Polycystic Ovaries Syndrome

Polycystic ovary syndrome is the most common female endocrinopathy, and could affect 3.5-11.2% of all women within their reproductive years (Knochenhour et al, 1998 (1). This wide variation in reporting the real incidence might reflect the different diagnostic criteria used. On the other hand ultrasonically diagnosed polycystic ovaries have been reported in 16-25% of women with regular menstruation (Polson 2, Abdel-Gadir 3, and Wong 4). Variable familial expression in sisters has been reported, and both autosomal dominant and sex linked transmission modes have been described. However, most women with PCOS showed normal 46XX chromosome, but some could also show dermatoglyphic male pattern. Though unusual dermatoglyphic patterns relate to genetic disorders (Shiono 1986 5, Katznelsan 1982 6), excessive intrauterine exposure to androgens was thought to be the cause in patients with PCOS. A distinction between PCOS and the mere ultrasound diagnosis of PCO is clinically necessary, but some evidence showed that such distinction might not be as strict as previously suggested in many cases (Abdel-Gadir et al 1992 3, Carmine and Lobo 1999 7 and Chang et al 2000 8).

Certain milestones exist in our current understanding of PCOS. The first description of polycystic ovaries was recorded by Antonio Vallisneri in 1721. The English translation of the Italian text was reported as follows: ‘Young married peasant women, moderately obese and infertile, with two larger than normal ovaries, bumpy, shiny and whitish, just like pigeon eggs’ (Cindy Farquhar et al 2000 9). This was followed by a description of sclerocystic ovarian changes by Lisfranc in 1830. A similar pattern was then described respectively by Chereau in 1844, Tilt in 1850 and Gallard in 1873, as reported by O’Dowd et al in 2000 (10). However, the syndrome had its name and fame in modern medicine by 1935 when Stein and Leventhal (11) described a combination of amenorrhoea, obesity and hirsutism in women with enlarged ovaries. Adhering to this combination of symptoms and signs before making a diagnosis, would exclude many of the patients with the syndrome as we know it now.

The endocrine revolution in the diagnosis of PCOS has been started by McArthur et al in 1958 (12), when they first described high LH levels in the urine of patients with polycystic ovaries. High androgens levels were first reported in patient with polycystic ovaries by De Vane et al in 1975 (13). This was followed by the documentation of normal luteinising hormone blood levels in women with PCOS by Rebar in 1976 (14). Ultrasound reporting of the ovarian polycystic pattern was first described by Swanson et al in 1981 (15), followed by Hann et al in 1984 (16), before been objectively quantified by Adams et al in 1985 (17). Finally the association of PCOS with insulin resistance was described by Kahn et al (1976, 18), and became a major issue in the aetiology and morbidity of PCOS. Further discussion of all these parameters would be found in the corresponding parts of this chapter.

Traditionally, an ovary was described as polycystic when it contains 10 or more small cystic areas less than 10 mm in diameter in one ultrasound plane (Adams et al 1985) (17). Most of the clinical, endocrine, biochemical and Doppler ultrasound research was conducted in patients diagnosed on the basis of this criterion, plus the other clinical and biochemical parameters of anovulation and hyperandrogenism. A new consensus has been agreed by the Rotterdam Study Group (19), that 12 or more cysts in the whole ovary should be used as a new parameter for the ultrasonic diagnosis of PCO. This new ultrasound criterion is less stringent than the previous one, and many normal patients would be included in such a diagnosis.

Normally, as many as 10-20 follicles are recruited each cycle, but most of them arrest before reaching full maturation, because of the dominant effect of the leading follicle. Through oestradiol’s negative feedback mechanism, FSH production by the pituitary gland is reduced. The dominant follicle would continue growing because it had already developed a good microvascular blood flow, and enough FSH receptors, with adequate aromatase enzyme activity to maintain an oestrogenic intrafollicular environment. Smaller follicles would stop growing because of their dominant androgenic endocrine milieu. Many researchers have documented 2-3 fold increased development of pre-antral and antral follicles in polycystic compared to normal ovaries (Hughesdon 1982 (20), Maciel 2004 (21) and Webber 2003 (22). Most of these follicles would stop growing because of the abnormal hormonal signalling and the high androgenic environment within the ovaries (Jonard and Dewailly 2004, (23). Accumulation of such small cysts (underdeveloped follicles) leads to the characteristic polycystic appearance which is more common under the age of 35 years than in older women (Abdel-Gadir et al (24). These cystic areas form a wide spectrum of growing follicles and atretic cysts. On one side there are many growing follicles and few atretic cysts. On the other end of the spectrum, there are many atretic cysts and few growing follicles. Most cases fall in between these two extremes. This could explain the differences in response shown by patients with polycystic ovaries during induction of ovulation. It is not possible to differentiate between these patterns during basic ultrasonography. Women with more follicles would be expected to respond more quickly than those with atretic cysts. The final destiny of the atretic cysts theca cells would be an addition to form extra secondary ovarian stoma.

Better pictures became available since the advent of transvaginal scan examination showing different patterns of cyst distribution within the ovaries:

A view was held previously that cysts distribution within the polycystic ovaries could be a reflection of different endocrine patterns (Yen SS, 1980 (25). This was not the case as Abdel-Gadir et al in 1991 (26) showed that basic and pulsatile patterns of FSH, LH and testosterone were not related to the pattern of cysts distribution. This later view was upheld by the recent consensus meeting on PCOS held in Rotterdam in 2003 (19), which also recommended that increased stromal echogenicity should be excluded from the diagnostic criteria. A volume ≥10 ml has been reaffirmed as a diagnostic criterion. Furthermore, they suggested that one ovary showing a polycystic pattern in enough to make the diagnosis in the presence of other diagnostic criteria. The significance of volume has been shown in a previous study which documented better and quicker response of patients with PCOS and larger ovaries to induction of ovulation with gonadotrophins than others with smaller yet polycystic ovaries (Abdel-Gadir, (27). However, no difference could be detected in the duration of symptoms, LH or testosterone blood levels or pulse pattern between the two groups (Abdel-Gadir, (26). This observation was in agreement with a previous statement made by Givens et al in 1976 (28) that normal size polycystic ovaries might have similar histological and biochemical abnormalities as enlarged ones. In this context volume could be more useful as a prognostic parameter for better response to induction of ovulation with gonadotrophins, rather than a reflection of the extent of derangement of the endocrine milieu, or severity of the PCOS condition.

Figures 18 and 19 shown below demonstrates two ovaries with classical subcapsular and universe cyst distribution. Figure 20 on polycystic and one normal ovary. Note the difference in the size between the two ovaries (11.0 vs 2.7 cc) All three patients were hyperandrogenic and had irregular periods.

Prolong use of a combined oral contraceptive pill can affect ovarian morphology and volume, which can mask the polycystic pattern. A dominant follicle or corpus luteum could have a similar effect on the corresponding ovary. Accordingly, patients should be scanned during the early follicular phase to guard against this artefact in patients with regular menstrual function. Otherwise, patients with oligomenorrhoea might need to be rescanned if the initial examination showed a dominant follicle or corpus luteum. However, the presence of polycystic changes in the non active ovary would be adequate to make the diagnosis.

Images 21 and 22 show the same ovaries before and 8 months after using a combined oral contraceptive pill respectively. Note the loss of PCO pattern in figure 22. Image 23 show tow polycystic ovaries. The mature follicle in the right ovary did not conceal the PCO pattern on that side in this case.

PCOS is characterised by ovarian dysfunction with the important features of hyperandrogenism, anovulation and polycystic ovaries on pelvic ultrasound scan examination. It appears to have a familial tendency as 40 % of sisters and 35% of mothers of affected women also have the syndrome (Kahsar-Miller et al 2001 (29). It should be seen as a life long general medical condition rather than just a fertility issue.

It usually has prepubertal onset especially in girls with premature pubarche before the age of 8 years. It could be triggered or unmasked by obesity, insulin resistance, and stress, or dopaminergic dysregulation. A hypothalamic dysfunction has been suggested, as pubertal women with PCOS had their LH surge at midday, rather than at midnight. On the other hand, a dysfunctional pituitary gland has also been suggested, because of the increased LH response to GnRH stimulation (Barnes et al 1989 – (30). Furthermore, an ovarian role has also been suggested related to abnormal activity of 17a-hydroxylase and 17/20 lyase (Rosenfield et al 1990 – (31).

Over the years the emphasis on the study of polycystic ovaries changed from a histological, to pure endocrine, then genetic and ultrasound oriented, and lately metabolic. This is a good indication that the exact cause or causes behind the development of PCOS are not yet well known. It also proved that the condition is a heterogeneous one with different causes in different patients. Nevertheless, once PCOS develops, the ovaries assume a prime role in producing androgens (Goldzieher JW 1962, (32), and Fisher et al 1974 (33). Logically one could think of polycystic ovaries in 2 different ways; as normal ovaries with abnormal gonadotrophins drive, or abnormal ovaries showing abnormal morphology and response, irrespective of the gonadotrophins drive.

Different reports described increased and normal levels of leptin in patients with PCOS. For a given BMI, leptin was not different in PCOS vs. normal control (Caro F 1997 (36). A direct correlation of leptin to insulin has also been reported and hyperleptinaemia has bee suggested to be part of the insulin resistance syndrome (de Courten et al 1997 37). A direct effect of leptin on liver function has been suggested as a cause for insulin resistance. This is affected through attenuation of tyrosine phosphorylation of the insulinreceptor substrate-1 (IRS-1) which is a major effect of insulin, and down-regulationof gluconeogenesis. In contrast, leptin increased the activityof IRS-1-associated phosphatidylinositol 3-kinase (Cohen et al 1996 38). Persistent activation of this enzyme causes insulin resistance due to accelerated insulin-induced insulin receptor substrate-1 degradation in adipocytes (Egawa et al 39). Furthermore subcutaneous fat has been shown to be more effective than intra-abdominal fat in causing high leptin levels (Vaulhonen et al 1998b 40). The variable results regarding the level of leptin in patients with PCOS could be understood as the insulin related leptin secretion is limited to insulin resistance in adipocytes in women with PCOS (Jacobs and Conway 1999 41) Another adipocytokine produced solely by adipose cells is adiponectin which is a 244 amino acid protein. It is thought to have a protective role against insulin resistance (Weyer C et al 2001 42). Its level has been reported to be low in women with PCOS, not related to obesity or hyperinsulinaemia (Carmina et al 2005, 43).

The role of insulin in ovarian function is essential, but has to be a balanced one. It is necessary for normal follicular growth and maturation, as well as oestradiol production by the granulosa cells. This phenomenon would be defective in cases of insulin deficiency. On the other hand, excessive insulin exposure could enhance the androgenic pathway of the theca cells to produce more androstenedione, creating a hyperandrogenic intraovarian environment which could lead to anovulation and PCO formation.

Women with PCOS might present with different problems at different age groups, and treatment usually is tailored differently to suit those different needs. This could be a reflection of the changes in the range and severity of the symptoms themselves over the years, or the emphasis on fertility needs by older women with irregular ovulation. In general terms, PCOS is associated with hyperandrogenic symptoms and signs, obesity, irregular periods and infertility. Some patients might have all the listed problems, but show concern to one or two of them without any concern about the other signs. This is especially so in teenage girls who are more concerned about their weight and skin problems than irregular periods. Older women might be more concerned about fertility issues, with minor if any concerns about excessive hair growth or acne. They often choose to have induction of ovulation to help them get pregnant rather than antiandrogens to help with their skin problems. Obesity could be a real issue in the older age group, yet they might try to neglect its implications, and seek advice to conceive nevertheless. This pattern is not rigid, and patients in different age groups might present with the same concerns

Though young women might have problems with acne, excessive weight gain and irregular periods, the first two problems rank higher in their own minds. The disfiguring facial spots could affect their lives, relationship with parents and peers and might lead to some psychological problems. Hirsutism is also frequently seen in both young and older women. It entails growth of dark terminal hair in a male distribution pattern, which is not socially acceptable. Different ethnic groups have different numbers of hair follicles per unit area of skin. Oriental women tend to have the least number, compared to other races. Furthermore, the perception of how much hair is unacceptable is different among different groups. However, a score of more than 8 in the Ferriman-Gallwey scoring system (44) is considered abnormal. Hirsutism is generally related to exposure of the hair follicles to excessive androgens, resulting in prolongation of the growth phase of the facial and body hair cycle. An opposite effect might be seen on scalp hair follicles, with androgenic alopecia being more common than the more severe frontal hair recession, which is one of the signs of virilization. Hirsutism must be differentiated from hypertrichosis which indicates excessive growth of ambosexual hair which is seen in both women and men, mainly in the arms and legs. This is more common as a familiar or genetic trait, but could be seen in patients on glucocorticoid therapy.

Other androgenic skin problems including greasy skin, androgenic alopecia and dandruff could also be seen. Pigmentation of the skin could be seen mainly in association with insulin resistance. Such problems include velvety dark patches called acanthosis nigricans behind the neck, in the axillae and under the breasts. Skin tags or flaps medically known as acrochordons or cutaneous papillomas could also be present. Other names used included cutaneous tags, fibroma molluscum and fibroepithelial polyps.

Recording the body mass index (BMI) is important in all patients presenting with anovulation or hyperandrogenic signs. It is a good reflector of the amount of body fat, but not a perfect one. Nevertheless, it is a good parameter to use in a clinical setup. All the same, its significance should be related to the fat distribution areas, and the presence of other cardiovascular diseases risk factors. The normal range is between 18.5 – 24.9 kg/m², overweight range is 25-29.9 kg/m², obesity 30-34.9 kg/m², and severe obesity >35.0 kg/m². Patients with PCOS are at risk of developing obesity and figures between 40-50% have been quoted (Goldzieher and Green 1963, 45, Lobo and Carmina 2000 46). It might even be a triggering factor during early puberty for the development of the syndrome itself. It is usually of the android type, which is a male characteristic with increased abdominal fat deposition. A waist : hip ratio >0.85 in women indicates increased risk of cardiovascular diseases. Alternatively, and more simply, the waistline could be used instead. A figure >88 cm (35 inches) would be consistent with abdominal obesity in women.

Women with PCOS are prone to metabolic problems related to obesity, high blood pressure, insulin resistance, high insulin level, type II diabetes, high LDL cholesterol and triglycerides levels, low HDL cholesterol, low fibrinolysis and alteration in plasminogen activator inhibitor 1 (PAI-1). All these factors are related to Syndrome-X which is known to increase the risks of cardiovascular accidents. However, many epidemiological studies did not show increased risk of cardiovascular fatality in women with PCOS (Pierpoint 1998 (47) and Wild et al 2000 (48)). This has been attributed to the high oestrogen environment with its vasodilatory effects and high levels of vascular endothelial growth factor in women with PCOS. Oestrogen acts on blood vessels wall eliciting release of nitric oxide which is a potent vasodilator and improves blood flow (Gisclard et al 1988 49). Furthermore, women with PCOS are 30 times more likely to experience obstructive sleep apnoea syndrome (OSAS), in comparison to matched controls (Vgontzas et al 2001 (50). Insulin resistance was a stronger risk factor of the condition than BMI or testosterone level. Women with OSAS are also more liable to snoring, interrupted night sleep, excessive daytime sleepiness and easy fatigability The last authors also suggested that progressive deterioration of PCOS leads to OSAS. High incidence of cholithiasis has also been found in women with PCOS, even at a young age.

Beside it effect on insulin level, obesity could aggravate the endocrine abnormalities in patients with PCOS through the following means:

It is important to remember that not all obese women with high insulin blood levels are hyperandrogenic. This could emphasise the importance of local ovarian abnormalities which could make them more liable to produce excessive amounts of androgens in response to hyperinsulinaemia.

The reproductive side in women with PCOS could also be compromised with increased risk of anovulation, menstrual abnormalities, ovarian hyperstimulation syndrome and cancer of the endometrium. Ovulatory problems could show as inadequate or short luteal phase, menorrhagia, polymenorrhoea, dysfunctional uterine bleeding, oligomenorrhoea and amenorrhoea. However, it is more often for these patients to present with oligomenorrhoea and dysfunctional uterine bleeding than any of the other mentioned problems (Abdel-Gadir et al (3). The major impact of PCOS on ovulation is affected through increased ovarian hyperandrogenic milieu, as well as the effect of increased circulating androgens on the hypothalamo-pituitary unit. Androgens are known to have a direct detrimental effect on ovulation at the level of the ovaries as they could:

All these problems could lead to disturbed menstrual function and reduced fertility potential. The reproductive capacity of patients with PCOS is also compromised by a higher risk of hyperstimulation and multiple pregnancies, after induction of ovulation. Furthermore, increased miscarriages rate has been documented by many authors in relation to obesity or high LH, androgens, PAI-1, insulin resistance and hyperinsulinaemia. Lower levels of glycodelin and insulin like growth factor binding protein 1 (IGFBP-1), have been reported in patients with PCOS during the first trimester of pregnancy (Jakubowicz et al 2004 51) and in the non-pregnant state (Suikkari et al, 1989 52). These two proteins are necessary for proper implantation by inhibiting the immune response of the endometrium to the embryos. However, no association has been firmly documented between PCOS and recurrent miscarriages, despite the over representation of the presence of PCO in these cases (Essah PA et al 2004 53). The same last authors related all the risk factors mentioned above to insulin resistance and hyperinsulinaemia. Reducing the insulin blood level resulted in reduction of LH, androgens and PAI-1, and increase in the level of glycodelin and IGFBP-1 blood levels. In addition, patients with PCOS were shown to be at greater risk of gestation diabetes and high blood pressure during pregnancy irrespective of being insulin resistant or not. However, the risk of pre eclampsia was high only in patient who were insulin resistant before getting pregnant. Newborns from PCOS pregnancies were significantly more often delivered by caesarean section and transferred to neonatal intensive care unit more often than controls (Bjercke S et al, 2002 54). Many studies documented reduced miscarriage rate with metformin (Khatab el al, 2006 55)

As many as 21% hyperandrogenic women with PCO and regular menstruation were found to have anovulatory cycles (Carmina and Lobo 7). Similarly asymptomatic women with ultrasonically diagnosed PCO and regular cycles had low luteal serum progesterone level (Abdel-Gadir 3). This later group might be the first stage in a continuous chain of events, passing through a phase of regular anovulatory cycles, before they develop irregular and anovulatory menstruation. This puts further emphasis on the point that the presence of PCO even in women with regular menstrual cycles should not be considered as a normal finding, and these patients might benefit of having regular follow up.

Women with PCOS are at risk of mood swings, anxiety and depression with impaired quality of life (Benson S et al 2009 (56). Obesity, hirsutism, irregular periods and subfertility were major sources of psychological morbidity. However obesity was reported in one study to be the most prevalent cause of mental distress, where as the impact of the other symptoms proved to be less well defined (Bishop et al 2009 (57) and Adali et al 2008 (58)). A positive correlation has been reported between the degree of insulin resistance, even before the outbreak of type 2 diabetes, and the severity of depression (Timonen et al, 2005 59). Such psychological difficulties might represent disturbed stress responses by patients with PCOS, as shown by enhanced hypothalamo-pituitary-adrenal axis and heart rate reactivity, as well as reduced upregulation of IL-6 in response to stress (Benson et al 2009, 60). This could be reflected biochemically with increased catecholamines response to provoked stress. To improve the compromised quality of life of this subgroup of patients, more attention should be paid to the psychological impact of the disorder

Management of patients with PCOS is usually directed towards their mode of presentation. However, the psychological impact of the problem should be assessed, especially in the younger age group. The reason why they are seeking advice should be ascertained, and dealt with though it might not be the more important medical problem. Related medical problems could be controlled but might not be totally cured. Furthermore, therapies would change with the age and needs of the patient. Accordingly, prolonged follow up is necessary to prevent long term medical problems. The management strategy should focus on:

It is not unusual for many patients to present with two or even all 4 problems together. Weight reduction is beneficial to all other 3 problems, and induction of ovulation would help with anovulatory abnormal uterine bleeding as well, for patients who are keen to get pregnant. However, treatment of skin hyperandrogenic signs usually clashes with the treatment of infertility, as it usually entails the use of drugs which block ovulation, or are contraindicated during pregnancy.

Excessive weight problems should be addressed as a priority, through significant changes in life style, including more physical activity and healthy eating. Self starvation should be avoided, as most women who lose weight through starvation will regain their initial weight within 2-3 years. Adequate weight loss could lead to significant improvement in insulin resistance, and reduces the level of circulating free androgens. It could also help in regulating ovulation, and improves the chances of getting pregnant. This could be done with a help of a dietician, and regular follow up over a long period of time to guarantee compliance.

Unfortunately, most women would find it difficult to lose weight, despite their serious attempts to do so. This might reflect the anabolic effects of the high insulin and androgens levels. This is affected through improved appetite at the level of the hypothalamus, reduced lipolysis and increased lipogenesis. Accordingly, sustained self motivation, and professional help would be needed. Metformin could be prescribed to patients with insulin resistance. It would help with insulin utilisation at tissues level, especially the liver and muscles. It could also reduce gluconeogenesis and glucose absorption from the gastrointestinal tract. It usually causes gastrointestinal side effects, and is better taken with food. The dose should be built up slowly to avoid side effects and to allow compliance. The usual dose is 500 mg twice daily, but 850 mg tablets could be taken with food 3 times every day by non-responsive patients. Metformin should be suspended few days before any major surgical procedure. Reports of liver damage have been published (Chaudhry and Simmons 2001 (61), Nammour et al 2003 (62), Kutoh E, 2005 (63)), and severe elevation of hepatic enzymes would give a good indication. Ideally, all patients should be tested few times during the first year of medication and annually thereafter. Rarely, it could be complicated with lactic acidosis.

Metformin use is contraindicated in patients with compromised hepatic or renal function tests. Other contraindications include severe infections, dehydration, alcoholism, heart failure, recent myocardial infarction and use of X-ray contrast media. An important side effect of metformin is the reduction of vitamin B12 absorption especially in patients who are at risk, mainly vegetarians. However, it does not cause hypoglycaemia, but could dos so if taken with alcohol. Nevertheless, it could normalise blood glucose level. To have a good impact on insulin resistance and obesity, change in life style and good feeding habits, as well as ample physical exercising are necessary. Metformin is not a slimming drug.

The 4 main strategies used in the treatment of female skin hyperandrogenisation are:

Any androgenic medication should be changed or suspended. Predisposing medical problems including adrenal hyperplasia, hyperprolactinaemia, or thyroid dysfunction should be addressed first. Ovarian androgens production could be reduced by blocking ovulation with an oral contraceptive pill. In certain circumstances a glucocorticoid might be necessary, especially when an adrenal enzymatic deficiency has also been diagnosed. The oestrogen fraction in any combined oral contraceptive pill would also stimulate the liver to produce more SHBG, to reduce the free fraction of androgens. However, the magnitude of this increase in patients using ethinyl oestradiol in a daily dose of 30 µg was found to be equivalent to SHBG level in women with regular menstrual cycles. A significant increase was reported after using 50 µg daily, which is a high dose, not usually used in most present day oral contraceptive pills. Currently, it is also believed that pills with androgenic progestogens should be avoided, especially so, as they could induce or worsen insulin resistance and might induce dislipidaemia. This is especially so, as many pills devoid of such androgenic progestogens are now available. Non androgenic progestogens include desogestrel, gestodene and norgestimate. Examples of PCOS friendly pills include mercilon, yasmin, cilest, marvelon, femodene, femodette and minulet. Metformin has also been shown to reduce androgens production by acting directly on the ovaries, and could help with skin problems even in patients who are not insulin resistant.

The most widely used antiandrogen nowadays is spironolactone which is an aldosterone antagonist. Though initially used as a diuretic, it proved to have excellent anti androgenic characteristics, with minimal side effects. It could take few months before seeing a significant effect, and it usually sustains its effect through the following mechanisms:

The main side effect of spironolactone is intermenstrual bleeding which usually settles with continued use. This is not a problem for patients using an oral contraceptive pill which would also improve the clinical response by inhibiting excessive ovarian androgens production. Using spironolactone during the early weeks of pregnancy could lead to feminization of male fetus genitalia. This follows its effect in reducing the concentration and activity of 5-dihydrotestosterone, which is necessary for the development of male external genitals. This is the idea behind the advice for using an oral contraceptive pill, or any other effective method of contraception, by sexually active women during their reproductive years, while on spironolactone. Changes in blood electrolytes are not common but should be kept in mind, as spironolactone is an anti aldosterone.

Other antiandrogens include cyproterone acetate, flutamide, finasteride and dutasteride. The most widely used one in this group is cyproterone acetate, either in a reversed sequential therapy as mentioned in chapter 4, or as part of an oral contraceptive pill as in dianette (Schering UK). It has a potent antigonadotrophic effect, and hence reduces ovarian androgens production. It also has a direct effect at the skin level by competing with 5-dihydrotestosterone for the receptor sites. It is important to combine it with an oestrogen, and should be used only during the first half of the treatment course. This is because of its long debo effect, which could cause menstrual dysfunction and excessive uterine bleeding. It is recommended that cyproterone acetate should not be used for a very long period of time after the symptoms have subsided. This is even true for dianette, despite the small dose of cyproterone acetate, as it has been shown to cause depression after prolonged use. Accordingly, dianette should not be used for contraception purposes only, by non hyperandrogenic women. Spironolactone in a daily dose of 100 mg has been shown to be more effective on the skin than dianette which contains 3 mg cyproterone and 35 µg ethinyl oestradiol. Other drugs are also potent but have significant hepatic toxicity and should be used only sparingly, and only when really necessary. Flutamide is an androgen receptor blocker given in a dose of 250 mg once or twice daily. It has hepatic toxicity, and could alter liver function tests. It could also cause anorexia, pruritis, dry skin and dark urine. It is mainly used for resistant cases of androgenic alopecia. Liver function tests should be performed before and regularly during the treatment. 5a-reductase inhibitors are not very popular in treating female hyperandrogenisation, and could be less effective than other antiandrogens. Finasteride (proscar) in a dose of 5 mg every day could be used for the treatment of hirsutism as it is mainly a type 2 isoeznyme inhibitor. On the other hand, dutasteride (avodart) in a dose of 0.5 mg every day could inhibit both type 1 and 2 isoenzymes and causes a dramatic reduction in dihydrotestosterone level within a short period of time. It has been portrayed as an effective treatment for androgenic alopecia.

Using any of these medications should be combined with wise use of good skin care and professional help for hair removal. Skin irritants should be avoided. It is always advisable that patients should take a polaroid (or digital) photograph before starting treatment, and at regular intervals thereafter, for comparison purposes, and to monitor response. Laser treatment proved to be effective in dealing with excessive hair growth, but should be part of a general management plan, involving medical treatment of the excessive androgens production.

A diagnosis of PCOS should also be considered in hyperandrogenic women with polycystic ovaries despite having regular menstruation. However, other causes of hyperandrogenism should be excluded first. Furthermore, a high level of LH is no longer considered necessary to confirm the diagnosis. It could be elevated in up to 60% of the patients, but its level could be affected by recent ovulation, ingestion of certain medications and BMI; being higher in leaner patients. Furthermore, it is secreted in 90- minute pulses and the level could depend on the timing of the sample within a pulse. In addition, the unreliability of single blood sample hormone estimations in representing the true endocrine milieu has been known since 1973 (Santen and Bardin 1973 64). The reliability of a single LH estimation was undermined by a variability of 38% and 92% in accuracy to represent the 6-hour mean value, following 20 minutes blood sampling during the follicular and luteal phases of the cycle respectively. The same authors suggested a minimum of 3-hour multiple sampling to detect changes of 40% or less in LH concentrations. Such variability has since been confirmed for LH as well as testosterone by Abdel Gadir et al in 1991 (25). In this respect, a high LH level is significant, but a normal level would not exclude LH hypersecretion. This is a reflection of a stronger positive predictive value, but a low negative one. However, the significance of measuring blood levels of LH in anovulatory hyperandrogenic women with PCOS lies in its prognostic value for selecting patients for ovarian electrocautery, as patients with high LH levels had a better response (Abdel Gadir 65).

This is a term used to describe excessive hair growth not accounted for by demonstrable excessive circulating androgens level. However, many of these patients might have a subtle adrenal enzymatic deficiency. In many patients only the free fraction of androgens is increased despite having normal total testosterone and androstenedione levels. This could follow low levels of the carrier molecule SHBG. Hepatic production of SHBG could be reduced by obesity, high blood insulin and androgens as well as low thyroid hormone levels.

Increased end tissue (skin) sensitivity has been described as the cause of excessive hair growth in cases of idiopathic hirsutism. This was related to rapid turnover of androgens at the skin level, due to increased numbers of androgen receptors, or increased conversion rate of testosterone to the more biologically active dihydrotestosterone. This is related to an increased 5a-reductase enzymatic activity, which is reflected by increased blood levels of dihydrotestosterone metabolite 3a-androstandiol glucoronide (3a-diol G). Oral contraceptives do not usually affect this end byproduct, which is usually reduced by spironolactone and cyproterone acetate, which act as antiandrogens at the skin level. This increased tissues turnover of androgens could explain why women with the same circulating levels of androgens could have different degrees of excessive hair growth. In this context, hirsutism is not a reflection of the circulating level of androgens, but rather an expression of the skin turnover of 5-dihydrotestosterone, as reflected by the increased level of 3a-diol glucuronide. However, this metabolite byproduct is produced by many tissues in the body, which dampened the initial enthusiasm for using it as a sole marker of idiopathic hirsutism.

Infertile obese women with PCOS should be offered fertility treatment only after a good effort has been invested in losing weight. This has been shown to improve ovulation and increase their chances of natural conception, even without any medication. The risk of increased miscarriage in obese women with PCOS after induction of ovulation should be kept in mind (Bohrer and Kemmann 1987 (66), Abdel-Gadir et al 1990, (26). This is on top of the real risk of gestational diabetes and associated fetal and maternal complications.

Induction of ovulation should be attempted first with clomiphene citrate under supervision because of the risk of multiple ovulations. A starting dose of 50 mg every day for 5 days could be started on the 3^rd day of withdrawal bleeding. Ovulation usually occurs about 5-7 days after the last tablet. A higher dose of 100 mg every day for 5 days might be needed. Higher doses would usually be ineffective and could affect the cervical mucus fluidity, increase the histological aging of the endometrium relative to the follicle, and interfere with tubal motility and fluid chemistry as an antioestrogen. Treatment with clomiphene should not be continued for more than 6 cycles. Gonadotrophins could be used in nonresponsive cases, but they need special expertise, and easy access to professional ultrasound monitoring. Attempts should be made to aim at monofollicular ovulation by starting medication with a single ampoule for 7-10 days, before increasing the dose in half an ampoule doses, at equal time spacing. The cycle should be abandoned if multiple follicles were recruited. The risk of multiple follicular development could exceed 50% and 80% with clomiphene citrate and gonadotrophins therapy respectively in patients with PCOS. Ovarian diathermy has been advocated as an alternative to gonadotrophins, with good outcome. In fact 52.1-84% of patient with the sole problem of PCOS conceived after such a procedure (Abdel-Gadir et al 1990 (26), Gjönnaess H, 1994 – (67)), which proved to be as effective as gonadotrophins in inducing ovulation and pregnancy rate, with no risk of hyperstimulation or multiple pregnancies (Abdel-Gadir 1990 - 26). A lower risk of miscarriages has also been reported after ovarian electrocautery (Abdel-Gadir - 26). The same last authors reported better endocrine response in patients with PCOS and high LH level compared to those with normal LH level but high LH:FSH ratio (Abdel Gadir 65). A further benefit of laparoscopy in these cases is that it offers a good chance to examine the pelvis for other infertility factors at the same time. The risk of developing pelvic adhesions after ovarian drilling should be weighed against the prospective benefits expected in these patients. Such risk could be reduced by adopting principles used during microsurgery (Abdel Gadir 1993 68):

Metformin has been shown to be highly effective in augmenting clomiphene citrate activity for induction of ovulation in previously resistant patients (Siebert TI et al 2006, 69). This could lead to reduction in androgen production by the ovaries, better ovulation and reduced miscarriage rates. Reduction of insulin level has been shown to reduce ovarian cytochrome P450c17a activity and serum free testosterone (Nestler et al 1996, 70). Furthermore, it has been shown to reverse the metabolic and endocrine risk factors associated with increased miscarriage rate in women with PCOS. Other than reducing the level of androgens, it also reduces the levels of PAI-1, and luteinising hormone. However, it might take 4 months before a full molecular effect is achieved. Recent reports suggested that it might have a direct effect on the ovaries, even in women who are not insulin resistant (Tan S et al, 71).

It has been reported that 16-25% of 'normal' women had polycystic-appearing ovaries without any specific symptoms or signs (Polson 2, Abdel-Gadir 3, and Wong 4). However, many of them might have increased risks and similar morbidity as related to PCOS. They have been reported to show subtle metabolic (Carmina et al 72) and endocrine (Abdel-Gadir et al 1992 3, and Chang et al 8) abnormalities including:

Furthermore, they have the same risk of excessive response to induction of ovulation and ovarian hyperstimulation syndrome as patients with PCOS. In addition, Doppler studies showed normal women with PCO had similar uterine and ovarian blood flow as patients with PCOS (Zaidi et al 73). Accordingly, the notion that the presence of PCO in this group of women is totally ‘normal’ should be revised.

Polycystic ovaries could be seen in women with a wide range of different endocrine problems including:

Accordingly, total reliance on ultrasonography alone would create a diagnostic problem, and wise utilisation of endocrine investigations would be necessary (Abdel-Gadir et al 1992 3).

As many as 36.4% patients with hypothyroidism have shown PCO on ultrasound scan examinations (Abdel-Gadir 3). This could be a good representation of normal ovaries which changed polycystic due to abnormal external impulses, as alluded to before. Thyroxine is needed for the production of SHBG, and patients with hypothyroidism have lower levels than normal. This would lead to increased level of free androgens in circulation, which could be aromatised in the hypothalamus leading to changes in the gonadotrophins pulse secretion, which could affect the ovaries negatively. Secondly, high androgens could affect the ovaries directly leading to reduction in FSH and LH receptors, reduced aromatase activity, and development of PCO. Furthermore, the effect of hypothyroidism could lead to TRH induced secondary hyperprolactinaemia in these patients.

Polycystic ovaries have been reported in 50.0 % of patients with hyperprolactinaemia (Abdel Gadir et al 1992 3). High prolactin could affect the ovaries in the following ways:

Partial enzymatic deficiencies have been discussed thoroughly in chapter 4. They could be seen at puberty or early adult life, and are described to be of ‘adult onset’. The most common variety is partial deficiency of the 21a-hydroxylase enzyme to different extent in different patients. All adrenal enzymatic deficiencies are autosomal recessive genetical problems, inherited from either or both parents. The pattern of presentation could include symptoms and signs similar to those described for PCOS. In fact the ovaries could be polycystic in virtually 100% of the cases (Abdel-Gadir et al 1992 3). The American College of Obstetrics and Gynaecology recommended that all women with a suspected diagnosis of PCOS should be screened for 17a-hydroxyprogesterone levels.

Hypothalamic dysfunction is a diagnosis of omission, when all known causes capable of causing ovulatory dysfunction have been excluded. Many non measurable factors could be involved in these cases, such as severe stress which is known to affect GnRH pulse generation. Also weight-related problems could affect the ovaries, but they have been associated more with multicystic rather than polycystic ovaries. Medication could have a similar effect, especially so for antiepileptic drugs and all other drugs capable of affecting the brain neurotransmitters, and accordingly GnRH pulse generation.

Despite the lack of an overt endocrine dysfunction that could be revealed by a peripheral blood test, polycystic ovaries were reported in 23.7% of anovulatory patients with hypothalamic dysfunction (Abdel Gadir et al 1992 3). Such ovaries would behave the same way as any other polycystic ones with increased risk of hyperstimulation during induction of ovulation. Treatment of patients in this group should be tailored to their needs and their symptoms. This could involve induction of ovulation to facilitate conception. An oral contraceptive pill could be used to induce regular withdrawal bleeding and protect against unwanted pregnancies. Regular progestogen withdrawal bleeding every 8 weeks would guard again endometrial hyperplasia in patients who are not sexually active.

With advancing age, women with PCO tend to have more regular periods, lower circulating androgens (Winters et al 2000, (74)), and loss of the PCO pattern (Abdel Gadir et al 2009, 23). These changes could be secondary to the age related reduction in the total number of recruitable follicles and the increase in FSH level leading to a more favourable LH / FSH ratio. However, obese patients with PCOS are more likely to develop type II diabetes. A figure of 80% risk of type II diabetes by the age 40 years has been quoted in this subgroup. They are also at more at risk of carcinoma of the endometrium, especially in the presence of other risk factors including amenorrhoea, endometrial hyperplasia, and high blood pressure.

PCOS is a heterogeneous condition involving interlinked metabolic, endocrine and reproductive problems. Its exact cause is not yet well known, but many theories have been put forward to explain its development. It has a familial predisposition, though an exact genetic or chromosomal cause has not been established. More evidence is accumulating relating it to abnormal insulin resistance and hyperinsulinaemia. It should be treated as a general medical problem, rather than just a fertility issue. Controlling the metabolic disorder by reducing body weight and reducing insulin resistance should be the primary management objective. This would impact favourably on the endocrine and reproductive sides of the syndrome. The presence of ultrasonically diagnosed PCO in patients with menstrual irregularity and hyperprolactinaemia, or thyroid and adrenal dysfunction, stresses the need for a proper endocrine assessment. This would help in making a definitive diagnosis before starting any form of medical or surgical treatment solely on the ultrasonic findings.

1. Knochenhauer E, Key TJ, Kahsar-Miller M, Waggoner w, Boots LR and Azziz R. Prevalence of polycystic ovary syndrome in unselected black and white women of southeastern United States: a prospective study. J Clin Endocrinol Metab 1998; 83(9): 3078-3082.

2. Polson DW, Adam J, Wadsworth J and Franks S. Polycystic ovaries – a common finding in normal women. Lancet 1988; 1: 870-871.

3. Abdel Gadir, A., Khatim MS. Mowafi RS, Alnaser HM, Muharib NS and Shaw RW. Implications of ultrasonically diagnosed polycystic ovaries (1)-Correlation with basal hormonal profiles. Human Reproduction 1992; 7 (4), 453-457.

4. Wong LI, Morris RS, Legro R, Paulson RJ and Sauer MV. Isolated polycystic morphology in ovum donors predicts response to ovarian stimulation. Hum Rep 1995; 10: 524-528.

5. Shiono H. Dermatoglyphics in medicine. Am J Forensic Med Pathol 1986; 7 (2): 120-6.

6. Katznelson M and Goldman B. Fetal dermatoglyphics. Clin Genet 1982; 21 (4): 237-42.

7. Carmina E and Lobo RA. Do hyperandrogenic women with normal menses have polycystic ovary syndrome? Fertil Steril 1999; 71: 319-322.

8. Chang PL, Lindheim SR, Lowre C et al. Normal ovulatory women with polycystic ovaries have hyperandrogenic pituitary-ovarian response to gonadotrophin-releasing hormone agonist testing. JCE & M 2000; 85(3):995-1000.

9. Farquhar C. History of polycystic ovary syndrome. In: Polycystic Ovary Syndrome. Edited by: Gabor T. Kovacs. Cambridge University Press 2000, Chapter 2, pp 4-22.

10. O’Dowd MJ, Philipp EE and Phillip EE. Chromosomal, hormonal and psychogenic amenorrhoea and oligomenorrhoea. In: History of Obstetrics and Gynaecology. Informa Healthcare 2000, pp 307-316.

11.Stein IF and Leventhal ML. Amenorrhoea associated with bilateral polycystic ovaries. Am J Obstet Gynecol 1935; 29: 181-186.

12.McArthur JW, Ingersoll FM and Worchester J. Urinary excretion of interstitial-cell and follicle stimulating hormone activity by women with diseases of the reproductive system. J Clin Endocrinol Metab 1958; 18: 1202-1215.

13.De Vane GW, Czekala NM, Judd HL et al. Circulating gonadotrophins, oestrogens and androgens in polycystic ovarian disease. Am J. Obstet Gynecol 1975; 121: 496-500.

14.Rebar R, Judd HL and Yen SS, Rakoff J, Vandenberg G and Naftolin F. Characterisation of the inappropriate gonadotropin secretion in polycystic ovary syndrome. J Clin Invest 1976; 57(5): 1320-9.

15.Swanson M, Sauerbrei EE and Cooperberg PL. Medical implications of ultrasonically detected polycystic ovaries. J Clin Ultras 1981; 9(5): 219-222.

16.Hann LE, Hall DA, McArdle CR and Seibel M. Polycystic ovarian disease: sonographic spectrum. Radiology 1984; 150(2): 531-534

17.Adam J, Franks S, Polson DW and Mason HD. Multifollicular ovaries: clinical and endocrine features and response to pulsatile gonadotropin releasing hormone. Lancet 1985; 2: 1375-9

18.Kahn CR, Flier J, Bar RS, Archer JA, Gorden P, Martin MM and Roth J. They syndrome of insulin resistance and acanthosis nigricans. Insulin receptor disorder in man. N Engl J Med 1976; 294(14): 739-745.

19.Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome. Fertil Steril 2004; 81(1):19 – 25.

20.Hughesdon P. Morphology of morphogenesis of the stein-leventhal ovary and so called hyperthecosis. Obstet Gynecol Surv 1982; 37: 59-77.

21.Maciel GA, Baracat EC, Benda JA, Markham SA, Hensinger K, Chang RJ and Erickson GF. Stockpiling of transitional and classic primary follicles in ovaries of women with polycystic ovary syndrome. J Clin Endocrinol Metab 2004; 89: 5321-5327.

22.Webber LJ, Stubbs S, Stark J, Trew GH, Margara R, Hardy K and Frank S. Formation and early development of follicles in the polycystic ovaries. Lancet 2003; 362: 1017-1021.

23.Jonard S and Dewailly D. The follicular excess in polycystic ovaries, due to intraovarian hyperandrogenism, may be the main culprit for the follicular arrest. Hum Reprod Update 2004; 10(2): 107-110.

24.Abdel-Gadir A, Oyawoye O and Chander B. Coexistence of polycystic ovaries and uterine fibroids and their combined effect on the uterine arteries blood flow in relation to age and parity. The Journal of Reproductive Medicine 2009; 54: 347-352.

25.Yen SSC. Chronic anovulation caused by peripheral endocrine disorders. In Reproductive Endocrinology, Physiology, Pathophysiology and Clinical Management. Ed. Yen SSC and Jaffe MD, WB Saunders Company. Philadelphia, pp 441-499.

26.Abdel Gadir A, Khatim MS, Mowafi RS, Alnaser HM, Alzaid HGN and Shaw RW. Polycystic ovaries: Do these represent a specific endocrinopathy? British Journal of Obstetrics and Gynaecology 1991, 98, 300-305.

27.Abdel Gadir A, Mowafi RS, Alnaser HM, Alrashid AH, Alonezi OM and Shaw RW. Ovarian electrocautery versus human menopausal gonadotrophins and pure follicle stimulating hormone therapy in the treatment of patients with polycystic ovarian disease. Clin Endocrinol 1990; 33: 585-592.

28.Givens RJ, Andersen RN, Umstol ES and Wiser WL. Clinical findings and hormonal responses in patients with polycystic ovarian disease with normal versus elevated LH levels. Obstet Gynecol 1976; 47: 388-394.

29.Kahsar-Miller MD, Nixon C, Boots LR, Gor RC and Aziz R. Prevalence of polycystic ovary syndrome (PCOS) in first degree relatives of patients with PCOS. Fertil Steril 2001; 75: 53-58.

30.Barnes R, Rosenfield RL. Burstein S and Ehrmann DA. Pituitary ovarian response to nafarelin testing in the polycystic ovary syndrome. N Engl J Med 1989; 320(9): 559-65.

31.Rosenfield RL, Barnes RB, Cara JF and Lucky AW. Dysregulation of cytochrome P450c 17 alpha as the cause of polycystic ovarian syndrome. Fertil Steril 1990; 53(5): 785-791.

32.Goldzieher J and Green J. The polycystic ovary: 1. Clinical and histological features. J Clin Endocrinol Metab 1962; 22: 325-338.

33.Fisher ER, Gregorio r, Stephen T, Nolan S and Donowski TS. Ovarian changes in women with morbid obesity. Obstet Gynecol 1974; 44: 839-844.

34.Dunaif A, Xia j, Book CB, Schenker E and Tang Z. Excessive insulin receptor phosphorylation in cultured fibroblasts and in skeletal muscle. A potential mechanism for insulin resistance in polycystic ovary syndrome; F Clin Invest 1985; 96: 801-810.

35.Nestler JE, Jakubowicz DJ, Reamer P, Gunn RD and Allan G. Ovulatory and metabolic effects of a D-chiro-inositol in the polycystic ovary syndrome. N Engl J Med 1999; 340: 1314-1320.

36.Caro J G. Editorial: Leptin is normal in PCOS, an editorial about three ‘negative’ papers. J Clin Endocrinol Metab 1997; 82(6): 1685-6.

37.de Courten M, Zimmet P, Hodge A, Collins V, Nicolson M, Staten M, Dowes G and Alberti KG. Hyperleptinaemia: the missing link in the metabolic syndrome? Diab Med 1997; 14(3): 200-208.

38.Cohen B, Novick D and Rubinstein M. Modulation of insulin activity by leptin. Science 1996; 274. (5290): 1185 – 1188.

39. Egawa K, Nakashima K, Sharma PM, Maegawa H, Nagai Y, Kashiwagi A, Kikkawa R and Olefsky JM. Persistent activation of phosphatidylinositol 3-Kinase causes insulin resistance due to accelerated insulin-induced insulin receptor substrate-1 degradation in 3T3-L1 adipocytes. Endocrinology 2000; 141(6): 1930-1935.

40. Vauhkonen I, Niskanen L, Haffner S, Kainulainen S, Uusitupa M and Laakso M. Insulin resistant is associated with high serum leptin levels in offspring of patients with non-insulin-dependent diabetes mellitus. Eur J Endocrinol 1998; 139(6): 598-604.

41.Jacobs HS and Conway GS. Leptin, polycystic ovaries and polycystic ovary syndrome. Hum Reprod Update 1999; 5(2): 166-171.

42.Weyer C, Funahashi T, Tanaka S, Hotta K, Matsuzawa Y, Pratley RE and Totaranni PA. Hyppoadiponectinemia in obesity and type 2 diabetes: close association with insulin resistance and hyperinsulinaemia. J Clin Endocrinol Metab 2001; 86: 1930-1935.

43.Carmina E, Orio F, Palomba S, Cascella T, Longo RA, Colao AM, Lombardi G and Lobo RA. Evidence for altered adipocyte function in polycystic ovary syndrome. Eur J Endocrinol 2005; 152: 389 – 394.

44.Ferriman D and Gallwey JD. Clinical assessment of body hair growth in women. J Clin Endocrinol 1961; 21: 1440-1447.

45.Goldzieher J and Green J. Clinical and biochemical features of polycystic ovarian disease. Fertil Steril 1963; 14: 631-653.

46.Lobo O and Carmina E. The importance of diagnosing the polycystic ovary syndrome. Ann Intern Med 2000; 132(12): 989-993.

47.Pierpoint T, McKeigue PM, Isaacs AJ, Wild SH and Jacobs HS. Mortality of women with polycystic ovary syndrome at long-term follow-up. J Clin Epidemiol 1998; 51(7): 581-586.

48.Wild S, Pierpoint T, McKeigue P and Jacobs H. Cardiovascular disease in women with polycystic ovary syndrome at long-term follow-up: a retrospective cohort study. Clin Endocrinol 2000; 52(5): 595-600.

49.Gisclard V, Miller VM and Vanhoutte PM. Effect of 17 beta oestradiol on endothelium-dependent responses in the rabbit. J Pharmacol Exp Ther 1988; 244: 19-22.

50.Vgontzaz AN, Legro RS, Bixler EO, Grayev A, Kales A and Chrousos GP. Polycystic ovary syndrome is associated with obstructive sleep apnoea and daytime sleepiness: Role of insulin resistance. J Clin Endocrinol Metab 2001; 86(2): 517-520.

51.Jakubowicz DJ, Essah PA, Seppäläl M, Jakubowicz S, Baillargeon JP, Koistinen R and Nestler JE. Reduced serum glycodelin and insulin like growth factor binding protein01 in women with polycystic ovary syndrome during the first trimester of pregnancy. J Clin Endocrinol Metab 2004; 89: 833-839.

52.Suikkari AM, Ruutianinen K, Erkkola R, Seppala M. Low levels of low molecular weight insulin-like growth factor-binding protein in patients with polycystic ovarian disease. Hum Reprod 1989; 4: 136-139.

53.Essah PA, Cheang KI and Nestler JE. The pathophysiology of miscarriages in women with polycystic ovary syndrome. Review and proposed hypothesis of mechanisms involved. Hormones 2004; 3(4): 221-227.

54.Bjercke S, Dale PO, Tanbo T, Storeng R, Ertzeid G and Abyholm T. Impact of insulin resistance on pregnancy complications and outcome in women with polycystic ovary syndrome. Gynecol Obstet Invest 2002; 54(2): 94-98.

55.Khatab S, Mohsen IA, Foutouh IA, Ramadan A, Moaz M and Al- Inany H. Metformin reduces abortion in pregnant women with polycystic ovarian syndrome. Gynecol Endocrinol 2006; 22(12): 680-684

56.Benson S, Hahn S, Tan S, Mann K, Janssen OE, Schedlowski M and Elsenbruch S. Prevalence and implications of anxiety in polycystic ovary syndrome: results of an internet-based survey in Germany. Hum Reprod 2009; 24(6): 1446-1451.

57.Bishop SC, Basch S and Futterweit W. Polycystic ovary syndrome, associated depression and affective disorders. Endocr Pract 2009; 6: 1-31.

58.Adali E, Yildizhan R, Kurdoglu M, Kolusari A, Edirne T, Sahin HG, Yildizhan B and Kamaci M. They relationship between clinico-biochmemical characteristics and psychiatric distress in young women with polycystic ovary syndrome. J Int Med Res 2008; 36(6): 1188-1196.

59.Timonen M, Laakso M, Jokelainen J, Rajala U, Meyer-Rochow VB and Keinänen-Kiukaanniemi S. Insulin resistance and depression: cross sectional study. BMJ 2005; 330: 17-18.

60.Benson S, Arch PC, Tan S, Mann K, Rifaie N, Janssen OE, Schedlowski M and Elsenbruch S. Disturbed stress response in women with polycystic ovary syndrome. Psychoneuroendocrinology 2009, 34(5): 727-735.

61.Chaudhry MU and Simmons DL. Case of the month, hepatic and renal failure in a patient taking troglitazone and metformin. J Ark Med Soc 2001; 98(1): 16-19.

62.Nammour FE, Fayad NF and Peikin SR. Metformin-induced cholestatic hepatitis; Endocr Pract 2003; 9(4): 3070309

63.Kutoh E. Possible metformin-induced hepatotoxicity. Am J Geriatr Pharmacother 2005; 3(4): 270-273.

64.Santen RJ and Bardin CW. Episodic luteinising hormone secretion in man. Pulse analysis, clinical interpretation, physiologic mechanisms. J Clin Invest 1973; 52: 2617-2628.

65.Abdel Gadir A, Khatim MS, Alnaser HM, Mowafi RS and Shaw RW. Ovarian electrocautery: Responders versus non-responders. Gynaecological Endocrinology 1993; 7, 43-48.

66.Bohrer M and Kemmann E. Risk factors for spontaneous abortion in menotropin-treated women. Fertil Steril 1987; 48: 571-575

67.Gjönnaess H. Ovarian electrocautery in the treatment of women with polycystic ovary syndrome (PCOS). Factors affecting results. Acta Obstet Gynecol Scand 1994; 73(5): 407-12.

68.Abdel Gadir A. Ovarian surgery. In: The control and stimulation of follicular growth. Advances in Reproductive Endocrinology. Volume 5, PP 111-124. Edit RW Shaw, The Parthenon Publishing Group, Casterton Hall, Carnforth, Lancs. 1993.

69.Siebert TI, Kruger TF, Steyn DW and Nosarka S. Is addition of metformin efficacious in the treatment of clomiphene citrate-resistant patients with polycystic ovary syndrome? A structured literature review. Fertil Steril 2006; 86(5): 1432 – 1437.

70.Nestler JE and Jakubowicz DJ. Decreases in ovarian cytochrome P450c17a activity and serum free testosterone after reduction of insulin secretion in polycystic ovary syndrome. New Eng J Med 1996; 335(9): 617-623.

71.Tan S, Hahn S, Benson S, Dietz T, Lahner H, Moeller LC, Schmidt M, Elsenbruch S, Kimmig R, Mann K and Janssen OE. Metformin improves polycystic ovary syndrome symptoms irrespective of pre-treatment insulin resistance. Eur J Endocrinol 2007; 157: 669–676.

72.Carmina E, Wong L, Chang L, Paulson RJ, Sauer MV, Stanczyk FS and Lobo RA. Endocrine abnormalities in ovulatory women with polycystic ovaries on ultrasound. Hum Rep 1997; 12(5): 905-909.

73.Zaidi J. Blood flow changes in the ovarian and uterine arteries in women with normal and polycystic ovaries. Hum Fertil (Camb) 2000; 3: 194-198.

74.Winters SJ, Talbott E, Guzick DS, Zborowski J and McHugh KP. Serum testosterone levels decrease in middle age in women with polycystic ovary syndrome. Fertil Steril 2000; 73: 724-729.

75.Takahashi K, Yoshino K, Nishigaki A, Eda Y and Kitao M. On the relationship between endocrine and ovulatory abnormalities and polycystic as diagnosed by ultrasonography. Int J Fertil 1992; 37(4): 222-6.

76.Gonzalez F, Chang L, Horab T and Lobo RA. Evidence for heterogeneous etiologies of adrenal dysfunction in polycystic ovary syndrome. Fertil Steril 1996; 6(3): 354-361.

77.Legro RS, Finegood D and Dunaif A. A fasting glucose to insulin ratio is a useful measure of insulin sensitivity in women with polycystic ovary syndrome. J Clin Endocrinol Metab 1998; 83(8): 2694-2698.

78.Chang RJ, Lindheim SR, Lowre C, Ferin M, Gonzalez F, Berglund L, Carmina E, Sauer MV and Lobo RA. Normal ovulatory women with polycystic ovaries have hyperandrogenic pituitary ovarian responses to gonadotropin-releasing hormone agonist testing. J Clin Endocrinol Metab 2000; 85(3): 995-1000.

79.Lobo RA and Carmina E. The importance of diagnosing polycystic ovary syndrome. Annals of Int Med 2000; 132(12): 989-93.

80.Ehrmann DA, Sturis J, Byrne MM, Karrison T, Rosenfield RL, Polonsky KS. Insulin secretory defects in polycystic ovary syndrome. Relationship to insulin sensitivity and family history of non-insulin-dependent diabetes mellitus. J Clin Invest 1995; 96: 520-527.