Effects of pioglitazone on menstrual frequency, hyperandrogenism and insulin resistance in adoloscents and young adults with polycystic ovary syndrome

J Pediatr Adolesc Gynecol (2008) 22:91e95 Effects of Pioglitazone on Menstrual Frequency,Hyperandrogenism and Insulin Resistance in Adoloscents andYoung Adults with Polycystic Ovary Syndrome Laddiperla Narsing Rao, MD, Jubbin Jagan Jacob, MD, Thomas V. Paul, MD, Simon Rajarathinam, MD,Nihal Thomas, MD, and Mandalam S. Seshadri, MDDepartment of Endocrinology, Christian Medical College and Hospital, Vellore, Tamil Nadu, India Study Objective: To study the clinical, meta- Key Words. Polycystic ovary syndrome—Adoles bolic and adverse effects of pioglitazone over a period of cents—Young adults—Pioglitazone—Fasting glucose 6 months in obese adolescent and young adults with poly- Design: This was an open labeled study. Each patient Setting: Outpatient department of a university affiliated Participants: Unmarried women (age 15e25 yrs) with Polycystic ovary syndrome (PCOS) is one of the most chronic anovulatory cycles and obesity, and with clinical common endocrine problem affecting adolescent girls. Adolescents present mainly symptoms related Interventions: Pioglitazone at a dose of 30 mg once daily to chronic anovulation (amenorrhea and oligomenor- for a period of 6 months along with dietary advice and rhea) and hyperandrogenism. In addition majority of Main Outcome Measures: Resumption of normal men- the girls are obese, insulin resistant, and have hyper- strual cycles, clinical improvement in hyperandrogenism insulinism. This combination predisposes these ado- and changes in insulin resistance measured by fasting glu- lescents to an increased risk of diabetes mellitus, cardiovascular disease, and infertility later in life.
Results: Twenty-two women were enrolled. At the end The pathogenesis of this syndrome is still unclear.
of the study period 91% of the subjects had regularization Over 60% of patients with PCOS are insulin resistant of menstrual cycles. There was no change in the modified and obese. Furthermore 20e40% of non obese PCOS Ferriman-Gallwey hirsutism scores. Decline in fasting insu- patients have evidence of insulin resistance.Chronic lin levels at the end of the study was 45.6% from baseline exposures to high insulin levels leads to development along with significant increase in the fasting glucose/insulin of acanthosis, increased body fat and finally glucose intolerance. Current data suggest that lowering insulin Conclusion: Administration of pioglitazone for 6 months along with advice about diet and physical activity in obese resistance with the use of sensitizers can ameliorate adolescents and young adult women with polycystic ovary menstrual abnormalities, improve ovulation rates, syndrome results in significant improvements in menstrual lower circulating androgen levels, and improve meta- frequency. There is a significant improvement in insulin resistance using the G/I ratio (!7.5 mg/10À4 U) as the nificant improvements in free testosterone levels,spontaneous ovulation rates, and improvement in met-abolic parameters were noted in various Thiazolidinediones as a new class of insulin sensi- Source(s) of support: The study was funded by a FLUID Research tizers were introduced in 1998 and initial studies with Grant of the Christian Medical College Vellore.
Troglitazone, one of the thiazolidinedione class, sug- Address correspondence to: Jubbin Jagan Jacob, Endocrine and Diabetes Unit, Department of Medicine, Christian Medical Col-lege, Ludhiana, Punjab, India 141 008; E-mail: Pioglitazone is a newer thiazolidinedione devel- oped for the treatment of type 2 diabetes. The drug Ó 2009 North American Society for Pediatric and Adolescent Gynecology Rao et al: Pioglitazone in Adolescents with PCOS is well tolerated in patients with diabetes in clinical All patients were given dietary advice, encouraged studies. Common adverse events include edema, to increase physical activity and asked to maintain a menstrual calendar. All patients were started on pio- al used pioglitazone for the first time in adult women glitazone at a dose of 30 mg once daily per oral after with PCOS. The 6-month study showed improve- breakfast. The medication was continued for a period ments in menstrual frequency, hirsutism, and insulin of 6 months. Patients were reviewed after 3 and 6 months. At each visit compliance of treatment was This present study was designed to study the clin- checked with a pill count and subjective evaluation ical (menstrual frequency, weight/body mass index, and hyperandrogenism), metabolic (fasting glucose At 6-month repeat visit, laboratory assessment of and insulin resistance) and adverse effects of pioglita- fasting insulin, glucose and total testosterone was zone over a period of 6 months in obese adolescent done. Menstrual frequency over a 6-month period and young adult patients with PCOS. To the best of was noted from the menstrual calendar. Modified Fer- our knowledge, no previous studies have looked at riman-Gallwey hirsutism score, grade of acanthosis the use of pioglitazone in this important subgroup of nigricans, and anthropometry was rechecked at the Hormonal AssaysFasting serum insulin was measured by radioimmuno- assay using the Coat-A-Count kits (DPC, Los An- This was an open labeled study. Each patient served geles, CA). The intraassay coefficient of variation as her own control and baseline characteristics were (CV) in our lab for serum insulin estimation is compared to those observed at the end of study.
9.3%, 5.1%, 3.5%, and 5% at mean concentration of The Institutional Ethics committee cleared the 17, 39, 80,117 and 278 mIU/ml. The interassay CV study protocol and an informed consent was obtained is 10%, 7.1%, 7.2% and 4.9% at mean concentration from each subject before entry into the study.
of 16, 35, 76, and 9 mIU/ml, respectively.
Serum total testosterone was measured on Immulite 2000 analyzer (DPC, Los Angeles, CA) by a competi- Unmarried young women in the age group of 15e25 tive chemiluminescence immunoassay. The normal years with clinical features suggestive of polycystic range for women in our laboratory was 0.5e1.2 ng/ ovary syndrome were screened on presentation at ml. The intraassay CV is 27%, 10.5%, 10% and the Endocrinology Out patient department of a univer- 9.5% at mean concentrations of 0.5, 1.0, 2.0, and 4.0 ng/ml. The interassay CV is 13.8% at a mean concen- Patients meeting the following inclusion criteria Fasting glucose was measured using glucose oxi- dase method on Hitachi 912 auto analyzer (Boeh- 1. Unmarried women (age 15e25), who were not ringer Mannheim). The intraassay CV at a mean planning to get married in the following 12 months concentration of 160 mg/dl was 1% and the interassay CV was 3.2% and 4.2% at mean concentrations of 99 2. Chronically anovulating defined as less than 6 menstrual cycles in the past 12 months.
3. Obese (defined as having a Body Mass Index Fasting glucose and insulin ratio (G/I ratio) were cal- 4. Clinical evidence of hyperandrogenism (defined as culated. A value of !4.5 mg/10À4 U was a strong a modified Ferriman-Gallweyhirsutism score predicator of Insulin resistance in adult women and a value !7.5 mg/10À4 U was a predictor of insulinresistance in adolescents.
Study ProtocolOn entry into the study, clinical details including de- tailed menstrual history including age of onset of The sample size was based on the previous study with menarche and menstrual pattern over the past year troglitazone where a 7.8 mIU/ml decrease in fasting were recorded. Fasting Insulin levels, fasting plasma Insulin and a decrease in 0.4 ng/dl of total testoster- glucose, and total serum testosterone values were one was The sample size required to detect measured. Hyperandrogenism was assessed by the a similar post treatment difference with 99% (P ! modified Ferriman-Gallwey (mFG)and acanthosis 0.01) confidence and 80% power was computed with nigricans was graded according to Burke et al.
the True Epistat statistical software. This worked out Rao et al: Pioglitazone in Adolescents with PCOS to 22 subjects. A Student t test was used to assess sta- persisted at the end of study. Decline in fasting insulin tistical significance. For the proportion of patients levels at the end of the study was 45.6% from base- with insulin resistance (IR) before and after interven- line. There was a significant increase in the fasting tion, the significance was assessed using the chi- glucose/insulin ratio (G/I ratio) from baseline. Using a value of G/I ratio ! 4.5 mg/10À4 U as significantlyassociated with IR in adolescents there were 10 (45.4%) subjects with IR at baseline and after treat-ment this declined to 6 (27.2%) subjects. This differ- Twenty-two women fulfilled the study criteria and ence was not statistically significant (P 5 0.23).
were enrolled in the study. All patients received 30 However, if a fasting G/I ratio of ! 7.5 mg/10À4 U mg of pioglitazone for 6 months and completed both was taken as a predictor of insulin resistance, 18 baseline and end of study assessments. The mean age (81.8%) subjects had IR at baseline. Only 11 (50%) of the study participants was 19.4 yrs (range 15 subjects continued to have IR at the end of the study yrs) and the mean age of menarche was 13 years period. This was statistically significant (P 5 0.02) Total serum testosterone was elevated (O 1.2 ng/ a mean BMI of 29.5 Æ 7.9 kg/m2 (mean Æ SD). All ml) in 32% of subjects. There was a minimal increase patients had severe menstrual irregularities. Average in the mean post treatment total testosterone values number of cycles in the last 6 months was 1.4 Æ which was not statistically significant.
0.5 (mean Æ SD) with mean duration of menstrual Pioglitazone was well tolerated by all subjects dur- symptoms of 44.6 Æ 30.2 months (mean Æ SD) (range ing the period of study. There were no drug-related chemical features before and after treatment with pio-glitazone for 6 months.
The major manifestations of PCOS in adolescence Though there was no significant change in weight or are chronic anovulation characterized by oligomenor- BMI from baseline (twelve subjects gained rhea and amenorrhea and by symptoms related to weight ranging from 1 to 6 kg during the course of hyperandrogenism. Most adolescent girls with PCOS this study. At the end of the study period 20 of 22 seek medical help for the menstrual irregularities, (91%) of the women had regularization of menstrual obesity, and/or hirsutism. Pioglitazone has been stud- cycles. There was, however, no change in the grade ied in adult women with PCOS and has shown of acanthosis and in the mFG hirsutism scores at the ovulatory rates and hyperandrogenism. The aim ofour study was to look at clinical and metabolic out- comes with the use of pioglitazone in adolescent girls Changes in the mean fasting plasma glucose (FPG) and young adults with PCOS over a period of 6 levels were significant at the end of study ().
None of the subjects experienced any hypoglycemic For the purpose of the study we recruited a subset symptoms on therapy with pioglitazone. Impaired of young women with PCOS who had clinically sig- fasting glucose (defined as FPG O110 mg %) was nificant manifestations of the disease. All patients seen in only one subject at initial evaluation and this had significant oligomenorrhea (less than 2 cycles in Table 1. Details of Important Clinical and Biochemical Variables Seen at Baseline and after 6 Months of Therapy with 30 mg PioglitazoneOnce Daily Number of menstrual cycles in previous six months Fasting Glucose/Insulin ratio (mg/10À4 U) Abbreviation: NS, not statistically significant*Statistically significant Rao et al: Pioglitazone in Adolescents with PCOS the preceding 6 months), were obese (BMI O 25 kg/ PCOS.The mean G/I ratio in adolescents was m2), had clinical evidence of insulin resistance (pres- 7.52 Æ 2.54 (mean Æ SD) compared to 4.28 Æ 0.44 ence of acanthosis nigricans), and had evidence of hy- (mean Æ SD) in adult women with PCOS. This study perandrogenism (mFG score of $ 7). In this subgroup suggests that there is worsening insulin resistance of patients with PCOS the use of pioglitazone 30 mg with advancing age and a different cut-off should be daily for a period of 6 months in addition to dietary used in adolescent girls (G/I ratio ! 7.5 mg/10À4 and exercise advice lead to significant improvements U). Applying these criteria 82% (18/22) of subjects in menstrual frequency, fasting insulin levels, and in- in our study had objective evidence of insulin resis- sulin resistance calculated by the fasting G/I ratio.
tance at the beginning of the study. After 6 months There was a tendency for weight gain that was not sta- of therapy with pioglitazone, the number of subjects tistically significant, with no clinical improvement in with objective insulin resistance declined to 50% hyperandrogenism. There was no decrease in total tes- (11/22). This difference in proportion was statistically Our clinical findings are in agreement with previ- Though some authors have questioned the relation- ous studies carried out in adult women with ship of insulin resistance and hyperinsulinism with Ninety-one percent of the study subjects had regulari- oligomenorrhea, in our study we have shown signifi- zation of menstrual cycles at the end of 6 months of cant association between fasting insulin levels and pioglitazone therapy. This was much higher than has fasting G/I ratio with menstrual frequency. This is par- been obtained in studies with adult women, probably ticularly true when an objective G/I ratio of ! 7.5 because all patients had severe menstrual irregulari- mg/10À4 U was used to document insulin resistance.
ties at baseline when they entered the study. Romualdi It seems reasonable to use this simple test for the di- et documented regularization of menstrual cycles agnosis of insulin resistance with a cut off of ! 7.5 in 83% of adult women (18e36 yrs) treated with pio- mg/10À4 U in adolescents to document insulin resis- glitazone for 6 months and Brettenthaler et al tance and initiate therapy with sensitizers.
showed cycle regularization in 43% of adult women For the purpose of assessing response of treatment at the end of 3 months of therapy with pioglitazone.
with pioglitazone on hyperandrogenism, we used the Both these studies have shown some increase in mFG hirsutism score to clinically grade the degree weight but as in our study, the results were not statis- of hyperandrogenism and total testosterone levels.
There was no significant change in the clinical scores Chronic insulin resistance and hyperinsulinemia are or in the total testosterone levels. This may be a reflec- implicated in the development of reproductive abnor- tion of the short duration of the trial and high interin- malities and hyperandrogenism in patients with PCOS.
dividual variability in total testosterone and sex Insulin resistance is also documented as a risk factor hormone binding globulin (SHBG) levels. The lack for clinical pathologies associated with PCOS, such as of availability of reliable free testosterone and SHBG gestational and type 2 diabetes mellitus, hypertension assays in our present setting limited us from docu- and cardiovascular In addition the elevated menting any change in the free hormone levels.
insulin levels may increase adrenocorticotrophic hor- Romualdi et al were, however, able to document sig- mone-stimulated steroidogenesis from the adrenal nificant changes in FG scores at the end of 6 months glands, while impairing 17,20-lyase activity of pioglitazone therapy in adult women which we In this study we used a simple fasting G/I ratio to objectively document insulin resistance in study sub- tenthaler and colleagues did not observe any signifi- jects and studied the effect of therapy with pioglita- cant change in FG scores at the end of 3 months of zone for 6 months. Dunaif et al had suggested that therapy with pioglitazone in adult women.
a fasting G/I ratio of less than 4.5 is a highly sensitive There were no significant adverse events which lead and specific test for the diagnosis of insulin resistance to the discontinuation of the drug in any patient during in adult women with PCOS.All subjects in our study period. Weight gain, though not statistically sig- study had obesity and acanthosis nigricans. However, nificant, was observed in more than half of the patients using the above criteria (G/I ratio ! 4.5 in mg/10À4 enrolled in the study. Subjects gained up to 6 kilograms U), only 46% (10/22) of our study group had objective with 23% (5/22) of patients gaining more than 3 kg of insulin resistance. There was also no significant weight during the course of the study. In a study from change in the percentage of patients with insulin resis- Mexico that randomized obese adult women with tance using the above criteria after 6 months of piogli- PCOS to either pioglitazone or metformin therapy tazone therapy. Macut et al first compared adolescent for over 6 months, an average of 4.7 kg weight gain PCOS with adult PCOS and they found that fasting was seen in the pioglitazone In contrast the plasma glucose is significantly lower and fasting G/I weight gain appears to be to a lesser extent in adoles- ratio is significantly higher in adolescent girls with cents with PCOS treated with pioglitazone.
Rao et al: Pioglitazone in Adolescents with PCOS Recent reports about a possible increase in cardio- 6. Unluhizarci K, Kelestimur F, Bayram F, et al: The effects vascular morbidity and mortality associated with the of metformin on insulin resistance and ovarian steroido- use of rosiglitazone in type 2 diabetes mellitus have genesis in women with polycystic ovary syndrome. Clin queered the pitch for all drugs in this Larger studies focusing on cardiovascular outcomes are un- 7. Dunaif A, Scott D, Finegood D, et al: The insulin-sensitiz- ing agent troglitazone improves metabolic and reproduc- derway. Unlike rosiglitazone, pioglitazone has been tive abnormalities in the polycystic ovary syndrome.
studied in a large prospective, randomized trial of cardiovascular outcomes called PROACTIVE. The 8. Ehrmann DA, Schneider DJ, Sobel BE, et al: Troglitazone primary end point, which was a broad composite that improves defects in insulin action, insulin secretion, ovar- ian steroidogenesis, and fibrinolysis in women with poly- events, showed a trend towards benefit with pioglita- cystic ovary syndrome. J Clin Endocrinol Metab 1997; zone. A secondary endpoint of myocardial infarction, stroke and death from any cause showed a significant 9. Hasegawa I, Murakawa H, Suzuki M, et al: Effect of tro- glitazone on endocrine and ovulatory performance in In summary our study shows that administration of women with insulin resistance-related polycystic ovarysyndrome. Fertil Steril 1999; 71:323 pioglitazone for 6 months along with advice about 10. Gillies PS, Dunn CJ: Pioglitazone. Drugs 2000; 60:333 diet and modest increase in physical activity in obese 11. Romualdi D, Guido M, Ciampelli M, et al: Selective ef- adolescents and young women with PCOS resulted in fects of pioglitazone on insulin and androgen abnormali- significant improvements in menstrual frequency and ties in normo- and hyperinsulinaemic obese patients with in over 90% of subjects it established cyclical men- polycystic ovary syndrome. Hum Reprod 2003; 18:1210 struation. There is a significant improvement in insu- 12. Hatch R, Rosenfield RL, Kim MH, et al: Hirsutism: impli- lin resistance using the G/I ratio (!7.5 mg/10À4 U) as cations, etiology, and management. Am J Obstet Gynecol the biochemical marker. Weight gain though present appears to be much lesser than what was seen in adult 13. Burke JP, Hale DE, Hazuda HP, et al: A quantitative scale women with PCOS. No significant adverse outcomes of acanthosis nigricans. Diabetes Care 1999; 22:1655 were noted in 6 months of therapy. No changes were 14. Brettenthaler N, De Geyter C, Huber PR, et al: Effect of noticed clinically in hyperandrogenism and no change the insulin sensitizer pioglitazone on insulin resistance, hy-perandrogenism, and ovulatory dysfunction in women with was noticed in the total testosterone values at the end polycystic ovary syndrome. J Clin Endocrinol Metab 2004; of 6 months. Long term studies are required to assess whether these short term improvements in insulin 15. Rajkhowa M, Glass MR, Rutherford AJ, et al: Polycystic resistance lead to reduction in new onset type 2 diabe- ovary syndrome: a risk factor for cardiovascular disease? tes and cardiac events on the long term.
16. Guido M, Romualdi D, Suriano R, et al: Effect of piogli- tazone treatment on the adrenal androgen response to cor- ticotrophin in obese patients with polycystic ovarysyndrome. Hum Reprod 2004; 19:534 1. Norman RJ, Dewailly D, Legro RS, et al: Polycystic ovary 17. Legro RS, Finegood D, Dunaif A: A fasting glucose to in- sulin ratio is a useful measure of insulin sensitivity in 2. Ciampelli M, Lanzone A: Insulin and polycystic ovary women with polycystic ovary syndrome. J Clin Endocrinol syndrome: a new look to an old subject. Gynecol Endocri- 18. Ortega-Gonzalez C, Luna S, Hernandez L, et al: Re- 3. Velazquez EM, Mendoza S, Hamer T, et al: Metformin sponses of serum androgen and insulin resistance to therapy in polycystic ovary syndrome reduces hyperinsuli- metformin and pioglitazone in obese, insulin-resistant nemia, insulin resistance, hyperandrogenemia, and systolic women with polycystic ovary syndrome. J Clin Endocrinol blood pressure, while facilitating normal menses and preg- 19. Nissen SE, Wolski K: Effect of rosiglitazone on the risk of 4. Nestler JE, Jakubowicz DJ: Decreases in ovarian cyto- myocardial infarction and death from cardiovascular chrome P450c17 alpha activity and serum free testosterone after reduction of insulin secretion in polycystic ovary syn- 20. Dormandy JA, Charbonnel B, Eckland DJ, et al: PROac- tive investigators. Secondary prevention of macrovascular 5. Nestler JE, Jakubowicz DJ, Evans WS, et al: Effects of events in patients with type 2 diabetes in the PROactive metformin on spontaneous and clomiphene-induced ovula- Study (PROspective pioglitAzone Clinical Trial In macro- tion in the polycystic ovary syndrome. N Engl J Med 1998; Vascular Events): a randomised controlled trial. Lancet

Source: http://www.cmcendovellore.org/publicationpdf/effects%20of%20pioglitazone.pdf