pituitary level to compensate for the hypoandrogenemia (similar to obesity but not as marked). The disproportionately higher LH observed in older men (70 yr) with comorbid conditions may be associated with deterioration in testicular function, but an additional hypothalamic defect, i.e. a reduction of negative feedback efficacy by T (37), may also contribute.
Previous studies have shown conflicting results on the relationships between smoking and total and free T (38). Our findings suggest that the apparent increase in total T in smokers occurs through a primary increase in SHBG with a compensatory rise in LH. After this resetting of the "gonadostat," smokers’ free T remained unchanged. This situation is the same as the response of the pituitary-thyroid axis to increased thyroid-binding globulin during pregnancy (39). Although chronic alcohol abuse is known to suppress LH (40), our data showed no significant association among the three hormones or SHBG and alcohol intake.
The current results support our hypothesis that age is not a simple chronological summator of lifestyle, stress, or chronic disease exposures. Age and other risk factors independently and conjointly exert their effects on HPT axis function during male midlife transition. Similar to other cross-sectional studies mentioned above, up to 21% of the variance in T was explained by the predictors considered in this analysis (Table 2). However, the effects of obesity (BMI or waist circumference) was by far the most important determinant of variance in total T, whereas age per se was important for SHBG, LH, and free T with comorbidity and smoking being comparatively minor contributors. It is noteworthy that these predisposing lifestyle and health factors are modifiable. This implies that the apparent age-related decline in T may constitute a barometer of health and thus be potentially preventable and/or reversible. Our cross-sectional data therefore do not favor the concept of an inevitable sex hormone deficiency syndrome of aging in men, in contrast to the menopausal transition in females.
There are a number of limitations in this study. The cross-sectional nature of the data cannot be definitive about the temporal and directional natures of the described relationships. In addition, there is some evidence in American men of complex age-period-cohort effects suggesting population data may not be at equilibrium over time (41). Nonetheless, longitudinal data showing that obesity and poor health predict declines in androgens (26, 42) lend support to our interpretations. The longitudinal follow-up data from EMAS will provide a good opportunity to substantiate our hypotheses in European men. Although only a single blood sample from each individual was analyzed, the key hormone of interest, T, is not a labile analyte (43) and single measurements of T on morning samples, can provide representative and reliable data in large epidemiological studies such as the EMAS. Our data are based on men randomly sampled from the general population. The response rate in the study was 45%. It is possible that levels of sex hormones among those who participated may have differed from those who declined. In EMAS, based on a sample of nonparticipants, there was some evidence that nonparticipants were slightly older and left education earlier, although the differences were small in absolute terms. Furthermore, there were no consistent differences in measured lifestyle factors or morbidities between participants and nonparticipants, providing some reassurance against major response bias, although we cannot exclude this. Accordingly, our results, accrued from a community-dwelling European population, should be extrapolated beyond this setting only with caution.
In conclusion, the present results show that multiple functional alterations in the HPT axis, specifically linked to distinct risk factors, are simultaneously superimposed on a background of progressive testicular impairment associated with increasing age. The risk factors operate largely independently of, but can also interact with, chronological age in contributing to the lowering of T levels in some middle-aged and elderly men. Because some of these risk factors are potentially modifiable, the present results, if confirmed, support possible preventative strategies to maintain T levels in aging men.
|
Acknowledgments |
|---|
Authors’ contributions included: F. C. W. Wu: study coordinator, principal investigator, chair of project coordinating committee, was responsible for study design, study management, data interpretation, and manuscript writing, and had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis; A. Tajar: statistician; responsible for data analysis, preparation of tables and figures; and contributed to manuscript writing; and had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis; S. R. Pye: statistician and contributed to data analysis and preparation of tables and figures; A. J. S. Silman: coprincipal investigator, was responsible for study design and study management, and contributed to data interpretation; T. W. O'Neill: member of the project coordinating committee and contributed to data interpretation; G. Bartfai: lead investigator, Szeged, Hungary, member of project coordinating committee, was responsible for data collection, and contributed to data interpretation; S. Boonen: coinvestigator, Leuven, Belgium, member of project coordinating committee, was responsible for data collection, and contributed to data interpretation; F. Casanueva: lead investigator, Santiago de Compostela, Spain, member of project coordinating committee, was responsible for data collection, and contributed to data interpretation; J. D. Finn: study administrative coordinator, member of project coordinating committee, was responsible for study management, data collection, data quality control, and contributed to data interpretation; G. Forti: Lead investigator, Florence, Italy, member of Project Coordinating Committee, was responsible for data collection, hormone analyses and contributed to data interpretation and manuscript writing; A. Giwercman: lead investigator, Malmö, Sweden, member of project coordinating committee, and was responsible for data collection; I. T. Huhtaniemi: lead investigator Turku, Finland, member of project coordinating committee, was responsible for study design, and contributed to data interpretation and manuscript writing; K. Kula: lead investigator Lodz, Poland, member of project coordinating committee, was responsible for data collection, and contributed to data interpretation; M. Punab: lead investigator Tartu, Estonia, member of project coordinating committee, was responsible for data collection, and contributed to data interpretation; and D. Vanderschueren: lead investigator, Leuven, Belgium, member of project coordinating committee, was responsible for data collection, and contributed to data interpretation.
We thank the men who participated in the eight countries; the research/nursing staff in the eight centers (C. Pott, Manchester; E. Wouters, Leuven.; M. Nilsson, Malmo; M. del Mar Fernandez, Santiago de Compostela; M. Jedrzejowska, Lodz; H.-M. Tabo, Tartu; A. Heredi, Szeged) for their meticulous data collection; the technical staff at the General Laboratory (Azienda Ospedaliero-Universitaria Careggi, Florence, Italy) for hormone assays; P. Steer (Manchester) for data management; C. Moseley (Manchester) for data entry and project coordination; and H. Heyes (Manchester) for preparation of figures. We also thank individual members of the project coordinating committee (G. Dunn, University of Manchester; M. Lean, University of Glasgow; and N. Pendleton, University of Manchester) for their expert advice.
|
Footnotes |
|---|
Part of this work was presented at the 88th Annual Meeting of The Endocrine Society, Boston, MA, June 26, 2006 (Abstract U6–1).
Disclosure Statement: F.C.W.W. consulted for Bayer-Schering Healthcare, Germany; Akzo-Nobel (Organon), The Netherlands; Pierre-Fabre Medicaments, France; Ardana Biosciences, U.K; Procter & Gamble, United States; and Lily-ICOS, United States, and has also received research grant funding support from Bayer-Schering Healthcare, Germany; Bayer Schering; Lilly-Icos; and other companies. All other authors have nothing to declare.
First Published Online February 12,2008
Abbreviations: BMI, Body mass index; CI, confidence interval; EMAS, European Male Aging Study; HPT, hypothalamic-pituitary-testicular; T, testosterone.
Received September 5, 2007.
Accepted February 4, 2008.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A.A. MacDonald, G.P. Herbison, M. Showell, and C.M. Farquhar The impact of body mass index on semen parameters and reproductive hormones in human males: a systematic review with meta-analysis Hum. Reprod. Update, November 4, 2009; (2009) dmp047v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. B Yeap, S A P. Chubb, Z. Hyde, K. Jamrozik, G. J Hankey, L. Flicker, and P. E Norman Lower serum testosterone is independently associated with insulin resistance in non-diabetic older men: the Health In Men Study Eur. J. Endocrinol., October 1, 2009; 161(4): 591 - 598. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Theodoraki and P.-M. Bouloux Testosterone therapy in men Menopause Int, June 1, 2009; 15(2): 87 - 92. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. T. Huhtaniemi, S. R. Pye, K. L. Limer, W. Thomson, T. W. O'Neill, H. Platt, D. Payne, S. L. John, M. Jiang, S. Boonen, et al. Increased Estrogen Rather Than Decreased Androgen Action Is Associated with Longer Androgen Receptor CAG Repeats J. Clin. Endocrinol. Metab., January 1, 2009; 94(1): 277 - 284. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Wang, E. Nieschlag, R. Swerdloff, H. M. Behre, W. J. Hellstrom, L. J. Gooren, J. M. Kaufman, J.-J. Legros, B. Lunenfeld, A. Morales, et al. Investigation, Treatment, and Monitoring of Late-Onset Hypogonadism in Males: ISA, ISSAM, EAU, EAA, and ASA Recommendations J Androl, January 1, 2009; 30(1): 1 - 9. [Full Text] [PDF]
|
|
|
|
|
|
C Wang, E Nieschlag, R Swerdloff, H M Behre, W J Hellstrom, L J Gooren, J M Kaufman, J-J Legros, B Lunenfeld, A Morales, et al. Investigation, treatment and monitoring of late-onset hypogonadism in males: ISA, ISSAM, EAU, EAA and ASA recommendations Eur. J. Endocrinol., November 1, 2008; 159(5): 507 - 514. [Full Text] [PDF]
|
|
|
|
|
|
P. J. Snyder Decreasing Testosterone with Increasing Age: More Factors, More Questions J. Clin. Endocrinol. Metab., July 1, 2008; 93(7): 2477 - 2478. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
