The current understanding of polycystic ovary syndrome

Transcription

1 DOI: /toag Keywords hyperandrogenism, insulin resistance, metformin, ovulation induction, polycystic ovary syndrome The current understanding of polycystic ovary syndrome Adam Balen The polycystic ovary syndrome (PCOS) is a heterogeneous condition which consists of the presence of two out of the following three criteria: oligo- and/or anovulation, hyperandrogenism (clinical and/or biochemical), polycystic ovaries, with the exclusion of other aetiologies. The association between abdominal obesity and hyperinsulinaemia plays a key role in the pathophysiology of PCOS and worsens both reproductive and long-term health. Management is directed towards an individual s needs. A variety of therapies may be used to regulate the menstrual cycle, induce ovulation and improve signs of hyperandrogenism. Lifestyle improvement, however, is the most successful modifier of long-term health. Author details Adam Balen MD FRCOG, Consultant Obstetrician and Gynaecologist and Subspecialist in Reproductive Medicine and Surgery, Clarendon Wing, The General Infirmary, Leeds, LS2 9NS, UK. adam.balen@leedsth.nhs.uk Introduction The polycystic ovary syndrome (PCOS) is a heterogeneous collection of signs and symptoms that, gathered together, form a spectrum of a disorder with a mild presentation in some and in others a severe disturbance of reproductive, endocrine and metabolic function (Box 1). The pathophysiology of the PCOS appears to be multifactorial and polygenic.the definition of the syndrome has been much debated. Key features include menstrual cycle disturbance, hyperandrogenism and obesity.there are many extra-ovarian aspects to the pathophysiology of PCOS, yet ovarian dysfunction is central.at a joint consensus meeting of the American Society for Reproductive Medicine and the European Society of Human Reproduction and Embryology (ASRM/ESHRE) a refined definition of the PCOS was agreed,namely the presence of two out of the following three criteria: oligo- and/or anovulation hyperandrogenism (clinical and/or biochemical) polycystic ovaries with the exclusion of other aetiologies. 1 The morphology of the polycystic ovary has been redefined as an ovary with 12 or more follicles measuring 2 9 mm in diameter and/or increased ovarian volume (> 10 cm 3 ). 2 There is considerable heterogeneity of symptoms and signs among women with PCOS and for an individual these may change over time. 3 The PCOS is familial and various aspects of the syndrome may be differentially inherited. Polycystic ovaries can exist without clinical signs of the syndrome, which may then become expressed over time. There are a number of factors that affect expression of PCOS; for example, a gain in weight is associated with a worsening of symptoms while weight loss may ameliorate the endocrine and metabolic profile and symptomatology. 4 Genetic studies have identified a link between PCOS and disordered insulin metabolism. This indicates that the syndrome may be the presentation of a complex genetic trait disorder. The features of obesity, hyperinsulinaemia and hyperandrogenaemia, which are commonly seen in PCOS, are also known to be factors which confer an increased risk of cardiovascular disease and type 2 (non-insulin-dependent) diabetes mellitus. 5 There are studies which indicate that women with PCOS have an increased risk for these diseases which pose long-term risks for health, and this evidence has prompted debate as to the need for screening women for PCOS. 6 Various factors influence ovarian function and fertility is adversely affected by an individual being overweight or having elevated serum concentrations of luteinising hormone (LH). Strategies to induce ovulation include weight loss, oral antioestrogens (principally clomiphene citrate), parenteral gonadotrophin therapy and 66

2 Box 1. Signs and symptoms of polycystic ovary syndrome Symptoms Hyperandrogenism (acne, hirsutism, alopecia not virilisation) Menstrual disturbance Infertility Obesity Sometimes asymptomatic, with polycystic ovaries visible on ultrasound scan Serum endocrinology Fasting insulin (not routinely measured; insulin resistance or impaired glucose tolerance assessed by glucose tolerance test) Androgens (testosterone and androstenedione) Luteinising hormone, usually normal follicle stimulating hormone Sex hormone-binding globulin, results in elevated free androgen index oestradiol, oestrone (neither measured routinely as wide range of values) Prolactin (Figure 1). Although the ultrasound criteria for the diagnosis of polycystic ovaries have not, until now, been universally agreed, the characteristic features are accepted as being an increase in the number of follicles and the amount of stroma as compared with normal ovaries, resulting in an increase in ovarian volume.the term polycystic ovary in some respects adds to the confusion that surrounds its diagnosis. The cysts are not cysts, in the sense that they do contain oocytes. So truly it should be called a polyfollicular ovary, to reflect the finding that the cysts are actually follicles whose development has been arrested. Indeed, the prerequisite of a certain number of cysts may be of less relevance than the volume of ovarian stroma or of the ovary itself, which has been shown to closely correlate with serum testosterone concentrations. Possible late sequelae Diabetes mellitus Dyslipidaemia Hypertension and cardiovascular disease Endometrial carcinoma Breast cancer laparoscopic ovarian surgery.there have been no adequately powered randomised studies to determine which of these therapies provides the best overall chance of a continuing pregnancy. Women with PCOS are at risk of ovarian hyperstimulation syndrome (OHSS) and so ovulation induction has to be carefully monitored with serial ultrasound scans. The realisation of an association between hyperinsulinaemia and PCOS has resulted in the use of insulin sensitising agents, such as metformin, which appear to ameliorate the biochemical profile and improve reproductive function. What is a polycystic ovary? Polycystic ovaries are commonly detected by ultrasound or other forms of pelvic imaging, with estimates of the prevalence in the general population being in the order of 20 33% 7,8 Jonard et al. 9 studied 214 women with PCOS (oligo- or amenorrhoea, elevated serum LH and/or testosterone, and/or ovarian area greater than 5.5 cm 2 ) and 112 with normal ovaries to determine the importance of follicle number per ovary (FNPO). A FNPO of 12 follicles 2 9 mm gave the best threshold for the diagnosis of PCOS (sensitivity 75%, specificity 99%). 9 The authors suggest that intraovarian hyperandrogenism promotes excessive early follicular growth up to 2 5 mm, with more follicles able to enter the growing cohort which then become arrested at the 6 9 mm size. At the joint ASRM/ESHRE consensus meeting, a refined definition of PCOS was agreed, encompassing a description of the morphology of the polycystic ovary. According to the available literature, the criteria fulfilling sufficient Figure 1. Transvaginal ultrasound scan of a polycystic ovary 67

3 specificity and sensitivity to define the polycystic ovary are the presence of 12 or more follicles measuring 2 9 mm in diameter and/or increased ovarian volume (> 10 cm 3 ). 2 If there is a follicle greater than 10 mm in diameter, the scan should be repeated at a time of ovarian quiescence in order to calculate volume and area.the presence of a single polycystic ovary is sufficient to provide the diagnosis. The distribution of the follicles and the description of the stroma are not required in the diagnosis. Increased stromal echogenicity and/or stromal volume are specific to polycystic ovaries but it has been shown that the measurement of the ovarian volume (or area) is a good surrogate for the quantification of the stroma in clinical practice. A woman having polycystic ovaries in the absence of an ovulation disorder or hyperandrogenism (asymptomatic polycystic ovaries) should not be considered as having PCOS, although she may develop symptoms over time, for example if she gains weight (Table 1). Genetics of PCOS PCOS has long been noted to have a familial component. 10 Genetic analysis has been hampered by the lack of a universal definition for PCOS. Most of the criteria used for diagnosing PCOS are continuous traits, such as degree of hirsutism, level of circulating androgens, extent of menstrual irregularity and ovarian volume and morphology. To perform genetic analyses, these continuous variables have to be transformed into nominal variables. Family studies have revealed that about 50% of firstdegree relatives have PCOS, suggesting a dominant mode of inheritance. Commonly firstdegree male relatives appear more likely to have premature baldness. As hyperandrogenism is a key feature of PCOS it is logical to explore the critical steps in steroidogenesis and potential enzyme dysfunction. A number of studies have found an abnormality with the cholesterol side chain cleavage gene (CYP11a), although this is debated. It has been hypothesised that polymorphisms in the INSR gene that induce mild changes in insulin receptor function may contribute to the development of PCOS, as it is unlikely that a major mutation is present given the wide variability of insulin resistance in women with PCOS. The insulin gene variable number of tandem repeats (VNTR) minisatellite locus which lies 5` to the insulin gene on chromosome 11p15.5 and regulates the expression of the insulin gene. The class III allele of the insulin VNTR gene is associated with anovulatory PCOS. While it is beyond the scope of this review to provide a more detailed account of the various genetic studies that have been performed to date, there is substantial evidence for genetic defects in steroidogenesis and insulin action being involved in the pathogenesis of PCOS. 10 The pathophysiology of PCOS therefore involves a combination of genetic abnormalities combined with environmental factors, such as nutrition and body weight, which affect expression of the syndrome. Table 1. Investigations for polycystic ovary syndrome Test Normal range a Additional points Pelvic ultrasound To assess ovarian morphology Transabdominal scan usually satisfactory in women who are not sexually and endometrial thickness active (depends upon body habitus) Testosterone (T) nmol/l A total testosterone is adequate for general screening. It is unnecessary to measure other androgens unless total testosterone is > 5 nmol/l, in which case referral is indicated Sex hormone-binding nmol/l Insulin suppresses SHBG, resulting in a high FAI in the presence globulin (SHBG) of a normal total T. Free androgen index (FAI): < 5 The measurement of SHBG is not required in routine practice and will not T x 100/SHBG affect management Oestradiol Measurement is unhelpful Oestrogenisation may be confirmed by endometrial assessment to diagnosis Luteinising hormone (LH) 2 10 iu/l FSH and LH best measured during days 1 3 of a menstrual bleed. If oligo- or amenorrhoeic then random samples are taken Follicle stimulating hormone (FSH) 2 8 iu/l Prolactin iu/l Measure if oligo- or amenorrhoeic Thyroid function, thyroid-stimulating < 500 mu/l hormone Fasting insulin < 30 mu/l Not routinely measured; insulin resistance assessed by glucose tolerance test a May vary with local laboratory assays 68

4 Table 2. Definitions of glucose tolerance after a 75-g glucose tolerance test (GTT) Diabetes mellitus Impaired glucose Impaired fasting tolerance glycaemia Fasting glucose (mmol/l) 7.0 < and < hour glucose (mmol/l) < 7.8 Action Refer to diabetic Dietary advice. Check Dietary advice. Check clinic fasting glucose annually fasting glucose annually Racial differences in expression of PCOS The highest reported prevalence of polycystic ovaries has been 52% among South Asian immigrants in Britain, of whom 49.1% had menstrual irregularity. 11 Rodin et al. 11 demonstrated that South Asian women with polycystic ovaries had a comparable degree of insulin resistance to controls with established type 2 diabetes mellitus. Insulin resistance and hyperinsulinaemia are common antecedents of type 2 diabetes, with a high prevalence in South Asians.Type 2 diabetes also has a familial basis, inherited as a complex genetic trait that interacts with environmental factors, chiefly nutrition, commencing from fetal life. We have found that South Asians with anovulatory PCOS have greater insulin resistance and more severe symptoms of the syndrome than anovulatory white women with PCOS. 12 Furthermore, we have found that women from South Asia living in the UK appear to express symptoms at an earlier age than their white British counterparts. Glucose tolerance Women who are obese, and also many slim women with PCOS, may have insulin resistance and elevated serum concentrations of insulin (normally less than 30 mu/l fasting). We suggest that a 75-g oral glucose tolerance test (GTT) be performed in women with PCOS and a body mass index (BMI) greater than 30 kg/m 2, with an assessment of the fasting and two-hour glucose concentration (Table 2). It has been suggested that South Asian women should have an assessment of glucose tolerance if their BMI is greater than 25 kg/m 2 because of the greater risk of insulin resistance at a lower BMI than is seen in the white population. Health consequences of PCOS Obesity and metabolic abnormalities are recognised risk factors for the development of ischaemic heart disease in the general population, and these are also recognised features of PCOS. The questions are whether women with PCOS are at an increased risk of ischaemic heart disease and whether this will occur at an earlier age than women with normal ovaries. The basis for the idea that women with PCOS are at greater risk of cardiovascular disease is that these women are more insulin resistant than weight-matched controls and that the metabolic disturbances associated with insulin resistance are known to increase cardiovascular risk in other populations. Insulin resistance is defined as a diminution in the biological responses to a given level of insulin. In the presence of an adequate pancreatic reserve, normal circulating glucose levels are maintained at higher serum insulin concentrations. In the general population, cardiovascular risk factors include insulin resistance, obesity, glucose intolerance, hypertension and dyslipidaemia. There have been many studies demonstrating the presence of insulin resistance and corresponding hyperinsulinaemia in both obese and non-obese women with PCOS. 5,6 Obese women with PCOS have consistently been shown to be more insulin resistant than weight-matched controls. It appears that obesity and PCOS have an additive effect on the degree and severity of the insulin resistance and subsequent hyperinsulinaemia in this group of women. The insulin resistance causes compensatory hypersecretion of insulin, particularly in response to glucose, so euglycaemia is usually maintained at the expense of hyperinsulinaemia. Insulin resistance is restricted to the extra-splanchnic actions of insulin on glucose dispersal. The liver is not affected (hence the fall in sex hormone-binding globulin, SHBG, and high-density lipoprotein, HDL), neither is the ovary (hence the menstrual problems and hypersecretion of androgens) nor the skin, hence the development of acanthosis nigricans. Women with PCOS who are oligomenorrhoeic are more likely to be insulin 69

5 resistant than those with regular cycles, irrespective of their BMI, with an intermenstrual interval correlating with the degree of insulin resistance. Women with PCOS have a greater truncal abdominal fat distribution as demonstrated by a higher waist to hip ratio. The central distribution of fat is independent of BMI and associated with higher plasma insulin and triglyceride concentrations and reduced HDL and cholesterol concentrations. From a practical point of view, if the measurement of waist circumference is greater than 88 cm, there will be excess visceral fat and an increased risk of metabolic problems. Thus, there is evidence that insulin resistance, central obesity and hyperandrogenaemia have an adverse effect on lipid metabolism, yet these are surrogate risk factors for cardiovascular disease. Pierpoint et al. 13 reported the mortality rate in 1028 women diagnosed as having polycystic ovary syndrome between 1930 and All the women were older than 45 years and 770 women had been treated by wedge resection of the ovaries. Seven hundred and eighty six women were traced; the mean age at diagnosis was 26.4 years and average duration of follow up was 30 years. There were 59 deaths, of which 15 were from circulatory disease. Of these 15 deaths, 13 were from ischaemic heart disease. There were six deaths from diabetes as an underlying or contributory cause compared with the expected 1.7 deaths. The standard mortality rate both overall and for cardiovascular disease was not higher in the women with PCOS compared with the national mortality rates in women, although the observed proportion of women with diabetes as a contributory or underlying factor leading to death was significantly higher than expected (OR 3.6, 95% CI ). Thus, despite surrogate markers for cardiovascular disease, in this study, no increased rate of death from cardiovascular disease could be demonstrated. PCOS in younger women The majority of studies which have identified the risk factors of obesity and insulin resistance in women with PCOS have investigated adult populations, commonly including women who have presented to specialist endocrine or reproductive clinics. However, PCOS has been identified in much younger populations, 8 in which women with increasing symptoms of PCOS were found to be more insulin resistant. These data emphasise the need for long-term prospective studies of young women with PCOS in order to clarify the natural history and to determine which women will be at risk of diabetes and cardiovascular disease later in life.a study of women with PCOS and a mean age of 39 years followed over a period of six years found that 9% of those with normal glucose tolerance developed impaired glucose tolerance and 8% developed type 2 diabetes 14 while 54% of women with impaired glucose tolerance at the start of the study had diabetes at follow up. The risks of disease progression, not surprisingly, were greatest in those who were overweight. Endometrial cancer Endometrial adenocarcinoma is the second most common female genital malignancy but only 4% of cases occur in women less than 40 years of age.the risk of developing endometrial cancer has been shown to be adversely influenced by a number of factors including obesity, long-term use of unopposed oestrogens, nulliparity and infertility. Women with endometrial carcinoma have had fewer births compared with controls and it has also been demonstrated that infertility per se gives a relative risk of two. Hypertension and type 2 diabetes mellitus have long been linked to endometrial cancer, conditions that are now known also to be associated with PCOS. The true risk of endometrial carcinoma in women with clearly defined PCOS is, however, difficult to ascertain. 15 Endometrial hyperplasia may be a precursor to adenocarcinoma, although the rate of progression is difficult to predict. While the degree of risk has not been clearly defined, it is generally accepted that for women with PCOS who experience amenorrhoea or oligomenorrhoea, the induction of artificial withdrawal bleeds to prevent endometrial hyperplasia is prudent management. 6 Indeed it is considered important that women with PCOS shed their endometrium at least every three months. For those with oligo- or amenorrhoea who do not wish to use cyclical hormone therapy an ultrasound scan to measure endometrial thickness and morphology is recommended every 6 12 months (depending upon menstrual history). An endometrial thickness greater than 10 mm in a woman with amenorrhoea warrants an artificially induced bleed, which should be followed by a repeat ultrasound scan and endometrial biopsy if the endometrium has not been shed. Another option is to consider a progestogen-secreting intrauterine system, such as the Mirena. 70

6 Breast cancer Obesity, hyperandrogenism and infertility occur frequently in PCOS and are features known to be associated with the development of breast cancer. However, studies examining the relationship between PCOS and breast carcinoma have not always identified a significantly increased risk.the study by Coulam et al. 16 calculated a relative risk of 1.5 (95% CI ) for breast cancer in the group of women with chronic anovulation, which was not statistically significant. After stratification by age, however, the relative risk was found to be 3.6 (95% CI ) in the postmenopausal age group. Pierpoint et al. 13 assessed mortality from the national registry of deaths and standardised mortality rates calculated for women with PCOS compared with the normal population. The average follow-up period was 30 years.the standardised mortality rate for all neoplasia was 0.91 (95% CI ) and for breast cancer 1.48 (95% CI ). Breast cancer was the leading cause of death in this cohort. Ovarian cancer In recent years there has been much debate about the risk of ovarian cancer in women with infertility, particularly in relation to the use of drugs to induce superovulation for assisted conception procedures. The risk of ovarian cancer appears to be inherently increased in women who have multiple ovulations, that is those who are nulliparous (possibly because of infertility) with an early menarche and late menopause. Thus, it may be that inducing multiple ovulations in women with infertility will increase their risk a notion that is by no means proven. Women with PCOS who are oligo- or anovulatory might therefore be expected to be at low risk of developing ovarian cancer if it is the lifetime number of ovulations rather than pregnancies that is critical. Ovulation induction to correct anovulatory infertility aims to induce unifollicular ovulation and so, in theory, should raise the risk of a woman with PCOS to that of a normal ovulating woman.the polycystic ovary, however, is notoriously sensitive to stimulation and it is only with the development of high-resolution transvaginal ultrasonography that the rate of unifollicular ovulation has attained acceptable levels.there are a few studies which have addressed the possibility of an association between polycystic ovaries and ovarian cancer.the results are conflicting and the ability to generalise is limited owing to problems with the study designs. In the large UK study of Pierpoint et al., 13 the standardised mortality rate for ovarian cancer was 0.39 (95% CI ). Management Obesity The clinical management of a woman with PCOS should be focused on her individual problems. Obesity worsens both symptomatology and the endocrine profile and so obese women (BMI greater than 30 kg/m 2 ) should therefore be encouraged to lose weight. Weight loss improves the endocrine profile, the likelihood of ovulation and a healthy pregnancy. Much has been written about diet and PCOS. The right diet for an individual is one that is practical, sustainable and compatible with her lifestyle. It is sensible to keep carbohydrate content down and to avoid fatty foods. It is often helpful to refer to a dietician. Anti-obesity drugs may help with weight loss. These can be prescribed by general practitioners and their use must be closely monitored. Menstrual irregularity Amenorrhoeic women with PCOS are not oestrogen deficient and are not at risk of osteoporosis. Indeed, they are oestrogen replete and at risk of endometrial hyperplasia (see above). The easiest way to control the menstrual cycle is the use of a low-dose combined oral contraceptive preparation. This will result in an artificial cycle and regular shedding of the endometrium. An alternative is a progestogen (such as medroxyprogesterone acetate or dydrogesterone) for 12 days every one to three months to induce a withdrawal bleed. It is also important once again to encourage weight loss. Hyperandrogenism and hirsutism The bioavailability of testosterone is affected by the serum concentration of SHBG. High levels of insulin lower the production of SHBG and so increase the free fraction of androgen. Elevated serum androgen concentrations stimulate peripheral androgen receptors, resulting in an increase in 5-alpha reductase activity directly increasing the conversion of testosterone to the more potent metabolite, dihydrotestosterone. Women with PCOS do not become virilised (that is, do not develop deepening of the voice, increased muscle mass, breast atrophy or clitoromegaly). A total testosterone of greater than 5 nmol/l or rapid onset of signs of hyperandrogenism requires further investigation. Late-onset congenital adrenal hyperplasia (CAH) is not common in the UK but is more prevalent in certain ethnic groups (e.g. Mediterranean, South American and some Jewish populations). 71

7 Screening for CAH by measuring 17- hydroxyprogesterone or performing adrenocorticotrophic hormone-stimulation tests is therefore not required in the absence of virilisation or a significantly elevated testosterone concentration. Hirsutism is characterised by terminal hair growth in a male pattern of distribution, including chin, upper lip, chest, upper and lower back, upper and lower abdomen, upper arm, thigh and buttocks. A standardised scoring system, such as the modified Ferriman and Gallwey score, may be used to evaluate the degree of hirsutism before and during treatments. Many women attend having already tried cosmetic techniques and so it may be difficult to obtain a baseline assessment. Drug therapies may take six to nine months or longer before any improvement of hirsutism is perceived. Physical treatments including electrolysis, waxing and bleaching may be helpful while waiting for medical treatments to work. Electrolysis is time-consuming, painful and expensive and should be performed by an expert practitioner. Regrowth is not uncommon and there is no really permanent cosmetic treatment. Laser and photothermolysis techniques are more expensive but may have a longer duration of effect. Comparative studies, however, have not been performed. Repeated treatments are required for a near permanent effect because only hair follicles in the growing phase are obliterated at each treatment. Hair growth occurs in three cycles so six to nine months of regular treatments are typical. Women should be appropriately selected (dark hair on fair skin is best) and warned that complete hair removal cannot be guaranteed and some scarring may occur. Medical regimens should stop further progression of hirsutism and decrease the rate of hair growth. Adequate contraception is important in women of reproductive age, as transplacental passage of anti-androgens may disturb the genital development of a male fetus. First-line therapy has traditionally been the preparation co-cyprindiol (Dianette, Schering Health),which contains ethinyloestradiol (35 g) in combination with cyproterone acetate (2 mg). Addition of higher doses of the synthetic progestogen cyproterone acetate ( mg) does not appear to confer additional benefit 17 but these are sometimes prescribed for the first ten days of each 21-day cycle for women who are particularly resistant to treatment with Dianette alone. The effect on acne and seborrhoea is usually evident within a couple of months. Cyproterone acetate can rarely cause liver damage and liver function should be checked regularly (after six months and then annually). Once symptom control has been obtained it is advisable to switch to a combined oral contraceptive pill containing a lower dose of ethinyloestradiol because of concerns about increased risk of thromboembolism with Dianette. Spironolactone is a weak diuretic with antiandrogenic properties and may be used at a daily dose of mg in women in whom the combined oral contraceptive pill is contraindicated. Drosperinone is a derivative of spironolactone and contained in the new combined oral contraceptive pill, Yasmin (ethinyloestradiol with drospirenone, Schering Health), which we are currently evaluating in women with PCOS. Other anti-androgens, such as ketoconazole, finasteride and flutamide, have been tried but are not widely used in the UK for the treatment of hirsutism in women, owing to their adverse effects. Furthermore, they are no more effective than cyproterone acetate. Metformin also appears to improve hirsutism. Infertility Weight loss is important to improve the prospects of both spontaneous and druginduced ovulation. In addition, overweight women with PCOS are at increased risk of obstetric complications, including gestational diabetes mellitus and pre-eclampsia. Ovulation can be induced with the anti-oestrogen clomiphene citrate ( mg) taken from days two to six of a natural or artificially induced bleed. While clomiphene is successful in inducing ovulation in over 80% of women, pregnancy only occurs in about 40% of them. Clomiphene citrate should only be prescribed in a setting where ultrasound monitoring is available (and performed) in order to minimise the 10% risk of multiple pregnancy and to ensure that ovulation is taking place. 18 A daily dose of more than 100 mg rarely confers any benefit and can cause thickening of the cervical mucus, which can impede passage of sperm through the cervix. Once an ovulatory dose has been reached, the cumulative conception rate continues to increase for up to ten to twelve cycles. 19 The therapeutic options for women with anovulatory infertility who are resistant to antioestrogens are either parenteral gonadotrophin therapy or laparoscopic ovarian diathermy. 20 The polycystic ovary is very sensitive to stimulation by exogenous hormones, so it is important to 72

8 start with very low doses of gonadotrophins. Follicular development must be carefully monitored by ultrasound scans. The advent of transvaginal ultrasonography has enabled the multiple pregnancy rate to be reduced to less than 5% because of its higher resolution and clearer view of the developing follicles. Cumulative conception and live birth rates after six months may be 62% and 54%, respectively, and after 12 months 73% and 62%, respectively. 21 Close monitoring should enable treatment to be suspended if more than two mature follicles develop, as the risk of multiple pregnancy increases. Women with PCOS are also at increased risk of developing OHSS. This occurs if too many follicles (greater than 10 mm) are stimulated and results in abdominal distension, discomfort, nausea, vomiting and sometimes difficulty in breathing.the mechanism for OHSS is thought to be secondary to activation of the ovarian renin angiotensin pathway and excessive secretion of vascular endothelial growth factor (VEGF). The ascites, pleural and pericardial effusions exacerbate this serious condition and the resultant haemoconcentration can lead to thromboembolism. The situation worsens if a pregnancy has resulted from the treatment as human chorionic gonadotrophin from the placenta further stimulates the ovaries. Hospitalisation is sometimes necessary in order for intravenous fluids and heparin to be given to prevent dehydration and thromboembolism. Although the OHSS is rare, it is potentially fatal and should be avoidable with appropriate monitoring of gonadotrophin therapy. Ovulation induction following ovarian diathermy is free of the risks of multiple pregnancy and ovarian hyperstimulation and does not require intensive ultrasound monitoring. Laparoscopic ovarian diathermy has taken the place of wedge resection of the ovaries (which resulted in extensive periovarian and tubal adhesions) and carries a reduced risk of multiple pregnancy compared with gonadotrophin therapy in the treatment of clomiphineinsensitive PCOS. A meta-analysis has shown that pregnancy rates are greater with six months gonadotrophin therapy compared with six months after laparoscopic ovarian diathermy, although by twelve months the pregnancy rates are similar. 22 Insulin-sensitising agents and metformin A number of pharmacological agents have been used to amplify the physiological effect of weight loss, notably metformin. This biguanide inhibits the production of hepatic glucose and enhances the sensitivity of peripheral tissue to insulin, thereby decreasing insulin secretion. It has been shown that metformin ameliorates hyperandrogenism and abnormalities of gonadotrophin secretion in women with PCOS and can restore menstrual cyclicity and fertility. The insulinsensitising agent troglitazone also appears to significantly improve the metabolic and reproductive abnormalities in PCOS. However, this product was withdrawn from the market in December 1997 because of reports of deaths from hepatotoxicity due to its metabolites. Newer insulin-sensitising agents, such as rosiglitazone and pioglitazone, are currently being evaluated. A Cochrane review has confirmed a beneficial effect of metformin in improving rates of ovulation when compared with placebo and also improving both rates of ovulation and pregnancy when used with clomiphene citrate compared with clomiphene citrate alone. 23 The data indicate that serum concentrations of insulin and androgens improve, although, contrary to popular belief, body weight does not fall. Metformin therapy may be commenced after appropriate screening and advice about diet, lifestyle and exercise for anovulatory women with PCOS who wish to conceive. The usual dose is either 850 mg twice daily or 500 mg three times daily. Baseline investigations should include an oral GTT, full blood count and assessment of renal and liver function. Adverse effects are predominantly gastrointestinal (anorexia, nausea, flatulence and diarrhoea) and may be reduced by taking metformin just before food and gradually increasing the dose from 850 mg at night to 850 mg twice daily after one week. Metformin therapy is not thought to cause lactic acidosis in women with PCOS who are not diabetic and who have normal renal and liver function. Metformin should be discontinued for three days after iodine-containing contrast medium has been given. Metformin is usually discontinued in pregnancy, although there is no evidence of teratogenicity and preliminary reports from retrospective studies show a reduced rate of gestational diabetes. More studies are required in order to assess the potential benefits of metformin and other insulin-sensitising agents on other symptoms of the PCOS and also for long-term use and potential reduction in progression of impaired glucose tolerance to type 2 diabetes. 73

9 Conclusions In summary, PCOS is a heterogeneous, familial condition. Ovarian dysfunction leads to the main signs and symptoms. The ovary is influenced by external factors in particular the gonadotrophins, insulin and other growth factors, which are dependent upon both genetic and environmental influences. There are longterm risks of developing diabetes and possibly cardiovascular disease. Therapy to date has been symptomatic but, through our improved understanding of the pathogenesis, treatment options are becoming available that strike more at the heart of the syndrome, most notably the use of insulin-sensitising agents. References 1. Fauser B,Tarlatz B, Chang J, Azziz R, Legro R, Dewailly D, et al.the Rotterdam ESHRE/ASRMsponsored PCOS Consensus Workshop Group. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS). Hum Reprod 2004;19: Balen AH, Laven JS,Tan SL, Dewailly D.The ultrasound assessment of the polycystic ovary: international consensus definitions. Hum Reprod Update 2003;9: Balen AH, Conway GS, Kaltsas G,Techatraisak K, Manning PJ,West C, et al. Polycystic ovary syndrome: the spectrum of the disorder in 1741 patients. Hum Reprod 1995;10: Clark AM, Ledger W, Galletly C,Tomlinson L, Blaney F,Wang X, et al. Weight loss results in significant improvement in pregnancy and ovulation rates in anovulatory obese women. Hum Reprod 1995;10: Rajkowha M, Glass MR, Rutherford AJ, Michelmore K, Balen AH. Polycystic ovary syndrome: a risk factor for cardiovascular disease? BJOG 2000;107: Royal College of Obstetricians and Gynaecologists. Long-term Consequences of Polycystic Ovary Syndrome. Guideline No. 33. London: RCOG; 2003 [ elineid=50]. 7. Polson DW, Adams J,Wadsworth J, Franks S. Polycystic ovaries: a common finding in normal women. Lancet 1988;1: Michelmore KF, Balen AH, Dunger DB,Vessey MP. Polycystic ovaries and associated clinical and biochemical features in young women. Clin Endocrinol (Oxf) 1999;51: Jonard S, Robert Y, Cortet-Rudelli C, Decanter C, Dewailly D. Ultrasound examination of polycystic ovaries: is it worth counting the follicles? Hum Reprod 2003;18: Legro RS, Strauss JF. Molecular progress in infertility: polycystic ovary syndrome. Fertil Steril 2002;78: Rodin DA, Bano G, Bland JM,Taylor K, Nussey SS. Polycystic ovaries and associated metabolic abnormalities in Indian subcontinent Asian women. Clin Endocrinol (Oxf) 1998;49: Wijeyaratne CN, Balen AH, Barth J, Belchetz PE. Clinical manifestations and insulin resistance (IR) in polycystic ovary syndrome (PCOS) among South Asians and Caucasians: is there a difference? Clin Endocrinol (Oxf) 2002;57: Pierpoint T, McKeigue PM, Isaacs AJ,Wild SH, Jacobs HS. Mortality of women with polycystic ovary syndrome at long-term follow-up. J Clin Epidemiol 1998;51: Norman RJ, Masters L, Milner CR,Wang JX, Davies MJ. Relative risk of conversion from normoglycaemia to impaired glucose tolerance or non-insulin dependent diabetes mellitus in polycystic ovary syndrome. Hum Reprod 2001;16: Balen AH. Polycystic ovary syndrome and cancer. Hum Reprod Update 2001;7: Coulam CB, Annegers JF, Kranz JS. Chronic anovulation syndrome and associated neoplasia. Obstet Gynecol 1983;61: Barth JH, Cherry CA,Wojnarowska F, Dawber RPR. Cyproterone acetate for severe hirsutism: results of a double-blind dose-ranging study. Clin Endocrinol (Oxf) 1991;35: National Collaborating Centre for Women s and Children s Health. Fertility:Assessment and Treatment for People with Fertility Problems. London: RCOG Press; Kousta E,White DM, Franks S: Modern use of clomiphene citrate in induction of ovulation. Hum Reprod Update 1997;3: Balen AH. Ovulation induction optimizing results and minimizing risks. Hum Fertil 2003;6:S Balen AH, Braat DDM,West C, Patel A, Jacobs HS: Cumulative conception and live birth rates after the treatment of anovulatory infertility. An analysis of of the safety and efficacy of ovulation induction in 200 patients. Hum Reprod 1994;9: Farquhar C,Vandekerckhove P, Lilford R. Laparoscopic drilling by diathermy or laser for ovulation induction in anovulatory polycystic ovary syndrome. Cochrane Database Syst Rev 2002;(4). 23. Lord JM, Flight IHK, Norman RJ. Metformin in polycystic ovary syndrome: systematic review and meta-analysis. BMJ 2003;327: Applications are invited for the post of Editor-in-Chief Qualified obstetricians and gynaecologists with experience of editing journals are invited to apply for the post of Editor-in-Chief. A handwritten application should be accompanied by a curriculum vitae and a short statement outlining the applicant s views on the future of the journal and proposed editorial policy. The closing date for applications is 9 July It is intended to hold interviews before the end of July to allow the successful candidate to take up the appointment from September Up to nine months handover period will be arranged, depending upon experience. Full details may be obtained from the Deputy College Secretary, RCOG 27 Sussex Place, Regent s Park, London NW1 4RG, ENGLAND. bstevens@rcog.org.uk 74