Nutrition, Metabolism & Cardiovascular Diseases (2014) xx, 1e6

Available online at www.sciencedirect.com

Nutrition, Metabolism & Cardiovascular Diseases

journa l homepage: www.e lsev ie r .com/ locate/nmcd

Central body fat changes in men affected by post-surgicalhypogonadotropic hypogonadism undergoing testosteronereplacement therapy are modulated by androgen receptor CAGpolymorphism

G. Tirabassi a, N. delli Muti a, E. Buldreghini a, A. Lenzi b, G. Balercia a,*a Division of Endocrinology, Department of Clinical and Molecular Sciences, Umberto I Hospital, Polytechnic University of Marche, Ancona, Italyb Andrology, Pathophysiology of Reproduction and Endocrine Diagnosis Unit, Policlinic Umberto I, University of Rome ‘La Sapienza’, Rome, Italy

Received 23 October 2013; received in revised form 5 February 2014; accepted 25 February 2014Available online - - -

KEYWORDSAndrogen receptor;Body composition;Hypogonadotropichypogonadism;Testosterone;Hypopituitarism;Dual-energy X-rayabsorptiometry

Abbreviations: AR, androgen receptorment therapy; DEXA, dual-energy Xinsulin-like growth factor-1; GH, grostimulating hormone; LH, luteinizingfree T4; PCR, polymerase chain reacti* Corresponding author. Division of E

Clinical and Molecular Sciences, Via CPolytechnic University of Marche, 6012596 3738; fax: þ39 071 887 300.

E-mail addresses: g.balercia@ospedaunivpm.it (G. Balercia).

Please cite this article in press as:hypogonadism undergoing testosteMetabolism & Cardiovascular Diseas

0939-4753/$ - see front matter ª 2014 Elseviehttp://dx.doi.org/10.1016/j.numecd.2014.02.016

Abstract Background and aims: Little is known about the effect of androgen receptor (AR) geneCAG repeat polymorphism in conditioning body composition changes after testosterone replace-ment therapy (TRT). In this study, we aimed to clarify this aspect by focussing our attention onmale post-surgical hypogonadotropic hypogonadism, a condition often associated with partial ortotal hypopituitarism.Methods and results: Fourteen men affected by post-surgical hypogonadotropic hypogonadismand undergoing several replacement hormone therapies were evaluated before and after TRT.Dual-energy X-ray absorptiometry (DEXA)-derived body composition measurements, pituitary-dependent hormones and AR gene CAG repeat polymorphism were considered. While testos-terone and insulin-like growth factor-1 (IGF-1) levels increased after TRT, cortisol concentrationdecreased. No anthropometric or body composition parameters varied significantly, except forabdominal fat decrease. The number of CAG triplets was positively and significantly correlatedwith this abdominal fat decrease, while the opposite occurred between the latter and D-testos-terone. No correlation of IGF-1 or cortisol variation (D-) with D-abdominal fat was found. At mul-tiple linear regression, after correction for D-testosterone, the positive association between CAGtriplet number and abdominal fat change was confirmed.Conclusions: In male post-surgical hypogonadotropic hypogonadism, shorter length of AR CAGrepeat tract is independently associated with a more marked decrease of abdominal fat after TRT.ª 2014 Elsevier B.V. All rights reserved.

; TRT, testosterone replace--ray absorptiometry; IGF-1,wth hormone; FSH, follicle-hormone; FT3, free T3; FT4,on; D-, variations.ndocrinology, Department ofonca 71, Umberto I Hospital,6 Ancona, Italy. Tel.: þ39 071

liriuniti.marche.it, g.balercia@

Tirabassi G, et al., Central bodyrone replacement therapy are mes (2014), http://dx.doi.org/10.10

r B.V. All rights reserved.

Introduction

It is well established that one of the mechanisms by whichtestosterone influences the metabolic profile is by itseffect on body composition [1]. Ageing-related testos-terone decline in men is one of the factors which causesabdominal fat accumulation, contributing to a higher riskof metabolic syndrome, type 2 diabetes and coronary heartdisease [2e5]. Similarly, in patients affected by hypo-gonadism, testosterone replacement therapy (TRT) resul-ted in total and visceral adiposity decrease [6,7] and in fat-free mass increase. [7]

fat changes in men affected by post-surgical hypogonadotropicodulated by androgen receptor CAG polymorphism, Nutrition,16/j.numecd.2014.02.016

2 G. Tirabassi et al.

Testosterone has an effect on target organs via theandrogen receptor (AR), which is also expressed in muscleand adipose tissue [8]. The AR gene (Xq11eq12) contains apolymorphic trinucleotide CAG repeat sequence in exon 1,which encodes a polyglutamine chain in its NH2-terminaltranscriptional activation domain [1]. It is generallybelieved that the number of CAG repeats (the CAG repeatpolymorphism) is inversely correlated to the transcrip-tional activity of the hormoneereceptor complex [1]. Thepossible underlying mechanisms between transcriptionalactivity and AR CAG polymorphism can be attributed tovariations in the basal activity of the AR, to the functionalinteraction of the polyglutamine stretch with coactivatorssuch as ARA24 and p160 and to the decreased expressionof AR messenger RNA (mRNA) [1]. However, resultsregarding the association of this polymorphism with bodycomposition are contradictory. In studies analysinghealthy, community-dwelling or randomly selected sub-jects, some authors identified a relationship of CAG repeatnumber with various measurements of body fat (i.e., total,trunk and thigh fat mass) [1,8,9] and lean mass (total, thighand trunk amount) [1,8,10], whereas others did not findany association [11,12]. Specifically, no longitudinal studieshave been carried out which have evaluated the impact ofAR CAG polymorphism in conditioning TRT effects on theamount of body composition in hypogonadal subjects.

Given these premises, the aim of our work is to studythe role of CAG repeat polymorphism in body compositionvariations in hypogonadal subjects undergoing TRT; wefocussed on post-surgical hypogonadotropic hypogonad-ism, a rare condition which is usually associated withpartial or total hypopituitarism.

Methods

Subjects

Fourteen men were retrospectively considered. Inclusioncriteria were as follows: a) hypogonadotropic hypo-gonadism [2,13] due to surgical removal of pituitary ade-noma; b) lack of hormone imbalance, includinghypogonadism, before surgery; and c) availability offollow-up data.

Study protocol

Instrumental and biochemical evaluation at the beginningof TRT (time 0) and in the recovery phase (before the eightundecanoate testosterone injections (74e84 weeks afterthe first)) was carried out. Undecanoate testosterone(1000 mg intramuscularly) was administered 6 weeksafter the first one (loading dose), followed by similar in-jections after 10e14 weeks depending on the clinical andbiochemical profile [14].

Pituitary function deficits were as follows: all patientshad a deficit of both somatotropic and gonadotropicfunctions [2,13,15], six also had thyrotropic function deficit[16], two also corticotropic deficit [17] and three had boththyrotropic and corticotropic deficits. Glucocorticoid

Please cite this article in press as: Tirabassi G, et al., Central bodyhypogonadism undergoing testosterone replacement therapy are mMetabolism & Cardiovascular Diseases (2014), http://dx.doi.org/10.10

(cortisone acetate, 37.5e50 mg daily), somatotropic (re-combinant human growth hormone (GH), 0.3e0.6 mgdaily) and thyroid (levothyroxine, 75e100 mg daily)replacement therapies were started depending on thespecific deficit. Replacement therapies with cortisone ac-etate and levothyroxine were initiated upon diagnosis ofthe deficit. Recombinant human GH and testosterone wereadministered between 6 and 12 months after surgery; inall our subjects, replacement therapy with recombinanthuman GH was started before TRT. The period betweensurgery and the beginning of TRT (time 0) was consideredas the duration of hypogonadism.

This study is a retrospective one and the data consid-ered were part of the diagnostic work-up, with theexception of genetic and body composition evaluation,which was carried out as part of a research protocol. Thestudy was conducted according to the principles of theHelsinki Declaration and was approved by the institutionalethics committee. Participants gave their written informedconsent.

Body composition measurement

Body composition measurement was carried out as pre-viously described [3]. Briefly, a whole-body dual-energy X-ray absorptiometry (DEXA) scanner (Lunar Prodigy, GEMedical Systems, Madison, WI, USA; software enCore 2007version 11.4.) was used. The entire body was scanned.Scanned images of the whole body were subdivided intohead, trunk, left and right arm and legs. The abdominalarea, which has been shown to exhibit a relatively highcontent of visceral fat and a low content of subcutaneousfat on magnetic resonance imaging [18], was alsomeasured by DEXA between vertebrae L2 and L4. All scanswere obtained and analysed by the same physician. Bodyfat and lean mass were expressed in grams.

Hormone evaluation

Blood samples were taken at 8 AM after fasting. Thefollowing parameters were considered: follicle-stimulating hormone (FSH), luteinizing hormone (LH),total testosterone, free T3 (FT3), free T4 (FT4), cortisol,insulin-like growth factor-1 (IGF-1) and prolactin. Allhormone assays were carried out by immunoassay com-mercial kits. The reference ranges for the hormone pa-rameters studied were: FSH, 1.7e6.9 IU/L; LH, 1.6e10.0 IU/L; total testosterone, 3e8.5 ng/mL; FT3, 2.3e4.2 pg/mL;FT4, 0.8e1.8 ng/dL; serum cortisol (8.00 AM), 7e27.5 mcg/dL; IGF-1, 66e251 ng/mL for males between 40 and 50years; 57e221 ng/mL for males between 50 and 60 years;46e211 ng/mL for males between 60 and 70 years; andprolactin, 2e15 ng/mL.

Polymerase chain reaction amplification and sequencing

We carried out a previously described protocol for theamplification of AR CAG tract [19]. DNAwas extracted fromblood and the AR CAG gene polymorphism was analysed

fat changes in men affected by post-surgical hypogonadotropicodulated by androgen receptor CAG polymorphism, Nutrition,16/j.numecd.2014.02.016

Central body fat changes in men affected by post-surgical hypogonadotropic hypogonadism undergoing testosterone replacement therapy 3

using polymerase chain reaction (PCR). The number of CAGrepeats was analysed by sequencing the appropriate PCRproducts, with the direct sequencing method (BigDyeTerminator Sequencing Kit, Applied Biosystem, CA, USA)on a CEQ 2000 XL sequencer (Beckman Coulter, Fullerton,CA, USA).

Statistical analysis

The ShapiroeWilks test was applied to verify the normaldistribution of the continuous variables. Data areexpressed as median (interquartile range) when not nor-mally distributed and as mean � standard deviation whennormally distributed. Variations of the parameters studiedbetween the two phases were calculated as the valuepresent at the recovery phase minus the value present attime 0. Statistical comparison between the two phases wasmade using Student’s paired t-test. In order to study theeffect of AR CAG polymorphism on body composition,independently from that of the co-administered hor-mones, a two-step approach was adopted. First, Pearson orSpearman correlations of significant hormone variationsand of the number of AR CAG triplets with significantvariations of body composition measurements were car-ried out. Then, multiple linear regression analysis wasperformed including, as independent variables, those pa-rameters that at bivariate correlation were found to besignificantly correlated with the significant variations ofbody composition indexes and as dependent variables, thesignificant variations of body composition measurements.The level of significance was set at P < 0.05. Statisticalanalyses were performed using SPSS 16 package (SPSS Inc.,Chicago, IL, USA).

Results

Table 1 shows the general and hormone characteristics ofthe studied subjects. Significant hormone variations

Table 1 General and hormone characteristics of studied subjects.

General characteristicsAge (years) 55 � 8.65Duration of hypogonadism (months) 8 � 1.61Number of AR CAG triplets 18.3 � 3.34Hormone characteristics

Time 0FSH (IU/L) 0.85 � 0.37LH (IU/L) 0.90 � 0.34Total testosterone (ng/mL) 1.55 � 0.53FT3 (pg/mL) 3.32 � 0.46FT4 (ng/dL) 1.23 � 0.29Cortisol (mcg/dL) 16.35 (11.97-18.62)b

IGF-1 (ng/mL) 34.28 � 8.29Prolactin (ng/mL) 6.95 � 2.63

Values are expressed as mean (standard deviation) or median (interquartilAR Z androgen receptor; D Z variation; FSH Z follicle-stimulating horm1 Z insulin-like growth factor-1; NS Z not significant.a Statistical comparison between time 0 and recovery phase.b These data are normally distributed but, in order to present the varia

range.

Please cite this article in press as: Tirabassi G, et al., Central bodyhypogonadism undergoing testosterone replacement therapy are mMetabolism & Cardiovascular Diseases (2014), http://dx.doi.org/10.10

between the two phases were observed for total testos-terone and IGF-1 (concentration increase) and cortisol(concentration decrease) (Table 1). Among anthropo-metric and body composition parameters, only waistcircumference and abdominal fat decreased after TRT,whereas no significant variations of the other measure-ments were evident (Table 2). The number of CAG tripletscorrelated positively and significantly (r: 0.828;P < 0.001) with variation (D-) of abdominal fat, whereasD-testosterone showed an inverse correlation with D-abdominal fat (r: �0.570;P: 0.033) (Fig. 1). D-abdominalfat did not correlate significantly either to D-cortisol or toD-IGF-1 (Fig. 1). Multiple linear regression analysis wascarried out including, as a dependent variable, D-abdominal fat and, as independent variables, both thenumber of CAG triplets and D-testosterone. Afteradjusting for D-testosterone, CAG repeat length waspositively and significantly associated with D-abdominalfat (Table 3).

Discussion

Testosterone is one of the hormones that most influencesbody composition. It is in fact known that testosteroneinhibits fat accumulation via its link with ARs which have ahigher density in visceral than in subcutaneous adiposetissue [20]. In addition, testosterone plays an inhibitoryrole on lipoprotein lipase activity and on glycerophosphatedehydrogenase by hindering the accumulation of lipidsand favouring lipolyses in visceral adipocytes [21].Conversely, the decrease in testosterone level could beinvolved in the lack of negative regulation both of thedifferentiation of preadipocytes into mature adipocytesand of the differentiation of stem cells into adipocytes,therefore leading to hypogonadism-related overweight [2].As far as lean mass is concerned, testosterone adminis-tration stimulates muscle protein synthesis, provokesmuscle fibre hypertrophy and increases myonuclear

Recovery phase D Pa

0.73 � 0.34 �0.11 � 0.53 NS0.77 � 0.21 �0.12 � 0.46 NS3.9 � 0.34 2.41 � 0.71 <0.0013.07 � 0.46 �0.24 � 0.66 NS1.32 � 0.15 0.09 � 0.39 NS12.05 (10.10e16.50)b �0.65 (�5.87 to 0.97) 0.043152.28 � 39.58 118 � 39.53 <0.0017.72 � 3.02 0.76 � 4.64 NS

e range) if they are respectively normally or not normally distributed.one; LH Z luteinizing hormone; FT3 Z free T3; FT4 Z free T4; IGF-

ble homogeneously, they are expressed as median and interquartile

fat changes in men affected by post-surgical hypogonadotropicodulated by androgen receptor CAG polymorphism, Nutrition,16/j.numecd.2014.02.016

Table 2 Anthropometric and body composition parameters before and after testosterone replacement therapy.

Time 0 Recovery phase D Pa

Anthropometric parametersWeight (kg) 80.7 � 10.4 79.5 � 10.8 �1.2 � 2.4 NSWaist (cm) 96.4 � 10.3 95.1 � 10.4 �1.2 � 1.9 0.028Body fat parametersTotal fat (g) 25 192.7 � 3494.8 25 096.6 � 3672.2 �96.1 � 1274.9 NSTrunk fat (g) 13 594.7 � 1783.3 13 412.4 � 1927.2 -�182.2 � 643.1 NSArm fat (g) 2654.5 � 612.6 2734.1 � 642.5 79.5 � 247.2 NSLeg fat (g) 7991.3 � 1313.2 7923.5 � 1372.7 �67.8 � 387.4 NSAbdominal fat (g) 3085 � 223.7 2811.6 � 216 �273.3 � 111.2 <0.001Body lean parametersTotal lean (g) 52 721 � 6204.5 52 235 � 7296.8 �485.9 � 1970 NSTrunk lean (g) 26 145 � 4097.7 25 415 � 4362.8 �730.7 � 1377.5 NSArm lean (g) 6913.6 � 612.7 6932 � 681.8 18.4 � 361.8 NSLeg lean (g) 16 314.7 � 1842 16 537.7 � 2927.9 223 � 1723.4 NSAbdominal lean (g) 3684.4 � 232.7 3550 � 282.3 �134.4 � 296 NS

Values are expressed as mean (standard deviation) or median (interquartile range) if they are respectively normally or not normally distributed.D Z variation.a Statistical comparison between time 0 and recovery phase.

4 G. Tirabassi et al.

number per fibre [22]. Furthermore, testosterone exertsinhibitory effects on the differentiation of mesenchymalmultipotent cells into adipogenic cells and promotes theirdifferentiation into cells of myogenic lineage. [22]

This work evaluated for the first time the impact of ARCAG polymorphism on the effects of TRT on bodycomposition in subjects affected by post-surgical hypo-gonadotropic hypogonadism. We observed that the shorter

Figure 1 Scatterplots with regression lines between D-abdominal fat andtriplets (D) in the whole sample.

Please cite this article in press as: Tirabassi G, et al., Central bodyhypogonadism undergoing testosterone replacement therapy are mMetabolism & Cardiovascular Diseases (2014), http://dx.doi.org/10.10

length of AR CAG repeat is independently associated with agreater decrease of visceral fat after TRT. Consistently withthe beneficial effects of testosterone on fat mass, D-abdominal fat was also found to be negatively and signif-icantly correlated with D-testosterone, an indirect index ofthe dosage of the administered hormone. However, in oursample, the relationship between CAG repeat number andabdominal fat decrease remained significant even after

D-testosterone (A), D-cortisol (B), D-IGF-1 (C) and number of AR CAG

fat changes in men affected by post-surgical hypogonadotropicodulated by androgen receptor CAG polymorphism, Nutrition,16/j.numecd.2014.02.016

Table 3 Influence of AR gene CAG repeat length on abdominal fat.

AR CAG repeat length

Unstandardized b coefficient (95% CI)a P

D-abdominalfat (g)

35.5 (15.3e55.7) 0.003

AR Z androgen receptor; CI Z confidence interval; D Z variation.a adjusted for D-Testosterone.

Central body fat changes in men affected by post-surgical hypogonadotropic hypogonadism undergoing testosterone replacement therapy 5

adjusting for D-testosterone (Table 3), thus furtherstrengthening the independent role of the polymorphismin body fat mass regulation. In this regard, Zitzmann et al.showed, in a cross-sectional study evaluating 106 healthy20e50-year-old males, that a low number of CAG repeatswas independently associated with low body fat mass,assessed by bioimpedance measurement [9]. Furthermore,Nielsen et al. reported that, in 393 men investigated bymagnetic resonance imaging, CAG number correlatedpositively with thigh and lower trunk relative subcutane-ous adipose tissue and with lower extremity and totalrelative fat mass [8]. In addition, Lapauw et al., in 159community-dwelling men aged 75e89 years, found thatlow levels of testosterone led to a greater amount of fatmass in subjects with a longer CAG repeat length than inthose with a shorter length [1]. However, it must also beacknowledged that other authors did not find any associ-ation between CAG repeat number and body fat [11,12].

In addition, in our subjects, TRT induced a decrease inthe amount of visceral fat evenwhen no variations of otherfat distribution parameters were present (Table 2).Regarding this aspect, some evidence shows that TRTprovokes a reduction of total body fat, diversely associatedwith its decrease in the trunk, abdomen, legs and arms[6,7,23,24]. In agreement with our results, Kapoor et al.found that TRT resulted in a reduction of visceral adiposity,as assessed by waist circumference and waist/hip ratio,despite a lack of significance in the decrease of body fatpercentage [6]. In our subjects, the negligible changes inbody fat parameters after TRT are probably due to the factthat TRT facilitates the loss of the small amount of visceralfat accumulated during the few months of hypogonadism(Table 1). The selective action of testosterone on visceralfat would therefore entail a decrease of this amount of fatthat, in the presence of unchanged calorie intake, would beredistributed to other body areas without varying the totalamount of body fat. Conversely, at variance with theliterature consistently reporting lean mass increase inhypogonadal subjects undergoing TRT [7], in our sample,this parameter did not significantly change in its total orregional amount (Table 2). Even in this case, it is likely thatthe short duration of hypogonadism would not havecaused significant variations of fat-free mass, thus pre-venting TRT from significantly affecting this parameter.

Furthermore, it is important to note that, due to partialor total concomitant hypopituitarism, hypogonadal sub-jects also underwent somatotropic, thyroid and cortico-tropic replacement therapies, besides TRT, treatmentsthat are theoretically able to affect body composition.

Please cite this article in press as: Tirabassi G, et al., Central bodyhypogonadism undergoing testosterone replacement therapy are mMetabolism & Cardiovascular Diseases (2014), http://dx.doi.org/10.10

Regarding this aspect, it is known that adults and childrenwith GH deficiency have increased fat mass, and hormonereplacement therapy especially decreases visceral fat area[25]. In addition, GH supplementation increases the vol-ume of visceral organs and muscle [26]. Instead, wherecortisol is concerned, this replacement therapy is able toincrease body weight, together with a redistribution ofadipose tissue at the abdominal level and a decrease ofsubcutaneous fat in the limbs [27e30]. Moreover, thishormone usually affects muscles by inducing atrophy viathe inhibition of protein synthesis and activation of muscleproteolysis [27]. Finally, hypothyroid patients, whoseweight is on average 15e30% higher than euthyroid ones,return to their normal ponderal values during replacementtherapy [31]; in addition, an influence of thyroid hormoneson lean mass is plausible [32]. Notwithstanding theseconsiderations, in our particular sample, the associationbetween CAG repeat length and TRT-related visceral fatdecrease was not influenced by concomitant pituitary-function replacement therapies.

In conclusion, our study has demonstrated that ashorter length of AR CAG repeat tract is independentlyassociated with a more marked visceral fat decrease afterTRT. In view of the role of abdominal fat as a cardiovascularrisk factor [33e36], if further investigations shouldconfirm this trend, the study of AR CAG repeat poly-morphism could become a variable which should be takeninto account when evaluating the metabolic profile ofsubjects undergoing TRT.

Acknowledgement

All authors report no conflicts of interest related to thisstudy. No external funding, apart from the support of theauthors’ institution, was available for this study.

References

[1] Lapauw B, Goemaere S, Crabbe P, Kaufman JM, Ruige JB. Is theeffect of testosterone on body composition modulated by theandrogen receptor gene CAG repeat polymorphism in elderlymen? Eur J Endocrinol 2007;156:395e401. http://dx.doi.or-g/10.1530/EJE-06-0607.

[2] Tirabassi G, Gioia A, Giovannini L, Boscaro M, Corona G, Carpi A,et al. Testosterone and cardiovascular risk. Intern Emerg Med2013;8(Suppl. 1):65e9. http://dx.doi.org/10.1007/s11739-013-0914-1.

[3] Faloia E, Tirabassi G, Canibus P, Boscaro M. Protective effect of legfat against cardiovascular risk factors in obese premenopausalwomen. Nutr Metab Cardiovasc Dis 2009;19:39e44. http://dx.doi.org/10.1016/j.numecd.2008.02.004.

[4] Costarelli L, Muti E, Malavolta M, Cipriano C, Giacconi R, Tesei S,et al. Distinctive modulation of inflammatory and metabolic pa-rameters in relation to zinc nutritional status in adult over-weight/obese subjects. J Nutr Biochem 2010;21:432e7. http://dx.doi.org/10.1016/j.jnutbio.2009.02.001.

[5] Marcellini F, Giuli C, Papa R, Tirabassi G, Faloia E, Boscaro M, et al.Obesity and body mass index (BMI) in relation to life-style andpsycho-social aspects. Arch Gerontol Geriatr 2009;49(Suppl. 1):195e206. http://dx.doi.org/10.1016/j.archger.2009.09.029.

[6] Kapoor D, Goodwin E, Channer KS, Jones TH. Testosteronereplacement therapy improves insulin resistance, glycaemic con-trol, visceral adiposity and hypercholesterolaemia in hypogonadal

fat changes in men affected by post-surgical hypogonadotropicodulated by androgen receptor CAG polymorphism, Nutrition,16/j.numecd.2014.02.016

6 G. Tirabassi et al.

men with type 2 diabetes. Eur J Endocrinol 2006;154:899e906.http://dx.doi.org/10.1530/eje.1.02166.

[7] Isidori AM, Giannetta E, Greco EA, Gianfrilli D, Bonifacio V,Isidori A, et al. Effects of testosterone on body composition, bonemetabolism and serum lipid profile in middle-aged men: a meta-analysis. Clin Endocrinol (Oxf) 2005;63:280e93. http://dx.doi.org/10.1111/j.1365-2265.2005.02339.x.

[8] Nielsen TL, Hagen C, Wraae K, Bathum L, Larsen R, Brixen K, et al.The impact of the CAG repeat polymorphism of the androgenreceptor gene on muscle and adipose tissues in 20-29-year-oldDanish men: Odense Androgen Study. Eur J Endocrinol 2010;162:795e804. http://dx.doi.org/10.1530/EJE-09-0763.

[9] Zitzmann M, Gromoll J, von Eckardstein A, Nieschlag E. The CAGrepeat polymorphism in the androgen receptor gene modulatesbody fat mass and serum concentrations of leptin and insulin inmen. Diabetologia 2003;46:31e9. http://dx.doi.org/10.1007/s00125-002-0980-9.

[10] Walsh S, Zmuda JM, Cauley JA, Shea PR, Metter EJ, Hurley BF, et al.Androgen receptor CAG repeat polymorphism is associated withfat-free mass in men. J Appl Physiol 2005;98:132e7. http://dx.doi.org/10.1152/japplphysiol.00537.2004.

[11] Voorhoeve PG, van Mechelen W, Uitterlinden AG. Delemarre-vande Waal HA, Lamberts SW. Androgen receptor gene CAG repeatpolymorphism in longitudinal height and body composition inchildren and adolescents. Clin Endocrinol (Oxf) 2011;74:732e5.http://dx.doi.org/10.1111/j.1365-2265.2011.03986.x.

[12] Ponce-González JG, Guadalupe-Grau A, Rodríguez-González FG,Dorado C, Olmedillas H, Fuentes T, et al. Androgen receptor genepolymorphisms and the fat-bone axis in young men and women. JAndrol 2012;33:644e50. http://dx.doi.org/10.2164/jandrol.111.014415.

[13] Bhasin S, Cunningham GR, Hayes FJ, Matsumoto AM, Snyder PJ,Swerdloff RS, , et alTask Force, Endocrine Society. Testosteronetherapy in men with androgen deficiency syndromes: an Endo-crine Society clinical practice guideline. J Clin Endocrinol Metab2010;95:2536e59. http://dx.doi.org/10.1210/jc.2009-2354.

[14] Saad F, Kamischke A, Yassin A, Zitzmann M, Schubert M,Jockenhel F, et al. More than eight years’ hands-on experiencewith the novel long-acting parenteral testosterone undecanoate.Asian J Androl 2007;9:291e7. http://dx.doi.org/10.1111/j.1745-7262.2007.00275.x.

[15] Gabellieri E, Chiovato L, Lage M, Castro AI, Casanueva FF. Testinggrowth hormone deficiency in adults. Front Horm Res 2010;38:139e44. http://dx.doi.org/10.1159/000318503.

[16] Persani L. Clinical review: Central hypothyroidism: pathogenic,diagnostic, and therapeutic challenges. J Clin Endocrinol Metab2012;97:3068e78. http://dx.doi.org/10.1210/jc.2012-1616.

[17] Grossman AB. Clinical Review: the diagnosis and management ofcentral hypoadrenalism. J Clin Endocrinol Metab 2010;95:4855e63. http://dx.doi.org/10.1210/jc.2010-0982.

[18] Kamel EG, McNeill G, Van Wijk MC. Change in intra-abdominaladipose tissue volume during weight loss in obese men andwomen: correlation between magnetic resonance imaging andanthropometric measurements. Int J Obes Relat Metab Disord2000;24:607e13.

[19] Delli Muti N, Agarwal A, Buldreghini E, Gioia A, Lenzi A, Boscaro M,et al. Have androgen receptor gene CAG and GGC repeat poly-morphisms an effect on sperm motility in infertile men? Andro-logia; 2013 Jun 3. http://dx.doi.org/10.1111/and.12119 [Epub aheadof print].

[20] Freedland ES. Role of a critical visceral adipose tissue threshold(CVATT) in metabolic syndrome: implications for controlling di-etary carbohydrates: a review. Nutr Metab (Lond) 2004;1:12. http://dx.doi.org/10.1186/1743-7075-1-12.

[21] Wajchenberg BL. Subcutaneous and visceral adipose tissue: theirrelation to the metabolic syndrome. Endocr Rev 2000;21:697e738. http://dx.doi.org/10.1210/er.21.6.697.

Please cite this article in press as: Tirabassi G, et al., Central bodyhypogonadism undergoing testosterone replacement therapy are mMetabolism & Cardiovascular Diseases (2014), http://dx.doi.org/10.10

[22] Miller KK. Androgen deficiency: effects on body composition. Pi-tuitary 2009;12:116e24. http://dx.doi.org/10.1007/s11102-008-0121-7.

[23] Permpongkosol S, Tantirangsee N, Ratana-Olarn K. Treatment of161 men with symptomatic late onset hypogonadism with long-acting parenteral testosterone undecanoate: effects on bodycomposition, lipids, and psychosexual complaints. J Sex Med2010;7:3765e74. http://dx.doi.org/10.1111/j.1743-6109.2010.01994.x.

[24] Snyder PJ, Peachey H, Hannoush P, Berlin JA, Loh L, Lenrow DA,et al. Effect of testosterone treatment on body composition andmuscle strength in men over 65 years of age. J Clin EndocrinolMetab 1999;84:2647e53. http://dx.doi.org/10.1210/jc.84.8.2647.

[25] Berryman DE, List EO, Sackmann-Sala L, Lubbers E, Munn R,Kopchick JJ. Growth hormone and adipose tissue: beyond theadipocyte. Growth Horm IGF Res 2011;21:113e23. http://dx.doi.org/10.1016/j.ghir.2011.03.002.

[26] Svensson J, Lönn L, Johannsson G, Bengtsson BA. Effects of GH andinsulin-like growth factor-I on body composition. J EndocrinolInvest 2003;26:823e31.

[27] Arnaldi G, Mancini T, Tirabassi G, Trementino L, Boscaro M. Ad-vances in the epidemiology, pathogenesis, and management ofCushing’s syndrome complications. J Endocrinol Invest 2012;35:434e48.

[28] Arnaldi G, Tirabassi G, Papa R, Furlani G, Trementino L,Cardinaletti M, et al. Human corticotropin releasing hormone testperformance in the differential diagnosis between Cushing’s dis-ease and pseudo-Cushing state is enhanced by combined ACTHand cortisol analysis. Eur J Endocrinol 2009;160:891e8. http://dx.doi.org/10.1530/EJE-09-0125.

[29] Tirabassi G, Faloia E, Papa R, Furlani G, Boscaro M, Arnaldi G. Use ofthe desmopressin test in the differential diagnosis of Pseudo-Cushing state from Cushing’s disease. J Clin Endocrinol Metab2010;95:1115e22. http://dx.doi.org/10.1210/jc.2009-1146.

[30] Tirabassi G, Papa R, Faloia E, Boscaro M, Arnaldi G. Corticotrophin-releasing hormone and desmopressin tests in the differentialdiagnosis between Cushing’s disease and pseudo-Cushing state: acomparative study. Clin Endocrinol (Oxf) 2011;75:666e72. http://dx.doi.org/10.1111/j.1365-2265.2011.04096.x.

[31] Krotkiewski M. Thyroid hormones in the pathogenesis and treat-ment of obesity. Eur J Pharmacol 2002;440:85e98. http://dx.doi.org/10.1016/S0014-2999(02http://dx.doi.org/10.1016/S0014-2999(02)01420-6.

[32] Roef G, Lapauw B, Goemaere S, Zmierczak HG, Toye K,Kaufman JM, et al. Body composition and metabolic parametersare associated with variation in thyroid hormone levels amongeuthyroid young men. Eur J Endocrinol 2012;167:719e26. http://dx.doi.org/10.1530/EJE-12-0447.

[33] Chien KL, Lin HJ, Lee BC, Hsu HC, Chen MF. Relationship ofadiposity and body composition to the status of metabolic syn-drome among ethnic Chinese Taiwanese. Nutr Metab CardiovascDis 2011;21:643e50. http://dx.doi.org/10.1016/j.numecd.2010.01.007.

[34] Oliveira A, Lopes C, Severo M, Rodríguez-Artalejo F, Barros H. Bodyfat distribution and C-reactive proteinea principal componentanalysis. Nutr Metab Cardiovasc Dis 2011;21:347e54. http://dx.doi.org/10.1016/j.numecd.2009.10.013.

[35] Talaei M, Sadeghi M, Marshall T, Thomas GN, Kabiri P, Hoseini S,et al. Impact of metabolic syndrome on ischemic heart disease - aprospective cohort study in an Iranian adult population: IsfahanCohort Study. Nutr Metab Cardiovasc Dis 2012;22:434e41. http://dx.doi.org/10.1016/j.numecd.2010.08.003.

[36] Walker SE, Gurka MJ, Oliver MN, Johns DW, DeBoer MD. Racia-l/ethnic discrepancies in the metabolic syndrome begin in child-hood and persist after adjustment for environmental factors. NutrMetab Cardiovasc Dis 2012;22:141e8. http://dx.doi.org/10.1016/j.numecd.2010.05.006.

fat changes in men affected by post-surgical hypogonadotropicodulated by androgen receptor CAG polymorphism, Nutrition,16/j.numecd.2014.02.016