Departments of 1 Endocrinology and 2 Molecular Genetics, Hospital Ramón y Cajal, Madrid, Spain
3 To whom correspondence should be addressed at: Department of Endocrinology, Hospital Ramón y Cajal, Carretera de Colmenar km 9'1, E-28034 Madrid, Spain. E-mail: hescobarm.hrc{at}salud.madrid.org
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: insulin resistance/insulin-receptor substrate/polycystic ovary syndrome/polymorphisms
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Considering that insulin resistance and an increased risk of type 2 diabetes mellitus are frequent in patients with PCOS (Dunaif, 1997; Ehrmann et al., 1999
), and that this disorder develops soon after puberty, PCOS may be considered to represent the earliest manifestation of the metabolic syndrome in some women.
Although the mechanisms underlying the defective insulin signalling in this disorder have not been fully elucidated (Legro, 1998; Dunaif et al., 2001
), several studies have shown that allelic variants in the insulin receptor substrate genes, IRS-1 Gly972Arg and IRS-2 Gly1057Asp, may play a functional role on the insulin-resistant component of PCOS (El Mkadem et al., 2001
; Sir-Petermann et al., 2001
; Ehrmann et al., 2002
).
Sir-Petermann et al. (2001) reported that the frequency of the IRS-1 Arg972 allele was higher in PCOS patients, compared with normal women, in the Chilean population. El Mkadem et al. (2001)
, studying Caucasian women of European extraction, did not find any difference between PCOS patients and controls in the distribution of IRS-1 Gly972Arg and IRS-2 Gly1057Asp alleles. The IRS-1 Arg972 allele was more prevalent in insulin-resistant PCOS patients compared with non-insulin-resistant patients or control subjects (El Mkadem et al., 2001
). Carriers of IRS-1 Arg972 alleles presented with increased fasting insulin levels and increased insulin resistance measured by homeostasis model assessment (HOMA-IR), compared with subjects homozygous for wild-type alleles (El Mkadem et al., 2001
). Also, carriers of IRS-2 Asp1057 alleles presented with increased 2 h glucose and insulin levels during an oral glucose tolerance test (OGTT) (El Mkadem et al., 2001
). Finally HOMA-IR was higher in carriers of both IRS-1 and IRS-2 variants than in those with IRS-2 mutations only or those with wild-type alleles (El Mkadem et al., 2001
).
On the contrary, Ehrmann et al. (2002) in their study of 227 non-diabetic PCOS white and African-American patients, found that non-diabetic subjects carrying one or two Asp1057 alleles of IRS-2 had significantly lower 2 h OGTT glucose levels compared with those homozygous for Gly1057 alleles (Ehrmann et al., 2002
), in sharp contrast to the results of El Mkadem et al. (2001)
.
Given that these controversies might have been influenced by the ethnic origin of the subjects included in previous studies (El Mkadem et al., 2001; Sir-Petermann et al., 2001
; Ehrmann et al., 2002
), we have evaluated whether IRS-1 Gly972Arg and IRS-2 Gly1057Asp influence insulin resistance, fasting glucose levels, obesity and hyperandrogenism, in Caucasian PCOS patients and controls from the Spanish population.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The control group of 48 women (age 31 ± 8 years, BMI 27.9 ± 7.5 kg/m2) was recruited from healthy female volunteers, and from consecutive patients attending the clinical practice of one of the authors (H.F.E.-M.) for dietary treatment of obesity. None of the controls had signs or symptoms of hyperandrogenism, menstrual dysfunction or history of infertility.
Data from some patients and controls, regarding different aspects of the pathophysiology of hirsutism, have been published previously (San Millan et al., 2004). The patients and controls had not taken hormonal medications, including contraceptive pills, for the last 6 months. All the subjects were Caucasian.
The ethics committee of the Hospital Ramón y Cajal approved the study, and informed consent was obtained from each patient and control.
Hormone profiles
Studies were performed between days 5 and 10 of the menstrual cycle, or during amenorrhoea, after excluding pregnancy by appropriate testing. Hirsutism was quantified by the modified FerrimanGallwey score (Hatch et al., 1981). Between 8 a.m and 9 a.m. after a 12 h overnight fast, an indwelling i.v. line was placed in a forearm vein, and, after 1530 min, basal blood samples were obtained for the measurement of total testosterone, dehydroepiandrosterone sulphate (DHEA-S), sex hormone-binding globulin (SHBG), glucose and insulin. In a subgroup of 72 PCOS patients and 34 controls, a standard 75 g OGTT was performed, and blood samples were obtained after 0, 30, 60, 90 and 120 min for glucose and insulin determinations. These subjects received a 300 g carbohydrate diet for 3 days before sampling. All samples were immediately centrifuged, and serum and plasma were separated and frozen at 20ºC until assayed.
The technical characteristics of the assays employed for hormone measurements have been reported elsewhere (Escobar-Morreale et al., 1997, 2000
). The free testosterone concentration was calculated from total testosterone and SHBG concentrations, assuming a serum albumin concentration of 4.3 g/l, and taking a value of 1 x 109 l/mol for the association constant of SHBG for total testosterone and a value of 3.6 x 104 l/mol for that of albumin for total testosterone (Vermeulen et al., 1999
). Insulin resistance in the fasting state was estimated from glucose and insulin levels using homeostasis model assessment (HOMA-IR) (Matthews et al., 1985
). Individuals were classified as having normal insulin sensitivity (HOMA-IR <3.10) or having insulin resistance (HOMA-IR
3.10). The HOMA-IR cut-off of 3.10 was established from the mean +2SD of our non-obese control group (n = 21). In the subjects submitted to an OGTT, the composite insulin sensitivity index was also determined (Matsuda and DeFronzo, 1999
).
DNA analysis
Genomic DNA was obtained from whole blood using a Nucleon BACC3 DNA isolation kit (Amersham, Buckingham, UK). Genotyping of allelic variants Gly972Arg of IRS-1 and Gly1057Asp of IRS-2 was conducted using PCR followed by digestion with BstNI and HaeII restriction enzymes, respectively (Mammarella et al., 2000; Baroni et al., 2001
).
Statistical analysis
Results are expressed as means ± SD unless otherwise stated. The KolmogorovSmirnov statistic was applied to continuous variables. Logarithmic transformation was applied as needed to ensure normal distribution of the variables. Unpaired t-tests were used to compare the central tendencies of the different groups.
To evaluate the association between discontinuous variables, we used the 2 test and Fishers exact test as appropriate. A priori power analysis of the differences in frequencies between PCOS patients and controls was conducted. Our overall 151 sample size permitted the detection of effect sizes for the difference between frequencies of genotypes of 0.23 for the
2 test with 1 degree of freedom (autosomal dominant or recessive model of inheritance), and 0.25 for the
2 test with 2 degrees of freedom (distribution of the three possible genotypes of a single nucleotide polymorphism), used here.
Taking into account the case to control ratio and the allelic frequencies previously reported by others (El Mkadem et al., 2001; Sir-Petermann et al., 2001
), we also calculated a priori power analysis of the evaluation of the difference in allelic frequencies between our patients and controls by the
2 test for proportions in two groups (Lemeshow et al., 1990
) using the PASS2000 software (Number Cruncher Statistical Systems, Kaysville, UT). Our sample size allowed detection of differences in the allelic frequencies of the IRS-1 gene variant of 0.19 (Sir-Petermann et al., 2001
) and 0.22 (El Mkadem et al., 2001
) depending on the report used as the data source, and a difference between IRS-2 gene variant frequencies of 0.21 (El Mkadem et al., 2001
).
By convention, effect sizes for the differences between frequencies are considered very small or trivial when <0.10, small from 0.10 to 0.30, moderate from 0.30 to 0.50, and large when >0.50 (Cohen, 1988). Therefore, our sample size permitted the detection of small differences between the frequencies in PCOS patients and controls. However, very small and minor differences between frequencies in both groups of subjects may have not been detected in our study because of the relatively small sample size.
The 2 test served to compare the expected genotypic frequencies based on the HardyWeinberg equilibrium with the actual frequencies observed in the PCOS and controls groups separately.
The differences among PCOS patients and controls in clinical and biochemical variables related to hyperandrogenism and to insulin resistance were evaluated by general linear model (GLM) analysis introducing age and BMI as covariates. The influence of the different genotypes on clinical and biochemical variables related to hyperandrogenism and to insulin resistance was analysed by one-way analysis of variance (ANOVA) followed by the least significant differences test for post hoc comparisons. Multivariate or univariate GLM analyses were used to evaluate the effects of IRS-1 genotypes on both BMI and insulin resistance, and the possible interaction of both IRS-1 and IRS-2 genotypes on glucose metabolism, on BMI and on insulin resistance, as needed. Analyses were performed using SPSS10 for the Macintosh (SPSS Inc, Chicago, IL). P < 0.05 was considered statistically significant.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Compared with controls, PCOS patients presented with increased hirsutism scores, total and calculated free testosterone, DHEA-S and androstenedione levels, increased OGTT glucose values, fasting and OGTT insulin concentrations and HOMA-IR values, and decreased insulin sensitivity index and SHBG levels.
No statistically significant differences in genotypes or allelic frequencies for both variants were observed between controls and PCOS patients (Table II). Moreover, no association was observed when considering the combination of both polymorphisms: 33.3% of controls and 47.6% of PCOS patients were homozygous for wild-type alleles of both polymorphisms, 60.4% of the controls and 46.6% of PCOS patients had mutated alleles of only one of the variants, and 6.3% of the controls and 5.8% of PCOS patients had mutated alleles for both polymorphisms (2 = 2.778, P > 0.200).
|
Given that IRS-1 and IRS-2 variants were not differently distributed among PCOS patients and controls, to investigate whether these polymorphisms affected any of the clinical or biochemical variables we considered initially patients and controls as a whole.
Carriers of the IRS-1 Arg972 alleles presented with increased fasting and 30 min OGTT insulin concentrations (P = 0.008 and P = 0.038, respectively), HOMA-IR (P = 0.009) and a near significant tendency (P < 0.1) towards higher BMI and 60 min OGTT insulin levels (Table III). When studied separately, most of these differences reached statistical significance in the PCOS group but not in the controls (Table III), although the small sample size in the latter might have precluded reaching statistical significance, despite the finding of very large differences in insulin levels (Table III).
|
When all individuals were classified depending on the presence or absence of insulin resistance (HOMA-IR 3.10 or <3.10, respectively), the prevalence of IRS-1 Arg972 alleles in insulin-resistant individuals was 2.7-fold higher compared with those with normal HOMA-IR values (20.7 versus 7.7%,
2 = 5.379, P = 0.025). The prevalence of IRS-1 Arg972 alleles reached a 6-fold increase in insulin-resistant PCOS patients compared with patients with normal insulin sensitivity (21.7 versus 3.6%,
2 = 7.841, P = 0.011), yet were equally distributed among controls with or without insulin resistance (16.7 versus 13.9%,
2 = 0.056, P > 0.200).
When the whole group was classified into lean (BMI <25 kg/m2) or overweight/obese subjects (BMI 25 kg/m2), the prevalence of IRS-1 Arg972 alleles was 3.3-fold higher in overweight/obese individuals than in normal weight individuals (17.2 versus 5.2%,
2 = 4.701, P = 0.042). The prevalence of IRS-1 Arg972 alleles reached a 6.2-fold increase in overweight/obese PCOS patients compared with lean patients (16.7 versus 2.7%,
2 = 4.491, P = 0.034), but they were not differently distributed among controls with or without overweight or obesity (18.5 versus 9.5%,
2 = 0.767, P > 0.200).
Because IRS-1 Arg972 alleles were associated with insulin resistance and with obesity, a multivariate GLM analysis, introducing BMI and HOMA-IR as dependent variables, and carrying IRS-1 Arg972 alleles versus homozygosity for wild-type alleles as the independent variable, was performed. Given that these influences were somehow different in the PCOS and control groups, being a patient or a control was introduced as a covariate in the model. The results showed that IRS-1 Arg972 alleles influenced HOMA-IR (F = 7.835, P = 0.006), and tended to influence the BMI (F = 3.556, P = 0.062), irrespective of patient or control status.
With respect to the IRS-2 Gly1057Asp variant, when considering PCOS patients and controls as a whole, 60 and 90 min OGTT glucose levels were lower in subjects with one or two Asp1057 alleles when compared with individuals homozygous for the wild-type Gly1057 allele, and a similar near-significant tendency was found for 120 min OGTT glucose levels (Table IV). However, these differences did not reach statistical significance when studying PCOS patients and controls separately (Table IV).
|
To evaluate the possible collaboration of the IRS-1 and IRS-2 variants on insulin resistance, a univariate GLM was used, introducing HOMA-IR as a dependent variable, and carrying IRS-1 Arg972 alleles or carrying one or two IRS-2 Asp1057 alleles as independent variables. PCOS patient or control status was introduced as a covariate in the model, given the differences in insulin resistance between both groups. The model retained only carrying IRS-1 Arg972 alleles as a significant factor (F = 7.529, P = 0.007), and showed no effect of IRS-2 Asp1057 alleles (F = 0.515, P > 0.200) and no interaction of both variants (F = 0.995, P > 0.200).
A similar model was used to study the influence on BMI, which was introduced as a dependent variable. The influence of carrying IRS-1 Arg972 alleles was near-significant (F = 3.310, P = 0.071), with no effect of IRS-2 Asp1057 alleles (F = 0.249, P > 0.200) or any interaction between both variants (F = 1.515, P > 0.200).
Similarly, a multivariate GLM analysis introducing 60 and 90 min OGTT glucose levels as dependent variable showed no effect of IRS-1 Arg972 alleles (F = 0.004, P = 0.948 and F = 0.637, P > 0.200), a near significant effect of IRS-2 Asp1057 alleles (F = 3.410, P = 0.068 and F = 2.859, P = 0.094) and no effect of the interaction between both variants (F = 0.743, P = 0.391 and F = 0.052, P > 0.200).
Finally, when controls and PCOS patients were considered as a whole and classified depending on the different combinations of alleles of both polymorphisms (homozygous for wild-type alleles on both polymorphisms, mutated alleles on only one of the variants or mutated alleles on both polymorphisms), no differences were found on any of the phenotypic features studied here (data not shown).
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Recently, the IRS-1 Gly972Arg and IRS-2 Gly1057Asp polymorphisms have been studied in PCOS (El Mkadem et al., 2001; Sir-Petermann et al., 2001
; Ehrmann et al., 2002
), as well as in related disorders such as premature pubarche and adolescent hyperandrogenism (Witchel et al., 2001
; Ibanez et al., 2002
). However, the results in PCOS patients were in considerable disagreement and, therefore, the role of these variants in the pathogenesis of insulin resistance and PCOS is debatable.
Our present study may contribute to enlighten these controversies. First, neither the IRS-1 Gly972Arg nor IRS-2 Gly1057Asp polymorphisms, nor any combination of both, was associated with PCOS. This result is in agreement with the study of El Mkadem et al. (2001), who did not find any difference in the distribution of IRS-1 Gly972Arg or IRS-2 Gly1057Asp alleles between PCOS patients and controls. Although, Sir-Petermann et al. (2001)
reported a small but statistically significant increase in the frequency of the IRS-1 Arg972 allele in PCOS patients compared with normal women in the Chilean population, as many as 87.9% of the PCOS patients in this series were homozygous for IRS-1 Gly972 alleles, compared with 94.5% of the controls, and, therefore, the IRS-1 Arg972 allele is unlikely to play a major role in the development of PCOS. Although there was a slight deviation from HardyWeinberg equilibrium in the IRS-2 Gly1057Asp genotype distribution in the PCOS group, which might have been related to the fact that our patients were not randomly selected from the population but from clinical practice, it should be highlighted that the non-hyperandrogenic controls were in equilibrium, and that there were no differences in allelic frequencies and in genotype distributions between PCOS patients and controls. Therefore, it is unlikely that the minimal departure from the HardyWeinberg equilibrium of IRS-2 Gly1057Asp genotype distribution in the PCOS group influenced our present results.
Secondly, we suggest here that the IRS-1 Gly972Arg and IRS-2 Gly1057Asp polymorphisms influence glucose homeostasis, and might influence obesity, irrespective of whether or not women presented with PCOS.
IRS-1 Arg972 alleles were more prevalent in insulin-resistant individuals, in agreement with the results of El Mkadem et al. (2001), and these alleles were also more prevalent in overweight/obese individuals, suggesting that the IRS-1 Gly972Arg polymorphism also influences body weight, especially when the analysis was restricted to PCOS patients.
Given that obesity increases insulin resistance, we performed a multivariate GLM analysis to study the possible effects of the IRS-1 genotype on both insulin resistance and BMI. Our present results further suggest that the main effect of the IRS-1 polymorphism is on insulin resistance, yet it might also have an additional less important effect on BMI.
When considering the effect of these polymorphisms on glucose homeostasis, we found that carriers of IRS-1 Arg972 alleles presented with higher fasting insulin levels and HOMA-IR values, compared with subjects homozygous for wild-type alleles. In conceptual agreement, in previous studies, this variant has been associated with higher fasting insulin levels and HOMA-IR values among PCOS women (El Mkadem et al., 2001) and with 2 h OGTT insulin levels in PCOS patients (Sir-Petermann et al., 2001
).
With respect to the IRS-2 Gly1057Asp polymorphism, in our series, carriers of one or two Asp1057 alleles presented a tendency towards lower OGTT glucose levels compared with subjects homozygous for Gly1057 alleles, reaching statistical significance at the 60 and 90 min sample. Our results replicate, in a Caucasian population from Spain, the previous finding of Ehrmann et al. (2002) in their series of White and African-American PCOS patients, in sharp contrast to the increase in 2 h OGTT glucose levels reported previously by El Mkadem et al. (2001)
in Caucasian women of French origin. We do not have an explanation for the discrepancy of our present result, and for the discrepancy of the results of Ehrmann et al. (2002)
with those of El Mkadem et al. (2001)
.
In conclusion, given the lack of an association between the IRS-1 Gly972Arg and IRS-2 Gly1057Asp polymorphisms and PCOS, our present results confirm that these loci should not be considered major contributors to the pathogenesis of this prevalent disorder. Yet these polymorphisms influence glucose homeostasis and body weight in PCOS patients and in non-hyperandrogenic women, and therefore might contribute to explain the pathogenesis of insulin resistance, glucose intolerance and/or obesity in premenopausal women.
![]() |
Acknowledgements |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Notes |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Baroni MG, Arca M, Sentinelli F, Buzzetti R, Capici F, Lovari S, Vitale M, Romero S and DiMario U (2001) The G972R variant of the insulin receptor substrate-1 (IRS-1) gene, body fat distribution and insulin-resistance. Diabetologia 44,367372.[CrossRef][ISI][Medline]
Carmina E and Lobo RA (1999) Polycystic ovary syndrome (PCOS): arguably the most common endocrinopathy is associated with significant morbidity in women. J Clin Endocrinol Metab 84,18971899.
Cohen J (1988) Statistical Power Analyisis for the Behavioral Sciences, 2nd edn, Lawrence Erlbaum, Hillsdale, NJ.
Dunaif A (1997) Insulin resistance and the polycystic ovary syndrome: mechanism and implications for pathogenesis. Endocr Rev 18,774800.
Dunaif A, Wu X, Lee A and Diamanti-Kandarakis E (2001) Defects in insulin receptor signaling in vivo in the polycystic ovary syndrome (PCOS). Am J Physiol 281, E392E399.[ISI]
Ehrmann D, Barnes R, Rosenfield R, Cavaghan M and Imperial J (1999) Prevalence of impaired glucose tolerance and diabetes in women with polycystic ovary syndrome. Diabetes Care 22,141146.[Abstract]
Ehrmann DA, Tang X, Yoshiuchi I, Cox NJ and Bell GI (2002) Relationship of insulin receptor substrate-1 and -2 genotypes to phenotypic features of polycystic ovary syndrome. J Clin Endocrinol Metab 87,42974300.
El Mkadem SA, Lautier C, Macari F, Molinari N, Lefebvre P, Renard E, Gris JC, Cros G, Daures JP, Bringer J et al. (2001) Role of allelic variants Gly972Arg of IRS-1 and Gly1057Asp of IRS-2 in moderate-to-severe insulin resistance of women with polycystic ovary syndrome. Diabetes 50,21642168.
Escobar-Morreale HF, Serrano-Gotarredona J, Varela C, García-Robles R and Sancho JM (1997) Circulating leptin concentrations in women with hirsutism. Fertil Steril 68,898906.[CrossRef][ISI][Medline]
Escobar-Morreale HF, Avila S and Sancho J (2000) Serum prostate-specific antigen concentrations are not useful for monitoring the treatment of hirsutism with oral contraceptive pills. J Clin Endocrinol Metab 85,24882492.
Hatch R, Rosenfield RL, Kim MH and Tredway D (1981) Hirsutism: implications, etiology, and management. Am J Obstet Gynecol 140,815830.[ISI][Medline]
Ibanez L, Marcos MV, Potau N, White C, Aston CE and Witchel SF (2002) Increased frequency of the G972R variant of the insulin receptor substrate-1 (IRS-1) gene among girls with a history of precocious pubarche. Fertil Steril 78,12881293.[CrossRef][ISI][Medline]
Legro RS (1998) Insulin resistance in polycystic ovary syndrome: treating a phenotype without a genotype. Mol Cell Endocrinol 145,103110.[CrossRef][ISI][Medline]
Legro RS, Driscoll D, Strauss JF 3rd, Fox J and Dunaif A (1998) Evidence for a genetic basis for hyperandrogenemia in polycystic ovary syndrome. Proc Natl Acad Sci USA 95,1495614960.
Legro RS, Bentley-Lewis R, Driscoll D, Wang SC and Dunaif A (2002) Insulin resistance in the sisters of women with polycystic ovary syndrome: association with hyperandrogenemia rather than menstrual irregularity. J Clin Endocrinol Metab 87,21282133.
Lemeshow S, Hosmer DW, Klar J and Lwanga SK (1990) Adequacy of Sample Size in Health Studies. World Health OrganizationJohn Wiley & Sons, Chichester, UK.
Mammarella S, Romano F, Di Valerio A, Creati B, Esposito DL, Palmirotta R, Capani F, Vitullo P, Volpe G, Battista P et al. (2000) Interaction between the G1057D variant of IRS-2 and overweight in the pathogenesis of type 2 diabetes. Hum Mol Genet 9,25172521.
Matsuda M and DeFronzo RA (1999) Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care 22,14621470.[Abstract]
Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF and Turner RC (1985) Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28,412419.[CrossRef][ISI][Medline]
San Millan JL, Corton M, Villuendas G, Sancho J, Peral B and Escobar-Morreale HF (2004) Association of the polycystic ovary syndrome with genomic variants related to insulin resistance, type 2 diabetes mellitus, and obesity. J Clin Endocrinol Metab 89,26402646.
Sesti G, Federici M, Hribal ML, Lauro D, Sbraccia P and Lauro R (2001) Defects of the insulin receptor substrate (IRS) system in human metabolic disorders. FASEB J 15,20992111.
Sir-Petermann T, Perez-Bravo F, Angel B, Maliqueo M, Calvillan M and Palomino A (2001) G972R polymorphism of IRS-1 in women with polycystic ovary syndrome. Diabetologia 44,12001201.[CrossRef][ISI][Medline]
Vermeulen A, Verdonck L and Kaufman JM (1999) A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab 84,36663672.
Witchel SF, Smith R, Tomboc M and Aston CE (2001) Candidate gene analysis in premature pubarche and adolescent hyperandrogenism. Fertil Steril 75,724730.[CrossRef][ISI][Medline]
Wu X, Sallinen K, Anttila L, Makinen M, Luo C, Pollanen P and Erkkola R (2000) Expression of insulin-receptor substrate-1 and -2 in ovaries from women with insulin resistance and from controls. Fertil Steril 74,564572.[CrossRef][ISI][Medline]
Yildiz BO, Yarali H, Oguz H and Bayraktar M (2003) Glucose intolerance, insulin resistance, and hyperandrogenemia in first degree relatives of women with polycystic ovary syndrome. J Clin Endocrinol Metab 88,20312036.
Zawadzki JK and Dunaif A (1992) Diagnostic criteria for polycystic ovary syndrome: towards a rational approach. In Merriam GR (eds), Polycystic Ovary Syndrome. Blackwell Scientific Publications, Boston, pp. 377384.
Submitted on February 27, 2005; resubmitted on May 17, 2005; accepted on June 16, 2005.
|