a Departments of Clinical Sciences in Lund and b Malmö, Lund University, Lund, Sweden

Transcription

1 Do sex hormones influence features of the metabolic syndrome in middle-aged women? A populationbased study of Swedish women: The Women s Health in the Lund Area (WHILA) Study Yasameen A. Shakir, M.D., Ph.D., a Göran Samsioe, M.D., Ph.D., a Per Nyberg, Ph.D., b Jonas Lidfeldt, M.D., Ph.D., b Christina Nerbrand, M.D., Ph.D., b and Carl-David Agardh, M.D., Ph.D. b a Departments of Clinical Sciences in Lund and b Malmö, Lund University, Lund, Sweden Objective: To outline perceived associations between various sex hormones and risk markers for cardiovascular disease in middle-aged women, with an emphasis on features of the metabolic syndrome (MS). Design: Cross-sectional analysis. Setting: Women s Health in the Lund Area Study. Patient(s): Population-based cohort. Intervention(s): A generic questionnaire, physical examinations, and laboratory assessments were completed by 6,917 women aged years living in the Lund area of southern Sweden. Women at or above defined cutoff limits for the MS were considered itively screened. After exclusion of women using hormone therapy (HT), 2,038 women with (MS ) and 2,054 women without features of the MS (MS ) were included. The ELISA techniques were used for the determination of serum androstendione (A), E 2, T, sex hormone-binding globulin (SHBG), cortisol, insulin, and leptin levels. Serum lipids and lipoproteins were determined by conventional methods. Multiple linear regression analyses were performed, controlling for age, body mass index (BMI), and smoking habits. Main Outcome Measure(s): Features of the MS, sex steroids, cardiovascular risk markers. Result(s): In the MS group, a itive association was seen between A and systolic blood pressure. Estradiol was atively associated with total cholesterol and diastolic blood pressure. The SHBG was atively associated with triglycerides, blood glucose, and diastolic blood pressure and itively with high-density lipoprotein (HDL). In the MS- group, there were itive associations between A, blood glucose, and systolic blood pressure. Testosterone was itively associated with HDL. Estradiol was atively associated with total cholesterol and itively with systolic blood pressure. The SHBG was itively associated with HDL and atively with triglycerides and diastolic blood pressure. There were itive associations between cortisol, low-density lipoprotein (LDL) cholesterol, blood glucose, and systolic blood pressure and a ative association with triglycerides in both MS and MS- groups. Conclusion(s): Androstendione, E 2, and T levels were associated with cardiovascular risk factors in middle-aged women. Effects by sex steroids on cardiovascular risk markers seem to be different in women with or without features of the MS. (Fertil Steril 2007;88: by American Society for Reproductive Medicine.) Key Words: Metabolic syndrome, sex hormones, SHBG, cortisol, middle-aged women Cardiovascular disease (CVD) is the leading cause of death in women in Western societies. Compelling evidence from both prospective and retrospective observational studies suggests that estrogen (E) monotherapy as well as sequential hormone therapy (HT) reduce the risk of CVD in healthy tmenopausal women (1). However, the Women s Health Initiative (WHI) and other randomized studies (2, 3) did not confirm these results and changed our understanding of risks Received December 19, 2005; revised November 10, 2006; accepted November 21, Supported by grants from the Skane County Council Foundation for Research and Development, and the Faculty of Medicine, Lund University. Reprint requests: Yasameen A. Shakir, M.D., Ph.D., Department of Gynecology and Obstetrics, Lund University Hospital, S Lund, Sweden (FAX: ; Yasameen.Shakir@med.lu.se). and benefits associated with HT. A tmenopausal woman with diabetes is three times more likely to develop CVD or stroke than a healthy woman (4, 5). Obesity, hypertension, hyperlipidemia, or hyperglycemia augment the risk to develop type 2 diabetes as well as CVD. A combination of these risk factors further increases this risk (6). Estrogen, T, and sex hormone-binding globulin (SHBG) have been suggested to further influence the risk pattern, but sible interactions with other risk factors are less well known. Data on associations between SHBG and CVD have been contradictory. Low SHBG levels have been associated with low serum high-density lipoprotein (HDL) cholesterol (7) as well as an increased risk of diabetes (8), whereas in the Rancho Bernardo Study, SHBG levels were not associated with cardiovascular mortality when adjusted for body mass index /07/$32.00 Fertility and Sterility Vol. 88, No. 1, July 2007 doi: /j.fertnstert Copyright 2007 American Society for Reproductive Medicine, Published by Elsevier Inc. 163

2 (BMI) (9). In another study, low SHBG levels were associated with a higher risk of CVD among women who did not use HT, but this relationship was not independent of BMI (10). In case-control studies, low plasma levels of SHBG were associated with a higher likelihood of atherosclerosis (11) and carotid intimal medial thickness (12). However, one angiographic case-control study failed to find an association (13). High androgen levels may increase CVD risk in women, sibly mediated by effects on lipids, blood pressure, and glucose metabolism (14, 15). In one cross-sectional study, T levels were associated with the amount of coronary atherosclerosis, as verified by angiography (16). The aim of the present study was to outline perceived associations between various sex hormones and risk factors for cardiovascular disease in middle-aged women with emphasis on subjects with features of the metabolic syndrome (MS). MATERIALS AND METHODS This report is an analysis from The Women s Health in the Lund Area (WHILA) Study. The WHILA project covered all women (n 10,766) born between December 2, 1935, and December 1, 1945, and living in the Lund area, Sweden, by December 1, 1995.The Lund area is located in southern Sweden and comed of a university town with about 100,000 inhabitants, and its surrounding rural areas, mainly farmland with a population of about 50,000 inhabitants. Women were invited to a screening procedure, which took place between 1996 and Details from the study have been published elsewhere (17). A population register comprising all inhabitants identified women eligible for study. Informed consent was obtained and the ethics committee at Lund University approved the study. A specially trained midwife nurse collected the questionnaires at the time of the examinations and personally interviewed each woman, and potential problems were addressed. At the interview, 19% of the subjects made some corrections in their written answers caused by mistakes or misunderstandings when filling out the forms. The questionnaires were answered before the laboratory results were obtained. The physical examination included measurements of body weight, height, minimal waist and maximal hip circumference (WHR), systolic and diastolic blood pressure, random capillary blood glucose, and a lipid profile. Blood pressure was measured twice in the right arm after 15 minutes of rest in the seated ition using a mercury sphygmomanometer with a cuff size adjusted to the circumference of the arm. Korotkoff phase V was taken as the diastolic blood pressure. The average of two recordings, measured to the nearest 2 mm Hg, was the blood pressure used for statistical calculations. TABLE 1 Number and percentage of women with screening variables at or above the cutoff levels (alone or in combination), (n 2,038). Screening variables and cutoff levels Positive screening outcome No. % of total group Random capillary blood glucose 8.0 mmol/l Nonfasting serum triglycerides 2.3 mmo/l BMI 30 kg/ m WHR SBP 160 or DBP mm Hg Family history of diabetes mellitus a Drug treatment of hypertension Drug treatment of hyperlipidemia a Women with family history of diabetes as only screening variable excluded. Shakir. Features of the metabolic syndrome and sex hormones. Fertil Steril Subjects The cutoff values for the primary screening procedure are shown in the Table 1. Women with one (women with family history of diabetes mellitus as only screening were excluded) or more of the eight factors for itive primary screening were defined as having features of the MS. Altogether 3,309 women (47.8%) had a itive screening outcome (18, 19). According to the hormonal situation, the participating women (n 6,917) were divided into three groups (i.e., 492 [7.1%] premenopausal [PM] with regular menstruation, 3,600 [52.1%] tmenopausal without HT [PM0], and 2,816 [40.8%] with use of HT [PMT]). Menopause was defined as a bleed-free interval of at least 12 months. After exclusion of the PMT group, 2,038 women with features of the MS (MS ) (994 women had one of those eight factors of itive primary screening and 1,044 women had a combination) and 2,054 women without features of the MS (MS ) were included (Fig. 1). Fasting serum insulin and leptin levels were analyzed in a random group of every third women in the itive screening group (570 women were included). Laboratory Analysis Blood glucose, as well as and serum levels of triglycerides, total cholesterol, HDL-cholesterol and low-density lipopro- 164 Shakir et al. Features of the metabolic syndrome and sex hormones Vol. 88, No. 1, July 2007

3 FIGURE 1 Population study. PM premenopause; PM0 tmenopause without using hormone therapy; PMT tmenopause with use of hormone therapy. Multiple linear regression analyses, controlling for age, BMI, and smoking habits (method stepwise) were performed to evaluate associations between variables of the MS and sex steroids. The P values.05 were regarded as statistically significant. Calculations were performed using the statistical program SPSS version 11.5 (SPSS Inc., Chicago, IL). RESULTS The population comprised 10,766 women, of whom 6,917 (64.2%) completed the examinations and formed the basis for the present report. Age distribution was similar between responders and nonresponders. More nonresponders than responders died during the period (2.6% vs. 0.2%, P.001), as well as during the next 2 years, (1.5% vs. 0.3%, P.001). The main causes of death were cancer and CVD. An analysis of nonresponders has been published previously (17). Shakir. Features of the metabolic syndrome and sex hormones. Fertil Steril tein (LDL)-cholesterol were measured by a Cholestech LDX instrument (Cholestech Corporation, Hayward, CA) on capillary whole blood. The ELISA techniques were used for the determination of serum androstendione (A), SHBG, cortisol, insulin, and leptin levels. Commercial ELISA methods using monoclonal antibodies were purchased from the DRG Instrument GmbH (Marburg, Germany). KRYPTOR-Testosterone, KRYPTOR- Estradiol 17 (E 2 ) are a kit designed for KRYPTOR automated immunofluorescent assays of T and E 2 in human serum. (B.R.A.H.M.S Ag., Hennigsdorf, Germany). KRYPTOR uses TRACE (Time Resolved Amplified Cryptate Emission) technology, based on a nonradioactive transfer of energy. This transfer takes place between two fluorescent tracers. Detection limits and coefficient of variations for hormones were as follow: E 2 : 3.5 pmol/l and 7.1%, T: 0.15 nmol/l and 6.4%, A: 0.15 nmol/l and 5.14%, cortisol: 6.9 nmol/l and 9.59%, SHBG: 0.2 nmol/l and 3.0%, insulin: 1.5 IU/mL and 4.11%. Insulin resistance was expressed through the homeostasis assessment model (HOMA-IR) and calculated by the equation of fasting insulin x fasting glucose/22.5 (20). A T index was defined as T/SHBG x 100. The E 2 index was calculated as E 2 /SHBG x 100. These indices were calculated to consider also free and protein bound steroids (21). STATISTICS For continuous variables, when normally distributed, Student s t-test was used for determination of differences between groups. When not normally distributed, the Mann- Whitney test was used. Median levels of E 2 (P.006) and SHBG (P.001) were lower in the MS group, whereas serum levels of E 2 index and T index (P.001 for both) were higher compared to the MS- group (Table 2). By definition, systolic and diastolic blood pressures, total serum cholesterol, LDL-cholesterol, triglycerides and blood glucose level were higher and HDL-cholesterol was lower in the MS compared to the MS- group (P.001 for all) (Fig. 2). Multiple Linear Regression Analyses Multiple linear regression analyses were controlled for age, BMI, and smoking habits. All subjects Multiple linear regression analyses revealed itive associations between A, blood glucose (P.05), and systolic blood pressure (P.001). Estradiol was atively associated with total cholesterol (P.001) and triglycerides (P.02). The SHBG was atively associated with triglycerides (P.001), blood glucose (P.004), and diastolic blood pressure (P.001), but itively with HDLcholesterol (P.001). Cortisol was itively associated with LDL, blood glucose, and systolic blood pressure, but atively with triglycerides (P.001 for all). Testosterone index was itively associated with triglycerides (P.001) and atively with HDL (P.003). We could not find any association with T and E 2 index (Table 3). Subjects with features of the metabolic syndrome In the MS group, a itive association was seen between A and systolic blood pressure (P.03). Estradiol was atively associated with total cholesterol (P.001) and diastolic blood pressure (P.05). The SHBG was atively associated with triglycerides (P.001), blood glucose (P.01), and diastolic blood pressure (P.007), and itively with HDL (P.001). Positive associations were also found between cortisol and LDL-cholesterol (P.037), systolic blood pressure and Fertility and Sterility 165

4 TABLE 2 Median levels (interquartile range) of sex hormone in the total, MS and MS groups. Sex hormones All groups (n 4092) MS (n 2038) MS (n 2054) P value Androstendione (nmol/l) 4.1 (3.1) 4.1 (3.2) 4.1 (2.9).6 Estradiol (pmol/l) 17.4 (37.3) 16.5 (31.8) 18.2 (41.5).006 Estradiol index 35.9 (81.7) 39.7 (82.5) 31.3 (79.5).001 Testosterone(nmol/L) 2.1 (1.8) 2.08 (1.8) 2.1 (1.9).6 Testosterone index 3.7 (4.6) 4.3 (5.5) 3.2 (3.8).001 SHBG(nmol/L) 52.4 (38.4) 44.0 (34.0) 60.3 (37.6).001 Cortisol (nmol/l) (182.1) (189.8) (173.4).09 Note: P value regarding differences between MS and MS (Mann-Whitney test). blood glucose (P.001 for both) and a ative association with triglycerides (P.001). Estradiol index was atively associated with total cholesterol (P.001). Testosterone index was itively associated with triglycerides, and atively with HDL (P.001for both) (Table 4). When we added insulin concentrations, insulin resistance and leptin levels to the multiple linear regressions in the MS group, a ative association between T index and leptin levels was found (P.007). Subjects without features of the metabolic syndrome There were itive associations between A, blood glucose (P.005) and systolic blood pressure (P.001). Testosterone was itively associated with HDL (P.02). Estradiol was atively associated with total cholesterol (P.004) and itively with systolic blood pressure (P.01). The SHBG was itively associated with HDL (P.001) and atively with triglycerides (P.001) and diastolic blood pressure (P.001). There were itive associations between cortisol and LDL-cholesterol, blood glucose (P.001 for both) and systolic blood pressure (P.006) and a ative association with triglycerides (P.001). There were itive associations between T index, LDL (P.02), and triglycerides (P.003). We could not find any associations with E 2 index (Table 5). FIGURE 2 Mean levels of lipids, glucose, systolic (SBP), and diastolic (DBP) blood pressure in MS (n 2,038) and MS- (n 2,054) groups. P.001. TC serum total cholesterol; HDL high-density lipoprotein; LDL low-density lipoprotein; TG triglycerides. 166 Shakir et al. Features of the metabolic syndrome and sex hormones Vol. 88, No. 1, July 2007

5 Fertility and Sterility TABLE 3 Associations of potential risk factors by different sex hormones, SHBG, and cortisol in the total group (n 3,672). Androstendione E 2 T SHBG Cortisol E 2 index T index B coefficient B coefficient B coefficient B coefficient B coefficient B coefficient B coefficient Serum total cholesterol (mmol/l) c Serum HDL cholesterol (mmol/l) c b Serum LDL cholesterol (mmol/l) c Serum triglycerides (mmol/l) a c 14.6 c c Glucose 0.06 a b 9.9 c Systolic blood pressure 0.01 c c Diastolic blood pressure c Note: Multiple linear regression analysis controlling for age, BMI, and smoking habits. a P.05; b P.01; c P.001. TABLE 4 Associations of potential risk factors by different sex hormones, SHBG, and cortisol in women with features of the metabolic syndrome (MS ) (n 1,972). Androstendione E 2 T SHBG Cortisol E 2 index T index B coefficient B coefficient B coefficient B coefficient B coefficient B coefficient B coefficient Serum total cholesterol (mmol/l) c 0, c 0.3 Serum HDL cholesterol (mmol/l) c c Serum LDL cholesterol (mmol/l) a Serum triglycerides (mmol/l) c 11.7 b c Glucose b 8.4 c Systolic blood pressure a c Diastolic blood pressure a b Note: Multiple linear regression analysis controlling for age, BMI, and smoking habits. a P.05; b P.01; c P

6 TABLE 5 Associations of potential risk factors by different sex hormones, SHBG, and cortisol in women without features of the metabolic syndrome (MS ) (n 1,748). Androstendione E 2 T SHBG Cortisol E 2 index T index B coefficient B coefficient B coefficient B coefficient B coefficient B coefficient B coefficient Serum total cholesterol (mmol/l) b Serum HDL cholesterol (mmol/l) a 7.4 c Serum LDL cholesterol (mmol/l) c a Serum triglycerides (mmol/l) c 29.4 c b Glucose 0.2 b c Systolic blood pressure 0.02 c 0.5 b b Diastolic blood pressure c Note: Multiple linear regression analysis controlling for age, BMI, and smoking habits. a P.05; b P.01; c P.001. DISCUSSION The present study demonstrated that E 2, A, and T were associated with risk factors for CVD in middle-aged women. Our results are in accordance with previous studies (22, 23) suggesting that women with the MS have lower levels of serum E 2 and SHBG and higher androgens. To our knowledge, another analysis of differences between women with and without features of the MS in relation to gonadal hormones has not been carried out. In the MS women, there was a ative association between E 2 and diastolic blood pressure. This finding indicates that women with MS would benefit more from E administration than the MSgroup. Also in the MS group, E 2 index was atively associated with total cholesterol, but there was no association in the MS group. The SHBG levels were lower in the MS than in the MS group. Hence the free fractions of all sex steroids were greater in the MS group compared to the MS group and the direct influence of hormones greater in the MS group. Blood glucose was the only factor that differed in its association with SHBG. There was a ative association between blood glucose and SHBG in the MS group, which was not found in the MS group (Table 6). Among nondiabetic Australian women (21) not using hormone or lipid-lowering medications, no relationship between serum E 2 and HDL-cholesterol, LDL-cholesterol, triglycerides, or diastolic blood pressure levels was found, whereas free androgen index was itively associated with LDLcholesterol. After controlling for confounding variables, we found a ative association between E 2 and total cholesterol and a itive one with systolic blood pressure in the MS group. Our data confirm other studies that found no association between T and HDL cholesterol (24). Also we could not find any association whether in the total group or in the MS or MS groups. Korhonen et al. (25) demonstrated that a hyperandrogenic hormone profile appeared to be a typical feature in premenopausal females with the MS, even without the polycystic ovary syndrome (PCOS). Previous studies did not determine which components of the MS are most strongly related to T levels in tmenopausal women, mainly due to a small sample size (26) or lack of data on certain components of the MS (27). In the MS- group, there was a itive association between T and HDL, which probably indicates a favorable effect of endogenous androgen on the risk marker for CVD. Also there were different influences by T index, that is, a ative association with HDL-cholesterol and a itive association with triglycerides in the MS, whereas in the MS group, there were itive associations with LDL and triglycerides. In women, compared to men, relationships between endogenous androgens and obesity, insulin, and cardiovascular risk factors have been contradictory. In cross-sectional studies, serum levels of T had itive correlations with BMI and 168 Shakir et al. Features of the metabolic syndrome and sex hormones Vol. 88, No. 1, July 2007

7 Fertility and Sterility TABLE 6 Comparison of findings by multiple linear regressions in the MS and MS groups. Androstendione E 2 T SHBG Cortisol E 2 index T index MS MS MS MS MS MS MS MS MS MS MS MS MS MS Serum total cholesterol (mmol/l) Serum HDL cholesterol (mmol/l) Serum LDL cholesterol (mmol/l) Serum triglycerides (mmol/l) Glucose Systolic blood pressure Diastolic blood pressure * P.05; P.01; P.001. * * 169

8 leptin levels in women (22, 28). Low serum levels of SHBG, which could be an indirect measure of female hyperandrogenism, were found to be associated with high BMI and WHR as well as with high serum levels of leptin and insulin and low serum levels of HDL-cholesterol (29, 30). Leptin has been shown to be itively associated with many metabolic variables and insulin resistance (31), but the levels of leptin have been similar in subjects with diabetes and those without (32, 33). In one study, plasma leptin was lower in women with IGT and type 2 diabetes compared to women with normal glucose tolerance test (34). We also found a ative association between T index and leptin levels in the MS group, but we have limited data in the MS group. Hence, T index seems to have a ative impact on risk factors for CVD in women with features of the MS. However, no relationship with A was found. One study showed that a high free androgen index was associated with insulin resistance (35). Moreover, in a large prospective study, 20% of women with SHBG levels below the fifth percentile developed type 2 diabetes during a 12-year follow-up period (36). Psychological stress has been linked to the MS and diabetes (37). Stress mechanisms have been suggested to activate central nervous centers, by hypothalamic stimulation and elevated cortisol, and may be involved in the pathogenesis of abdominal obesity and metabolic abnormalities (38, 39). Another study (40) showed that hypercortisolemic patients with depression display resistance to insulin and increased visceral fat. We found the same associations with cortisol (i.e., a itive association with LDL-cholesterol, systolic blood pressure, and blood glucose) but a ative one with triglycerides in the MS and MS- groups. The pathophysiological mechanisms and perceived clinical significance need to be investigated further. In conclusion, sex hormone levels seem to be of importance for features of the MS in middle-aged women. Impacts by sex steroid on cardiovascular risk markers were different in women with or without features of the MS. REFERENCES 1. Grodstein F, Stampfer MJ, Manson JE, Colditz GA, Willett WC, Rosner B, et al. Postmenopausal estrogen and progestin use and the risk of cardiovascular disease. N Engl J Med.1996;335: Grady D, Herrington D, Bittner V, Blumenthal R, Davidson M, Hlatky M, et al, for the HERS Research Group. Cardiovascular disease outcomes during 6 8 years of hormone therapy: Heart and Estrogen/ progestin Replacement Study follow-up (HERS II). JAMA 2002;288: The Writing Group for the Women s Health Initiative investigators. Risks and benefits of estrogen plus progestin in healthy tmenopausal women. JAMA 2002;288: Goldberg RJ, Larson M, Levy D. Factors associated with survival to 75 years of age in middle aged men and women. The Framingham Study. Arch Intern Med 1996;156: Kannel WB, McGee DL. Diabetes and cardiovascular disease. The Framingham Study. JAMA 1979;241: Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: Diagnosis and classification of diabetes mellitus: provisional report of a WHO consultation. Diabet Med 1998;15: Haffner SM, Moss SE, Klein BE, Klein R. Sex hormones and DHEA- SO4 in relation to ischemic heart disease mortality in diabetic subjects: the Wisconsin Epidemiologic Study of Diabetic Retinopathy. Diabetes Care 1996;19: Haffner SM, Valdez RA, Morales PA, Hazuda HP, Stern MP. Decreased sex hormone-binding globulin predicts noninsulin-dependent diabetes mellitus in women but not in men. J Clin Endocrinol Metab 1993;77: Goodman-Gruen D, Barrett-Connor E. A prospective study of sex hormone-binding globulin and fatal cardiovascular disease in Rancho Bernardo men and women. J Clin Endocrinol Metab 1996;81: Rexrode KM, Manson JE, Lee IM, Ridker PM, Sluss PM, Cook NR, et al. Sex hormone levels and risk of cardiovascular events in tmenopausal women. Circulation 2003;108: Reinecke H, Bogdanski J, Woltering A, Breithardt G, Assmann G, Kerber S, et al. A Relation of serum levels of sex hormone binding globulin to coronary heart disease in tmenopausal women. Am J Cardiol 2002;90: Bernini GP, Sgro M, Moretti A, Argenio GF, Barlascini CO, Cristofani R, et al. Endogenous androgens and carotid intimal-medial thickness in women. J Clin Endocrinol Metab 1999;84: Hauner H, Bognar E, Blum A. Body fat distribution and its association with metabolic and hormonal risk factors in women with angiographically assessed coronary artery disease: evidence for the presence of a metabolic syndrome. Atherosclerosis 1994;105: Liu Y, Ding J, Bush TL, Longenecker JC, Nieto FJ, Golden SH, et al. Relative androgen excess and increased cardiovascular risk after menopause: a hypothesized relation. Am J Epidemiol 2001;154: Haffner SM, Newcomb PA, Marcus PM, Klein BE, Klein R. Relation of sex hormones and dehydroepiandrosterone sulfate (DHEA-SO4) to cardiovascular risk factors in tmenopausal women. Am J Epidemiol 1995;142: Phillips GB, Pinkernell BH, Jing TY. Relationship between serum sex hormones and coronary artery disease in tmenopausal women. Arterioscler Thromb Vasc Biol 1997;17: Lidfeldt J, Nerbrand C, Samsioe G. The women s health in the Lund area (WHILA) study. In: Schneider HFG, Naftolin F, eds. Climacteric medicine where do we go? London & New York: Taylor and Francis, 2005: Lidfeldt J, Nerbrand C, Samsioe G, Schersten B, Agardh C-D. A screening procedure detecting high-yield candidates for OGTT. The Women s Health in the Lund Area (WHILA) study: a population based study of middle-aged Swedish women. Eur J Epidemiol 2001;17: Shakir YA, Samsioe G, Nerbrand C, Lidfeldt J. Combined hormone therapy in tmenopausal women with features of metabolic syndrome. Results from a population-based study of Swedish women: Women s Health in the Lund Area study. Menopause 2004:11: Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28: Shelley JM, Green A, Smith AM, Dudley E, Dennerstein L, Hopper J, et al. Relationship of endogenous sex hormones to lipids and blood pressure in mid-aged women. Ann Epidemiol 1998;8: Acien P, Quereda F, Matallin P, Villarroya E, Lopez-Fernandez JA, Acien M, et al. Insulin, androgens, and obesity in women with and without polycystic ovary syndrome: a heterogeneous group of disorders. Fertil Steril 1999;72: Björntorp P. Hyperandrogenicity in women a prediabetic condition? J Intern Med 1993;234: Noyan V, Yucel A, Sagsoz N. The association of androgenic sex steroids with serum lipid levels in tmenopausal women. Acta Obstet Gynecol Scand 2004;83: Shakir et al. Features of the metabolic syndrome and sex hormones Vol. 88, No. 1, July 2007

9 25. Korhonen S, Hippelainen M, Vanhala M, Heinonen S, Niskanen L. The androgenic sex hormone profile is an essential feature of metabolic syndrome in premenopausal women: a controlled community-based study. Fertil Steril 2003;79: Kumagai S, Kai Y, Sasaki H. Relationship between insulin resistance, sex hormones and sex hormone-binding globulin in the serum lipid and lipoprotein profiles of Japanese tmenopausal women. J Atheroscler Thromb 2001;8: Oh JY, Barrett-Connor E, Wedick NM, Wingard DL. Rancho Bernardo Study. Endogenous sex hormones and the development of type 2 diabetes in older men and women. Diabetes Care 2002;25: Rouru J, Anttila L, Koskinen P, Penttila TA, Irjala K, Huupponen R, et al. Serum leptin concentrations in women with polycystic ovary syndrome. J Clin Endocrinol Metab 1997;82: Hergenc G, Schulte H, Assmann G, von Eckardstein A. Associations of obesity markers, insulin, and sex hormones with HDL-cholesterol levels in Turkish and German individuals. Atherosclerosis 1999;145: Sherif K, Kushner H, Falkner BE. Sex hormone-binding globulin and insulin resistance in African-American women. Metabolism 1998;47: Ruigi JB, Dekker JM, Blum WF, Stehouwer CD, Nijpels G, Mooy J, et al. Leptin and variables of body adiity, energy balance, and insulin resistance in a population-based study. The Hoorn Study. Diabetes Care 1999;22: Saladin R, De Vos P, Guerre-Millo M, Leturque A, Girard J, Staels B, et al. Transient increase in obese gene expression after food intake or insulin adminstration. Nature 1995;377: Flier JS. Clinical review 94. What s in a name? In search of leptin s physiologic role. J Clin Endocrinol Metab 1998;83: Panarotto D, Ardilouze JL, Tessier D, Maheux P. The degree of hyperinsulinemia and impaired glucose tolerance predicts plasma leptin concentrations in women only: a new exploratory paradigm. Metabolism 2000;49: Golden SH, Ding J, Szklo M, Schmidt MI, Duncan BB, Dobs A. Glucose and insulin components of the metabolic syndrome are associated with hyperandrogenism in tmenopausal women: the atherosclerosis risk in communities study. Am J Epidemiol 2004; 160: Lindstedt G, Lundberg P, Lapidus L, Lundgren H, Bengtsson C, Björntorp P. Low sex hormone binding globulin concentration as an independent risk factor for development of NIDDM: 12 yr follow up of population study of women in Gothenburg. Diabetes 1991:40: Raikkonen K, Keltikangas-Jarvinen L, Adlercreutz H, Hautanen A. Psychological stress and the insulin resistance syndrome. Metabolism 1996;45: Björntorp P. Behaviour and the metabolic disease. Int J Behav Med 1996;3: Björntorp P. Do stress reactions cause abdominal obesity and comorbidities? Obes Rev 2001;2: Weber-Hamann B, Hentschel F, Kniest A, Deuschle M, Colla M, Lederbogen F, et al. Hypercortisolemic depression is associated with increased intra-abdominal fat. Psychosom Med 2002;64: Fertility and Sterility 171