Insulin and free oestradiol are independent risk factors for benign prostatic hyperplasia

Transcription

1 (2009) 12, & 2009 Nature Publishing Group All rights reserved /09 $32.00 ORIGINAL ARTICLE Insulin and free oestradiol are independent risk factors for benign prostatic hyperplasia J Hammarsten 1, J-E Damber 2, M Karlsson 3,4, T Knutson 2,ÖLjunggren 5, C Ohlsson 6, R Peeker 2, U Smith 7 and D Mellström 6,8 1 Department of Urology, Skaraborg Hospital, Skövde, Sweden; 2 Department of Urology, Institute of Clinical Sciences, Sahlgrenska University Hospital, Göteborg, Sweden; 3 Department of Clinical Sciences, Malmö General Hospital, Lund University, Malmö, Sweden; 4 Department of Orthopedics, Malmö General Hospital, Lund University, Malmö, Sweden; 5 Department of Medical Sciences, Uppsala University Hospital, Uppsala, Sweden; 6 Department of Internal Medicine at the Sahlgrenska Academy, Center for Bone Research, Sahlgrenska University Hospital, Göteborg, Sweden; 7 The Lundberg Laboratory for Diabetes Research, Department of Internal Medicine, The Sahlgrenska Academy at Göteborg University, Göteborg, Sweden and 8 Department of Geriatrics at the Sahlgrenska Academy, Center for Bone Research, Sahlgrenska University Hospital, Göteborg, Sweden The aetiology of benign prostatic hyperplasia (BPH) remains unclear. The objective of the present study was to test the insulin, oestradiol and metabolic syndrome hypotheses as promoters of BPH. The design was a risk factor analysis of BPH in which the total prostate gland volume was related to endocrine and anthropometric factors. The participants studied were 184 representative men, aged years, residing in Göteborg, Sweden. Using a multivariate analysis, BPH as measured by the total prostate gland volume correlated statistically significantly with fasting serum insulin (b ¼ 0.200, P ¼ 0.028), free oestradiol (b ¼ 0.233, P ¼ 0.008) and lean body mass (b ¼ 0.257, P ¼ 0.034). Insulin and free oestradiol appear to be independent risk factors for BPH, confirming both the insulin and the oestradiol hypotheses. Our findings also seem to confirm the metabolic syndrome hypothesis. The metabolic syndrome and its major endocrine aberration, hyperinsulinaemia, are possible primary events in BPH. (2009) 12, ; doi: /pcan ; published online 30 September 2008 Keywords: benign prostatic hyperplasia; insulin; lean body mass; oestradiol; the metabolic syndrome Introduction The aetiology of benign prostatic hyperplasia (BPH) is unknown, although this disorder is the most common benign neoplasm in men. 1,2 In the past, most interest has been focused on the steroid hormones, especially testosterone and oestradiol and their metabolites as promoters of BPH. 2,3 The majority of studies have failed to show an association between androgen levels and the prostate gland volume. However, many reports have shown a link between oestradiol, or the quotient between oestradiol and testosterone on the one hand, and BPH as measured by the total prostate gland volume on the other. 2,4,5 All these findings suggest that oestradiol is a promoter of BPH. We are referring to this notion as the oestradiol hypothesis. In recent years, attention has been more and more focused on the metabolic syndrome and its major Correspondence: Dr J Hammarsten, Department of Urology, Skaraborg Hospital, Skövde, Sweden, Leopardgatan 3, Verberg S , Sweden. jan.hammarsten@telia.com Received 1 May 2008; revised 13 August 2008; accepted 27 August 2008; published online 30 September 2008 endocrine aberration hyperinsulinaemia Our group started its research prompted by the clinical observation that diabetic and/or obese men seem to have a larger prostate gland than men without these disorders. In our reports, 11 out of 11 components of the metabolic syndrome were linked to fast-growing BPH. This indicates that BPH also is a component of the metabolic syndrome. We are referring to this as the metabolic syndrome hypothesis. The major endocrine aberration in connection with the metabolic syndrome is hyperinsulinaemia. 13 It seemed biologically plausible that insulin is the link between the metabolic syndrome and BPH. This relationship has been confirmed in all our previous studies using univariate statistics. 6 9 Moreover, multivariate statistics showed that fasting serum insulin was an independent risk factor for BPH. All these findings suggest that insulin is a promoter of BPH. Other authors have reached the same conclusion. 10,11,14 We refer to this as the insulin hypothesis. Thus, there seems to be a link between oestradiol, the metabolic syndrome, and insulin on the one hand, and BPH on the other. In the present report, these three hypotheses will be tested once again on a representative group of men, aged years, living in Göteborg, Sweden.

2 Methods Participants The Mr OS Study is a multicentre study on a representative group of elderly men (age years) in Sweden (n ¼ 3014), Hong Kong (n ¼ E2000) and the United States (n ¼ E6000). This investigation is an international study of osteoporosis in men that deals with clinical, anthropometric, metabolic, endocrine and genetic factors. The men in Sweden were randomly selected using national population registers and were contacted and asked to participate in the study. 15 To be eligible for participation the men had to be able to walk without aids and were not allowed to have a bilateral hip prosthesis. There were no other exclusion criteria. In this report, a subgroup of the Mr OS Study participants involving men living in Göteborg, Sweden, were examined with respect to the association between endocrine and anthropometric factors on the one hand and BPH as measured by the prostate gland volume, on the other. This subgroup included 184 men, years of age. Men previously diagnosed with prostate cancer or subjected to treatment due to BPH were excluded, after which 159 men remained to be evaluated. Pertinent participant data are given in Table 1. The Mr OS Study in Sweden was approved by the ethics committee of Göteborg University. Informed consent was obtained from all study participants. Assessment of anthropometric variables Height and weight were measured using standard equipment. Two consecutive measurements of height were performed in the same session and the average of these measurements was calculated. If there was a discrepancy of X5 mm between the first two measurements, a third measurement was performed and the average of the two values with the least mutual discrepancy was calculated. Lean body mass and fat mass were assessed using the DXA Hologic QDR 4500/ A-Delphi (Hologic). Pertinent median anthropometric variables are given in Table 1. Assessment of sex hormones and fasting insulin All serum samples were collected at 0800 hours in the morning after at least 10 h of fasting and non-smoking. Table 1 Anthropometric and endocrine variables Variables n ¼ 159 Age (years) 72 (1) Body height (cm) 176 (6) Body weight (kg) 83 (12) Prostate gland volume (ml) 47 (24) BMI 26.5 (3.5) Fat mass (kg) 18.3 (5.7) Lean body mass (kg) 61.2 (6.9) Fasting serum insulin (mu/l) 9.7 (0.9) Oestradiol (pm) 105 (41) Free oestradiol (pm) 1.8 (0.7) Testosterone (nm) 18.8 (6.6) Free testosterone (nm) 0.35 (0.12) SHBG (nm) 43 (22) Abbreviations: BMI, body mass index; SHBG, sex hormone-binding globuline. Values are given as mean±s.d. Serum insulin samples were transported directly to the laboratory. Serum samples for sex hormones were stored at 80 1C. Total oestradiol (E2) was measured using an ultrasensitive radioimmunoassay (RIA; Orion Diagnostics, Esboo, Finland; limit detection 5 pm (140 pg per 100 ml), intra-assay CV 3% and inter-assay CV 6%). Total testosterone (T) was measured using RIA (Orion Diagnostics; limit detection 0.1 nm (3 ng per 100 ml), intra-assay CV 6% and inter-assay CV 6%). Sex hormonebinding globuline (SHBG) was measured using immunoradiometric assay (Orion Diagnostics; limit of detection 1.3 nm, intra-assay CV 3% and inter-assay CV 7%). Two subjects had undetectable E2 levels and one had undetectable T levels. Free testosterone (FT) and free oestradiol (FE2) were calculated in accordance with the method described by Vermeulen et al. 16 and Van den Beld et al., 17 taking the concentrations of total T, total E2 and SHBG into account and assuming a fixed albumin concentration of 43 g l 1. All samples were analysed in duplicate in one laboratory and the duplicates were averaged for further analyses. Serum insulin was measured using an immunometric method based on chemiluminescense technology on an ADIVA Centaur (Bayer AB, Gothenburg, Sweden) with a detection limit of 0.1 muml 1 and intra- and inter-assay CV below 10%. Assessment of the prostate gland volume The prostate gland was examined using digital rectal examination and ultrasound equipment (B-K Medical Panther 2002 ADI). The prostate gland volume was determined by means of ultrasound using the ellipsoid method. 18,19 Statistical analysis Univariate associations among variables were examined using Pearson s correlation after log transformations of markedly non-normally distributed variables (Tables 2 and 3). The independent predictors of the prostate gland volume were tested using linear regression analyses including height, serum insulin, total body fat mass, total body lean mass, FT and free oestradiol (Table 4). Results Pertinent data on the study participants are given in Table 1. The mean prostate gland volume was 47 ml ( ml). Table 2 Univariate analysis of the correlation between the prostate gland volume and the endocrine variables listed below Variables Pearson s correlation coefficient P-value Fasting serum insulin Oestradiol Free oestradiol Testosterone Free testosterone Testosterone/oestradiol SHBG Abbreviation: SHBG, sex hormone-binding globuline. 161

3 162 Table 3 Univariate analysis of the correlation between the prostate gland volume and the anthropometric variables listed below Variables Pearson s correlation coefficient P-value BMI Body weight Lean body mass o0.001 Fat body mass Abbreviation: BMI, body mass index. Table 4 Multivariate analysis of the correlation between the prostate gland volume and the endocrine and anthropometric variables listed below The following endocrine variables suggested to be established components of the metabolic syndrome were positively correlated to BPH: fasting serum insulin and free oestradiol. The quotient testosterone/oestradiol and SHBG was negatively correlated to BPH as measured by the prostate gland volume. Total oestradiol and total FT did not reach statistical significance (Table 2). The following anthropometric variables suggested to be components of the metabolic syndrome were linked to BPH as measured by the prostate gland volume: body mass index (BMI), body weight and lean body mass. Fat body mass did not reach statistical significance in this relationship (Table 3). Using multivariate analysis, fasting serum insulin, free oestradiol and lean body mass were linked to BPH (Table 4). Discussion Variables b-coefficient P-value Fasting serum insulin Free oestradiol Lean body mass b-coefficients are given after adjustment for the following variables in the model: height, serum insulin, total body fat mass, total body lean mass, free testosterone and free oestradiol. The most important findings in the present report are that both insulin and free oestradiol were independent risk factors for BPH, suggesting that these hormones are promoters of BPH. Another important finding was that BPH is linked to several established endocrine and anthropometric components of the metabolic syndrome. This suggests that BPH is a component of the metabolic syndrome. In the present report, insulin was found to be an independent risk factor for BPH as measured by the prostate gland volume. Insulin has also been shown previously to be a risk factor for BPH. 6 11,14 All these data support the hypothesis that insulin is a promoter of BPH. Insulin resistance may change the risk of BPH through several biological pathways. The most obvious pathway is by the insulin itself, as shown in the present report. Insulin is a mitogen and a growth factor for prostate epithelial cells and also has an anti-apoptotic effect. 20 Moreover, hyperinsulinaemia stimulates the liver to produce more insulin-like growth factor 1 (IGF-1), another mitogen and an anti-apoptotic agent. 20 Serum IGF-1 has been associated with BPH in several studies, 21,22 but not in all. 23 Although stimulating liver production of IGF-1, insulin also suppresses the production of IGFBP-1 of the liver, which might result in an even higher increase of the free biologically active IGF-1 level. 20 Insulin resistance may also change the risk of BPH through the obesity sex hormone pathway. 24 Insulin resistance generates obesity and with increasing fatty tissue accumulation, there is an increase in aromatase activity, associated with a greater conversion of testosterone to oestradiol in men. This, in turn, increases the oestradiol level and reduces the testosterone/oestradiol quotient as shown in the present report. There are both clinical and experimental data linking oestradiol to BPH. 2,4,5 However, in many of these clinical reports, poor surrogate measurements of BPH have been used. In the present report, the free oestradiol level was shown to be independently linked to the total prostate gland volume, which has been shown to correlate highly significantly with the transition zone volume. 8 The transition zone is the exclusive site of the origin of BPH. 25 All these data taken together support the hypothesis that free oestradiol is a promoter of BPH. However, some data are inconsistent with the oestradiol hypothesis. In a report on the effect of stilbestrol therapy on the size of the benignly hypertrophied prostate gland, it was found that the size of the prostate gland was reduced in 10 of 13 cases. 26 Thus, as early as in the 40s, researchers summarized that a relatively slight excess of oestrogenic hormone produces an indirect castration effect on the secondary sex organs. This was largely due to inhibition of the pituitary gonadotropic hormone and subsequent testicular atrophy. 27 Other evidence that is inconsistent with the oestradiol hypothesis are the reports that treatment with aromatase inhibitors with observed reductions of the oestradiol level has no effect on clinically established BPH. 28,29 In the present report, 8 out of 10 established endocrine and anthropometric components of the metabolic syndrome were linked to BPH. In previous reports, 11 out of 11 established components of the metabolic syndrome were linked to BPH (Table 5). Thus, 19 out of 21 established components of the metabolic syndrome have been shown to be risk factors for BPH in these reports. However, the strongest evidence in the present report in support of the hypothesis that BPH is a component of the metabolic syndrome is the fact that BPH and components of the metabolic syndrome share the same endocrine and anthropometric aberrations. When it comes to endocrine factors, the following are linked to components of the metabolic syndrome: hyperinsulinaemia, 13 an increased level of oestradiol, 30 a reduced level of testosterone, 31,32 a reduced testosterone/oestradiol quotient 33 and a reduced level of SHBG. 34 All these endocrine aberrations, with the exception of a reduced testosterone level, which did not reach statistical significance, were linked to BPH as measured by the total prostate gland volume in the present report. It may be questioned why an inverse testosterone level did not come out as a predictor of BPH in the present report, as a reduced testosterone level has been linked to the metabolic syndrome in several reports. 31,32 Speculatively, the prostate gland tissue is a testosterone-dependent organ and, consequently, when the testosterone level decreases as a result from an increasingly pronounced metabolic syndrome, the

4 Table 5 Compilation of established components of the metabolic syndrome, which have been shown to be risk factors for BPH in the present and our previous reports Fasting serum insulin (present report 6 9 ) Free oestradiol (present report) Testosterone/oestradiol quotient (inverse; present report) Sex hormone-binding globuline (inverse; present report) Body length 7,9 Body mass index 6 9 Body weight 6 9 Waist measurement 6 9 Hip measurement 6 9 Waist/hip ratio 6 8 Lean body mass (present report) Type II diabetes 6 9 Atherosclerotic disease manifestations 9 Treated hypertension 5,7,8 Systolic blood pressure 6,9 Diastolic blood pressure 7 9 HDL-cholesterol (inverse) 6 8 Uric acid 9 Alanine aminotransferase 9 Abbreviation: HDL, high-density lipoprotein. growth-stimulating effect on the prostate gland by other aberrations might possibly be reduced. As for anthropometric factors, lean mass was found to be an independent risk factor for BPH. In two reports, lean mass has been shown to be related to the metabolic syndrome in women, 35,36 but has not been shown in men. On the other hand, another well-known anthropometric component of the metabolic syndrome, that is obesity, as measured by fat mass, did not come out significantly as a risk factor for BPH. In contrast to this, two other expressions of obesity, that is body weight and BMI, were related to BPH. The reason for this contradictory outcome could be that BPH as most components of the metabolic syndrome is primarily linked to visceral obesity rather than to the overall obesity. 7,14,37 In the clinical setting, waist measurement is probably a better risk factor for BPH than BMI. To sum up, 19 out of 21 established components of the metabolic syndrome turned out to be risk factors for BPH in our reports (Table 5). Reduced testosterone level and increased fat body mass were not shown to be predictors of BPH, but there are understandable causes for this outcome. Hence, all our data indicate that BPH is a new component of the metabolic syndrome. If this is true, BPH might be another Western civilization disorder in addition to type II diabetes, atherosclerosis, hypertension, obesity and dyslipidaemia. The classic metabolic syndrome in men has been defined in different ways by several health organizations. 38 The practical use of the composite definitions of the metabolic syndrome focuses on its potential value as a risk factor for the development of cardiovascular diseases. However, it has been recently claimed that the classic metabolic syndrome is imprecisely defined and appears to be of limited independent value as a marker of risk for cardiovascular diseases. 38 Given these limitations of the classic metabolic syndrome, it is reasonable not to use composite definitions of the metabolic syndrome in BPH research. In the present report, none of these established definitions have been used. In clinical practice, single or multiple components of the metabolic syndrome associated with BPH in our reports may be looked upon as alert signs for BPH. Thus, in the clinical setting, the results of our reports indicate that any physician confronted by a patient with type II diabetes, hypertension, visceral obesity and dyslipidaemia or some other component of the metabolic syndrome associated with BPH should consider the possibility that the patient also has lower urinary tract symptoms due to BPH. Conversely, a physician who is confronted by a patient with lower urinary tract symptoms due to BPH should consider the possibility that the patient also may have any of the above-mentioned conditions. In the past, many different definitions of BPH have been used. Most definitions have been based on the Hald s concept, which combines prostate volume, lower urinary tract symptoms and objective proof of difficult micturition. 39 These definitions seem to be inadequate tools when it comes to exploring aetiological factors in BPH. More recently, BPH has been defined as a noncancerous enlargement of the prostate gland focusing on the fact that BPH constitutes a benign tumour of the prostate gland. 40 It has been recently concluded that the prostate gland volume is an objective, quantitative measure of BPH. 41 Thus, the total prostate gland volume, used as a surrogate measure of BPH in the present study, seems to be an adequate measure of BPH. It may be argued that the associations in the present report between BPH on the one hand, and endocrine and anthropometric factors on the other, are largely confirmatory. However, previous studies have involved clinical studies including men referred because of lower urinary tract symptoms. In the present report, the same associations have been confirmed for the first time in a group of men representative for a whole population. It may also be argued that the associations are modest, although statistically significant. However, it must be kept in mind that the prostate gland volume, used as a proxy of BPH in the present report, is the result of a process going on for more than 30 years, whereas many of the endocrine and anthropometric measurements are highly volatile variables one morning at the end of this period of time. On the basis of the present report and other current knowledge, the following hypothesis on the pathogenesis of BPH is suggested. As a consequence of unhealthy lifestyle factors, including poor diet, low physical activity, obesity and genetic factors, a defective insulinstimulated glucose uptake, that is insulin resistance, is generated, mostly in the muscle tissue. To compensate for this defective insulin-stimulated glucose uptake, secondary hyperinsulinaemia is generated to maintain the glucose homoeostasis as shown in the present report. Hyperinsulinaemia, in turn, stimulates the liver which results in an increase of the free biologically active IGF-1 level. 20 As another consequence of constant hyperinsulinaemia, a cluster of disorders, including visceral obesity, is generated over the years. One consequence of obesity is an increased aromatase activity that results in an increased oestradiol level and a decreased testosterone/oestradiol balance, as shown in the present report. However, as mentioned above, other reports disqualify oestradiol as a promoter of BPH. Thus, this new hypothesis on the pathogenesis of BPH states that an increased insulin and IGF-1 level increase the prostate gland volume. The hypothesis also states that the metabolic syndrome and its major endocrine aberration, 163

5 164 hyperinsulinaemia, are primary events for BPH. This hypothesis will be tested in a prospective study now in progress. In summary, insulin and free oestradiol appear to be independent risk factors for BPH, confirming both the insulin and the oestradiol hypothesis. Our findings also seem to confirm the metabolic syndrome hypothesis stating that BPH is a new component of the metabolic syndrome. The metabolic syndrome and its major endocrine aberration, hyperinsulinaemia, are possible primary events for BPH. Acknowledgements This study was supported by an unconditional grant from Sanofi-Aventis, the Swedish Research Council, the Swedish Foundation for Strategic Research, the ALFA/ LUA research grant in Gothenburg, the Novo Nordisk Foundation, Wilhelm and Martina Lundgrens Forskningsfond and Alice Swenzons Stiftelse. References 1 Guess HA. Benign prostatic hyperplasia: antecedents and natural history. Epidemiol Rev 1992; 14: Partin AW, Oesterling JE, Epstein JI, Horton R, Walsh P. Influence of age and endocrine factors on the volume of benign prostatic hyperplasia. JUrol1991; 145: Walch PC, Retic AB, Stamey TA, Vaughan ED. Campbell s Urology, 6th edn. W.B. Saunders: Philadelphia, 1992, pp Suzuki K, Ito K, Ichinose Y, Kurokawa K, Suzuki T, Imai K et al. Endocrine environment of benign prostatic hyperplasia: prostate size and volume are correlated with serum estrogen concentration. Scand J Urol Nephrol 1995; 29: Schatzl G, Brossner C, Schmid S, Kugler W, Roehrich M, Treu T et al. Endocrine status in elderly men with lower urinary tract symptoms: correlation of age, hormonal status, and lower urinary tract function. The Prostate Study Group of the Austrian Society of Urology. Urology 2000; 55: Hammarsten J, Högstedt B, Holthuis N, Mellström D. Components of the metabolic syndrome risk factors for the development of benign prostatic hyperplasia. Prostate Cancer Prostatic Dis 1998; 1: Hammarsten J, Högstedt B. Clinical, anthropometric, metabolic and insulin profile of men with fast annual growth rates of benign prostatic hyperplasia. Blood Press 1999; 8: Hammarsten J, Högstedt B. Hyperinsulinaemia as a risk factor for developing benign prostatic hyperplasia. Eur Urol 2001; 39: Hammarsten J, Högstedt B. Calculated fast-growing benign prostatic hyperplasia a risk factor for developing clinical prostate cancer. Scand J Urol Nephrol 2002; 36: Nandeesha H, Koner BC, Dorairajan LN, Sen SK. Hyperinsulinaemia and dyslipidaemia in non-diabetic benign prostatic hyperplasia. Clin Chim Acta 2006; 370: Ozden C, Ozdal OL, Urgancioglu G, Koyuncu H, Gokkaya S, Memis A. The correlation between metabolic syndrome and prostatic growth in patients with benign prostatic hyperplasia. Eur Urol 2007; 51: Rohrmann S, De Marzo AM, Smit E, Giovannucci E, Platz EA. Serum C-reactive protein concentration and lower urinary tract symptoms in older men in the Third National Health and Nutrition Examination Survey (NHANES III). Prostate 2005; 62: DeFronzo RA, Ferrannini E. Insulin resistance. A multifaceted syndrome responsible for NIDDM, obesity, hypertension, dyslipidemia, and atherosclerotic cardiovascular disease. Diabetes Care 1991; 14: Dahle SE, Chokkalingam AP, Gao Y-T, Deng J, Stanczyk FZ, Hsing AW. Body size and serum levels of insulin and leptin in relation to the risk of benign prostatic hyperplasia. JUrol2002; 168: Mellström D, Johnell O, Ljunggren Ö, Eriksson A-L, Lorentzon M, Mallmin H et al. Free testosterone is an independent predictor of BMD and prevalent fractures in elderly men: MrOS Sweden. J Bone Miner Res 2006; 4: Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab 1999; 84: Van den Beld AW, De Jong FH, Grobbee DE, Pols HAP, Lamberts SWJ. Measures of bioavailable serum testosterone and estradiol and their relationships with muscle strength, bone density, and body composition in elderly men. J Clin Endocrinol Metab 2000; 85: Littrup PJ, Williams CR, Egglin TK, Kane RA. Determination of prostate volume with transrectal US for screening Part II. Accuracy of in vitro and in vivo techniques. Radiology 1991; 179: Terris MK, Stamey TA. Determination of prostate volume by transrectal ultrasound. JUrol1991; 145: Giovannucci E. Nutrition, insulin, insulin-like growth factors and cancer. Horm Metab Res 2003; 35: Stattin P, Kaaks R, Riboli E, Ferrari P, Dechaud H, Hallmans G. Circulating insulin-like growth factor-i and benign prostatic hyperplasia a prospective study. Scand J Urol Nephrol 2001; 35: Chokkalingam AP, Gao YT, Deng J, Stanczyk FZ, Sesterhenn IA, Mostofi FK et al. Insulin-like growth factors and risk of benign prostatic hyperplasia. Prostate 2002; 52: Roberts RO, Jacobson DJ, Girman CJ, Rhodes T, Klee GG, Lieber MM et al. Insulin-like growth factor I, insulin-like growth factor binding protein 3, and urologic measures of benign prostatic hyperplasia. Am J Epidemiol 2003; 157: Cohen PG. The hypogonadal-obesity cycle: role of aromatase in modulating the testosterone estradiol shunt a major factor in the genesis of morbid obesity. Med Hypothesis 1999; 52: McNeal J. Pathology of benign prostatic hyperplasia: insight into etiology. Urol Clin North Am 1990; 17: Peirson EL. A study of the effect of stilbestrol therapy on the size of the benignly hypertrophied prostate gland. JUrol1946; 55: Campbell M, Harrison J. Urology, 6th edn. W.B. Saunders: Philadelphia, 1970, pp Gingell JC, Knonagel H, Kurth KH, Tunn UW. Placebo controlled double-blind study to test the efficacy of the aromatase inhibitor atamestane in patients with benign prostatic hyperplasia not requiring operation. The Schering Study Group. JUrol1995; 154: Radlmaier A, Eickenberg HU, Fletcher MS, Fourcade RO, Reis Santos JM, van Aubel OG et al. Estrogen reduction by aromatase inhibition for benign prostatic hyperplasia: results of a doubleblind, placebo-controlled, randomized clinical trail using two doses of the aromatase-inhibitor atamestane. Atamestane Study Group. Prostate 1996; 29: Tivesten Å, Hulthe J, Wallenfeldt K, Wikstrand J, Ohlsson C, Fagerberg B. Circulating estradiol is an independent predictor of progression of carotide artery intima-media thickness in middleaged men. J Clin Endocrin Metab 2006; 91: Laaksonen DE, Niskanen L, Punnonen K, Nyyssönen K, Tuomainen T-P, Valkonen V-P et al. Sex hormones, inflammation and the metabolic syndrome: a population-based study. Eur J Endocrinol 2003; 149: Laaksonen DE, Niskanen L, Punnonen K, Nyyssönen K, Tuomainen T-P, Valkonen V-P et al. Testosterone and sex hormone-binding globulin predict the metabolic syndrome and diabetes in middle-aged men. Diabetes Care 2004; 27:

6 33 Fejes I, Koloszar S, Zavaczki Z, Daru J, Szollosi J, Pal A. Effect of body weight on testosterone/estradiol ratio in oligozoospermic patients. Arch Androl 2006; 52: Muller M, Grobbee DE, den Tonkelaar I, Lamberts SWJ, van der Schouw YT. Endogenous sex hormones and metabolic syndrome in aging men. J Clin Endocrinol Metab 2005; 90: Brochu M, Tchernof A, Dionne IJ, Sites CK, Eltabbakh GH, Sims EAH et al. What are the physical characteristics associated with a normal metabolic profile despite a high level of obesity in postmenopausal women? J Clin Endocrinol Metab 2001; 86: You T, Ryan AS, Nicklas BJ. The metabolic syndrome in obese postmenopausal women: relationship to body composition, visceral fat, and inflammation. J Clin Endocrinol Metab 2004; 89: Lee S, Min HG, Choi SH, Kim YJ, Oh SW, Kim YJ et al. Central obesity as a risk factor for prostatic hyperplasia. Obesity (Silver Spring) 2006; 14: Kahn R, Buse J, Ferrannini E, Stern M. The metabolic syndrome: time for a critical appraisal. Joint statement from the American Diabetes Association and the European Association for the Study of Diabetes. Diabetolgia 2005; 48: Hald T. Urodynamics in benign prostatic hyperplasia: a survey. Prostate Suppl 1989; 2: Leveillee RJ, Patel VR, Bird VG. Prostate hyperplasia, Benign Parsons JK, Carter HB, Partin AW, Windham BG, Metter EJ, Ferrucci L et al. Metabolic factors associated with benign prostatic hyperplasia. J Clin Endocrinol Metab 2006; 91: