• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Screening for Insulin Resistance in Women With Polycystic Ovarian Syndrome

Posted on: Thursday, 26 May 2005, 03:00 CDT

Abstract

Introduction Insulin resistance is implicated in the pathogenesis of polycystic ovarian syndrome (PCOS). Insulin-sensitizing agents are increasingly used in the treatment of infertility and hirsutism in PCOS. However, not all women with PCOS are insulin-resistant.

Objective To assess the degree of insulin resistance within a clinic population of women referred for treatment of oligomenorrhoea or infertility.

Design We evaluated 25 consecutive PCOS outpatients referred for treatment of menstrual dysfunction/infertility and a matched control group. All underwent a standard oral glucose tolerance test (OGTT) with serial insulin measurements. Insulin sensitivity was calculated using homeostasis model assessment (HOMA).

Results Five of the 25 clinic patients had abnormal glucose handling (two had previously unknown type 2 diabetes and three had impaired glucose tolerance). Fasting and 2-h insulin levels were significantly higher in the PCOS women. Mean HOMAS (insulin sensitivity) was even lower for PCOS women with normal GTT status (mean (95% confidence interval): 0.53 (0.34-0.72)) than for controls (0.94 (0.84-1.04)) (F=4.2, p < 0.001). HOMA-B (pancreatic β- cell function) was nearly tripled for normal GTT status PCOS women at 273 (205-342) versus 105 (70-139) for controls (F=6.8, p < 0.001).

Conclusions The results suggest a role for routine measurement of HOMA-S in identifying women with PCOS with insulin resistance with a view to targeting them with insulin-sensitizing agents.

Keywords: Polycystic ovarian syndrome, insulin resistance, homeostasis model assessment modelling

Introduction

Polycystic ovarian syndrome (PCOS) is one of the commonest endocrine disorders affecting women of reproductive age. The prevalence of PCOS in the United Kingdom is between 4 and 12% [I]. It is characterized by hyperandrogenism and chronic anovulation [2]. PCOS is associated with abnormal carbohydrate metabolism and insulin resistance, with insulin sensitivity being decreased by 35-40% in PCOS patients, regardless of body mass index (BMI) [3]. Women with PCOS have a significantly increased risk of glucose intolerance and the prevalence of undiagnosed diabetes mellitus in women with PCOS has been suggested to be seven-fold that of the normal population [4]. Retrospective studies have also shown that PCOS increases the risk of hypertension, coronary vascular disease and myocardial infarction in later life [5].

The lean woman with PCOS seems to have a form of insulin resistance that is intrinsic (and perhaps unique) to the syndrome and poorly understood [69]. The obese woman with PCOS possesses not only the form of insulin resistance intrinsic to the syndrome, but also has an added burden of insulin resistance that is related to excess adiposity [1O]. Most PCOS patients have basal and glucose- stimulated hyperinsulinaemia [3] and the associated increased androgen production and disordered gonadotrophin secretion cause chronic anovulation. Insulin-sensitizing agents such as metformin are increasingly being used to treat infertility and hirsuitism in PCOS patients [9,11,12]. However not all PCOS subjects are insulin- resistant. Plasma insulin concentrations are usually measured experimentally, and rarely clinically. There is no clear reference range for insulin levels, but fasting hyperinsulinaemia is usually classified as an insulin level greater than 17-20 mU/1 [U]. Accurate and reproducible methods of measuring insulin resistance are also needed, with standardization of results to improve comparability between studies [13].

Metformin has been shown to reduce fasting serum insulin, androgen and luteinizing hormone (LH) concentrations and increase sex hormonebinding globulin (SHBG) levels [U]. The drug also improves ovarian function with resumption of spontaneous ovulation in a proportion of women [14]. A reduction in hirsutism and weight, especially central obesity, may occur in obese women treated with metformin [15]. As insulin resistance is associated with development of future glucose intolerance, there are potential long-term benefits of metformin in this group. Not all PCOS women who are candidates for such treatment are insulinresistant [16]. Given the importance of identifying these individuals appropriately, and to assess their degree of insulin resistance, we examined the utility of fasting insulin/glucose measurement alone compared with full oral glucose tolerance testing with serial insulin measurements, within a clinic population of women referred for treatment of oligomenorrhoea or infertility.

Materials and methods

Selection and definition of study subjects

The study was approved by the ethical committee of Salford Royal Hospitals NHS Trust, and each woman gave informed consent. Twenty- five consecutive Caucasian women referred for treatment of menstrual disturbance or infertility to the gynaecology/endocrine clinic at Hope Hospital, Salford, UK, were included. All subjects were being considered for treatment with metformin. All had been diagnosed to have PCOS, on the basis of at least three of the following criteria: polycystic ovaries on ultrasound examination, oligo/amenorrhoea, hirsuitism or elevated serum testosterone concentration. Patients with disorders which may present in a manner similar to PCOS, i.e., Cushing's syndrome, lateonset congenital adrenal hyperplasia, thyroid dysfunction, hyperprolactinaemia and androgen-secreting tumours, were excluded. None of the patients had evidence of virilization, and previous abnormal thyroid and prolactin levels were recorded.

Controls were selected from a population of non-PCOS women in Manchester, UK, of normal glucose tolerance status, as reported elsewhere [17]. They were matched according to age and BMI. Women on the oral contraceptive pill were excluded. None of the control women had oligo/amenorrhoea, hirsutism or elevated serum testosterone concentration.

Experimental protocol

All patients underwent a standard (75 g) oral glucose tolerance test (OGTT) with venous blood taken for measurement of plasma glucose and insulin levels at O h, 30 min, 60 min, 90 min and 120 min. Controls all had 2-h glucose tolerance values establishing their 2-h normoglycaemia, but only fasting glucose and insulin are reported here. Blood samples were centrifuged and aliquots of plasma (for measurement of glucose and insulin levels) or serum (for all other assays) were frozen at -70C until analysis. Glucose tolerance criteria were those of the World Health Organization's June 2000 guidelines [18].

Menstrual pattern was classified as regular, oligomenorrhoea (cycles lasting more than 35 days) or amenorrhoea (no menstrual period for over 6 months). The severity of hirsutism was graded using the Ferriman-Gallwey score [19], and grouped as mild (< 10/ 44), moderate (> 10 to < 25/44) and severe (≥ 25/44). The presence of acne, male-pattern baldness and acanthosis nigricans was recorded as were any previous pregnancies, and if there were concerns about fertility. Any medications taken by the patients at the time of the OGTT were noted. Evidence of polycystic ovaries on ultrasound scan was based on the radiologists' opinion.

Assays

Measurement of serum testosterone (normal range 0.5-2.4 nmol/1), LH (< 1-90 U/l), follicle-stimulating hormone (< 1-21 U/l), prolactin (non-pregnant female 25-629 mU/1) and thyroid function tests (free thyroxine 9-22 pmol/1) were made on an Abbott Architect analyser (Abbott Diagnostics, Maidenhead, Berkshire, UK).

Serum concentrations of SHBG (18-114 nmol/1) and oestradiol (100- 1430 pmol/1) were measured using a Wallac Deifia fluorometer (Perkin Elmer Life Sciences, Cambridge, UK). Androstenedione (1-11.5 nmol/ 1) was measured by the Incstar (Stillwater, MN, USA) radioimmunoassay (RIA) method and dehydroepiandrosterone sulfate (DHEA-S) (0.94-11.6 mol/l) was measured by DSL (Webster, TX, USA) RIA. 17-Hydroxyprogesterone (< 10 nmol/1) was measured using an in- house RIA utilizing ether extraction. Intra-assay coefficients of variation (CVs) for 17-hydroxyprogesterone were 12, 11 and 9.6% at a concentration of 2, 6.1 and 19.6 mmol/1, respectively. Plasma glucose concentrations were measured on a Roche Integra analyser (Roche Diagnostics Ltd, Lewes, Sussex, UK). Plasma insulin was measured using the Mercodia (Uppsala, Sweden) enzyme-linked immunosorbent assay specific for intact insulin. The detection limit of < 1 mU/1 was calculated as two standard deviations above the zero standard. Withinand between-assay CVs were ≤ 5% across the range of measured insulin concentrations. Cross-reactivity of the insulin assay for proinsulin is < 0.1%.

Homeostasis model assessment (HOMA) was applied. HOMA-S (insulin sensitivity) and HOMA-B (pancreatic β-cell function) were calculated from fasting insulin and glucose concentrations. This was done using an iterative computer program which found the values for HOMA-S and HOMA-B at which the non-linear model for glucose and insulin homeostasis was at equilibrium in steady state for the observed fasting insulin and glucose values [20-23]. HOMA modelling is considered to be valid in type 2 diabetes, as in individuals with impaired glucose tolerance and impaired fasting glucose [24]. A cutpoint for HOMA-S of < 0.4 was taken as compatible with significant insulin resistance (HOMA-R). HOMA-R isdefined as the reciprocal of HOMA-S (1/HOMA-S).

The area under the curve insulin was calculated using the trapezoid rule.

Data analysis

Statistical analysis was performed using SPSS for Windows version 10.0 (SPSS Inc., Chicago, IL, USA). Results for normally distributed variables are reported as arithmetic means with 95% confidence intervals (CIs). Non-normally distributed variables were logarithmically transformed prior to analysis. Comparison between groups was by one-way analysis of variance. A p value of < 0.05 was regarded as statistically significant. Univariate correlation between continuous variables for the whole group used Spearman coefficients (two-tailed test), with p < 0.05 being regarded as statistically significant.

Results

Clinical features

The majority of patients (20/25, 80%) suffered from menstrual disturbances, with five women (20%) complaining of secondary amenorrhoea. Concern about impaired fertility was recorded in the medical records of 11 of the women. Twelve of the 25 (48%) PCOS women were parous. Nine of the women (36%) had acne and four (16%) had scalp hair loss. Two (8%) women had acanthosis nigricans. At the time of the OGTT, one woman was taking spironolactone. Subject characteristics (cases vs. controls) are shown in Table I. Of the 25 subjects with PCOS, six (24%) were in the non-obese (BMI < 30 kg/ m^sup 2^) range and the remaining 19 (76%) were in the obese range.

Metabolic results

Two women (8%) were found to have type 2 diabetes mellitus (2-h venous plasma glucose ≥ 11.1 mmol/ 1). Three patients (12%) had impaired glucose tolerance (IGT) (2-h glucose 7.8-11.0 mmol/1 and fasting glucose < 7.0 mmol/1) but none had impaired fasting glucose (6.1-6.9 mmol/1). Twenty out of 25 women (80%) had normal glucose tolerance. Two of the control group of women had IGT. None had diabetes.

All PCOS subjects with normal BMI had HOMA-S > 0.4. Six patients with HOMA-S > 0.4 had BMI > 30 kg/m^sup 2^, i.e., in the obese range (32% of obese subjects). Thus some subjects with BMI > 30 kg/m^sup 2^ (in the obese range) were not insulin-resistant. The correlations between metabolic variables are shown in Table II.

The mean fasting glucose concentration for PCOS women was not significantly different from the control group of non-PCOS women (see Table I). Mean serum fasting insulin concentration was higher than in controls (.F=IO.1, p < 0.001) as was 2-h insulin concentration (F= 12.2, p < 0.001). For normal GTT status PCOS women, mean HOMA-S (Figure Ia) was significantly lower (0.53 (0.34- 0.72)) than for matched normal GTT status controls (0.94 (0.84- 1.04)) (F =4.2, p < 0.001). HOMA-B (Figure Ib) was correspondingly higher: 273 (205-342) for normal GTT status PCOS women versus 105 (70-139) for matched controls (F= 6.8, p < 0.001). Mean peak glucose and insulin levels occurred 60 min after the glucose load, but with considerable individual variation: only 12/25 PCOS patients reached their peak glucose concentration at 60 min, and eight with their peak insulin concentration at that time.

Table I. Subject characteristics (arithmetic means with 95% confidence intervals unless otherwise stated).

Table II. Spearman correlation coefficients for metabolic variables.

Figure 1. (a) Insulin sensitivity (HOMA-S) and (b) pancreatic β-cell function (HOMA-B) in women with polycystic ovarian syndrome (PCOS) and normal glucose tolerance status compared with matched controls; (c) HOMA-S and (d) HOMA-B in PCOS women with normal versus abnormal glucose tolerance status (World Health Organization 2000 criteria). Data are expressed as means with 95% confidence intervals. GTT, glucose tolerance test; IGT, impaired glucose tolerance.

Within the group of PCOS women there was a significant difference in HOMA-S by GTT status, with normal glucose-tolerant PCOS women having a significantly greater HOMA-S than the women denned as having IGT or diabetes (Figure Ic). There was no difference in HOMA- B by GTT status for the PCOS group (Figure Id).

For PCOS women, a positive correlation was found between insulin resistance (HOMA-R) (inverse of HOMA-S), based on fasting measurements, and insulin concentration during the OGTT (area under the curve) (Figure 2a; data available for 21/25 PCOS patients) and peak insulin (Figure 2b; data available for 21/25 PCOS patients). This demonstrates the close negative relationship between a oneoff measurement of HOMA-R and insulin secretion for the whole OGTT.

SHBG had a strong negative correlation with HOMA-R in PCOS women (p=-0.52, p < 0.05) (Figure 3). As insulin resistance increases, SHBG concentration decreases. SHBG correlates negatively with fasting insulin (p =0 -0.34, p < 0.05) and peak insulin (p =-0.64, p < 0.01) but not significantly with 2-h insulin (ρ=-0.18, NS).

The relationship between BMI and HOMA-R for the PCOS women and the control group is shown in Figure 4. For a given incremental increase in BMI, there was a greater increase in HOMA-R for the PCOS women (equation for line: HOMA-R=6.8 + 0.34 BMI, r^sup 2^ = 0.19) than for the control group (HOMA-R =-2.5+ 0.18 BMI, r^sup 2^ = 0.27) (p < 0.01 for difference between gradients of lines).

The mean testosterone concentration of this sample was 2.47 (2.14- 2.79) nmol/1. Fourteen women (56%) had elevated testosterone levels (> 2.4 nmol/1). The highest serum testosterone concentration in this group was 4.6 nmol/1. Nine patients had recent measurements of serum androstenedione greater than the normal reference range (1.8-12.2 nmol/1). All DHEA-S and 17-hydroxyprogesterone measurements were within the normal reference ranges.

Discussion

We have shown that not all women with PCOS being considered for treatment with metformin are insulin-resistant. A very high proportion - a fifth - of this group of otherwise healthy young women referred for investigation of confirmed or suspected PCOS were found to have impaired glucose tolerance or type 2 diabetes. This has significant health implications for the individuals concerned and raises the issue of screening for impaired glucose handling in PCOS women. These subjects, once identified, could be managed with early intervention. Although the three subjects with impaired glucose tolerance would not have been identified by fasting measurements, HOMA-S for these subjects who are at risk of developing type 2 diabetes were all < 0.4. They may benefit significantly from treatment with metformin or other insulin- sensitizing agents.

Figure 2. Scatter plot of (a) area under the curve insulin versus insulin resistance (HOMA-R) (data available for 21/25 patients with polycystic ovarian syndrome (PCOS)) and (b) peak insulin concentration versus HOMA-R (data available for 23/25 PCOS patients).

Figure 3. Scatter plot of sex hormone-binding globulin concentration versus insulin resistance (HOMA-R) in women with polycystic ovarian syndrome. NS, not significant.

The different relationship between HOMA-R as a measure of insulin resistance and BMI in PCOS women compared with controls (Figure 4), with a greater increase in HOMA-R for a given change in BMI in the PCOS group, would suggest that any reduction of BMI in PCOS women would have a greater potential benefit in terms of improvement in HOMA-R than for comparable non-PCOS subjects.

HOMA-S has been shown to be a valid measure of insulin sensitivity in comparison with suggested gold standard methods, such as the euglycaemic clamp [22,23]. There is good agreement between HOMA-S and insulin secretory response as measured by area under the curve insulin for the whole OGTT. Thus, in this group, HOMA-S provides a valid measure of the total insulin response to a glucose load in relation to peripheral insulin sensitivity. We accept that those subjects with low HOMA-S will tend to have higher peak and area under the curve insulin values for the whole GTT but this close relationship indicates that a one-off measure of fasting insulin and glucose is adequate, rather than carrying out a full GTT with serial insulin measurement. While all the cases with BMI < 25 kg/m^sup 2^ had HOMA-S > 0.4, BMI was an insufficient criterion to predict low HOMA-S with several of the relatively insulin-sensitive subjects having a BMI within the Obese range', i.e., > 30 kg/m^sup 2^.

HOMA modelling is primarily an epidemiological tool but this easily-calculated estimate could be simply used in a clinic population where the carrying out of insulin clamp studies is not feasible. The variability in all measurements of insulin sensitivity makes such measurements more useful in epidemiological comparisons between groups than for diagnosis in individuals. However, the degree of insulin resistance may be such that even an imprecise measure can have discriminatory value. Given the difficulty of doing clamp studies in a clinic setting, we would suggest that HOMA-S is a valid way of screening for those patients who are insulin- resistant. Further studies are required to test this proposition.

Lean women with PCOS seem to have a form of insulin resistance that is intrinsic (and perhaps unique) to the syndrome and poorly understood [6-9]. All but five PCOS women in this study were overweight or obese, as were all the controls. Consequently the findings here cannot be extrapolated to a non-obese PCOS population.

The literature on the prevalence of an abnormal OGTT in women with PCOS describes diabetes and impaired glucose tolerance in 20- 40% of patients [4,25,26]. Although the numbers in this study were small, the prevalence of glucose intolerance in this study of 20% agrees with these previous findings. From the relationship between BMI and HOMA-R as shown in Figure 4, it is clear that insulin sensitivity may be widely distributed from normality to the resistance state even in the presence of excess body weight for women with and without PCOS.

As expected, SHBG had a negative univariate association with HOMA- R, in keeping with the negative regulation o\f SHBG production by hepatic portal insulin [27]. This agrees with recent studies that have shown a clear negative relationship between insulin resistance and SHBG levels [28-30]. One of the binding proteins for the insulin- like growth factors (IGFs), insulin-like growth factor binding protein-1 (IGFBP-I) has been closely linked with peripheral and hepatic insulin sensitivity [31] and is significantly lower in PCOS women than in those without PCOS [32]. It may well be that measurement of IGFBP-I would also be a useful tool to discriminate those women with PCOS who could benefit from insulin-sensitizing agents.

Determining the degree of insulin resistance of PCOS patients has important implications. Insulin-sensitizing agents have been shown to cause a significant reduction in LH and androgen concentrations, and an increase in SHBG levels in hyperinsulinaemic patients with a normal OGTT. Studies also suggest alterations in BMI, waist-hip ratio, fat distribution, blood pressure and regulation of menstruation in women taking metformin [15]. While it could be argued that our HOMA-S of < 0.4 to define significant insulin resistance was arbitrary, there are no hard and fast definitions that can be utilized in a clinic population. We propose that it is quite reasonable to attempt to quantify insulin resistance in this group of patients prior to commencement on insulin-sensitizing agents, so that at least for a given patient the response to treatment can be monitored with serial measurement of insulin sensitivity.

Figure 4. Scatter plot of insulin resistance (HOMA-R) versus body mass index (BMI) for subjects with polycystic ovarian syndrome (HOMA- R =-6.8+ 0.34 BMI, r^sup 2^ = 0.19) and for the control group (HOMA- R =-2.5 + 0.18 BMI, r^sup 2^ = 0.27; ρ < 0.01 for difference between gradients of lines).

Larger, randomized, placebo-controlled trials are now needed to examine the potential impact of metformin in PCOS women. Metformin has been proved to be active as a pro-ovulatory compound even in the presence of normal fasting insulin [28], which suggests a direct action of this drug at the ovarian level. Metformin has recently been shown to reduce serum C-reactive protein (CRP) levels in women with PCOS [3O]. The decrease of serum CRP levels during metformin therapy is in accordance with the known beneficial metabolic effects of this drug and suggests that CRP or other inflammation parameters could be used as markers of treatment efficiency in women with PCOS.

Importantly, these trials should be extended to include lean PCOS women, some of whom will not necessarily be hyperinsulinaemic or insulin-resistant, to clarify whether metformin has any mechanism of action in PCOS other than by inducing weight loss. Furthermore, extrapolation of the changes seen so far suggests that improvements can be expected in the clinical signs of androgen excess, such as hirsutism and acne, as has recently been reported [33]. It is also important to investigate the effects of metformin and other insulin sensitizers in combination with other medications used to treat PCOS, e.g. the oral contraceptive pill, because of their potential effects on metabolism.

Confirmation of the beneficial effect of metformin on hormonal and metabolic variables in women with PCOS could have implications not only for the treatment of the common gynaecological presenting features, but also for the burden of type 2 diabetes and vascular disease in these women. Identifying those women suitable for treatment with metformin is integral to any planned intervention strategy. The results of this study suggest a role for routine measurement of HOMA-S in identifying which women with PCOS can benefit from longer-term treatment with insulin sensitizers, in order to ensure that these agents are used in the appropriate clinical context. Given the high proportion of PCOS women with significant insulin resistance, this may impact on the potential risk for these women of developing type 2 diabetes in the future.

References

1. Knochenhauer ES3 Key TJ, Kahsar-Miller M, Waggoner W, Boots LR, Azzizet R. Prevalence of the polycystic ovary syndrome in unselected black and white women of the southeastern United States: a prospective study. J CHn Endocrinol Metab 1998;83:3078-3082.

2. Franks S. Polycystic ovary syndrome. N Engl J Med 1995; 333:853-861.

3. Burghen GA, Givens JR3 Kitabehi AE. Correlation of hyperandrogenism with hyperinsulinaemia in polycystic ovarian disease. J CHn Endocrinol Metab 1980,50:113-116.

4. Legro RS, Kunselman AR, Dodson WC, Dunaif A. Prevalence and predictors of risk for type 2 diabetes mellitus and impaired glucose tolerance in polycystic ovary syndrome: a prospective, controlled study in 254 affected women. J CHn Endocrinol Metab 1999;84:165- 169.

5. Lobo RA, Carmina E. The importance of diagnosing the polycystic syndrome. Ann Intern Med 2000;132:989-993.

6. Ciaraldi TP, el Roeiy A, Madar Z, Reichart D, Olefsky JM, Yen SSC. Cellular mechanisms of insulin resistance in polycystic ovarian syndrome. J CHn Endocrinol Metab 1992;75:577-583.

7. Dunaif A, Xia J, Book CB, Schenker E, Tang Z. Excessive insulin receptor serine phosphorylation in cultured fibroblasts and in skeletal muscle. A potential mechanism for insulin resistance in the polycystic ovary syndrome. J Clin Invest 1995;96:801-810.

8. Jahanfar S, Eden JA, Warren P, Seppala M, Nguyen TV. A twin study of polycystic ovary syndrome. Fertil Steril 1995; 63:478-486.

9. Diamanti-Kandarakis E, Baillargeon JP, Iuorno MJ, Jakubowicz DJ, Nestler JE. A modern medical quandary: polycystic ovary syndrome, insulin resistance, and oral contraceptive pills. J CHn Endocrinol Metab 2003;88:1927-1932.

10. Campbell PJ, Gerich JE. Impact of obesity on insulin action in volunteers with normal glucose tolerance: demonstration of a threshold for the adverse effect of obesity. J CHn Endocrinol Metab 1990;70:1114-1118.

11. Glueck CJ, Wang P, Fontaine R, Tracy T, Sieve-Smith L. Metformin-induced resumption of normal menses in 39 of 41 (91%) previously amenorrhoeic women with the polycystic ovary syndrome. Metabolism 1999;48:511-519.

12. Unluhizarci K, Kelestimur F, Bayram F, Sahin Y, Tutus A. The effects of metformin on insulin resistance and ovarian steroidogenesis in women with polycystic ovary syndrome. CHn Endocrinol 1999;51:231-236.

13. Chevenne D, Trivin F, Porquet D. Insulin assays and reference values. Diabetes Metab 1999;25:459-476.

14. Nestler JE, Jakubowicz DJ, Evans WS, Pasquali R. Effects of metformin on spontaneous and clomiphene-induced ovulation in the polycystic ovarian syndrome. N Engl J Med 1998;338:1876-1880.

15. Pasquali R, Gambineri A, Biscotti D, Vicennati V, Gagliardi L, Colitta D, Fiorini S, Cognigni GE, Filicori M, Morselli-Labate AM. Effect of long-term treatment with metformin added to hypocaloric diet on body composition, fat distribution, and androgen and insulin levels in abdominally obese women with and without the polycystic ovary syndrome. J CHn Endocrinol Metab 2000;85:2767- 2774.

16. Taylor AE. Insulin-lowering medications in polycystic ovary syndrome. Obstet Gynecol CHn North Am 2000;27:583-595.

17. Riste L, Khan F, Cruickshank K. High prevalence of type 2 diabetes in all ethnic groups, including Europeans, in a British inner city: relative poverty, history, inactivity or 21st century Europe. Diabetes Care 2001;24:1377-1383.

18. World Health Organization. Guidelines for diabetes. Geneva: WHO: 2000 Jun.

19. Ferriman D, Gallwey JD. Clinical assessment of body hair growth in women. J CHn Endocrinol Metab 1961;21:1440-1447.

20. Matthews DR, Hosker JB, 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;28:412-419.

21. Rudenski AS, Matthews DR, Levy JC, Turner RC. Understanding 'insulin resistance': both glucose resistance and insulin resistance are required to model human diabetes. Metabolism 1991;40:908-917.

22. Matthews JNS, Altman DG, Campbell MJ, Royston P. Analysis of serial measurements in medical research. Br Med J 1990;300:230-235.

23. Ferrannini E, Mari A. How to measure insulin sensitivity. J Hypertens 1998;16:895-906.

24. Rudenski AS, Hadden DR, Atkinson AB, Kennedy L, Matthews DR, Merrett JD, Pockaj B, Turner RC. Natural history of pancreatic islet B-cell function in type 2 diabetes mellitus studied over six years by homeostasis model assessment. Diabetic Medicine 1988;5:36-41.

25. Dunaif A, Graf M, Mandeli J, Laumas V, Dobrjansky A. Characterization of groups of hyperandrogenic women with acanthosis nigricans, impaired glucose tolerance, and/or hypcrinsulinemia. J Clin Endocrinol Metab 1987;65:499-507.

26. Poison DW, Adams J, Wadsworth J, Franks S. Polycystic ovaries - a common finding in normal women. Lancet 1988;l:870-872.

27. Nestler JE. Sex hormone-binding globulin: a marker for hyperinsulinemia and/or insulin resistance? J CHn Endocrinol Metab 1993;76:273-274.

28. Sattar N, Hopkinson ZE, Greer IA. Insulin-sensitising agents in polycystic-ovary syndrome. Lancet 1998;354:305-307.

29. Morin-Papunen L, Rautio K, Ruokonen A, Hedberg P, Puukka M, Tapanainen JS. Metformin reduces serum C-reactive protein levels in women with polycystic ovary syndrome. J CHn Endocrinol Metab 2003;88:4649-4654.

30. Jayagopal V, Kilpatrick ES, Jennings PE, Hepburn DA, Atkin SL. The biological variation of testosterone and sex hormone- binding globulin (SHBG) in polycystic ovarian syndrome: implications for SHBG as a surrogate marker of insulin resistance. J CHn Endocrinol Metab 2003;88:1528-1533.

31. Heald AH, Cruickshank JK, Ristc LK, Cade JE, Andersen S, Greenhalgh A, Sampayo J, Taylor W, Fraser W, White A, et al. Close relationship of fasting insulin-like growth factor binding protein- 1 (IGFBP-I) with glucose tolerance and cardiovascular risk in 2 populations. Diabetologia 2001;44: 333-339.

32. Morris DV, Falc\one T. The relationship between insulin sensitivity and insulin-like growth factor-binding protein-1. Gynecol Endocrinol 1996; 10:407-412.

33. Kelly CJ, Gordon D. The effect of metformin on hirsutism in polycystic ovary syndrome. Eur J Endocrinol 2002;147:217-221.

ADRIAN H. HEALD1, SHARON WHITEHEAD1, SIMON ANDERSON2, KENNEDY CRUICKSHANK2, LISA RISTE2, IAN LAING3, ARAM RUDENSKI1, & HELEN BUCKLER1

1 Department of Endocrinology, University of Manchester, Salford NHS Trust, Salford, UK, 2 Clinical Epidemiology Group, University of Manchester, Manchester, UK, and 3 Clinical Biochemistry Department, Manchester Royal Infirmary, Manchester, UK

(Received 25 April 2004; revised 9 August 2004; accepted 27 August 2004)

Correspondence; A. H. Heald, Department of Diabetes and Endocrinology, University of Manchester, Salford Royal Hospitals University Trust, Hope Hospital. Stott Lane, Salford, Greater Manchester M6 8HD, UK. Tel: 44 161 7875146. Fax: 44 161 7875989. E- mail: aheald@fsl.ho.man.ac.uk

Copyright CRC Press Feb 2005

Source: Gynecological Endocrinology

More News in this Category