Cholesterol Center, Jewish Hospital, ABC Building, 3200 Burnet Avenue, Cincinnati, OH 45229, USA
1 To whom correspondence should be addressed. e-mail: glueckch{at}healthall.com
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: insulin resistance/metformin/pioglitazone/polycystic ovary syndrome
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Metformin and other insulin-sensitizing drugs (thiazolidinediones), by reducing insulin resistance and hyperinsulinaemia, reduce insulin-driven ovarian and adrenal hyperandrogenism, usually restoring normal LH and FSH secretion, facilitating normal ovulatory cycles and pregnancy (Glueck et al., 2002a;b;c; Nestler, 2002
; Nestler et al., 2002
; Ghazeeri et al., 2003
). There have been 38 open label and/or placebo-controlled studies which have documented metformins effectiveness in reducing hyperinsulinaemia and improving the endocrinopathy of PCOS with resultant normalization of menstrual cyclicity and restoration of ovulation in 2390% of both adults and adolescents (Glueck et al., 2002a
;b; Nestler, 2002
). In the largest randomized controlled trial of metformin in PCOS to date, ovulation frequency was 23% in 45 metformin-treated cases versus 13% in 47 placebo-treated controls (P < 0.01), and time to first ovulation was shorter (23.6 days versus 41.8 days, P < 0.05) (Fleming et al., 2002
). Benefits of metformin were not observed in morbidly obese subjects with body mass index (BMI) >37 kg/m2 (Fleming et al., 2002
).
In PCOS, insulin-sensitizing drugs as a class increase spontaneous ovulation, enhance induction of ovulation with clomiphene, and increase clinical pregnancy rates (Nestler, 2002; Nestler et al., 2002
; Ghazeeri et al., 2003
). Although the peroxisome proliferator activator receptor (PPAR)-gamma binding agent, troglitazone, has been withdrawn because of hepatotoxicity, it was effective in women with PCOS in increasing spontaneous ovulation, and increasing clomiphene induction of ovulation (Azziz et al., 2001
). In a 2 month, double-blind, placebo-controlled trial, rosiglitazone (8 mg/day another PPAR-gamma binding agent) enhanced both spontaneous and clomiphene-induced ovulation by improving insulin sensitivity and reducing hyperandrogenaemia in overweight and obese women with PCOS, whose mean BMI were 35.538.5 kg/m2 (Ghazeeri et al., 2003
). To date, no data have been published on the use of a third PPAR-gamma binding agent, pioglitazone, in women with PCOS.
In an observational study of 13 women with PCOS not optimally responsive to metformin and diet, our specific aim was to assess efficacy and safety of pioglitazone when added to antecedent metformin diet, and to compare these 13 women with 26 women with PCOS, responsive to metformin and diet, matched by age and by pre-treatment menstrual history.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Study protocol
This prospective, open label, single centre, consecutive case series study included 39 women from the Midwestern United States who were referred for a 2 year study of the efficacy and safety of metformin in PCOS (Glueck et al., 1999a;b;c). They came from a larger cohort of 743 women, referred for the diagnosis and therapy of PCOS from 6/12/97 to 1/23/02. None of the 39 women (13 non-optimal responders to metformin, 26 responders) had previous laparoscopic diathermy. The percentage of women having previously received clomiphene ovulation stimulation did not differ between non-responders (2/13, 15%) and the matched responders (7/26, 27%), Fishers P = 0.7 (non-significant).
At pre-metformin baseline, and after an overnight fast, blood was obtained for measurements of serum insulin and glucose, lipids and lipoprotein cholesterols, sex hormones, and sex hormone-binding globulin (SHBG) using previously reported methods (Glueck et al., 2002b). The homeostatic model assessment (HOMA) for insulin resistance (IR) and beta cell function was used as per Haffner et al. (1996
). A detailed history of menstrual status from menarche to the time of study entry was obtained, with special focus on menstrual frequency in the year before study entry (Table I and Table II). Weight and seated systolic and diastolic blood pressure were recorded.
|
|
When using pioglitazone in patients with PCOS, following United States FDA guidelines (Murray and Kelly, 2002), we obtained pre-treatment liver function tests, proceeding with caution if the baseline alanine aminotransferase (ALT) levels were mildly elevated, and not using the drug if baseline ALT was >2.5 times the upper limit of normal. We also rechecked liver function every 2 months during the first year of therapy and stopped pioglitazone if ALT levels increased to >3 times the upper normal limit or if clinical hepatitis developed (Murray and Kelly, 2002
).
At study entry, all women with BMI 25, categorized as overweight by Flegal et al. (2002
), were instructed on a 1500 calorie diet with 26% of the calories as protein, 44% as carbohydrates (42% complex), and 30% of calories as fat (polyunsaturate:saturate ratio 2:1). For women with normal BMI <25 (Flegal et al., 2002
), a 2000 calorie diet was given. For all women, diet was maintained throughout follow-up, with re-instruction every 46 months.
The five women who wished to conceive (nos. 3, 4, 6, 8, 11; Table I) were given folic acid, 1 mg/day, and were instructed to stop pioglitazone as soon as pregnancy was verified (subject no. 6, Table I). hCG was quantified during the pioglitazone + metformin treatment period if menses failed to appear within 6 weeks of the previous menses.
We knew, based on previous studies (Glueck et al., 1999b, 2001a;b; 2002;c; Fleming et al., 2002
), that up to 23% of women with PCOS receiving diet plus metformin (2.55 g/day) would be not optimally responsive to treatment, defined by failure to significantly reduce weight, androstenedione, testosterone, DHEAS, IR, insulin secretion, to increase SHBG and HDL cholesterol, and/or by failure to resume regular menstrual cycles. In 13 women who did not achieve these positive outcomes, after a median of 12 months on metformin plus diet, while maintaining the same diet and metformin 2.55 g/day, pioglitazone (45 mg/day) was added. Provision of concurrent pioglitazone and metformin was based on the hypothesis that addition of a second insulin-sensitizing drug (Smith, 2001
) in women non-optimally responsive to metformin diet might lead to more favourable endocrine outcomes. Eleven of the 13 non-optimally responsive women were evaluated every 2 months on pioglitazone + metformin diet for a median of 10 months, and their response compared with metformin diet alone. Two of the 13 women (Nos. 8 and 10), to date, have received pioglitazone + metformin <4.5 months and their responses to combined medication are not included.
In an attempt to characterize any unique features of the 13 women with PCOS who were not optimally responsive to metformin diet, we matched two control women with PCOS by age and by pre-treatment menstrual history to each non-responder (Table II and Table III). These 26 control women with PCOS were responsive to metformin diet over the same 12 month treatment period and were recruited into the study at the same time as the 13 non-responders.
|
|
Although there was a broad range of age among the 13 non-responders (two teenagers aged 14 and 19 years, six women aged 2030, four aged 3139, and one aged 39.6 years), each non-responder was age-matched with two responders (26 total responders), to reduce the likelihood that age or age range could distort the comparison of response to metformin in the two groups. Similarly, although there was a broad range of pre-treatment BMI in the 13 non-responders (one <25 kg/m2, two from 25 to <30, 3 from 30 to <40, and 7 40), neither BMI nor its categorical distribution differed (P = 0.7) from the 26 agemenstrual-matched controls, reducing the likelihood that obesity or the range of obesity could distort the comparison of metformin response. In the paired comparison of metabolic and menstrual response to metformin versus metformin + pioglitazone in 13 women, with each subject as her own control, age and weight should not have affected differences in response to the drugs.
Statistical analysis
Pre-treatment, baseline variables in non-responders versus responders were compared by Wilcoxon tests (SAS/STAT, 2002). We also used stepwise logistic regression where group [metformin responders (n = 26) and non-responders (n = 13)] was the dependent variable, and explanatory variables (at pre-treatment baseline) were BMI, insulin, glucose, IR, beta cell secretion, SHBG, androstenedione, DHEAS, and total testosterone. Paired Wilcoxon tests (SAS/STAT, 2002
) were used to compare pre-treatment baseline levels against the mean levels of each patient during therapy on metformin diet.
To further assess the contribution of weight to metformin response, the 13 non-responders and 26 responders were pooled, and stepwise regression analyses were done separately with the following variables (separately) as dependent variables: change on metformin in systolic and diastolic blood pressure, androstenedione, DHEAS, SHBG and testosterone (total and free). Explanatory variables in each model were group (responders, non-responders), pre-treatment values for weight, insulin, and glucose, and change on metformin in weight, insulin, and glucose. A separate stepwise regression model was run with change on metformin in insulin as the dependent variable and explanatory variables including group (responders, non-responders), pre-treatment values for weight, insulin, and glucose, as well as change in glucose and weight on metformin.
To judge the success of metformin + pioglitazone, mean values of each patient on metformin were compared with mean values on metformin + pioglitazone using paired Wilcoxon tests, (Table IV).
Menstrual status pre-treatment and the observed/expected percentage of menses on treatment with metformin (13 non-responders versus 26 responders) was compared by Wilcoxon tests, and for metformin versus metformin + pioglitazone, by paired Wilcoxon tests (SAS/STAT, 2002). We tested by trend analysis whether any increase in menstrual frequency and/or regularity was achieved early after starting metformin, and after adding pioglitazone, and persisted thereafter (Figure 1). Our hypothesis was that demonstration of an early increase in menses on metformin, with stability of menstrual status in the final 36 months of metformin, followed by a change when pioglitazone was started, would make a stronger case for the increase during pioglitazone treatment being the result from pioglitazone rather than a carryover or continuation of a progressive increase in menses from metformin.
|
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
At pre-treatment baseline, the 13 non-responders did not differ from the 26 responders by weight, by BMI category, systolic or diastolic blood pressure, sex hormones, insulin, glucose, IR, or insulin secretion (Wilcoxon P > 0.07 for all). By stepwise logistic regression, at pre-treatment baseline, the 13 non-responders did not differ (P > 0.15) from the 26 responders for BMI, insulin, glucose, IR, beta cell secretion, DHEAS, or total testosterone, but the 13 non-responders had higher SHBG (P = 0.016) and androstenedione (P = 0.04) than the 26 responders.
The 13 non-responders and the 26 responders did not differ from the remaining 704 women with PCOS also referred to our centre from December 6, 1997 to January 23, 2002 by BMI category (Flegal et al., 2002) (normal <25, overweight 2530, obese 3040, severely obese
40) (
2 = 8.36, 6 df, P = 0.21). The 13 non-responders and 26 responders differed from the remaining 704 women with PCOS by having no women age
40 years, versus 15%,
2 = 17.5, 6 df, P = 0.008. Thus, although there was a broad range of BMI and age in the 13 non-responders and 26 responders, BMI was representative of our PCOS referral study cohort, and age range was tighter.
Outcomes on metformin plus diet
In the non-optimally responsive cohort of 13 women, on metformin plus diet for a median of 12 months, median insulin fell from 21 to 16 µIU/ml (P = 0.048), and insulin secretion fell from 251 to 200 (P < 0.01) (Table III).
In the responsive cohort of 26 women, on metformin plus diet for a median of 13 months, weight fell, median DHEAS unexpectedly rose (P < 0.01) (within the normal range), testosterone fell, insulin fell, IR fell, and insulin secretion fell (all P 0.017, Table III), and diastolic blood pressure fell (P = 0.02). We do not know why DHEAS rose within the normal range in the 26 responders on metformin (Table III).
By stepwise regression with all 39 women pooled, group status was a significant co-variate for change (on metformin) in SHBG which fell more in non-responders (R2 = 11%, P = 0.042), and for change in free testosterone which rose more in non-responders (R2 = 20%, P = 0.007). Pre-treatment weight and weight change on metformin were significant co-variates for change in insulin on metformin. Insulin fell more on metformin when pre-treatment insulin was high (partial R2 = 57%, P < 0.0001), when pre-treatment weight was lower (partial R2 = 9.1%, P = 0.004), and when weight fell more on metformin (partial R2 = 7.3%, P = 0.005).
After 3 months on metformin, 46% of expected menses had occurred in the 13 non-responders, better than 14% at pre-treatment baseline, P = 0.05 (Figure 1). At months 6, 9 and 12, 38% (P = 0.07), 27% (P = 0.9) and 24% (P = 0.5) of menses occurred, versus 14% at pre-treatment baseline (Figure 1). Trend analysis did not reveal any significant trend in menses on treatment (P = 0.9).
By selection, the pre-treatment menstrual history in the 26 PCOS responders did not differ from the 13 non-responders (13 versus 14%, P = 0.9, Figure 1). During the first 3 months on metformin, 78% of expected menses occurred in the responders, versus 46% in the non-responders (P = 0.022, Figure 1). At months 6, 9 and 12, 83%, 91% and 94% of menses occurred in the responders, versus 38% (P = 0.0006), 27% (P = 0.0005) and 24% (P < 0.0001) in the non-responders (Figure 1). Over the 12 months, the occurrence on expected menses in the 26 responders was 2.5-fold higher than in the 13 non-responders (P < 0.0001, Figure 1). In the responders, trend analysis revealed a significant trend (P < 0.0001) towards increased menstrual frequency over time on treatment (Figure 1).
Outcomes on metformin and pioglitazone
In 11 women, non-optimally responsive to metformin diet alone, on pioglitazone + metformin diet over 10 months, com pared with their antecedent 12 months on metformin diet, median DHEAS fell (211 to 171 µg/dl, P = 0.02), SHBG rose from 31 to 43 nmol/l (P = 0.006), insulin fell (16.2 to 10.2 µIU/ml, P = 0.001), IR fell from 3.37 to 1.73 (P = 0.002), beta cell function fell from 217 to 124 (P = 0.004), glucose fell from 89 to 86 mg/dl (P < 0.05), and HDL cholesterol rose from 38 to 42 mg/dl (P = 0.003) (Table IV). Weight did not change (Table IV).
On pioglitazone + metformin diet, the percentage of menses over 9 months increased 2-fold when compared with that on metformin diet alone (Figure 1), P < 0.0001. During the first 3 months on pioglitazone + metformin, 67% of expected menses occurred versus 46% on metformin alone (P = 0.2), 77 versus 38% on metformin alone during the second 3 months (P = 0.017), and 73 versus 27% on metformin alone during the third 3 months (P = 0.032) (Figure 1). Trend was not significant (P = 0.61).
None of the women developed abnormal liver function tests while taking metformin + pioglitazone. There was no lactic acidosis, and no hypoglycaemia. In the one subject who conceived on pioglitazone + metformin, pioglitazone was immediately stopped and metformin was continued throughout (no. 6, Table I). Her pregnancy was uneventful, without gestational diabetes, and with delivery of a normal 5 lb 12 oz male infant at gestation week 36.
Power and sample size estimates
To determine whether outcomes on pioglitazone + metformin diet differed from those on metformin diet at P = 0.05, the current studys sample size had a power = 90% based on menstrual frequency in the second 3 month treatment period, 80% if based on the third 3 months, and 70% based on changes in insulin.
To perform a successful randomized controlled clinical trial at a significance level of 5% with power of 80%, comparing pioglitazone + metformin versus placebo + metformin, based on the current studys changes in insulin, there would need to be 14 in each group. Based on IR there would need to be 14 in each group, and based on menses in the first 3, 6 and 9 months on therapy, 40, 9 and 10 respectively, in each group.
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In the current report, the major characteristics which differentiated the 26 responders from the 13 non-responders were greater weight loss on metformin diet, and more marked resumption of regular menses, with progressive improvement over time. Neither pre-treatment BMI nor obesity category by BMI differed between non-responders and responders. With all 39 women pooled, by stepwise regression, on metformin insulin fell more when pre-treatment insulin was high, when pre-treatment weight was less, and when weight fell more on metformin. Since change in insulin on metformin is probably central to its beneficial metabolic and endocrine effects, this finding is consistent with the report of Fleming et al. (2002) that metabolic benefits of metformin were not observed in morbidly obese women with PCOS whose BMI was >37.
Based on the current study, in treatment of PCOS with metformin, if menstrual cyclicity does not improve by 36 months, and if endocrinopathy is not ameliorated, then consideration should be given to adding a second insulin-sensitizing drug, pioglitazone. We postulated that pioglitazone, with a different locus of insulin sensitization than metformin (Ikeda and Sugiyama, 2001; Smith et al., 2001
; Pittas and Greenberg, 2002
), would further ameliorate hyperinsulinaemia and resolve endocrinopathy, when added to metformin diet in obese women with PCOS who had not responded optimally to metformin diet. In the current study, when pioglitazone was added, there were significant incremental benefits beyond metformin diet; DHEAS, glucose, insulin, IR and insulin secretion fell, and HDL cholesterol and SHBG rose (all P < 0.01). On pioglitazone + metformin diet, the occurrence of expected menses was 2-fold higher than on metformin diet (P < 0.0001).
As in the current study, the normal response to reduced IR is reduced beta cell insulin output (falling insulin), which is essentially what models measure in relation to fasting glucose (Buchanan et al., 2002). In women with prior gestational diabetes, a fall in insulin output (-beta cell function) induced by troglitazone was protective for beta cells and prevented development of type 2 diabetes (Buchanan et al., 2002
).
We speculate that the apparently safe (no hepatotoxicity, no hypoglycaemia), added benefit of pioglitazone in the current study rose from further lowering of IR (Ikeda and Sugiyama, 2001; Smith et al., 2001
; Pittas and Greenberg, 2002
), with consequent lowering of insulin and insulin secretion, and with subsequent improvement in hyperandrogenaemia manifested by reduction in DHEAS and increments in SHBG. One potential drawback of addition of pioglitazone to metformin diet is cessation of weight loss. Pioglitazone apparently promotes weight gain by inhibition of glycerol kinase, with reduction of circulating fatty acids and subsequent reduction in insulin and retention of the fatty acids as triglycerides in adipose tissue (Guan et al., 2002
).
Studies with rosiglitazone and pioglitazone suggest that hepatotoxicity is not a class effect of the glitazones, an issue that rose from the hepatotoxicity of troglitazone (Hanefeld and Belcher, 2001; Lebovitz et al., 2002
; Ghazeeri et al, 2003
). Overall, despite two case reports of apparent drug-related hepatotoxicity (Maeda, 2001
; May et al., 2002
), clinical experience with pioglitazone has not revealed any trend towards unacceptable hepatotoxicity (Hanefeld et al., 2001
; Scheen, 2001
; Lebovitz et al., 2002
; Scherbaum et al., 2002
). In the current report, there were no elevations of ALT >3 times the upper normal limit, and no clinical hepatitis developed.
Rosiglitazone and pioglitazone increase insulin sensitivity by acting on muscle and liver to increase glucose utilization and decrease glucose production (Nolan et al., 1994; Iwamoto et al., 1996
). Rosiglitazone and pioglitazone have both been classified in category C. In animal models, treatment during midlate gestation was associated with fetal death and growth retardation. There are no case reports or systematic clinical studies of use of thiazolidindeones during pregnancy. In the current study, the single woman who conceived on pioglitazone + metformin stopped pioglitazone at first confirmation of pregnancy and delivered a normal infant at 36 weeks gestation. Given their pregnancy category C, neither rosiglitazone (Cataldo et al., 2001
; Ghazeeri et al., 2003
) nor pioglitazone should be continued after conception or used in the treatment of gestational diabetes. This differs from metformin which is pregnancy category B, appears to be safe during pregnancy for mother and fetus, reduces first trimester miscarriage (Glueck et al., 2002c
; Jakubowicz, 2002
) and reduces gestational diabetes (Glueck et al., 2002c
).
The current report was an observational, non-blinded pilot study to evaluate whether addition of pioglitazone (45 mg/day) to metformin 2.55 g/day in obese women who failed to optimally respond to metformin would further improve the endocrinopathy of PCOS. A randomized, placebo-controlled, double-blind trial will be required to confirm the improved outcomes on pioglitazone + metformin versus metformin alone in women not optimally responsive to metformin.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Azziz, R., Ehrmann, E., Legro, R., Whitecomb, R., Hanley, R., Fereshetian, A. and Al, E. (2001) Troglitazone improves ovulation and hirsutism in the polycystic ovary syndrome: a multicenter, double-blind, placebo-controlled trial. J. Clin. Endocrinol. Metab., 86, 16261632.
Buchanan, T., Xiang, A., Peters, R., Kjos, S., Marroquin, A., Goico, J. and Al, E. (2002) Preservation of pancreatic beta cell function and prevention of type 2 diabetes by pharmacological treatment of insulin resistance in high-risk Hispanic women. Diabetes, 51, 27962803.
Cataldo, N., Abbasi, F., Mclaughlin, T., Lamendola, C. and Reaven, G. (2001) Improvement in insulin sensitivity followed by ovulation and pregnancy in a woman with polycystic ovary syndrome who was treated with rosiglitazone. Fertil. Steril., 76, 10571059.[CrossRef][ISI][Medline]
Ehrmann, D., Cavaghan, M., Imperial, J. and Al, E. (1997) Effects of metformin on insulin secretion, insulin action, and ovarian steroidogenesis in women with polycystic ovary syndrome. J. Clin. Endocrinol. Metab., 82, 524530.
Ferriman, D, and Gallwey, J. (1961). Clinical assessment of body hair growth in women. J. Clin. Endocrinol. Metab., 21, 13401347.
Flegal, K.M., Carroll, M.D. and Ogden, C.L. (2002) Prevalence and trends in obesity among US adults, 19992000. J. Am. Med. Assoc., 288, 17231727.
Fleming, R., Hopkinson, Z. and Wallace, R. (2002) Ovarian function and metabolic factors in women with oligomenorrhea treated with metformin in a randomized double blind, placebo-controlled trial. J. Clin. Endocrinol. Metab., 87, 569574.
Gambineri, A., Pelusi, C., Vicennati, V., Pagotto, U. and Pasquali, R. (2002) Obesity and the polycystic ovary syndrome. Int. J. Obes. Relat. Metab. Disord., 26, 883896.[CrossRef][Medline]
Ghazeeri, G., Kutteh, W.H., Bryer-Ash, M., Haas, D. and Ke, R.W. (2003) Effect of rosiglitazone on spontaneous and clomiphene citrate-induced ovulation in women with polycystic ovary syndrome. Fertil. Steril., 79, 562566.[CrossRef][ISI][Medline]
Glueck, C., Wang, P., Fontaine, R., Sieve-Smith, L., Tracy, T. and Moore, K. (1999a) Plasminogen activator inhibitor activity: an independent risk factor for the high miscarriage rate during pregnancy in women with polycystic ovary syndrome. Metabolism, 48, 15891595.[ISI][Medline]
Glueck, C., Wang, P., Fontaine, R., Tracy, T. and Sieve-Smith, L. (1999b) Metformin-induced resumption of normal menses in 39 of 43 (91%) previously amenorrheic women with the polycystic ovary syndrome. Metabolism, 48, 511519.[ISI][Medline]
Glueck, C., Awadalla, S., Phillips, H., Cameron, D., Wang, P. and Fontaine, R. (2000) Polycystic ovary syndrome, infertility, familial thrombophilia, familial hypofibrinolysis, recurrent loss of in vitro fertilized embryos, and miscarriage. Fertil. Steril., 48, 15891595.
Glueck, C., Phillips, H. and Cameron, D. (2001a) Continuing metformin throughout pregnancy in women with polycystic ovary syndrome appears to safely reduce first-trimester spontaneous abortion: a pilot study. Fertil. Steril., 75, 4652.[CrossRef][ISI][Medline]
Glueck, C., Wang, P., Fontaine, R., Tracy, T. and Sieve-Smith, L. (2001b) Metformin to restore normal menses in oligo-amenorrheic teenage girls with polycystic ovary syndrome (PCOS). J. Adolesc. Health, 29, 160169.[CrossRef][ISI][Medline]
GlueckC., Streicher, P. and Wang, P. (2002a) Treatment of polycystic ovary syndrome with insulin-lowering agents. Expert Opin. Pharmacother., 3, 11771189.[Medline]
Glueck, C., Wang, P., Goldenberg, N. and Sieve-Smith, L. (2002b) Pregnancy outcomes among women with polycystic ovarian syndrome treated with metformin. Hum. Reprod., 17, 28582864.
Glueck, C., Wang, P. and Koyabashi, S. (2002c) Metformin therapy throughout pregnancy reduces development of gestational diabetes in women with polycystic ovary syndrome. Fertil. Steril., 77, 520525.[CrossRef][ISI][Medline]
Guan, H., Li, Y., Jensen, M., Newbard, C., Steppan, C. and Lazar, M. (2002) A futile metabolic cycle in adipocytes by antidiabetic agents. Nat. Med., 8, 11221128.[CrossRef][ISI][Medline]
Guzick, D. (1998) Polycystic ovary syndrome: symptomatology, pathophysiology, and epidemiology. Am. J. Obstet. Gynecol., 179, S89S93.[ISI][Medline]
Haffner, S., Gonzalez, C., Miettinen, H. and Al, E. (1996) A prospective analysis of the HOMA model. Diabetes Care, 19, 11381131.[Abstract]
Hanefeld, M. and Belcher, G. (2001) Safety profile of pioglitazone. Int. J. Clin. Pract., 121 (Suppl.), 2731.
Homburg, R. (2002) Should patients with polycystic ovarian syndrome be treated with metformin? Hum. Reprod., 17, 853856.
Ikeda, H. and Sugiyama, Y. (2001) Pharmacological effects of thiazolidinedione derivative, pioglitazone. Nippon Rinsho, 59, 21912194.[Medline]
Iwamoto, Y., Kosaka, K., Kuzuya, T., Akanuma, Y., Shigeta, Y. and Kaneko, T. (1996) Effects of troglitazone: a new hypoglycemic agent in patients with NIDDM poorly controlled by diet therapy. Diabetes Care, 13, 365370.
Jakubowicz, D.J., Iuorno, M.J., Jakubowicz, S., Roberts, K.A. and Nestler, J.E. (2002) Effects of metformin on early pregnancy loss in the polycystic ovary syndrome. J. Clin. Endocrinol. Metab., 87, 524529.
Kawadzki, J. and Dunaif, A. (1992) Diagnostic criteria for polycystic ovary syndrome: a rational approach. In Dunaif, A., Givens, J.R., Haseltine, F. and Merriam, G.R. (eds), In Polycystic Ovary Syndrome. Blackwell, Cambridge, MA, pp. 377384.
Kiddy, D., Hamilton-Fairley, D., Bush, A., Short, F., Anyaoku, V., Reed, M. and Al, E. (1992) Improvement in endocrine and ovarian function during dietary treatment of obese women with polycystic ovary syndrome. Clin. Endocrinol. (Oxf.), 36, 105111.[ISI][Medline]
Knochenhauer, E., Key, T., Kashar-Miller, M., Waggoner, W. and Boots, L. (1998) Prevalence of the polycystic ovary syndrome in unselected black and white women of the southeastern United States: a prospective study. J. Clin. Endocrinol. Metab., 83, 30783082.
Lebovitz, H. (2002) Differentiating members of the thiazolidinedione class: a focus on safety. Diabetes Metab. Res., 18, S2329.[CrossRef][ISI]
Maeda, K. (2001) Hepatocellular injury in a patient receiving pioglitazone [Letter]. Ann. Intern. Med., 135136.
May, L., Lefkowich, J., Kram, M. and Rubin, D. (2002) Mixed hepatocellularcholestatic liver injury after pioglitazone therapy. Ann. Intern. Med., 136, 449452.
Moran, L.J., Noakes, M., Clifton, PM., Tomlinson, L., and Norman, R.J. (2003) Dietary composition in restoring reproductive and metabolic physiology in overweight women with polycystic ovary syndrome. J. Clin. Endocrinol. Metab., 88, 812819.
Murray, L. and Kelly, G.L. (eds) (2002) Physicians Desk Reference. Medical Economics, p. 3277.
Nestler, J. (2002) Should patients with polycystic ovarian syndrome be treated with metformin? An enthusiastic endorsement. Hum. Reprod., 17, 19501953.
Nestler, J., Stovall, D., Akhter, N., Iuorno, M. and Jakubowicz, D. (2002) Strategies for the use of insulin-sensitizing drugs to treat infertility in women with polycystic ovary syndrome. Fertil. Steril., 77, 209215.[CrossRef][ISI][Medline]
Nolan, J., Judvik, B., Beerdsen, P., Joyce, M. and Olefsky, J. (1994) Improvement in glucose tolerance and insulin resistance in obese subjects treated with troglitazone. N. Engl. J. Med., 331, 12261227.
Pittas, A. and Greenberg, A. (2002) Thiazolidinediones in the treatment of type 2 diabetes. Expert Opin. Pharmacother., 3, 529540.[Medline]
SAS/STAT (2002) In Changes and Enhancements Through Release 8.02. SAS Institute, SAS Campus Drive, Cary, NC, USA, p. 415.
Scheen, A. (2001) Hepatotoxicity with thiazolidinediones: is it a class effect? Drug Safety, 24, 873888.[Medline]
Scherbaum, W. and Goke, B. (2002) Metabolic efficacy and safety of once daily pioglitazone monotherapy in patients with type 2 diabetes: a double-blind, placebo-controlled study. Horm. Metab. Res., 34, 589595.[CrossRef][ISI][Medline]
Smith, U. (2001) Pioglitazone: mechanism of action Int. J. Clin. Pract., 121 (Suppl.), 1318.
Sokol, R. and Rohlf, J. (1995) Biometry. W.H.Freeeman, New York, pp. 524530.
Velazquez, E., Acosta, A. and Mendoza, S. (1997a) Menstrual cyclicity after metformin therapy in polycystic ovary syndrome. Obstet. Gynecol., 90, 392395.
Velazquez, E., Mendoza, S., Hamer, T., Sosa, F. and Glueck, C. (1994) Metformin therapy in polycystic ovary syndrome reduces hyperinsulinemia, insulin resistance, hyperandrogenemia, and systolic blood pressure, while facilitating normal menses and pregnancy. Metabolism, 43, 647654.[ISI][Medline]
Velazquez, E., Mendoza, S., Wang, P. and Glueck, C. (1997b) Metformin therapy is associated with a decrease in plasma plasminogen activator inhibitor-1, lipoprotein(a), and immunoreactive insulin levels in patients with the polycystic ovary syndrome. Metabolism, 46, 454457.[ISI][Medline]
Submitted on February 27, 2003; resubmitted on April 7, 2003; accepted on May 9, 2003.