| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Centre de Recherche en Nutrition Humaine (P.-H.D., M.L.), and Fédération de Biochime de l’Hôpital Eouard Herriot (E.M., H.B.) Hospices Civils de Lyon, 69437 Lyon Cedex 08; Institut National de la Santé et de la Recherche Médical U449 (H.V., M.L.), Faculté de Médecine Laennec 69373 Lyon Cedex 08; and Fédération d’Endocrinologie de l’Hôpital de l’Antiquaille (P.C., M.P.), 69321 Lyon Cedex 05, France
Address all correspondence and requests for reprints to: Michel Pugeat, M.D., Fédération d’Endocrinologie, Hôpital Neuro-Cardiologique, Hospices Civils de Lyon, Bâtiment HGPO–59 boulevard Pinel, 69349 Lyon Cedex 03, France. E-mail: michel.pugeat{at}chu-lyon.fr.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This study confirmed that insulin resistance, despite the lack of obesity as such, is clearly present in many PCOS women, and demonstrated that GIR is the best predictor for insulin resistance. It was also shown that adiponectin level is a good indicator of abdominal fat mass and is associated to insulin resistance. Finally, low SHBG levels in PCOS are intimately associated with BMI, suggesting that some signal(s) from the adipose tissue, independent of adiponectin and leptin, may regulate liver production of SHBG.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SHBG is the main transport protein for testosterone and estradiol that modulates their biological activity (9, 10, 11). Within the normal population, SHBG level has a great variability that is explained in part by a genetic background and/or by changing hormone environment and nutrition (9, 11). It has been reported that a part of the variability in SHBG levels in women is associated with polymorphisms and mutations located within the coding and noncoding sequences of the human Shbg gene (12, 13, 14, 15, 16). On the other hand, direct and indirect evidence also suggests that insulin is a negative regulator of liver secretion of SHBG (17, 18). The concept has emerged that SHBG could be an interesting marker of insulin sensitivity in humans (19, 20), and SHBG level has therefore been included as such among the biochemical markers for the risk of developing type 2 diabetes (21, 22, 23) or cardiovascular diseases (24). Interestingly, low SHBG levels in normal obese adolescent girls (25)—as well as in many premature pubarchic girls, of whom it has been argued that they may be at risk of subsequent development of androgen disorder and insulin resistance (26)—have been shown to be negatively associated with increased body mass index (BMI) and insulin secretion (27). These data suggest that SHBG level could be an interesting screening tool for adolescents at risk of metabolic syndromes.
In the present study, it was found that GIR and waist circumference were the most useful predictors of insulin resistance in nonobese PCOS patients, and that low SHBG levels were strongly associated with BMI, suggesting that some signals from the adipose tissue might regulate SHBG.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study population consisted of 18- to 30-yr-old patients referred to our clinic for irregular menstrual cycle, hirsutism, and/or recurrent acne. They had not been taking any medication, including oral contraceptives, for at least 6 months. Each patient was examined in the morning, after a 12-h fast. Height, weight, and waist and hip circumferences were measured to calculate BMI (kilograms per square meter) and waist-to-hip ratio (WHR). Blood was sampled to measure serum androgen and lipid levels. The morphology of the ovaries was observed by pelvic or transvaginal ultrasound scans (Hitachi EUB-415 CFM). Androgen secreting tumors and nonclassic forms of 21-hydroxylase deficiency were ruled out by a basal dehydroepiandrosterone sulfate level lower than 15 µmol/liter and by an ACTH challenge test showing a 17-hydroxyprogesterone level lower than 30 nmol/liter at 30 min, in agreement with the international consensus.
We enrolled 16 nonobese hirsute women, with mean BMI of 23.6 ± 3.1 (ranging from 20–25) kg/m2, oligomenorrhea (six to eight menses during the last year), and increased plasma concentration of at least one androgen level (non-SHBG-bound testosterone >190 pmol/liter and/or androstenedione >7 nmol/liter). They had no evidence of recent spontaneous ovulation as indicated by a progesterone level lower than 3.2 nmol/liter (28). These patients also had increased ovary volume (>7 cm3) and presence on ultrasound scan of at least 10 small cysts/follicles (2–8 mm diameter) around a dense core of stroma. Informed written consent was obtained from each patient.
For determining steady-state glucose disposal references values for nonmenopausal women, we enrolled 10 nonobese women, less than 30 yr (20–25 yr old, mean 23), with no personal or familial history of type 2 diabetes, obesity (BMI, 20–25 kg/m2) or cardiovascular disease and with regular menses since their normal puberty. These women were not taking drug including oral contraception. They have normal fasting glucose, insulin, lipid, and androgen levels (Table 1
). Glucose clamp was performed at any day, except menses, of their menstrual cycle. The presence of spontaneous ovulation was not checked.
|
Study design
Background laboratory investigations. If menses were present, follicular phase plasma lipids and androgen levels were measured from a single blood sample.
An oral glucose (75 g) tolerance test was performed with glucose and insulin measures at 0, 30, 60, 90, 120, and 180 min, and the areas under the glucose and insulin level curves (AUCglucose and AUCinsulin) were calculated by the trapezoidal rule, using absolute values. The glucose (millimoles/liter) to insulin (picomoles/liter) ratio (GIR), proposed as an index of peripheral insulin sensitivity (29), was calculated as well as AUC glucose to AUC insulin ratio and simplified homeostasis model of assessment (HOMA) ([glycemia (mmol/liter) x insulin (picomoles/liter)] ÷ 22.5) (29).
Euglycemic hyperinsulinemic clamp. Basal glucose turnover rate was determined during the last 30 min of a 3-h basal period, by tracer dilution methodology using a primed [6,6–2H2] glucose (Eurisotop, St Aubain, France) infusion (0.02 mg/kg·min). A 3-h euglycemic hyperinsulinemic clamp was then started by the infusion of insulin (Actrapid Novo, Copenhagen, Denmark) at a rate of 450 pmol/m2·min. Primed [6,6–2H2] glucose was infused (0.1 mg/kg·min) during the clamp to determine the glucose turnover rate, whereas any decrease in blood glucose was prevented by adapted infusion of 20% glucose solution (Aguettant, Lyon, France). For the determination of metabolites, hormones and [6,6–2H2] glucose isotopic enrichment, blood samples were taken every 10 min during the last 30 min of the basal and the hyperinsulinemic periods. Plasma isotopic enrichment of [6,6–2H2] glucose was determined by gas chromatography mass spectrometry (5971 MSD, Hewlett-Packard, Palo Alto, CA), and glucose turnover rates were calculated using steady-state equations as previously described (30). The rate of glucose infusion during the clamp was used to estimate whole-body glucose disposal rate because it has been established (31) that endogenous glucose production is suppressed with the level of hyperinsulinemia reached during the clamp. To estimate glucose and lipid oxidation rates, respiratory exchange measurements were performed during the final 30 min of both the basal and hyperinsulinemic periods, using a flow-through canopy gas-analyzer system (Deltatrac Metabolic Monitor, Datex, Helsinki, Finland) (32). Because most PCOS patients had long and irregular menstrual cycle women, on the day of the glucose clamp study, progesterone level was measured and low value (<3.2 nmol/liter) indicated no evidence of ovulation.
Assays
Plasma glucose levels were measured by the glucose oxidase method, and total cholesterol, high-density lipoprotein (HDL) cholesterol and triglyceride levels by enzymatic techniques (Hitachi automatic analyzer 717, BioMerieux, Marcy l’Etoile, France). Serum insulin levels were determined by a double antibody RIA (Oris Industries, Gif sur Yvette, France) interassay coefficient of variation < 7%; intraassay coefficient of variation < 10%. SHBG levels were measured by an immunoradiometric assay (Coatria-SHBG, BioMerieux) and plasma concentrations of non-SHBG-bound-testosterone and 
4-androstenedione by specific RIAs as we previously described (33). A new RIA kit (Linco Research, St. Charles, MO) was used to measure leptin and adiponectin levels.
Statistical analysis
All data in the text and figures are presented as mean ± SD. Comparison of variables between PCOS women subgroups was performed using unpaired Student’s t test. Relationships between variables were sought by nonparametric correlation analysis (Spearman’s rank correlation coefficient) and by stepwise or multiple linear regression analysis with forward selection. The threshold for statistical significance was set at P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. BMI ranged from 18.9–29.1 kg/m2 (mean value, 23.6). None of these nonobese PCOS patients had diabetes according to the glucose tolerance test (on American Diabetes Association criteria), but two did have glucose intolerance. In eight of the 16 patients, glucose disposal rate, as assessed by the euglycemic hyperinsulinemic clamp, was lower than 8.5 mg/kg·min, the lowest limit for our age- and weight-matched nonobese control women (Table 1). Therefore, the eight patients with glucose disposal rates <46.75 mmol/kg·min (8.5 mg/kg·min) were identified as having insulin resistance. The insulin-resistant patients had, on average, a significantly higher BMI, waist circumference, and insulin level and a lower HDL cholesterol level than non-insulin-resistant patients. The GIR was highly discriminative between the two groups (P = 0.008, Table 1). In contrast, triglyceride and circulating free fatty acid levels were similar in both groups of patients. Fasting hepatic glucose production tended to be lower in the insulin-resistant group. Spearman correlations for glucose disposal to anthropometric and metabolic parameters are given in Table 2. In the whole PCOS group, the decrease in glucose disposal rate was inversely related to BMI and waist circumference, WHR alone explaining 57% of the glucose disposal rate variability. Single regression analysis showed that GIR was the best predictor of insulin resistance (R2 = 0.61), compared with fasting insulin level, HOMA, or AUCglycemia-to-AUCinsulin ratio (Table 2).
|
). Spearman correlations for SHBG level to parameters related to insulin sensitivity and circulating insulin concentration are shown in Table 3. The two best parameters related to SHBG level were BMI and fasting insulin level. Regarding the single regression coefficient, shown in Table 3, BMI alone accounted for 68% of plasma SHBG concentration variation, and basal insulin level for only 46%. Using a multiple linear regression with SHBG level as the dependent parameter, and a model including BMI, waist circumference, basal insulin level, glucose disposal rate, leptin and adiponectin levels (Table 4), BMI was the first and strongest outcome parameter to be independently related to SHBG level. Figure 1 shows that an SHBG level lower than 30 mmol/liter, the lowest normal value for normal control women, was observed in patients with a BMI higher than 25 kg/m2. In our 16 PCOS patients, leptin level correlated significantly with BMI (r = 0.62, P = 0.01), but its relation to SHBG level did not reach significance (r = -0.41, P = 0.109).
|
|
|
). To evaluate the respective roles of total fat mass and android type distribution in adiponectin level, a stepwise regression analysis, using adiponectin as an independent parameter, was performed with a model including BMI, waist circumference, WHR and GIR. Only waist circumference was predictive of adiponectin level (adjusted R2 = 0.36, P = 0.01). This negative correlation of adiponectin to WHR is shown on Fig. 2. A multiple regression analysis found that the best model to appraise adiponectin level included waist circumference and total weight (adjusted R2 = 0.66, P < 0.0006). Adiponectin level, like leptin level, was not related to SHBG level.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In our nonobese PCOS patients, insulin resistance, HOMA index, and GIR were strongly associated to SHBG levels, suggesting that the severity of insulin resistance is a determinant factor in SHBG level. Our results are in agreement with a recent study by Cibula et al. (36). However, these authors reported that SHBG was the most significant (even the single) predictor of insulin sensitivity (36). Because circulating SHBG levels has a strong genetic background (11, 12, 13, 14, 15, 16) their findings should be pondered, but it might reflect a fairly ethnically homogenous group of patients. In addition, the strong negative relationship of fasting insulin with SHBG levels that has been reported by most studies, suggests that indeed insulin has inhibitory activity on liver SHBG production as shown by human hepatoma cell line in vitro studies (17, 18), and in vivo by peritoneal infusion of insulin in nonobese type 1 diabetes (37). It has been proposed that SHBG may constitute an index of insulin resistance only in a hyperinsulinemic state (38). However, patients with advanced type 2 diabetes have low SHBG despite impaired insulin secretion (39). A central role for adipose tissue might conciliate these apparently discordant observations. Indeed, women with anorexia nervosa have increased SHBG levels (40), and in these patients a slight increase in BMI with no evidence of insulin change is effective in normalizing SHBG level. Our present study, showing that, in nonobese PCOS women, BMI was the only independent predictive parameter for SHBG levels, strongly suggests that adipose tissue is in some way involved in SHBG level variability. In agreement with this concept, we have shown elsewhere, in a population of nondiabetic but morbidly obese patients, that a profound reduction in fat mass, as measured by DEXA, following intensive diet or biliopancreatic diversion, had the effect of dramatically increasing SHBG levels (24). In this study, the percentage body fat, but not insulin, was an independent predictor of change in SHBG level.
Leptin level, as expected, correlated with BMI in nonobese PCOS women, confirming that leptin is an appropriate marker for total fat mass (41, 42). However, leptin and SHBG levels were not correlated.
Adiponectin, a recently discovered adipose-specific adipokine, has been said to be involved in obesity and diabetes-associated insulin resistance (43, 44). Despite the small number of patients involved in our study, we found a close relation between low adiponectin level and glucose disposal rate (insulin resistance). Unlike leptin, adiponectin level was not associated with BMI but was more specifically associated with WHR and with waist circumference, accounting alone for about 35% of adiponectin level variation. We therefore hypothesize that, in nonobese PCOS women, there is close association of intra-abdominal fat mass, decreased adiponectin level and insulin resistance, but that total fat mass is the best predictor of a low SHBG level. Taken together, these results suggest that some signal or signals from the adipose tissue, independent of adiponectin and leptin, may regulate SHBG liver production.
In conclusion, in nonobese PCOS women, GIR is a useful tool for predicting glucose disposal and insulin resistance, whereas adiponectin level is an interesting marker of abdominal fat tissue.
|
Footnotes |
|---|
Abbreviations: AUC, Area under the curve; BMI, body mass index; GIR, glucose-to-insulin ratio; HDL, high-density lipoprotein; HOMA, homeostasis model of assessment; PCOS, polycystic ovary syndrome; WHR, waist-to-hip ratio.
Received February 10, 2003.
Accepted May 5, 2003.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D. Shi, M. K. Dyck, R. R. E. Uwiera, J. C. Russell, S. D. Proctor, and D. F. Vine A Unique Rodent Model of Cardiometabolic Risk Associated with the Metabolic Syndrome and Polycystic Ovary Syndrome Endocrinology, September 1, 2009; 150(9): 4425 - 4436. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F Bonnet, B Balkau, J M Malecot, P Picard, C Lange, F Fumeron, R Aubert, V Raverot, H Dechaud, J Tichet, et al. Sex hormone-binding globulin predicts the incidence of hyperglycemia in women: interactions with adiponectin levels Eur. J. Endocrinol., July 1, 2009; 161(1): 81 - 85. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.A. Toulis, D.G. Goulis, D. Farmakiotis, N.A. Georgopoulos, I. Katsikis, B.C. Tarlatzis, I. Papadimas, and D. Panidis Adiponectin levels in women with polycystic ovary syndrome: a systematic review and a meta-analysis Hum. Reprod. Update, May 1, 2009; 15(3): 297 - 307. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Aigner, N. Bachofner, K. Klein, C. De Geyter, F. Hohla, W. Patsch, and C. Datz Retinol-Binding Protein 4 in Polycystic Ovary Syndrome--Association with Steroid Hormones and Response to Pioglitazone Treatment J. Clin. Endocrinol. Metab., April 1, 2009; 94(4): 1229 - 1235. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Kajaia, H. Binder, R. Dittrich, P. G Oppelt, B. Flor, S. Cupisti, M. W Beckmann, and A. Mueller Low sex hormone-binding globulin as a predictive marker for insulin resistance in women with hyperandrogenic syndrome Eur. J. Endocrinol., October 1, 2007; 157(4): 499 - 507. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Palomba, A. Falbo, T. Russo, F. Manguso, A. Tolino, F. Zullo, P. De Feo, and F. Orio Jr. Insulin Sensitivity after Metformin Suspension in Normal-Weight Women with Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 3128 - 3135. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.F. Escobar-Morreale, G. Villuendas, J.I. Botella-Carretero, F. Alvarez-Blasco, R. Sanchon, M. Luque-Ramirez, and J.L. San Millan Adiponectin and resistin in PCOS: a clinical, biochemical and molecular genetic study Hum. Reprod., September 1, 2006; 21(9): 2257 - 2265. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-J. Chen, W.-S. Yang, J.-H. Yang, C. K. Hsiao, Y.-S. Yang, and H.-N. Ho Low sex hormone-binding globulin is associated with low high-density lipoprotein cholesterol and metabolic syndrome in women with PCOS Hum. Reprod., September 1, 2006; 21(9): 2266 - 2271. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Meyer, B. P. McGrath, and H. J. Teede Overweight Women with Polycystic Ovary Syndrome Have Evidence of Subclinical Cardiovascular Disease J. Clin. Endocrinol. Metab., October 1, 2005; 90(10): 5711 - 5716. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Palomba, A. Falbo, F. Orio Jr, F. Manguso, T. Russo, A. Tolino, C. Annamaria, B. Dale, and F. Zullo A randomized controlled trial evaluating metformin pre-treatment and co-administration in non-obese insulin-resistant women with polycystic ovary syndrome treated with controlled ovarian stimulation plus timed intercourse or intrauterine insemination Hum. Reprod., October 1, 2005; 20(10): 2879 - 2886. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Yilmaz, N. Bukan, R. Ersoy, A. Karakoc, I. Yetkin, G. Ayvaz, N. Cakir, and M. Arslan Glucose intolerance, insulin resistance and cardiovascular risk factors in first degree relatives of women with polycystic ovary syndrome Hum. Reprod., September 1, 2005; 20(9): 2414 - 2420. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Codner, D. Mook-Kanamori, R. A. Bazaes, N. Unanue, H. Sovino, F. Ugarte, A. Avila, G. Iniguez, and F. Cassorla Ovarian Function during Puberty in Girls with Type 1 Diabetes Mellitus: Response to Leuprolide J. Clin. Endocrinol. Metab., July 1, 2005; 90(7): 3939 - 3945. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.A. Checa, A. Requena, C. Salvador, R. Tur, J. Callejo, J.J. Espinos, F. Fabregues, J. Herrero, and (Reproductive Endocrinology Interest Group of the Insulin-sensitizing agents: use in pregnancy and as therapy in polycystic ovary syndrome Hum. Reprod. Update, July 1, 2005; 11(4): 375 - 390. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Sepilian and M. Nagamani Adiponectin Levels in Women With Polycystic Ovary Syndrome and Severe Insulin Resistance Reproductive Sciences, February 1, 2005; 12(2): 129 - 134. [Abstract] [PDF]
|
|
|
|
 |
|
 S. T. Page, K. L. Herbst, J. K. Amory, A. D. Coviello, B. D. Anawalt, A. M. Matsumoto, and W. J. Bremner Testosterone Administration Suppresses Adiponectin Levels in Men J Androl, January 1, 2005; 26(1): 85 - 92. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. B. Tanko, J. M. Bruun, P. Alexandersen, Y. Z. Bagger, B. Richelsen, C. Christiansen, and P. J. Larsen Novel Associations Between Bioavailable Estradiol and Adipokines in Elderly Women With Different Phenotypes of Obesity: Implications for Atherogenesis Circulation, October 12, 2004; 110(15): 2246 - 2252. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Ibanez and F. de Zegher Ethinylestradiol-Drospirenone, Flutamide-Metformin, or Both for Adolescents and Women with Hyperinsulinemic Hyperandrogenism: Opposite Effects on Adipocytokines and Body Adiposity J. Clin. Endocrinol. Metab., April 1, 2004; 89(4): 1592 - 1597. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. T. Sheehan Polycystic Ovarian Syndrome: Diagnosis and Management Clin. Med. Res., February 1, 2004; 2(1): 13 - 27. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Orio, S. Palomba, F. Zullo, A. Colao, and G. Lombardi Are serum adiponectin levels really reduced in obese women with polycystic ovary syndrome? Hum. Reprod., January 1, 2004; 19(1): 215 - 215. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
