1 Department of Obstetrics and Gynecology, Università Cattolica del Sacro Cuore, Largo A.Gemelli 8, 00168 Roma, 2 Istituto Scientifico Internazionale Paolo VI and 3 OASI Institute for Research, Troina, Italy
4 To whom correspondence should be addressed. e-mail: maurizioguido{at}libero.it
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: adrenal glands/insulin/PCOS/pioglitazone
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In this regard, several lines of evidence seem to indicate a relationship between adrenal abnormalities and PCOS: for instance, this syndrome often occurs in women with congenital adrenal hyperplasia (Horrocks et al., 1982); Cushings syndrome and androgen-producing tumours are associated with PCOS (Kase et al., 1963
); rats given excess of dehydroepiandrosterone (DHEA) develop polycystic ovaries (Roy et al., 1962
). Furthermore, a hyper-responsiveness to adrenocorticotrophic hormone (ACTH) has been described to varying degrees in most PCOS women (Lucky et al., 1986
; Azziz et al., 1998
). These findings have led several authors to propose an exaggerated and/or prolonged adrenarche as a primary aetiopathogenic mechanism for the development of the syndrome (Lucky et al., 1986
; Lazar et al., 1995
).
It is widely recognized that insulin resistance and hyperinsulinaemia, which affect a large proportion of PCOS patients, may play a role in the aetiology of hyperandrogenism: at ovarian level, insulin promotes ovarian androgen secretion, playing a synergistic role with gonadotrophins both directly and stimulating insulin like-growth factor I (IGF-I) secretion (Cara and Rosenfield, 1988); in the liver it also decreases serum sex hormone-binding globulin (SHBG) synthesis (Nestler et al., 1991
), thus increasing free androgen concentrations; at the adrenal level, we previously demonstrated that hyperinsulinaemia is able to potentiate in vivo ACTH-stimulated androgen production in women with PCOS (Lanzone et al., 1992
); the same result was obtained by in vitro studies (Bianchi et al., 1993
); more recently, it was proposed that this effect of insulin was mediated by a relative impairment of 17,20-lyase activity (Moghetti et al., 1996
). In turn, modifications of insulin plasma concentration both by the opioid antagonist naltrexone (Lanzone et al., 1994) and by metformin administration (La Marca et al., 1999
; Arslanian et al., 2002
) led to a reduction in the ACTH-stimulated adrenal steroidogenesis in women affected by PCOS. Moreover, in a recent study evaluating the basal adrenal androgen levels, the thiazolidinedione troglitazone was reported to reduce dehydroepiandrosterone sulphate (DHEAS) circulating levels (Azziz et al., 2003
).
Pioglitazone, a new insulin-sensitizing agent belonging to the thiazolidinediones class, is able to enhance insulin action with a post-insulin receptor mechanism of action (Lehmann et al., 1995). The different chemical structure of this compound, which exhibits a higher affinity to the specific receptor peroxisome proliferator-activated receptor (PPAR)
, allows a more potent insulin-sensitizing effect with a much lower hepatotoxicity compared with troglitazone and rosiglitazone (Gillies and Dunn, 2000
). In a recent study from our group, pioglitazone administration to obese PCOS women induced an amelioration of the metabolic assessment, with a parallel improvement of several clinical and biochemical parameters typical of the syndrome (Romualdi et al., 2003
). In particular, the decrease in insulin secretion obtained with pioglitazone treatment was associated with a significant reduction of basal 17
-hydroxyprogesterone (17OHP) levels in hyperinsulinaemic PCOS patients, whereas the adrenal production of DHEAS remained unaffected.
On the basis of these studies, we wanted to investigate further the effect of pioglitazone on the adrenal steroidogenesis of women with PCOS, by subjecting them to ACTH stimulation before and after 6 months of pioglitazone treatment.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
PCOS was diagnosed according to the following (Homburg, 2002): the presence of clinical findings (at least two of these signs: amenorrhoea or oligomenorrhoea, hirsutism and/or acne, chronic anovulation), plasma androgen levels at the upper limit of, or above, the normal range [at least one of: free androgen index (FAI) >5; androstenedione
6.98 nmol/l; testosterone
2.0 nmol/l], and the presence of bilaterally normal or enlarged ovaries containing
710 microcysts (<5 mm in diameter) on ultrasonography, with an augmented stromal area:total area ratio (Adams et al., 1985
; Fulghesu et al., 2001
). A normal LH/FSH ratio was not considered an exclusion criterion. Pregnancy or possibility of pregnancy and nursing, significant liver [aspartate aminotransferase (AST), alanine aminotransferase (ALT), total bilirubin or alkaline phosphatase >2 times the upper limit of normal] or renal impairment (serum creatinine >1.8 ng/dl), neoplasm, cardiovascular disease or unstable mental illness were considered exclusion criteria.
Informed consent was obtained from each patient, and the study protocol was approved by our Institutional Review Board.
Study design
Studies were conducted during the early follicular phase of spontaneous or induced [medroxyprogesterone acetate (MPA) 10 mg/day for 7 days] menstrual cycles (day 37).
After following a standard carbohydrate diet (300 g/day) for 3 days and fasting overnight for 1012 h, blood samples were collected in order to perform the following laboratory assays: basal hormone assessment, hepatic and renal chemistries. On the same day, patients underwent an oral glucose tolerance test (OGTT). The OGTT was performed as follows: at 09:00 after overnight fasting, an indwelling catheter was inserted into the antecubital vein of one arm. Blood samples were collected basally and after ingestion of 75 g glucose in 150 ml water, within 5 min, and at 30, 60, 90, 120, 180 and 240 min. Insulin, glucose and C-peptide were assayed in all samples.
On the following day, after a 10 h overnight fast, a hyperinsulinaemiceuglycaemic clamp was performed to estimate peripheral insulin sensitivity. At 08:00, an i.v. catheter was placed in the antecubital vein for the infusion of glucose and insulin. Another catheter was placed in the dorsal vein of the contralateral hand for blood withdrawal and warmed to 65°C with a warming box. A constant infusion of insulin (Actrapid HM; Novo Nordisk, Denmark) 40 mIU/m2 per min was started (De Fronzo et al., 1979). After reaching the steady-state velocity for the insulin infusion within 10 min in order to achieve steady-state insulin levels of
717 pmol/l during the clamp (range 574897 pmol/l), a variable infusion of 20% glucose was begun via a separate infusion pump and the rate was adjusted, on the basis of plasma glucose samples drawn every 5 min, to maintain plasma glucose between 4.4 and 4.99 mmol/l. The plasma glucose level was determined by the glucose oxidase technique with a glucose analyser (Beckam Instruments, USA). The glucose infusion rate during the last 60 min of a 2 h infusion was then taken as the estimate of peripheral insulin sensitivity and measured as M (mg/kg/min).
On the last day of hospitalization, after overnight bed rest and fasting, an ACTH test was performed as follows: at 07:00 an indwelling catheter was inserted into the antecubital vein and saline solution was infused slowly throughout the test in order to keep the vein patent. Blood samples were collected just before and 60 min after the injection of 250 µg of ACTH (Synacthen; CibaGeigy, Italy). Plasma cortisol DHEAS, androstenedione, testosterone and 17OHP were assayed.
The first day of the following menstruation, therapy with pioglitazone was started: one pill of 45 mg daily in the morning for 6 months. During the study, chronically stabilized therapies not interfering with the parameters under evaluation were permitted. The use of antidiabetic and/or estroprogestinic drugs was not allowed. Patients were recommended not to modify their usual diet.
The main parameters of liver function were monitored each month during the treatment.
Following pioglitazone treatment, all patients had a second hospitalization at menstrual days 37 and repeated the same protocol study.
Assays
Plasma samples for glucose concentration were collected in tubes containing an inhibitor of glycolysis (sodium fluoride) and were analysed within 5 h. Plasma glucose concentrations were determined by the glucose oxidase technique with a glucose analyser (Beckam, USA). Plasma samples for insulin and C-peptide concentrations were placed in tubes standing in ice, centrifuged for 10 min at 1000 g using a 4226 ALC Centrifuge (ALC, Italy) and remained frozen at 30°C until assayed.
All hormone assays were performed with commercial radioimmunoassay kits (Radim, Italy). The intra-assay and inter-assay coefficients of variation for all hormones were <8% and <15% respectively. For each determination, all samples from the same patient were assayed simultaneously.
Data analysis
OGTT data were analysed as area under the curve (AUC) after the glucose ingestion, calculated by the trapezoidal rule and expressed as pmol/l/240 min for insulin and C-peptide and nmol/l/240 min for the glucose plasma levels.
The glycaemic responses were defined in accordance with the criteria of the National Diabetes Data Group (1979).
The ratio of testosteronex100:(SHBG) was used to calculate the (FAI).
The response to ACTH was evaluated on the basis of plasma hormone levels detected 60 min after the injection. The apparent activities of 17,20-lyase and 17-hydroxylase were calculated with the product/precursor ratios (androstenedione 17OHP and 17OHP progesterone respectively), as previously described (La Marca et al., 1999
).
All data are presented as mean ± SD.
The significance of differences among the pre- and post-treatment measures was determined with the use of one-way analysis of variance and any significant difference was identified by using the Bonferroni correction for multiple comparisons. For all analyses, P < 0.05 was considered statistically significant.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
The insulin response to OGTT was markedly decreased after 6 months drug treatment (AUC of insulin pre-treatment compared with post-treatment: 119.97 ± 76.79 and 76.39 ± 34.55 pmol/l per 240 min respectively; P < 0.02) (Figure 1). In the same group of patients, pioglitazone induced an improvement in insulin sensitivity, documented by the significant increase of M value during the euglycaemichyperinsulinaemic clamp (3.07 ± 1.29 versus 3.79 ± 1.82 mg/kg/min; P = 0.03) (Figure 2).
|
|
|
No significant changes occurred in the basal ratio of 17OHP progesterone, which indicates 17-hydroxylase activity, and in the basal ratio of androstenedione 17OHP, which indicates 17,20-lyase activity, after 6 months of pioglitazone treatment (1.63 ± 0.78 versus 1.32 ± 0.51 and 1.53 ± 0.76 versus 1.25 ± 0.56 respectively; P > 0.05). Nevertheless, analysing the same parameters after ACTH stimulation, we observed a significant variation in the apparent enzymatic activity of 17,20-lyase (androstenedione 17OHP ratio), which rose from the baseline mean value of 0.90 ± 0.41 to 1.11 ± 0.42 at the end of the therapy (P < 0.05) (Figure 3).
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
However, circulating levels as well as diurnal rhythms of ACTH are generally similar in normal and PCOS subjects (Horrocks et al., 1983). These findings suggest that the increased adrenal androgen production in patients with PCOS derives from an altered adrenal responsiveness or from an abnormal adrenal stimulation by factors other than ACTH.
In this context, insulin is believed to constitute a candidate for the stimulation of the adrenal cortex. We have previously reported an increase in serum 17OHP and androstenedione responses to ACTH in hyperinsulinaemic PCOS subjects in respect of PCOS women with normal insulin levels and controls (Lanzone et al., 1992). This contention was indirectly supported by the evidence that a normalization of insulin secretion, obtained with a long-term treatment with the opioid antagonist naltrexone, was able to decrease the adrenal 17OHP and androstenedione response to ACTH in PCOS hyperinsulinaemic patients (Lanzone et al., 1994
). In the present study, 6 months therapy with pioglitazone at a dose of 45 mg/day was effective in decreasing the insulinaemic response to OGTT and, in line with previous reports in the literature, in ameliorating the steroid milieu of our patients. The same treatment was able to significantly decrease the adrenal secretion of 17OHP in response to an ACTH bolus and to reduce the production of androstenedione under the same experimental conditions. No differences in ACTH-stimulated DHEAS levels seem to emerge from the comparison between pre- and post-treatment values.
Cytochrome P450-17 (cP450-17
) is a key enzyme required for the synthesis of all androgens and is expressed in both ovarian and adrenal cells (Miller et al., 1988
; Fevold et al., 1989
). It catalyses the conversion of pregnenolone to DHEA and, at least in the rat, of progesterone to androstenedione trough two sequential steps: 17 hydroxylation and 17,20-lyase. Hence, an abnormal regulation of the cP450-17
activity, perhaps through the insulin/IGF system (Rosenfield et al., 1990
), may represent the common pathway leading to the hyperactive steroidogenesis in the adrenals and gonads of many PCOS women.
The acute in vivo effects of insulin on the activities of 17,20-lyase and 17-hydroxylase, calculated with the precursor/product ratios, was elegantly evaluated by Moghetti et al. Experimentally induced hyperinsulinaemia, within the high physiological range, was able to shift the ACTH-stimulated adrenal steroidogenesis toward the production of the 17
-hydroxycorticosteroid intermediates 17OHP and 17OH-pregnenolone. This effect was likely due to a stimulation of the cP450c17
leading to an increase of the 17
-hydroxylase and 17,20-lyase activity, with the former being enhanced markedly more than the latter (Moghetti et al., 1996
). In the present study, the 17
-hydroxylase activity was calculated only before ACTH stimulation, as progesterone was not assayed during the test. However, in line with the previous hypotheses, the 17OHP progesterone basal ratio was slightly, though not significantly, lower after pioglitazone treatment and such decrease was entirely due to a reduction of 17OHP plasma levels. Concerning the 17,20-lyase activity, the basal androstenedione 17OHP ratio remained the same before and after pioglitazone treatment; interestingly, when this was calculated from the ACTH-stimulated steroid levels, pioglitazone treatment led to an overall decrease in the response of both the steroids, but with a more consistent reduction of the 17-ketosteroid intermediate 17OHP, thus resulting in slight relative increase in the apparent 17,20-lyase activity. These data are in line with the above-mentioned previous studies and support the hypothesis that insulin might represent the trigger factor responsible not only for the abnormal hyperstimulation of the adrenal cP450c17 (Ehrmann et al., 1992
), but also for its dysregulation (relative impairment of 17,20-lyase activity) typical of PCOS.
Whether this mechanism might constitute an intrinsic characteristic of all the patients affected by the syndrome or whether it might explain only in part the pathophysiology of the PCOS-related hyperandrogenism, remains controversial.
Few studies exist on the impact of insulin-lowering drugs on the adrenal steroid biosynthesis in patients with PCOS. La Marca et al. (1999) reported a generalized significant reduction in the response of all adrenal steroids to corticotrophin as well as in the activity of 17
-hydroxylase and 17,20-lyase after a single month of metformin treatment in unselected PCOS subjects. In contrast, we failed to observe any modification in testosterone and DHEAS response to ACTH injection in our pioglitazone-treated patients. This discrepancy is not easily explained. In fact, the authors did not provide any data regarding the change in the metabolic assessment of the subjects studied, thus the correlation between the decrease in insulin levels and the reduction in the adrenal steroidogenesis remains unclear. It could be hypothesized that the different chemical structure and pharmacological properties of pioglitazone in respect of metformin could influence in a different and, perhaps, more selective manner the steroidogenic pathway. Nevertheless, Arlt et al. recently studied the effect of different classes of insulin-lowering drugs on the humanized yeast that express the P450c17
in microsomal environments: metformin seems to be able to inhibit androgen synthesis only indirectly, most likely by decreasing insulin circulating levels; by contrast thiazolidinediones also display a direct inhibitory effect on the enzymatic activities of cP450-17
and 3
-hydroxysteroid dehydrogenase type 2 (Arlt et al., 2001
). Among the members of this last drug family, troglitazone is the compound with the most relevant enzymatic inhibitory activity, whereas rosiglitazone and pioglitazone were reported to exert a direct but weaker interference on both cP450-17
and 3
-hydroxysteroid dehydrogenase type 2. A more recent report on the effect of metformin in obese PCOS adolescents with impaired glucose tolerance documented a selective reduction of 17OHP and androstenedione adrenal ACTH-induced synthesis after 3 months of treatment; in line with our findings, this effect was parallel to an amelioration of insulin sensitivity and secretion (Arslanian et al., 2002
).
The only report in the literature on the in vivo effect of thiazolidinediones on the adrenal steroidogenesis in PCOS patients is represented by a study from Azziz et al. (2003), who demonstrated a significant reduction in basal DHEAS levels after 20 weeks of treatment with troglitazone.
In conclusion, the present study is the first trial testing the effect of pioglitazone on the adrenal response to ACTH in obese adult PCOS patients with a complete assessment of the hormonal and metabolic changes that occurred during the treatment. Our findings point towards an overactivity and a dysregulation of the adrenal P450c, which seems to be attenuated by the pharmacological reduction of insulin levels; however, a direct inhibitory effect of pioglitazone on this enzyme cannot be excluded.
Beside the intrinsic complexity of the endocrinemetabolic nature of PCOS, most of the disagreement on the relationship between insulin and adrenal function probably originates from the lack of standardized inclusion criteria, study designs, duration and interventions. Further studies are needed to enhance our understanding in this field.
![]() |
Acknowledgements |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Arlt W, Auchus RJ and Miller WL (2001) Thiazolidinediones but not metformin directly inhibit the steroidogenic enzymes P450c17 and 3-hidroxysteroid dehydrogenase. J Biol Chem 276,1676716771.
Arslanian SA, Lewy V, Danadian K and Saad R (2002) Metformin therapy in obese adolescents with polycystic ovary syndrome and impaired glucose tolerance: amelioration of exaggerated adrenal response to adrenocorticotropin with reduction of insulinemia/insulin resistance. J Clin Endocrinol Metab 87,15551559.
Azziz R, Black V, Hines GA, Fox LM and Boots LR (1998) Adrenal androgen excess in the polycystic ovary syndrome: sensitivity and responsivity of the hypothalamicpituitaryadrenal axis. J Clin Endocrinol Metab 83,23172323.
Azziz R, Ehrmann DA, Legro RS, Fereshetian AG, OKeefe M and Ghazzi MN (2003) Troglitazone decreases adrenal androgen levels in women with polycystic ovary syndrome. Fertil Steril 79,932937.[CrossRef][ISI][Medline]
Benjamin F, Deutsch S, Saperstein H and Seltzer VL (1996) Prevalence of the markers for the attenuated form of congenital adrenal hyperplasia and hyperprolactinemia masquerading as polycystic ovarian disease. Fertil Steril 46,215221.
Bianchi P, Toscano V, Monaco G, Marano G, Lubrano C and Sciarpa F (1993) Effects of ACTH, insulin and insulin-like growth factor-I on guinea pig adrenals fasciculataglomerulosa cells [Abstract]. Endocrinol Invest 16(Suppl 1),88.
Cara JF and Rosenfield R (1988) Insulin-like growth factor 1 and insulin potentiate luteinizing hormone induced androgen synthesis by rat ovarian thecalinterstitial cells. Endocrinology 123,733739.[Abstract]
DeFronzo JP, Tobin JD and Andrei R (1979) Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 237,E214E223.[ISI][Medline]
Ehrmann DA, Rosenfield RL, Barnes RB, Brigell DF and Sheikh Z (1992) Detection of functional ovarian hyperandrogenism in women with androgen excess. N Engl J Med 327,157162.[Abstract]
Fevold HR, Lorence MC, McCarthy JL, Trant JM, Kagimoto M, Waterman MR and Mason JL (1989) P450 17 from testis: characterization of a full-length cDNA encoding a unique steroid hydroxilase capable of catalyzing both
4 and
5 steroid17, 20-lyase reactions. Mol Endocrinol 3,968975.[Abstract]
Fulghesu AM, Ciampelli M, Belosi C, Apa R, Pavone V and Lanzone A (2001) A new ultrasound criterion for the diagnosis of polycystic ovary syndrome: the ovarian stroma/total area ratio. Fertil Steril 76,326331[CrossRef][ISI][Medline]
Gillies PS and Dunn CJ (2000) Pioglitazone. Drugs 60,333343.[ISI][Medline]
Gonzalez F (1997) Adrenal involvement in polycystic ovary syndrome. Semin Reprod Endocrinol 15, 137157.[Medline]
Homburg R (2002) What is polycystic ovarian syndrome? A proposal for a consensus on the definition and diagnosis of polycystic ovarian syndrome. Hum Reprod 17,24952499.
Horrocks PM, Franks S, Hockley AD, Rolfe EB, Van Noorden S and London DR (1982) An ACTH secreting pituitary tumor arising in a patient with congenital adrenal hyperplasia. Clin Endocrinol (Oxf) 17,457468.[ISI][Medline]
Horrocks PM, Kandeel FR, London DR, Butt WR, Lynch SS and Holder G (1983) ACTH function in women with polycystic ovarian syndrome. Clin Endocrinol 19,143150.[ISI][Medline]
Kase N, Kowal J, Perloff W and Soffer LG (1963) In vitro production of androgens by a virilizing adrenal adenoma and associated polycystic ovaries. Acta Endocrinol (Copenh) 44,1519.[Medline]
LaMarca A, Morgante G, Paglia T, Ciotta L, Cianci A and De Leo V (1999) Effects of metformin on adrenal steroidogenesis in women with polycystic ovary syndrome. Fertil Steril 72,985989[CrossRef][ISI][Medline]
Lanzone A, Fulghesu AM, Guido M, Caruso A and Mancuso S (1992) Differential androgen response to adrenocorticotropic hormone stimulation in polycystic ovarian syndrome: relationship with insulin secretion. Fertil Steril 58,296301.[ISI][Medline]
Lanzone A, Fulghesu AM, Guido M, Ciampelli M, Caruso A and Mancuso S (1994) Differential androgen response to adrenocorticotrophin hormone stimulation and effect of opioid antagonist on insulin secretion in polycystic ovarian syndrome. Hum Reprod 9,22422246.[Abstract]
Lazar L, Kauli R, Bruchis C, Nordemberg J, Galatzer A and Pertzelan A (1995) Early polycystic ovary-like syndrome in girls with central precocious puberty and exaggerated adrenal response. Eur J Endocrinol 133,403406.[ISI][Medline]
Lehmann J, Moore L, Smith-Oliver T, Wilkinson W, Wilson T and Kliewer S (1995) An antidiabetic thiazolidindiones is a high affinity ligand for peroxisome proliferator-activated receptor (PPAR). J Biol Chem 270,1295312956.
Lucky AW, Rosenfield RL, McGuire S, Rudy S and Helke S (1986) Adrenal androgen hyperresponsiveness to adrenocorticotropin in women with acne and/or hirsutism: adrenal enzyme defects and exaggerated adrenarche. J Clin Endocrinol Metab 62,84048.[Abstract]
Meikle AW, Worley RJ and West CD (1984) Adrenal corticoid hyper-responsiveness in hirsute women. Fertil Steril 41,575579.[ISI][Medline]
Miller WL (1988) Molecular biology of steroid hormone synthesis. Endocr Rev 9,295318.[ISI][Medline]
Moghetti P, Castello R, Negri C, Tosi F, Spiazzi GG, Brun E, Calducci R, Toscano V and Muggeo M (1996) Insulin infusion amplifies 17-hydroxycorticosteroid intermediates response to adrenocorticotropin in hyperandrogenic women: apparent relative impairment of 17, 20-lyase activity. J Clin Endocrinol Metab 81,881886.[Abstract]
Moran C and Azziz R (2001) The role of the adrenal cortex in polycystic ovary syndrome. Obstet Gynecol Clin North Am 28,6375.[ISI][Medline]
National Diabetes Data Group (1979) Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 28,10391057.[ISI][Medline]
Nestler JE, Powers LP, Matt DW, Steingold KA, Plymate SR, Rittmaster RS, Clore JN and Blackard WG (1991) A direct effect of hyperinsulinemia on serum sex-hormone-binding globulin levels in obese women with the polycystic ovary syndrome. J Clin Endocrinol Metab 72,8389.[Abstract]
New MI, Lorentzen F, Lerner AJ, Kohn B, Oberfeld SE, Pollack MS, Du Pont B, Stoner L, Levy DJ, Pang S and Levin LS (1983) Genotyping steroid 21-hydroxylase deficiency: hormonal reference data. J Clin Endocrinol Metab 57,320326.[Abstract]
Romualdi D, Guido M, Ciampelli M, Giuliani M, Leoni F, Perri C and Lanzone A (2003) Selective effects of pioglitazone on insulin and androgen abnormalities in normo and hyperinsulinemic obese patients with polycystic ovary syndrome. Hum Reprod 18,12101219.
Rosenfield RL, Barnes RB, Cara JF and Lucky AW (1990) Dysregulation of cytochrome P450c17 as the cause of polycystic ovary syndrome. Fertil Steril 53,785791.[ISI][Medline]
Roy S, Mahesh VB and Greenblatt RB (1962) Effect of dehydroepiandrosterone and androstenedione on the reproductive organs of rats: production of cystic changes in the ovary. Nature 68,4243.
Submitted on September 22, 2003; accepted on November 28, 2003.