Table-2: Effect of metformin and /slilymarin on Inusiln ,glucose ,HOMA-IR ratio,total testosterone and progester-one in women with PCOS Iraqi J Pharm Sci, Vol.19(2) 2010 Polycystic ovary syndrome and silymarin 11 Improving an Ovulation Rate in Women with Polycystic Ovary Syndrome by Using Silymarin Mohammed A.Taher* ,1 , Yaser A.Atia** and Manal K.Amin*** *Department of Clinical Laboratory Sciences, College of Pharmacy, University of Baghdad,Baghdad,Iraq ** College of Alkindney Medicine ,University of Baghdad, Baghdad, Iraq . *** Ministry of Health, Baghdad, Iraq. Abstract Polycystic ovary syndrome(PCOS) is a heterogeneous disorder of uncertain etiology , it is the most common endocrinopathy in women and most common cause of anovulatery infertility ,characterized by chronic anovulation and hyperandrogenemia .The present study was designed to investigate the effect of silymarin which is known to have antioxidant and insulin sensitivity effects on the levels of glucose, insulin ,testosterone ,leutinizing hormone(LH) and progesterone .Ovulation rate and Homeostasis Model Assessment of insulin Resistance (HOMA) ratio were determined .A 3-months of treatment were con- ducted in 60 PCOS patients in three well-matched groups .The first one (n=20),received si- lymarin(750mg/day) .The second group received metformin(1500mg/day) while the third group treated by combination of metformin (1500mg/day )and silymarin (750mg/day). All these groups had taken the drugs in divided doses. The results showed significant improvement in all parameters at the end of treat- ment .The percentage of increment in progesterone levels after completion of treatment were 12.12, 15.9, and 17.51 in groups 1, 2, and 3 respectively and the number of patients ovulated after 3 months of treat- ment were 4,5, and 10 in groups 1,2, and 3 respectively. However they are more better in group of pa- tients who were treated with combination of silymarin with metformin.In conclusion the addition of si- lymarin to metformin in treatment of PCOS patients has improving effect on disturbed hormones and ovulation rate. Key words: Polycystic ovary syndrome, silymarin, ovulation rate,metformin الخالصة طثة ٔأكثزعُد انُظاء غٛز يؤكدج.اَّ اعرالل ْٕريَٕٙ شائع ألطثابْٕ اضطزاب يرغاٚز األكٛاصيرالسيح انًثٛض يرعدد إٌ اتاضح يشيٍ ٔارذفاع انٕٓريٌٕ انذكز٘ .أٌ اندراطح انحانٛح صًًد نفحص ذأثٛز التاضح ٔٚرًٛش تعدواشائع نهعقى انًصحٕب تعدو , األَظٕنٍٛ, عهٗ يظرٕٚاخ انظكز نألَظٕنٍٛٔذقهٛم انًقأيح األَظٕنٍٛض خاصٛح ضد انرأكظد ٔ سٚادج نرحظ انظٛهًارٍٚ ْٕٔ انذ٘ ًٚهك HOMA) ) نألَظٕنٍٛ َظثح االتاظح ٔدنٛم يقأيح ذى احرظاب , انٕٓريٌٕ انهٕذُٛٙ ٔانثزٔخٛظرظٛزٌٔ. انٕٓريٌٕ انذكز٘ 06يكَٕح يٍ األٔنٗفٙ ثالثح يدايٛع . انًدًٕعح األكٛاصيزٚضح تًرالسيح انًثٛض يرعدد 06ل أعطٙيٍ انعالج أشٓزالثح ث.أٚضا يزٚضح عٕندٍ تعقار انًرفٕريٍٛ ٔتدزعح 06انًدًٕعح انثاَٛح يكَٕح يٍ يهغى ٕٚيٛا ٔ 056ظٛهًٛارٍٚ تدزعح يزٚضح عٕندٍ تعقار ان يهغى ٕٚيٛا ٔانظٛهًارٍٚ 0566عٕندٍ تًشٚح انًٛرفٕريٍٛ أٚضايزٚضح 06تًُٛا انًدًٕعح انثانثح ٔانًكَٕح يٍ يهغى ٕٚيٛا 0566 ذحظٍ انًظرٕٖ نكم انرحانٛم تعد َٓاٚح انعالج.اٌ أظٓزخيهغى ٕٚيٛا .كم ْذِ انًدايٛع اخذخ انعالج تدزع يدشئح .انُرائح 056تدزعح فٙ يدًٕعح انًزٚضاخ انهٕاذٙ عٕندٍ تًشٚح انظٛهًٛارٍٚ أفضمسٚادج َظثح يظرٕٚاخ انثزٔخٛظرٛزٌٔ َٔظثح االتاضح كاَد نّ ذأثٛز األكٛاصانظٛهًٛارٍٚ نهًرفٕريٍٛ نعالج انًزٚضاخ انًصاتاخ تًرالسيح انًثٛض يرعدد إضافح إٌانًٛرفٕريٍٛ .ٔفٙ االطرُراج ٔ حظٍ عهٗ انٕٓيَٕاخ انًضطزتح َٔظثح االتاضح. Introduction Polycystic ovary syndrome (PCOS) is a heterogeneous disorder of uncertain aetiology; it is the most common endocrinopathy in wom- en and most common cause of anovulatory in- fertility,affecting 5-10% of population of repro- ductive age. (1) It is characterized by chronic anovulation and hyperandrogenism. (2). Insulin resistance and associated hyperinsulinemia also have been recognized as important pathogenic factors in determining the majority of PCOS women particularly when obesity is present . (3) Most but not all women with PCOS have hyper- insulinemia with insulin resistance (4) .The asso- ciation between hyperinsulinemic insulin re- sistance and PCOS well recognized and play an import role in the development of PCOS (5) .Hyperinsulinemia has been shown to reduce sex hormone binding globuline (SHBG) synthesis in liver (6) and insulin has a direct ef- fect on ovarian steroidogenesis in theca cell. (7) Metformin is the oldest and still most insulin sensitizer world wide in the treatment of type2 diabetes mellitus and PCOS-associated with insulin resistance. It is a biguanide derivative and considered as an insulin sensitizer since it lowers glucose levels without increasing insulin secretion . (8) 1Corresponding author E- mail : Mohammed_taher34@yahoo.com Received : 16/1/2010 Accepted 26/5/2010 Iraqi J Pharm Sci, Vol.19(2) 2010 Polycystic ovary syndrome and silymarin 12 Silymarin is an active polyphenolic flavenoid extracted from fruits(seeds) of medicinal plant silybum marianum (milk thistle), extracts were standardized to contain 70-80% silymarin com- plex which comprised mainly of three major flavolignans , silybinin silychristin and silyd- ianin of which silybinin is the most biological active. Silymarin is considered to be very safe and there are only few reports on its adverse effects,mainly a mild laxative effect has been observed in occasional instances and there are no known contraindications or side effects re- ported during its regular use. (9) According to the multiple pharmacological actions of silymarin, silybinin have been clinically evaluated in dia- betics for their therapeutics value reduces the lipoperoxidation of cell membrane and insulin resistance significantly, decreasing endogenous insulin overproduction and the need for exoge- nous insulin administration. (10) So this study was designed to evaluate the efficacy of si- lymarin as insulin sensitizer improving an ovu- lation rate by treatment of PCOS and conse- quently its effect on hormonal and biochemical profile of the patients and comparing it with a classical one, metformin. Materials and Methods Patients This study was conducted into Baghdad city ,in al-Elwia maternity teaching hospital from 12/2006-6/2007.The study groups includ- ed 80 women selected randomly, 60 patients with PCOS aged (19-39) years with a mean age ( 27.5) years and 20 healthy control women aged (21-32) years with mean age (24) years.The diagnosis of PCOS was made by the gynaecologists depending on ultrasound exam- ination ,clinical features and laboratory tests according to diagnosis criteria of ( Rotterdam 2003) (11) . Table-1 shows that the clinical presentations of patients in present study like those reported in other studies of polycystic ovary syndrome in that it is a heterogeneous disorder Investigations included : serum fasting glucose levels, fasting insulin levels, serum testosterone, serum progesterone and serum leutinizing hormone (LH).All patients participi- tated in this study were diagnosed having PCOS and were non-diabetic, not hypertensive, not pregnant , and not taking any medications that affect the reproductive or metabolic func- tions with 90 days of study. The patients were followed weekly regularly under gynecologist supervision during the period of treatment.The women were grouped into 4 groups as follow: Group 1: included 20 PCOS patients ,with BMI (31.22±1.138 Kg/m2),and age (19-31) years. They received Sylimarin tablets (750mg/day) in 3 divided doses after meals for 3 months. Group 2: included 20 patients with BMI 30.84±1.23kg/m2) and age (20-35) years. The treatment was including metformin tablets 1500mg/day in 3 divided doses (500mg after meals for 3 moths. Group 3: included 20 patients with BMI 32.83±1.37 kg/m2), age (22-39) years. The treatment was consisting of combination of 2 drugs (sylimarin 750 mg/day ) and metformin ( 1500 mg /day) in 3 divided doses for 3 months. Group 4: included 20 healthy women with BMI 28.4±1.01kg/m2) ,age (21-32) years and these women were with regular cycle (21-32 days) who were taken from outside of the hos- pital and selected as controls. Sample collection Venous blood sample withdrew after overnight fasting ( at least 12 hours of fasting ) from PCOS women and the control group . For the subjects with regular cycles ,the samples were taken at 3-5 days after the cycle for de- termination of serum LH and the sample for progesterone were taken at 21 days of the cycle .The other patients with irregular cycle ,the samples were taken randomly. The samples were taken from the patients before starting and after one month of treatment . Biochemical analysis Determination of serum glucose and insulin levels Fasting serum glucose and insulin levels were measured by commercial kit obtained from Randox using enzymatic method (12,13) . Determination of Homeostasis Model As- sessment of insulin Resistance (HOMA-IR) , HOMA - IR was calculated using the following formula (14) : HOMA-IR=Fasting glucose (mmol/L)× Fasting insulin (pmol/ml)/22.5. Insulin resistance patients were defined as hav- ing HOMA>2.7. Determination of serum testosterone (15) and LH levels (16) : Serum testosterone and LH levels were determined by radioimmunoassay(RIA) method using a kit provided by Sigma-Aldrich. Determination of serum progesterone & Ovu- lation Rate Serum progesterone levels were deter- mined using kit obtained from Sigma-Aldrich , using (RIA) method, and the ovulation rate was determined according to mid-luteal phase pro- gesterone level that was equal to or more than 16nmol/L (5ng/ml) . (17) Determination of body mass index(BMI) BMI was calculated using standard formula : BMI= weight (kg)/high (m2). Iraqi J Pharm Sci, Vol.19(2) 2010 Polycystic ovary syndrome and silymarin 13 Obese patients were defined as having MBI > 27kg/m2 (18). Ultrasound study Transvaginal ultrasound study scan is performed for each patient at about day 12 of the cycle in order to to confirm follicular changes that appear through biochemical and hormonal changes , also it was repeated for each patient who had serum progesterone levels higher than or equal to 16nmol/L in order to confirm improvement of fertility and response of patients to treatment and follow up follicular development . (19) Diagnosis Hyperandrogenism Based on criteria of Androgen Excess Society (AES 2006), which recommended the following diagnostic criteria for PCOS hyper- androgenemia. (20) 1. Hyperandrogenism ( hirsutism and /or hy- perandrogenemia ) 2. Ovarian dysfunction (oligo-anovulation and /or PCOS). 3. Exclusion of related disorders such as hy- perprolactenemia and congenital adrenal hyperplasia. Hirsutism Based on Ferriman-Gallwey score , evalu- ates nine body sites including the face, chest , areolae , linea alba , upper back, lower back, buttocks, inner thighs and external genetalia . (21) Infertility Inability of any couple to conceive a child within a 12 months period of unprotected coitus ( sexual intercourse) . (22) Statical analysis Student t-test was used to examine the quantitative differences in the mean parameters. The results are expressed as mean±SD and the P-values <0.05were considered statically signif- icant. Results Table-1 shows that 43.3% of the patients were with hirsutism and 36.6% with acne .Most patients were obese 68.3% and 31.6% were lean.The percentage of infertility among the patients were 31% and only 7% were with regu- lar cycle while the percentage of amenorrhea and oligomenorrhea were 19% and 34% respec- tively.The percentage of insulin resistance was 78.3%, morover the androgenemia feature was the highest (85%).Table 2 shows a significant elevation ( P<0.05) in mean serum insulin lev- els (pmol/L) of base line levels in the three study groups compared with control group and it declined significantly (p<0.05) after 1 st ,2 nd and 3 rd month of treatment in all groups of pa- tients. There is significant increment (p<0.05) in mean serum glucose levels (mmol/L) of base line levels in three groups compared with con- trol group and it declined significantly (p<0.05) after 1 st ,2 nd , and 3 rd month of treatment in all groups of patients except in 1 st month of first group , it was non-significant (P>0.05).Also the same Table illustrated significant increment (P<0.05) in mean HOMA-IR of baseline level in 3 PCOS groups compared with control and it declined significantly (p<0.05) after 1 st ,2 nd , and 3 rd month of treatment in all groups of patients. There was significant increment (p<0.05) in mean serum testosterone levels (nmol/L) of base line levels in 3 groups compared with con- trol group and it declined significantly (p<0.05)after 1 st ,2 nd and 3 rd month of treatment in all groups of patients. Mean serum proges- terone levels (nmol/L) of baseline levels in three groups decreased significantly (p<0.05) compared with control group and elevated sig- nificantly (p<0.05) after 1 st ,2 nd , and 3 rd month of treatment in all groups of patients except 1 st first group , it was non-significant (p>0.05).This table also demonstrate significant increase(p<0.05) in mean serum LH levels (U/L) of base line levels compared with the control group and it declined significantly (p<0.05) after 1 st .2 nd . and 3 rd month of treatment in all groups of patients except in 1 st month of group 1 and 2 ,it was non-significant (p>0.05).Table-3 illustrated that, the percentage of increment in mean serum progesterone levels (nmol/L) was 4.28 % ,8.72 and 12.22 in group1 ,also 4.324%,8.42% and 15.9% in group2 and 4.179% ,8.79% and 17.51 in group 3 after 1 st ,2 nd .and 3 rd month of treatment for each group respectively. The numbers of women who had ovulated were 4,5 and 10 in group 1,2,3 respec- tively. Table 1: Demographic data of 60 women with polycystic ovary syndrome. Feature No. of patients (% ) Hirsutism 26(43.3) Acne 22(36.6) Obesity 41(68.3) Lean 19(31.6) Infertility 31(51.6) Amenorrhea 19(31.6) Oligomenorrhea 34(56.6) Regular cycle 7(11.6) Insulin resistance 47(78.3) Hyperandrogenemia 51(85) Iraqi J Pharm Sci, Vol.19(2) 2010 Polycystic ovary syndrome and silymarin 14 Table 2: Effect of metformin and /or silymarin on Insulin ,glucose ,HOMA-IR ratio, total testos- terone and progesterone in women with PCOS. G r o u p s Analytes Control Base line After 1 M After 2M After 3M 1 Insulin(pmol/L) 57.5±0.359 92.18±4.73 89.35±0.35* 85.65±4.28* 81.44±3.66* Glucose(mg/dl) 5.1±0.17 5.29± 0.29a 5.01± 0.192NS 4.88±0.128* 4.73±0.128* HOMA 2.13±0.015 3.11± 0.244a 2.865±0.233* 2.673±0.178* 2.02±0.178* Testosterone(nmol/L) 1.45±0.03 4.59± 0,223a 4.427±0.242* 4.242±0.303* 3.396±0,318 Progesterone(nmol/L) 17.15±0,02 12.84±0,612a 13.39±0.682NS 13.96±0.804* 14.41±0.942* LH(u/L) 5.2±0.365 9.38± 0.317a 9.18± 0.284NS 9±0.245* 8.71±0,376* 2 Insulin(pmol/L) 57.5±0.359 83.7±4.49a 82.1±3.468 80.8±3.01* 74.5±4.73* Glucose(mg/dl) 5.1±0.17 5.35± 0.362a 4.49± 0.209* 4.63±0.35* 4.25±0.229* HOMA 2.13±0.015 2.68± 0.226a 2.59± 0.212* 2.39±0.199* 2.02±0.178* Testosterone(nmol/L) 1.45±0.03 4.07± 0.199a 3.938±0.213* 3.765±0.185 3.9±0.167* Progesterone(nmol/L) 17.15±0,02 12.95±0.967a 13.517±0.941* 14.04±1.01* 15.01±1.33* LH(u/L) 5.2±0.365 111.13±0.87a 10.56±0.839NS 10.10±0.644* 9.52±0.741* 3 Insulin(pmol/L) 57.5±0.359 106±6.6 94.05±4.26* 84.16±4.50* 77.22±3.09 Glucose(mg/dl) 5.1±0.17 5.12±0.301a 4.58±0.330* 4.27±0.369* 3.87±0.22* HOMA 2.13±0.015 3.55±0.172a 2.75±0.144* 2.298±0.245* 1.91±0.135* Testosterone(nmol/L) 1.45±0.03 4.59±0.942a 4.18±0.176* 4.06±0.159* 3.9±0.167* Progesterone(nmol/L) 17.15±0,02 13.88±0.875a 14.46±0.792* 15.10±0.673* 16.31±0.916* LH(u/L) 5.2±0.365 10.08±0.510a 9.33±0.480* 8.89±4.22* 8.54±0.515* Values are Mean±SD , a P< 0.05 for comparison with control group ,*P<0.05 for comparison with baseline , NS non-significant P>0.05 Table 3: Comparison among mean % of in- creament in progesterone levels (nmol/L) and number of women who had ovulated during the study in all groups of PCOS patients. Discussion The percentage of patients with hirsutism and acne was 43.3% and 36% respectively (ta- ble-1) and this finding was consistence with other study performed in diagnosis of PCOS.Cutaneous manifestations of hyper- androgenism in PCOS include hirsutism,acne or combination , and male –pattern hair loss ( an- drogenic alopecia); whereas acanthosis nigri- gans is a cutaneous marker of hyperinsulinemia . (23) The study demonstrated that percentage of obese patients was 68.6% while it was 31.6% for lean , this is common in PCOS and it is in line with other studies which demonstrated that 40-60% of women with PCOS are obese (BMI>27 kg/m2). (24,25) The present study showed that (51.6%) of the patients were in- fertile , 31.6% with amenorrhea, 56% with oli- gomenorrhea ,11.6% with regular cycle,78.3% with insulin resistance and 85% with hyper- androgenemia ,these results are in agreement partly with other results which demonstrate the presence of infertility by (55-75%) , amenor- rhea (26-15%) ,oligomenorrhea (50-90%) regu- lar cycle (22%) and hirsutism (60-90%) in women with PCOS . (24,26) The high levels of androgens lead to chronic anovulation , men- strual disturbances and hirsutism. PCOS pa- tients typically have elevated LH levels and LH:FSH ratios. (27) because hyperandrogenism leads to abnormal folliculogenesis and endome- % of 1 st Month % of 2 nd Month % of 3 rd Month No.of women ovulated Group1 (sylimarin) 4.28 8.72 12.22 4 Group2 ( Metformin) 4.324 8.42 15.9 5 Group 3 (combination) 4.179 8.79 17.51 10 Iraqi J Pharm Sci, Vol.19(2) 2010 Polycystic ovary syndrome and silymarin 15 trial development . (28,29) Hyperandrogenemia is a key feature of the syndrome; but it is not al- ways linked to hyperandrogenic symptoms such as acne or hirsutism; indeed ,ethnic groups such as Asian shown insulin resistance and associat- ed hyperinsulinemia are also now recognized as important pathogenic factors in determining hyperandrogenism in the majority of PCOS women ,particularly when obesity is present . (30) The present study illustrated a significant (P<0.05) increase in serum insulin and glucose baseline levels and HOMA-IR index baseline value compared with control group, the results were compatible with those observed by Laure C.,et al. (31) , as characteristic features of women with PCOS.During three months of treatment with metformin and /or silymarin a significant (P<0.05) reduction in these parameters in all groups was observed except the effect of si- lymarin on glucose levels in the first month, was non-significant (P>0.05) , as shown in table (2). Metformin leads to increase glucose utiliza- tion, decrease hepatic glucose production, imcrease insulin receptor binding and insulin receptor tyrosin kinase activity, but it has ad- verse effect on gastrointestinal tract and liver function (32,33) , while silymarin, represents a new possibility in the treatment of PCOS, the underlying mechanistic links for this effects may be due to different possible mechanismas; silymarin increases, normalized and stimulated pancreatic activity of antioxidant and free radi- cal quenching enzymes, ( glutathion peroxi- dase,superoxide dismutase and catalase). (34.35) Silymarin may produce its effect on glucose and insulin levels by another mechanism through blockage of TNF-α where that serum TNF-α concentration have been high in normal- weight PCOS women and even higher levels in obese women with PCOS. (36) When combina- tion of silymarine and metformin were used , a powerful synergism effect occurred and led to best results as illustrated in third group , be- cause each drug act by different pharmacologi- cal mechanism and different receptor sites which means that they may not compete one with each other to get same response, so that pronounced reduction in glucose , insulin and HOMA-IR values was occurred.The present study demonstrated a significant (P<0.05) in- crease in baseline testosterone levels compared with control group ,this result was compatible with other studies which demonstrate that se- rum concentration of testosterone and andros- tenedion are elevated in women with PCOS ( the mean concentration are 50%-150% higher than controls). (37) During the 3 months of treat- ment with metformin and /or sylimarin, a sig- nificant reduction(P<0.05)from baseline of tes- tosterone was observed ( table -2 ).These results partly in agreement with Velazqwz et al. con- cerning metformin effect who reported that in an uncontrolled study, treatment with metfor- min for 8 weeks results in reduction of serum free testosterone in 29 non-diabetic women with PCOS , mostly overweight. (38) Most studies on this subject suggest that insulin lowering agents may affect the entire spectrum of endocrine, metabolic , and reproductive abnormalities in PCOS patients.However not all studies have assessed the effects of metformin in hyper- androgenic women have confirmed these find- ings. Interestingly, where insulin levels were reduced by treatment , serum androgens were lowered as well. (39) In an uncontrolled trial that assessed 26 obese women with PCOS before and after treatment with 1500 mg metformin /day for 8 weeks , a reduction in insulin concen- trations and in serum free testosterone were reported and SHBG increased by 23% (38).The combination of silymarin and metformin result- ed in a more remarkable reduction in testos- terone levels than group 1 and 2 , this may be contributed to additive effect of these two drugs.It has been reported in this study signifi- cant decrease in baseline progesterone levels compared with control group, this result was compatible partly with other study. (40) Treat- ment with metformin and /or silymarin for 3 months , demonstrated a significant increament (P<0.05) in serum progesterone levels in all groups , except in first month of group 1 ,it was non-significant (P>0.05) (table 2 ). The im- provement in ovulation rate ( as assessed by measurement of mid-luteal phase progesterone level (>5ng/ml or >16nmol/L) was evaluated according to the percent of increment in base- line progesterone levels and number women who had ovulated , ( table 3) which reflect that third group showed highest percentage of in- crement in progesterone levels and number of women who had ovulated. However , other re- searchers found a significant enhancement in luteal progesterone levels in PCOS women treated with metformin and they suggested that insulin resistance and hyperinulinemia may be responsible for low progesterone levels during luteal phase in PCOS (41) ; therefore the luteal progesterone levels may be enhanced in PCOS by decreasing insulin levels with metformin. It had been reported that an improvement in men- strual pattern or ovulation with only modest improvement in insulin resistance and hyperin- sulinemia is sufficient to promote preovulatory follicular maturation. (42) Silymarin was not dif- ferent entirely from metformin concerning its effect on ovulation rate and progesterone levels as a result of their effect on insulin resistance and hyperinsulinemia .There is a significant negative correlation between insulin and pro- Iraqi J Pharm Sci, Vol.19(2) 2010 Polycystic ovary syndrome and silymarin 16 gesterone, and between progesterone and LH concentration. (41) Therefore it is probaple that effect of silymarin on progesterone levels were consequences of its effect on insulin resistance and hyprinsulinemia. There was remarkable response to combination treatment because each drug act by its own mechanism and higher in- crement in progesterone and ovulation rate ex- erted by each drug alone may be enhanced by their combination . The base line LH levels in this work increases significantly (P<0.05) com- pared with control group, and it was compatible with other studies which demonstrate that women with PCOS , have 55-75% of a high LH to FSH ratio due to increased levels of LH than low levels of FSH. (43) Elevated serum concen- trations of LH are common in all reported series of women with PCOS. (44) Typically , PCOS associated with increased LH and androgens but with normal or low serum concentrations of FSH.Most investigations have also documented an increased LH pulse amplitude and frequency as characteristic feature of PCOS. (45) During three months of treatment with metformin /and or silymarin a significant reduction (P<0.05) in serum LH levels were observed in all groups except in first month of group 1 and 2 , it was non-significant (P>0.05), (table 2 ). The reduc- tion of plasma levels of LH are not a primary event in the reduction of hyperandrogenism induced by metformin because many studies have reported a reduction in plasma androgens but not concomitant reduction in LH, indicating that in these cases the reduction of steroid syn- thesis cannot be secondary to reduced stimula- tion of LH also. It is possible that spontaneous or induced ovulation or reduction in androgens may lead to a secondary reduction in LH.Therefore androgens returned to pretreat- ment levels when metformin was suspended and that rise preceded the rise in LH, sustaining the hypothesis that a primary disorder of andro- gen hypersecretion is the cause of LH hyperse- cretion. (46) The effect of silymarin can be ex- plained in the same manner, although its action on insulin levels are more pronounced when compared with metformin in current study, however there is a positive correlation between hyperinsulinemia and LH levels (41) , the possible effect of silymarin on LH though its action on hyperinsulinemia and insulin resistance , indeed improvement in hyperinsulinemia may lead to decrease response of LH to GnRH. As expected from above mechanism of each drug , a highest reduction in LH levels were observed when combination used ,( table2 ) which indicates that each drug may improve the other. References 1. Vincenzo De Leo,Antonio Ia Marca and Felice Petraglia.Insulin-lowering agents in the management of polycystic ovary syn- drome.Endocrine Reviews.2003;24(5):633- 667. 2. Acien P,Quereda F,Matalin P,et al.Insulin,androgen& obesity in woman with polycystic ovary syndrome:a hetero- genous group of disorder.Fertil Steril.1999;7:32-40. 3. Poretsky L,Cataldo NA,Rosenwaks Z & Giudice L C.The insulin- related ovarian regulatory system in health and dis- ease.Endocrine Reviws.1999;20:535-582. 4. Meirow D,Yossepowitch O,Rosler A,et al.Insulin resistance &non-resistance poly- cystic ovary syndromerepresent tow clini- cal &endocrinological subgroups.Hum Re- prod.1995;10:1951-1956. 5. Andrea Dunaif.Inuslinresistance and the polycystic ovary syndrome:mechanism and implications for pathogenesis.Endocrine Reviews.1997:18(6):774-800. 6. Nestler J,Powers L, Matt D, et al. A direct effect of hyperinsulinemia on SHBG levels in obese women with polycystic ovary syndrome.J Clin Endocrinol Metab.1991;72:83-89. 7. Nestler J,. Role of hyperinsulinemia in the pathogenesis of polycystic ovary syn- drome& its clinical implication.Sem Re- prod Endocrinol.1997;15:11-22. 8. Renato Pasquali and Alessandra Gam- bineri. Insulin –sensetizing agents in polu- cystic ovary syndrome, European Journal ofEnocrinology.2006;154(6):763-775. 9. Blumenthal M, Busse,W.The complete german commission E monographs: Therapuetic guide to herbal medicines, American botanical council and integrative medicine communica- tions:Austin,TX,1998; 685-698. 10. induced neurotoxicity by inhibiting micro- glia activation.Eur J Neuro Sci.2002 ; 16: 2103-2112. 11. Rotterdam ESHRE/ASRM-sponsored Pol- ycystic ovary syndrome consensus work- ship group. Revised 2003 consensus on di- agnostic criteria &long term health risks re- lated to polycystic ovary syndrome. Hum Reprod.2003;19:41-47. 12. Trinder P. Determination of glucose in blood using glucose oxidase with alterna- tive oxygen acceptor. Ann Clinical Bio- chemistry.1969;6:24-27. 13. Robbins D.C,Andersn L, Bowsher R, et al. Report of the American Diabetes associa- tions-Task Force on standardization of in- sulin assay.DIABETES.1996;45:242-256. Iraqi J Pharm Sci, Vol.19(2) 2010 Polycystic ovary syndrome and silymarin 17 14. Matthews DR,Hosker JP,Rudensker A, et al.Homoestasis model assessment :insulin reisstance &beta cell fuction from fasting plasma glucose &insulin concentration in man.Diabetolgia.1985;28:412-419. 15. Ratccliffe WA, Corrie JE Dalziel et al.Clinical chem..J.1982;28:1341-1318.As cited by sigma-Aldrich Diagnostic. 16. ThomasCM,Segers MF.Clin Chem.Apr.1988;34:768.As cited by Sigma- Aldrich Diagnostic. 17. Yilmas S,Unaly Y, et al.The effect of met- formin on insulin resistance ,clomiphen- induced ovulation &pregnancy rate in women with polycystic ovary syndrome .European J.Obst.&Gynecol,&Reprod.Biology.2004;1 13:214-220. 18. Peter n.Herbert.Eating disorders.In: Andeo- li,Carpenter (editor).Cecil essentials of medicine.Saunders Company,5 th end. 2001, 515-521. 19. Yeh HC,Futterweit W,Thornton JC. Poly- cystic ovarian disease:US features in 140 patients.Radiology.19987;163:111-116. 20. TaskForce on phenotype of PCOS of the Androgen Excess Society.Psition state- ment:The androgen Excess Society evi- dence-based criteria for defining PCOS as apredominantly hyperandrogenic syn- drome.J.clinical Endeocr Meta 2006;91: 9237-9245. 21. Ferriman D &Gallway J.D. Clincal assess- ment of body hair growth in women.J Clin- cal Endocr Meta.1961;21:1440-1447. 22. Text book of gynecology,L. Copel and, W.B.Saunders Co.,1993. 23. Tsilchorozidou T,Overton C, Conway GS.The pathophysiology of polycystic ova- ry syndrome.Clin Endocrinol (0xf). 2004;60:1-17. 24. Futterweit W.Polycystic ovary syndrome clinical perspective&management.Ostet Gynecol Surv.1999;54:402-413. 25. Campbell PJ, Geriich JE.Impact of obesity on isulin action in volunteers with normal glucose tolerance demonstration of a threshold for the adverse effect of obesity.J Clin Endocrnol Metabb.1990;70:1114- 1118. 26. Aziz R,Dunaif A,Giudice LC, et al.Diagnosis & management of polycystic ovary syndrome.Contemp Obstet Gyne- col.1998;43(supp11):1-29. 27. Legro RS,Kunselman AR,Dodson WC,et al.Prevenlance and predictors of risk for type 2 diabetes mellitus and impaired glu- cose tolerance in polycystic ovary syn- drome :a prospective, controlled study in 254 affected women.J clin Endocrinol Metab.1999;84:165-169. 28. Robert D.Utiger.Insulin &the polycystic ovary syndrome.N Engl J Med.1996; 9(335):657-658. 29. Franks S.Polylstic ovary syndrome.N.Eng J Med. 1989;333:853-861. 30. Clifford K,Rai R,Watson H, et al. An in- formative protocol for the investigation of recurrent miscarriage:preliminary experi- ence of 500 consecutive cases.Hum Re- prod.1994;9:1328-1332. 31. Laure C.Morin-Papunen,1lka Vauhkonen, et al.Insulin sensitivity ,insulin secretion & metabolic & hormonal parameters in healthy women & women with PCOS. Human Reproduction.2000; 15(6): 1266- 1274. 32. Landian K,Tengborn L,Smith U.Metformin and metoprolol CR treatment in non-obese men.J Intern Med.1994;235:335-341. 33. Hundal RS&Inzucchi SE.Metformin : new understanding, new uses. Drugs. 2003; 63: 1879-1894. 34. 4)Soto C, Recoba R,Barron H et al. Si- lymarin increases antioxidant enzymes in alloxan-induced diabetes in rat pancre- ase.Comp Biochem Physiol Toxicol Phar- macol.2003;136:205-212. 35. Maddux BA,See W, et al. Protection against oxidative stress-induced insulin re- sistance in rat L6 muscle cells by micro molar concentrations of alpha-lipoic ac- id.Diabetes . 2001;50:404-410. 36. Gonzalez F, Thusu K, abdel-Rahman E, et al. Elevated serum levels of tumor necrosis factor in normal –eight women with poly- cystic ovary syndrome.Metabolism. 1999; 48:437-441. 37. Conway GS , Honour JW,Jacobs HS. Het- erogenety of polycystic ovary syndrome: clinical , endocrine and ultrasound feaures in 556 patients.Clin Endocrinol (Oxf) 1989;30:459-470. 38. Valezquezz EM, Mendoza S, Hamer T, et al. Metformin therapy in PCOS reduces hyperinsulinemia , insulin resistancd, hy- perandrogenemia , and systolic blood pres- sure , while facilitating normal menses and pregnancy.Metabolism. 1996;43:647-654. 39. Acbay O, Gundogdu S. Can metformin resuce insulin reistance in polycystic ovary syndrome? Fertil Steril.1996;65:946-949. 40. Velazqquez E,Acosta A, Mendoza SG. Menstrual cylicity after metformin therapy in PCOS.Obsest Gynecol.1997;90:392-395. 41. Meenakumari KJ,Agarwal S, Krishna A, et al. Effect of metformin in treatment of lute- al phase progesterone concentration in Iraqi J Pharm Sci, Vol.19(2) 2010 Polycystic ovary syndrome and silymarin 18 PCOS.Brazilin J Medical & Biological Re- search .2004;37:1637-1644. 42. Derelli D, Derelli T, Bayraktar F, et al. Endocrine & metabolic effects of rosiglita- zone in non- obese women with PCOS,Endocrinol J.2005;52(3):299-308. 43. Karlo BN, Loucks TL, Begra SL.Neuromodulation in polycystic ovary syndrome.Obset Gynecol Clin North Am.2001; 28:35-62. 44. Frank S: polycystic ovary syndrome . N Engl J Med. 1989; 333: 853-861. 45. Waldstreicher J, Santoro NF, Hall JE,Filicari M, Crowley WF Jr.Hyperfunction of the hypothalamic- pituitary axis in women with polycystic ovarian disease.Idirect evidence for partal gonadotrophin desensitization . J Clin En- docrinol Metab . 1988; 66:165-172. 46. Vincenzo De Leo, Antonio La Marca and Felice petraglia. Insulin lowering agents in the management of polycystic ovary syn- drome .2003;24(5):633-667.