| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1Clinical Research Branch, National Institute on Aging, National Institutes of Health, Baltimore, Maryland; 2Tuscany Regional Health Agency, Florence; 3Department of Internal Medicine and Biomedical Sciences, Section of Geriatrics, School of Endocrinology, University of Parma; 4Geriatric Rehabilitation Unit, Azienda Sanitaria di Firenze, Florence, Italy; 5Department of Medicine, Division of Endocrinology, Johns Hopkins University School of Medicine, Bayview Medical Center, Baltimore, Maryland; 6Department of Geriatric Medicine and Metabolic Diseases II, University of Naples, Naples; 7Department of Clinical and Experimental Medicine, Section of Internal Medicine, Gerontology and Geriatrics, University of Ferrara, Ferrara, Italy; and 8Laboratory of Epidemiology, Demography, and Biometry, National Institute on Aging, National Institutes of Health, Bethesda, Maryland
Submitted 12 July 2006 ; accepted in final form 30 August 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
65 yr) had complete data on testosterone, cortisol, DHEA-S, SHBG, fasting insulin, total and free IGF-I, IL-6, and C-reactive protein (CRP). MetS was defined according to ATP-III criteria. Insulin resistance was calculated according to HOMA. MetS was found in 145 women (28.3%). Participants with vs. those without MetS had higher age-adjusted levels of bioavailable testosterone (P < 0.001), IL-6 (P < 0.001), CRP (P < 0.001), and HOMA (P < 0.001) and lower levels of SHBG (P < 0.001). After adjustment for potential confounders, participants with decreased SHBG had an increased risk of MetS (P < 0.0001) vs. those with low SHBG. In a further model including all hormones and confounders, log SHBG was the only independent factor associated with MetS (OR: 0.44, 95% CI 0.21–0.91, P = 0.027). In older women, SHBG is negatively associated with MetS independently of confounders, including inflammatory markers and insulin resistance. Further studies are needed to support the notion that raising SHBG is a potential therapeutic target for prevention and treatment of MetS. sex hormone-binding globulin; hormonal dysregulation; insulin-like growth factor I; androgens
In the elderly, MetS is probably multifactorial (27). In particular, age-associated changes in hormonal levels might contribute to the development of MetS. Higher androgen levels have been linked to MetS or its components in women (26). Moreover, MetS or its individual components are common features in young women with polycystic ovary syndrome (PCOS), characterized by increased levels of androgens and inflammatory markers (10, 11). Sex hormone-binding globulin (SHBG), the circulating protein that binds testosterone and estradiol, has low levels in patients with hyperandrogenism (2) and may play a role in the pathogenesis of MetS (16). Furthermore, previous studies have suggested that low insulin-like growth factor I (IGF-I) and high cortisol are significant correlates of MetS (5, 31, 34, 35).
Overall, this literature suggests that hormonal changes that occur as a part of the aging process may predispose to MetS; however, no study has yet addressed the relationship between multiple hormonal dysregulation and MetS in older women. To diminish this gap in the literature, we studied the relationship between different hormones and MetS in a representative cohort of older Italian women. We hypothesized that higher androgens and cortisol and lower levels of SHBG and IGF-I would be associated with MetS independently of inflammatory state and other potential confounders.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Definition of MetS.
In accordance with the National Cholesterol Education Program's ATP-III criteria, the diagnosis of MetS was established as the presence of three or more of the following features: 1) fasting blood glucose levels 126 mg/dl or use of hypoglycemic medications, 2) fasting serum triglycerides 150 mg/dl or use of hypolipemic medications, 3) serum HDL <50 mg/dl, 4) blood pressure 130/85 mmHg or use of antihypertensive medications, and 5) waist circumference >88 cm (39).
Waist circumference was measured at the midpoint between the lower rib margin and the iliac crest (normally umbilical level). Weight and height were measured using standard techniques. Body mass index (BMI) was calculated as weight divided by height (kg/m2). Baseline blood pressure was recorded using a standard mercury sphygmomanometer. All blood pressure measurements were performed with the participant in a supine position on three occasions separated by intervals of 2 min, and the average of the last two measurements was used in the analysis. History of hypertension or use of antihypertensive medications was considered to define hypertension.
Hormone assays. Fasting blood samples were drawn between 7:00 and 8:00 AM and were stored at –80°C until analysis. Cortisol, DHEA-S, total testosterone, and total IGF-I were assayed using commercial kits (Diagnostic Systems Laboratories, Webster, TX). For cortisol, the minimum detection limit was 0.5 µg/dl; intra-assay and interassay coefficients of variation (CVs) were <10%. For DHEA-S, the minimum detection limit was 1.7 µg/dl; intra- and interassay CVs for three different concentrations ranged between 4.1 and 5.3% and between 4.6 and 7.0%, respectively. For total testosterone, the minimum detection limit was 0.03 nmol/l; intra- and interassay CVs for three different concentrations were <9.6 and 9.1%, respectively. Total IGF-I was measured by IRMA. Inter- and intra-assay CVs for three concentrations (low, medium, high) were all <10%. Free IGF-I was measured by commercial radioimmunoassay (Diagnostic System Laboratories) Inter- and intra-assay CVs were 6.2 and 7.3%, respectively.
SHBG was measured by a radioimmunoassay (Diagnostic Products, Los Angeles, CA) with a minimum detected concentration of 0.04 nmol/l and inter- and intra-assay CVs for three concentrations <6.9 and 3.6%, respectively (36).
Concentration of bioavailable testosterone was calculated using the Vermeulen formula (23, 37). Plasma insulin level was determined with a double-antibody, solid-phase radioimmunoassay (intra-assay CV = 3.1 + 0.3%; Sorin Biomedica, Milan, Italy). Cross-reactivity with human proinsulin was 0.3% (1).
Serum glucose level was determined by using an enzymatic colorimetric assay (Roche Diagnostics, Mannheim, Germany) and a Roche-Hitachi 917 analyzer. Insulin resistance (IR) was calculated according to homeostasis model assessment (HOMA): IR = fasting insulin x fasting glucose/22.5 (6, 24).
Inflammatory markers. Serum IL-6 was measured by high-sensitivity ELISA (BIOSOURCE, Camarillo, CA). The lowest detectable concentration was 0.1 pg/ml, with an interassay CV of 7%. High-sensitivity CRP was measured in duplicate by enzyme-linked immunosorbent assay using purified protein and polyclonal CRP antibodies (Calbiochem, San Diego, CA) with standardization according to the World Health Organization's reference standard. The minimum detectable concentration was 0.03 mg/l, and the interassay CV was 5%.
Plasma HDL-cholesterol and trygliceride concentrations were assessed using commercial kits (Roche Diagnostics, Mannheim, Germany). The interassay CV was <3.8% for HDL-cholesterol and <2.5% for triglycerides.
Assessment of covariates. Glutamic oxaloacetic transaminase (GOT) and glutamic pyruvic transaminase (GPT) levels were determined using commercial kits (Roche Diagnostics). The measurement unit was expressed as units per liter, the analytic sensitivity was 4 U/l for both tests, and the intra- and interassay CVs were <2.0 and 4.0% for GOT and GPT, respectively. Serum albumin was measured with agarose electrophoretic technique [Hydragel Protein (E) 15/30; Sebia, Issy-les-Moulineaux, France] and was expressed in percent.
Physical activity during the year prior to the interview was coded as 1) sedentary: completely inactive or light-intensity activity <1 h/wk; 2) light physical activity: light-intensity activity 2–4 h/wk; 3) moderate-high physical activity: light activity 5 h/wk or more or moderate activity 1–2 h/wk. Daily alcohol (g) intake was estimated by the European Prospective Investigation Into Cancer and Nutrition Food Frequency Questionnaire (28). Smoking was assessed by self-report and expressed as pack-years (packs smoked per day x years of smoking). Sociodemographic variables included age, sex and educational level.
Statistical analysis.
Because of skewed distributions, log-transformed values for SHBG, CRP, IL-6, DHEA-S, bioavailable testosterone, and insulin were used in the analyses. Differences in hormonal levels and other parameters among participants with and without MetS were tested by age-adjusted ANCOVA. SHBG levels were divided into quartiles to better describe their relationship with MetS. Differences in the prevalence of MetS according to specific hormone quartiles were formally tested by Pearson 
2 tests.
Odds ratios (ORs) of having MetS associated with hormonal levels adjusted for confounders (age, smoking, alcohol consumption, physical activity in the year before the visit, log CRP, log IL-6, HOMA, GOT, GPT, albumin) were estimated from logistic regression models. A multiple logistic regression analysis with all hormones was used to assess the association between each hormone and the presence of MetS. The SAS 8.2 statistical package (SAS Institute, Cary, NC) was used for all analyses.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The most intriguing finding was the strong negative relationship of SHBG with MetS and its components, namely HDL-cholesterol and triglycerides. This is the first evidence of an independent association between SHBG and MetS in an older female population, as previous studies were restricted to premenopausal or younger postmenopausal women (11, 16, 26).
Our findings are in contrast with those reported by Hajamor et al. (16), who failed to detect a significant correlation between SHBG and MetS in postmenopausal women. However, the limited number of subjects (n = 46) and the younger age of participants with MetS in that study may account for the different results. It is known that SHBG concentrations are tightly regulated by sex hormones. In women, SHBG concentrations are increased by estradiol (which has higher affinity than testosterone for SHBG) and decreased by testosterone. Thus low SHBG levels per se are considered an indirect index of androgenicity (2). It has also been demonstrated that SHBG concentration is influenced by insulin levels; both in vitro studies in human hepatoma cell lines (Hep G2) and in vivo observations, for example during hyperinsulinemic euglycemic clamp, provide evidence for this notion (20, 29). However, even when HOMA (which reflects insulin levels) and the effect of different hormones were considered in the analysis with other potential confounders, the relationship between SHBG and MetS was still significant.
These findings confirm a recent meta-analysis suggesting that SHBG predicts the development of MetS and type 2 diabetes independently of fasting insulin levels (8).
Interestingly, in our study, participants with MetS had higher GPT and lower albumin levels, suggesting the existence in MetS of a liver dysfunction that potentially affects SHBG synthesis (17). However, after adjustment for liver function (GOT, GPT, and albumin), the strength of the association between SHBG and MetS was not affected.
Although our study does not provide clues for the mechanisms of this effect of SHBG on MetS, it is interesting to speculate that SHBG may be much more than a mere transport protein. Despite failures to clone a specific SHBG membrane receptor, signaling for SHBG has been detected in different tissues including prostate, testis, breast, liver, and possibly muscles (18–19). It has also been suggested that SHBG may reflect the integration of several hormonal and nutritional stimuli (16). Of note, in our study, SHBG levels were negatively and independently associated with low HDL (P < 0.05) and triglycerides (P < 0.001), suggesting that SHBG might have a regulatory effect on hepatic lipase activity (30). Our results are concordant with those reported by Mudali et al. (26) and Bataille et al. (4) in postmenopausal women of younger age.
Low IGF-I levels have been associated with some features of MetS and insulin resistance and have been shown as an independent predictor of incident diabetes (1, 34–35). In our study, we found no significant difference in free and total IGF-I levels between participants with and without MetS. However, in the multivariate analysis, free IGF-I (the biologically active fraction of IGF-I), but not total IGF-I, levels were almost significantly associated with MetS, partially confirming what Sesti et al. (34) found in a younger population using different criteria of MetS (World Health Organization criteria). The discrepancy between the age adjusted and the multivariate analyses provides ground for the hypothesis that the hormonal dysregulation rather than alteration of one single hormone is associated with MetS.
In accord with previous literature, we found that total and bioavailable testosterone were higher in women with MetS (7, 21, 26). However, after adjustment for potential confounders, the strength of this association was substantially weakened and was no longer statistically significant. The lack of association between bioavailable testosterone and MetS in the multivariate analysis is not surprising, since the analysis was adjusted for HOMA and inflammatory markers, factors that may impact testosterone concentration (11).
Although the causality of the relationship between inflammation and hyperandrogenism is still uncertain, recent studies in subjects with PCOS suggest that hyperandrogenism is the consequence of the activation of nuclear factor-
B and the subsequent inflammatory cascade (14). Our data support this hypothesis in the older population.
In fact, when we incorporated IL-6 and CRP in the multivariate analysis, the relationship between bioavailable testosterone and MetS was no longer statistically significant. To test the association between testosterone and MetS in older women, we excluded those with history of oophorectomy (because ovaries are an important source of testosterone in postmenopausal women) and those who were on hormone replacement therapy (because estradiol raises SHBG and reduces free androgens). Hence, the comparability between our study and other reports in the literature that focused on younger women is limited (21, 26).
We did not find any significant difference in serum DHEA-S levels between participants with and without MetS in either univariate or multiple logistic analyses. Although there is growing evidence of a protective role for DHEA in MetS in men, the role of DHEA-S in women is controversial (3, 38). Barrett-Connor et al. (3) showed a positive association between DHEA-S and single components of MetS, namely waist circumference and glucose tolerance. In contrast, Villareal et al. (38) reported an improvement in insulin sensitivity after DHEA treatment. We also found no differences in cortisol between subjects with and without the MetS. However, serum cortisol levels do not necessarily reflect tissue cortisol (especially abdominal adipose tissue) (9). More importantly, we only measured serum cortisol levels measured at 8 AM and not the 24-h urine levels of free cortisol. This limited our capability of laying out the association between cortisol and MetS.
The main limitation of our study is its cross-sectional design. Given the nature of the study, we cannot establish whether altered hormonal levels are a cause or consequence of MetS. Second, we did not measure estradiol, and, rather than obtaining a direct measure of bioavailable testosterone, we used estimated levels using a well-accepted formula. Third, information on IGF-I-binding proteins was not available, which might have provided more detail on the biological activity of IGF-I. Finally, in this study, we had only serum cortisol levels measured at 8 AM, which reflect neither 24-h nor adipose tissue levels. However, these limitations are offset by important strengths. This is the first large representative sample of older women with complete information on hormones and ATP-III criteria for MetS who were screened for multiple potential confounders, such as inflammatory markers, insulin resistance, smoking, physical activity, and alcohol intake. By performing a multivariate analysis including all the hormones, we were able to test for the first time in older women the impact of the hormonal dysregulation rather than a single hormone on MetS. In addition, by excluding participants with a history of oophorectomy and those on hormone replacement therapy, we removed the interferences that may be attributed to these conditions.
In conclusion, our data suggest that SHBG is inversely associated with MetS in older women independently of inflammatory markers, insulin resistance, and other confounders. Further studies are needed to clarify whether SHBG represents a simple marker of hormonal dysregulation or possibly a therapeutic target for MetS (5, 10, 18, 19).
|
ACKNOWLEDGMENTS |
|---|
None of the sponsoring institutions interfered with the collection, analysis, presentation, or interpretation of the data reported herein.
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
