Introduction:
PCOS is the most common cause of anovulatory infertility which affects 12-21% of the population. This percentage may vary according to the criteria used and population studied.Hyperandrogenism, chronic anovulation, and occasionally obesity are the characteristics of this syndrome. Currently, CC is the first-line infertility treatment in women with the PCOS .However, 20-25% of PCOS women fail to ovulate with incremental doses of CC and and do not respond due to resistance to CC .IR and hyperinsulinemia are considered as important factors in the pathogenesis of PCOS .Hence, lean and obese women with PCOS may have IR.Hyperinsulinism has been shown to play a part in ovarian hyperandrogenism and may directly influence and even prevent ovulation.Metformin, an insulin sensitizing agent, reduces levels of luteinizing hormone, hyperinsulinemia and also decreases ovarian production of androgens .For this reason a number of studies have been performed and evaluated the possible beneficial role of metformin in managing women with PCOS. However, the studies performed evaluate the possible benefit of metformin alone or in combination with CC as first-line treatment for infertility in women with PCOS have reported conflicting results .Recently a meta-analysis concluded that metformin was more burdensome than placebo, hence CC alone and not CC plus metformin should be the drug of choice in these patients.A high-quality Cochrane systematic review has reported on randomized controlled trials(RCTs) comparing metformin with CC in women with PCOS and demonstrated that CC had a significantly higher pregnancy and live birth rate in women with a BMI \>30 kg/m2 whereas metformin had better results in women with a BMI \<30 kg/m2.Lately, a meta-analysis reported that due to conflicting findings and heterogeneity across the included RCTs, there is insufficient evidence to establish a difference between metformin and CC in terms of ovulation, pregnancy, live birth, miscarriage and multiple pregnancy rates in nonobese women with PCOS. It was proposed that caution should be proceeded when prescribing metformin as first line pharmacological therapy in this group of women until further rigorous trials are done.
These conflicting results from the published studies may be due to the large differences and heterogeneity in study populations, particularly the possible presence or absence of IR. Metformin decreases IR, however there is not a single RCTwith metformin in IR PCOS patients. Therefore, the aim of current study is to compare CC with coadministration of metformin and with CC with coadministration of placebo in IR PCOS patients to induce ovulation.
Methods and materials:
A total of 388 IR women with PCOS will be enrolled in this study from infertility clinic at Amir-Al-Momenin Hospital, Semnan, Iran from november 2015 to april 2020 . All patients will be informed about the study and possible complications of the drugs by a consultant and informed consents will be signed. This study will be supervised and approved by the Research Council and Ethical Committee of Semnan University of Medical Sciences. For randomization random block allocation method was used, in which six blocks with four components will be present: AABB, BBAA, ABAB, BABA, ABBA, and BAAB. A random digit number table or generator to select and sort the blocks will be applied. The allocation sequence will be concealed in sequentially numbered, sealed envelopes.
This process will be prepared by a statistician from the Clinical Epidemiology Unit of Research Center. The corresponding envelope will be opened by a nurse who will be blinded to the study , after a patient was enrolled.
Sample size :
By taking the results of the study of Moll et al into account, which showed metformin plus CC resulted in 39.5 % pregnancy rate compared with 27.1% pregnancy rate with CC; and assuming that the results of the investigators' proposed study, which is done on IR women, are consistent with the results obtained from that study, the sample size calculated 388 by using the following formula, which compares the two proportions, and supposes the maximum error type 1 and 2 to be 5% and 20% respectively, provided that the two groups are equal and the trial is one sided.
Estimated sample size for two-sample comparison of proportions Test Ho: p1 = p2, where p1 is the proportion in population 1 and p2 is the proportion in population 2
Assumptions:
alpha = 0.0500 (one-sided) power = 0.8000 p1 = 0.3950 p2 = 0.2710 d=p1-p2=0.1240 n2/n1 = 1.00
Estimated required sample sizes:
n1 = 194 n2 = 194
Women will randomly be allocated to CC (100mg daily beginning on cycle day 3 for 5 days after spontaneous menses or withdrawal bleeding induced by progestin administration (medroxyprogesterone acetate \[Provera\], 5 mg per day for10 days and for up to 3 menstrual cycles.There is at least 5 days of pill free interval between progesterone(P) and clomid . The dose will be increased in subsequent cycles in cases of nonresponse (follicle \<12mm by day 16 in CC cycle.) The maximum daily dose of clomiphene will be 150 mg (three pills), given for 5 days plus metformin (1500mgdaily has been starting before) (group A; n=194;) or CC plus placebo (group B; n=194).
Study Design:
All women will be examined clinically and their weight, height , body mass index (BMI) ,pulse and blood pressure will be recorded . These measurements will be taken by a nurse who will be unaware of the admission number of the patients. BMI will be calculated as weight divided by height squared (kg/m2). After an overnight fasting of 10 to 12 h, blood samples will be obtained for the determination of fasting blood glucose (FBG), insulin, FSH, LH, estradiol (E2), total and free testosterone (T), DHEAS, PRL, TSH, and 17OH progesterone for exclusion of 21-hydroxylasedeficient nonclassic adrenal hyperplasia (NCAH) on the 3rd day of a spontaneous menses or withdrawal bleeding induced by Provera.
All blood samples will be centrifuged and stored at -20°C until assayed together. Serum levels of FSH, LH, TSH and PRL will be measured using Enzyme-linked immunosorbent assay (ELISA,Wernecce,3200) (Ideal, Tehran, Iran). Total T, FreeT, DHEAS and insulin will be also measured using ELISA assay (Monobind,Inc,USA) . FBG will be measured using an enzymatic colorimetric method with Glucose Oxidase.Women with hypothyroidism included in this study after treatment and normal TSH(because in women with PCOS, a significant association between thyroid function, as reflected by TSH \>2 mIU/l, and IR was found and the association appeared to be independent of age and BMI).
IR is determined with homeostatic model assessment( HOMA) method calculated by fasting insulin (mU/L) × fasting glucose (mmol/L)/ 22.5. The value of HOMA-IR ≥2.3 will be assumed as IR according to the Hosseinpanah et al studies´ in iranian women .After baseline studies are completed, the patients will be divided into two groups. In group1 metformin (Metformine, Chemidaro, Tehran, Iran) will be given at a dose of 500 mg three times a day for 8weeks. The dose will be increased from one to three tablets a day over a period of seven days to minimize side effects. The patients in group two will be receiving placebo (identical in appearance to metformin from the same factory).
The P levels in all subjects will be measured every other week in order to document ovulation. A P level \> 5 ng/ml (16 nmol per liter) will considered as confirmatory of ovulation.
If pregnancy occurred, metformin will be continued for another 12 weeks. In case of failure of ovulation after the end of this period, metformin and placebo were continued and patients were given 100 mg CC (Clomid, Iran Hormone, Iran) for 5 days starting from day 3 of their spontaneous menses or withdrawal bleeding induced by Provera.Ovarian follicular response will be monitored by transvaginal sonography every other day from day 10 of the cycle by a single sonographist. When at least one follicle reached ≥18 mm in diameter, 5000 IU of HCG (Pregnyl; N.V. Organon, OSS, Netherlands) will be given intramuscularly and timed intercourse will advised (every other day for one week starting after receiving HCG). If there is no follicle ≥12mm by day 16 in CC cycle, the cycle will be presumed to be anovulatory and monitoring will be discontinued.
In cases of ovulation with CC without pregnancy, the patients will be advised to participate in other two similar cycles of therapy with 100 mg CC (group1 and 2). When ovulation will not occur with 100 mg CC, its dose will be increased to 150 mg and the same treatment protocol will be used for this dose. In group 1 metformine and in group 2 placebo will be continued through CC cycles.
Clinical pregnancy will be defined as the detection of at least one gestational sac on transvaginal ultrasound examination starting one week after the missed period.Ongoing pregnancy will be defined as the detection of fetal heart activity after 12 weeks of gestations.
Statistical analysis:
All the data will be entered to the SPSS software (Version 11.5.0, © SPSS Inc.), T-test (or Mann-Whitney test if needed),ANOVA for quantitative variables , Chi-square test and Fisher exact test if necessary for qualitative variables are used. P\<0.05 will be considered significant in all tests.
