- Email: [email protected]
Reproductive consequences of polycystic ovarian syndrome
ARTICLE IN PRESS Current Obstetrics & Gynaecology (2006) 16, 273–280
Available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/curobgyn
Reproductive consequences of polycystic ovarian syndrome Saad Amer Department of Obstetrics and Gynaecology, University of Nottingham, The Medical School, Derby City General Hospital, Uttoxeter Road, Derby DE22 3DT, UK
KEYWORDS Polycystic ovarian syndrome; Anovulation; Ovulation induction; Recurrent miscarriage; Laparoscopic ovarian diathermy; Hyperandrogeneaemia; Insulin resistance
Summary Polycystic ovarian syndrome (PCOS) is a common endocrine disorder affecting women in their reproductive years. It is frequently associated with reproductive dysfunction, including anovulatory infertility and early pregnancy loss. The underlying pathophysiology of PCOS is not fully understood, although there is considerable evidence to suggest that an excess of ovarian androgen production, either genetically determined or due to hyperinsulinaemia or hypersecretion of luteinising hormone (LH), remains central in the pathogenesis of PCOS. Chronic anovulation seems to be the result of abnormal folliculogenesis characterised by follicular arrest at the small antral phase with escape from atresia. Hypersecretion of LH, hyperandrogenaemia and/or hyperinsulinaemia has been postulated as the possible underlying mechanism of early pregnancy loss in women with PCOS. Anovulatory infertility in PCOS women can be treated with insulin-sensitising measures (such as weight reduction and metformin), clomifene citrate, laparoscopic ovarian diathermy (LOD) and ovarian stimulation with follicle-stimulating hormone. LOD and metformin may help to reduce the risk of miscarriage in women with PCOS, although the effectiveness of these measures remains to be established. & 2006 Elsevier Ltd. All rights reserved.
Introduction Polycystic ovarian syndrome (PCOS) is one of the most common endocrinopathies, affecting 5–10% of women of reproductive age, and is a major cause of anovulatory infertility, accounting for about 75% of cases. According to the 2003 Rotterdam Consensus, PCOS is defined as a syndrome of ovarian dysfunction along with the cardinal features of hyperandrogenism and polycystic ovary morphology. The Corresponding author. Tel.: +44 01332 724 612.
E-mail address: [email protected]. 0957-5847/$ - see front matter & 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.curobgyn.2006.06.002
exact pathophysiology of PCOS remains uncertain. An increasing body of evidence suggests, however, that an excess of ovarian androgen production, either genetically determined or due to extra-ovarian factors such as hyperinsulinaemia or disturbances of the hypothalamic– pituitary–ovarian axis, remains central in the pathogenesis of PCOS. Clinically, PCOS is associated with short- and long-term consequences. The short-term consequences include three groups of disorder: hyperandrogenic, reproductive and metabolic. These manifestations co-exist in variable combinations in different women with PCOS. The longterm sequelae include diabetes mellitus, dyslipidaemia,
ARTICLE IN PRESS 274 hypertension, cardiovascular disease and endometrial carcinoma. The reproductive problems associated with PCOS relate to anovulatory subfertility and early pregnancy loss. The purpose of this review is to give an overview of the pathogenesis of these two common reproductive disorders and to present the different treatment options available.
Anovulatory infertility Chronic anovulation is very common in women with PCOS, affecting more than 80% of them, and is often associated with menstrual irregularities, mainly oligomenorrhoea or amenorrhoea (about 66%), which characteristically date from the time of the menarche. Polymenorrhoea, menorrhagia and dysfunctional uterine bleeding are relatively less common, affecting 4–14% of these women.
Pathogenesis Chronic anovulation in PCOS seems to be the result of two main underlying ovarian disorders: abnormal folliculogenesis and steroidogenesis. Although these two disorders are interlinked, it is difficult to determine the initiating disorder. Abnormal folliculogenesis Follicular development normally starts before birth with the daily recruitment of a cohort of primordial follicles. Under an unknown stimulus, these follicles are transformed into primary, secondary and then small antral follicles of 2–5 mm diameter. This initial development requires low levels of follicle-stimulating hormone (FSH) and takes about 70–80 days. Once the follicles reach that stage, they become FSHdependent. In the absence of an adequate FSH stimulus, these follicles will undergo atresia by default. At puberty and with the maturation of the hypothalamo-pituitary system, FSH rises to levels which intiate ovulatory cycles. In the late luteal phase of normally cycling women and with the intercyclical elevation of FSH above a certain threshold, several of these small antral follicles are recruited (i.e., rescued from atresia) and undergo further growth. Once a leading follicle reaches a diameter of 9–10 mm, the granulosa cells acquire luteinising hormone (LH) receptors, and further follicular development becomes LH-dependent. The rising oestrogen secretion by the leading follicle will result in a negative-feedback decline of FSH and a positive-feedback increase in LH. As a result, the dominant follicle continues to mature owing to the rising level of LH, while all the other follicles undergo atresia owing to the fall in FSH. In PCOS, despite a normal stock number of primordial follicles and a normal early FSH-independent follicular development, follicular growth becomes arrested at the small antral phase, with failure of dominance and escape from the natural process of atresia. This results in an increased number of primary, secondary and small antral follicles (2–8 mm in diameter). The mechanism of this disturbed folliculogenesis in PCOS remains largely unknown. Several theories have been postulated to explain the maturation arrest and the escape from atresia of the antral follicles in PCOS. Theories explaining follicular arrest
S. Amer include relative FSH deficiency, abnormal LH stimulus, a deficiency of certain local growth factors and abnormal ovarian steroidogenesis. The hypothesised relative FSH deficiency may be due to an abnormally increased inhibin B secretion by the increased number of small antral follicles and/or increased ovarian and/or peripheral oestrogen production as a result of hyperandrogenaemia. An abnormal LH stimulus has also been postulated to explain the premature follicular arrest in PCOS. There is evidence that the granulosa cells of small antral follicles in anovulatory women with PCOS acquire LH receptors prematurely (at follicular diameter of 4 mm), possibly due to hyperinsulinaemia. LH receptor acquisition of the granulosa cells seems to switch the follicle from proliferation to differentiation, resulting in a suppression of granulosa cell growth and ultimately inducing an arrest of follicular development and a failure of dominance. Some growth factors have also been implicated in the arrest of follicular growth in PCOS, for example a deficiency of growth differentiation factor 9 and increased activity of epidermal growth factor and transforming growth factor a. Increased secretion of the cytokine tumour necrosis factor a in women with PCOS has also been associated with disordered folliculogenesis. Attenuated apoptosis due to disregulation of the antiapoptotic and proapoptotic factors has been postulated as a possible mechanism of escape of the antral follicles from atresia in women with PCOS. Although the exact mechanism of disordered apoptosis is not well understood, it seems to be due to the abnormal expression of certain growth factors.
Abnormal steroidogenesis Normally, the secondary follicle acquires a theca layer characterised by LH receptors and steroidogenic capacity, whereas the granulosa cells contain receptors for FSH. According to the two-cell, two-gonadotrophin model, LH stimulates the theca cells to produce androgens, which are the precursors for oestrogen synthesis. The androgens then diffuse to the granulosa cells, where FSH stimulates the expression of cytochrome P450 aromatase, which converts the androgens to oestrogens. The rising intraovarian oestrogen and inhibin B concentrations result in a negative feedback on FSH secretion and a positive feedback on LH secretion. The resulting increase in LH, together with the rising inhibin B level, leads to an increase in thecal androgen production. The granulosa cells of the dominant follicle gradually acquire LH receptors, and most of the physiological actions of FSH on granulosa cells can be exerted by LH. In the presence of increasing levels of androgen precursors, the granulosa cells of the dominant follicle, stimulated by the rising LH, continue to produce increasing levels of oestrogens despite decreasing FSH levels. LH results in an increase in steroidogenesis, early progesterone production and luteinisation. Through a positive-feedback mechanism, progesterone, together with the high oestrogen levels, induces the mid-cycle LH surge, which results in ovulation. In PCOS, excess ovarian androgen production appears to be central in the pathogenesis of PCOS. Whether hyperandrogenaemia is the cause or the result of disordered folliculogenesis remains to be elucidated. Although it is
ARTICLE IN PRESS Reproductive consequences of polycystic ovarian syndrome
275
possible that the increased number of small antral follicles produces excess androgens, it is also possible that a genetically determined hypersecretion of androgens is responsible for the disordered folliculogenesis. The excess androgen secretion by the theca cells results in an increased availability of precursors for oestrogen production in the granulosa cells. The granulosa cells in anovulatory women with PCOS show LH-induced aromatase activity in the small antral follicles (secondary to hyperinsulinaemia), resulting in enhanced oestrogen production. The increased levels of circulating oestrogens result in an increased positive feedback on LH and a negative feedback on FSH secretion, thus causing disordered folliculogenesis, abnormal steroidogenesis and abnormal gonadotrophin secretion.
6 months before commencing clomifene citrate treatment. A weight loss of just 5–10% has been shown to reverse the deleterious effects of obesity on ovarian function and can restore reproductive function in a majority of patients within 6 months of weight reduction. Although it is effective, cheap and safe, weight loss represents a major challenge to clinicians as only a small minority of obese women manage to achieve significant weight reduction. Metformin has recently been increasingly used in many centres to help obese women with PCOS to reduce their weight, although controversy remains over its effectiveness in achieving weight reduction. When lifestyle measures fail, consideration should be given to other weight-reducing measures such as the lipase inhibitor orlistat or minimally invasive stomach surgery, especially in morbidly obese women, although the effectiveness and safety of these methods remain to be established.
Ovulation induction Several methods have been successfully used to restore ovulation and thereby fertility in anovulatory women with PCOS, including weight reduction, clomifene citrate, laparoscopic ovarian diathermy (LOD), metformin and gonadotrophin therapy (Fig. 1). Weight reduction Central obesity, which is very common in women with PCOS, accentuates insulin resistance, hyperinsulinaemia and hyperandrogenaemia. As a result, obesity not only exaggerates the disordered ovarian function in PCOS women, but also increases ovarian resistance to various methods of ovulation induction. It is therefore universally agreed that women with an elevated body mass index (BMI; 430 kg/m2) should first be advised to undergo a weight-loss programme for
Weight reduction
Clomifene citrate Clomifene citrate, which is an antioestrogenic drug, is widely used globally as the first-line medical method for ovulation induction in women with PCOS owing to its simplicity of use, low cost and relative safety and efficacy. It is a non-steroidal synthetic oestrogen that is related to the synthetic oestrogen diethylstilboestrol and is known for its antioestrogenic and weak oestrogenic properties. By binding to the oestrogen receptors of the hypothalamic– pituitary system, clomifene blocks the negative-feedback effect of oestradiol on gonadotrophin-releasing hormone (GnRH) secretion, resulting in an increase in the GnRH pulse amplitude that leads to increased gonadotrophin secretion from the pituitary. The resultant increase in the FSH stimulus to the ovary triggers an ovulatory cycle.
All overweight/obese women with PCOS should first be encouraged to lose weight before medical ovulation induction
First line Clomifene citrate
Clomifene citrate is the standard firstline method of medical ovulation induction in anovulatory women with PCOS
Second line LOD
FSH
Metformin
The second-line treatment after clomifene citrate resistance or failure is debatable, with competition between laparoscopic ovarian diathermy (LOD), gonadotrophin ovarian stimulation and metformin to be the preferred choice
Figure 1 Management options for anovulatory infertility associated with polycystic ovarian syndrome (PCOS). FSH, folliclestimulating hormone.
ARTICLE IN PRESS 276
S. Amer
Clomifene has also been shown to exert antioestrogenic activity in other oestrogen-dependent tissues, including the ovary, uterus and vaginal mucosa, via the same mechanism. This unwanted effect of clomifene citrate may explain the discrepancy between the high ovulation and low pregnancy rates achieved with clomifene. It may also explain why the pregnancy rate decreases with higher doses of clomifene, owing to the increased antioestrogenic activity. The prolonged antioestrogenic effects of clomifene citrate on the uterus result in delayed endometrial maturation, poor-quality cervical mucus and an alteration in uterine blood flow. Clomifene citrate is given for 5 days in the early follicular phase (days 2–6) of the menstrual cycle, initially at a daily dose of 50 mg, which is increased in a stepwise progression if ovulation does not occur up to a maximum daily dose of 150 mg. This achieves an ovulation rate of 60–85% and a pregnancy rate of 30–40%. Once ovulation has been achieved on a certain dose, treatment is continued with that dose for 6–12 months. Provided that all other subfertility factors have been excluded, the cumulative conception rate with clomifene citrate continues to increase until it reaches a plateau at treatment cycle 12. Prolonging clomifene treatment beyond 12 cycles has been linked with an increased risk of borderline or invasive ovarian tumour and should therefore be discouraged. Clomifene should be discontinued if it fails to induce ovulation on the maximum dose (clomifene citrate resistance) or if it fails to achieve pregnancy after 6–12 cycles despite ovulation (clomifene citrate failure). It is recommended that women receiving clomifene citrate should be monitored for follicular development with serial pelvic ultrasound scans, at least in their first cycle of treatment, to exclude multifollicular development and to minimise the risk of multiple pregnancies. If mono-ovulation is confirmed, further clomifene citrate cycles could just be monitored with a mid-luteal phase serum progesterone concentration. A progesterone level of 25 nmol/l or more is indicative of ovulation. Women with PCOS who are likely to be clomifeneresistant are those with an elevated BMI, amenorrhoea, marked hyperandrogenaemia and insulin resistance. Age and duration of infertility are the most important determinants of the probability of conception. The younger the woman and the shorter the duration of her infertility, the higher are her chances of conceiving. Other important factors determining the chances of pregnancy are the dose and duration of clomifene citrate. Most pregnancies (50%) occur with
Table 1
clomifene citrate 50 mg, and only about 10% occur with 150 mg. Drawbacks to clomifene treatment include a miscarriage rate of up to 40%, an increased risk of multiple pregnancies and a risk of ovarian hyperstimulation syndrome (OHSS). A possible increase in the risk of ovarian cancer has been suggested if more than 12 cycles are used. In addition, there are common but less serious side effects such as hot flushes, headaches and nausea. Laparoscopic ovarian diathermy Women with PCOS who experience clomifene citrate resistance or failure can be offered LOD, gonadotrophin ovarian stimulation or metformin. LOD, which is as effective as gonadotrophin therapy, offers several advantages over gonadotrophins (Table 1). Importantly, LOD results in monoovulation, with an incidence of multiple pregnancies no higher than the background rate. LOD corrects the endocrine abnormalities associated with PCOS, such as the excess secretion of LH and androgens. These favourable endocrine effects of LOD may reduce the risk of miscarriage known to be increased in women with PCOS. A recent longitudinal long-term follow-up study in our centre has shown that about two-thirds of women with PCOS undergoing LOD respond to the treatment, as evidenced by the resumption of a regular menstrual cycle. In about 50% of the initial responders, the effects are only transient (up to 1 year), whereas the remaining 50% (i.e., a third of the total number of women) continue to have regular ovulatory cycles for many years. As far as conception is concerned, women can expect an approximately 1 in 2 chance of conception within the first year, with a reduction in the incidence of miscarriage to a rate comparable to that of the general population. In about a third of cases, the improvement in reproductive performance seems to last for many years. Several techniques of LOD have been described, of which the most widely used technique is described here (Table 2). A specially designed monopolar electrocautery needle (Fig. 2) is used to penetrate the ovarian capsule at a number of points. The site of application should be away from the ovarian hilum (to minimise the risk of ovarian atrophy) and the fallopian tube (to reduce the risk of mechanical infertility). As the needle is pushed into the ovarian capsule, electricity is activated for 5 s using a monopolar coagulating current set at 30 W. The amount of energy applied per puncture is 5 s 30 W ¼ 150 J. The optimum number of punctures has been shown to be four per ovary.
Advantages of laparoscopic ovarian diathermy over gonadotrophin therapy.
It corrects the endocrine abnormalities of polycystic ovarian syndrome, such as increased levels of luteinising hormone and
androgens. It is less costly. It avoids complex monitoring. A single treatment produces ovulatory cycles for several months or years. It avoids ovarian hyperstimulation syndrome. There is no increase in multiple pregnancy. The incidence of miscarriage may be reduced. It allows concomitant testing of tubal patency and pelvic anatomy.
ARTICLE IN PRESS Reproductive consequences of polycystic ovarian syndrome
Table 2 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
277
Summary of the technique of laparoscopic ovarian diathermy.
Three-puncture laparoscopy is employed. The utero-ovarian ligament is grasped with non-toothed grasping forceps. The ovary is lifted up away from the bowel and stabilized. A specially designed monopolar electrocautery needle is used (Fig. 2). The needle is applied at a right angle to the surface of the ovary. The needle should situated be away from ovarian hilum and the fallopian tube. The power is set at 30 W (coagulating). As the needle penetrates the capsule, the electricity is activated for 5 s. The capsule of the ovary is penetrated to a depth of 6–8 mm. Four punctures are made in each ovary. The ovary is cooled with Hartmann’s solution after drilling. A crystalloid solution is instilled at the end of the procedure.
Figure 2
Laparoscopic ovarian drilling.
Women with PCOS and marked obesity (BMIX35 kg/m2), marked hyperandrogenism (testosteroneX4.5 nmol/l or free androgen indexX15) and/or a long duration of infertility (43 years) appear to be resistant to LOD. It is therefore recommended that alternative methods of treatment should be considered for this group of patients. On the other hand, high pretreatment serum LH concentrations in LOD responders appear to predict a higher probability of pregnancy. The main complications of LOD include adhesion formation and a theoretical risk of excessive ovarian damage leading to ovarian failure. These complications are, however, rare and appear to be of little clinical significance.
Metformin The link between insulin resistance and PCOS led many authors to consider insulin-sensitising agents for the management of this syndrome. These agents, which have been used for many years in type 2 diabetes, have recently been increasingly used worldwide in women with PCOS. The most commonly used agent is metformin, which is the only currently available biguanide drug. Despite its therapeutic benefits in PCOS, the mechanism of action of metformin in women with this syndrome remains uncertain. It improves insulin sensitivity by increasing peripheral glucose uptake in response to insulin at post-
receptor level. This in turn results in correction of the associated hyperinsulinaemia, which is responsible for the hypersecretion of ovarian androgens. In theory, the resulting decrease in androgen production improves the intraovarian microenvironment, which leads to a normalisation of ovarian follicular development. Metformin does not cause hyperinsulinaemia and is therefore not associated with hypoglycaemia. Hypoglycaemia could, however, occur when caloric intake is deficient or when strenuous exercise is not compensated by caloric supplementation. The exact role of metformin in PCOS has still to be established. The most common indication for metformin in PCOS is to induce ovulation in women seeking fertility treatment. Most gynaecologists reserve metformin use for women with PCOS who are resistant to clomifene citrate, although an increasing number of reproductive medicine specialists use it as a first-line treatment for ovulation induction in overweight/obese women with PCOS. If ovulation is not achieved after 3 months of metformin therapy, clomifene citrate could be added. The effective dose range of metformin in women with PCOS has yet to be determined. The drug should be started at a low dose, with gradual dose escalation, both to reduce gastrointestinal side effects and to permit identification of the minimum dose required to achieve regular ovulatory cycles. The usual starting dose is 500 mg twice a day or 850 mg once a day, given with meals. Dosage increases should be made in increments of 500 mg weekly or 850 mg every 2 weeks, up to a total of 2550 mg per day, given in divided doses. Patients should be warned against excessive alcohol intake while receiving metformin as this could precipitate lactic acidosis. Women with PCOS receiving metformin should be monitored for gastrointestinal side effects and for achievement of ovulation. If, after 3 months of treatment, ovulation has not been achieved, consideration should be given to increasing the dose of metformin, adding clomifene citrate or changing to an alternative therapy. If metformin is to be given for a long duration (41 year), initial and periodic monitoring of renal and hepatic functions should also be performed annually. The exact success rate of metformin in infertile women with PCOS is still uncertain. Most studies reported ovulation rates of 30–45% using metformin alone, although more recent reports revealed higher success rates.
ARTICLE IN PRESS 278
S. Amer
Metformin should be temporarily discontinued in patients undergoing radiological studies involving the intravascular administration of iodinated contrast materials, which can lead to an acute alteration of renal function and have been associated with lactic acidosis in patients receiving metformin. Metformin therapy should also be temporarily suspended for any surgical procedure associated with a restricted intake of food and fluids. Other contraindications to the use of metformin include renal dysfunction, hepatic impairment and congestive heart failure. The most common adverse reactions (45%) reported during metformin therapy include diarrhoea (50%), nausea/vomiting (25%), flatulence (12%), asthenia (9%), indigestion (7%), abdominal discomfort (6%) and headache (5%). Lactic acidosis is a very rare (0.003%), but serious, metabolic complication that can occur due to metformin accumulation; when it occurs, it is fatal in approximately 50% of cases. Patients at risk include those with renal insufficiency, hepatic impairment or congestive heart failure. Excessive alcohol intake is also known to potentiate the effects of metformin hydrochloride on lactate metabolism. Gonadotrophin therapy In many centres, gonadotrophin therapy is the preferred second-line treatment for induction of ovulation after clomifene citrate failure in anovulatory women with PCOS. Theoretically, the preferred gonadotrophin preparation for induction of ovulation in women with PCOS is one that does not contain LH, in view of the high levels of endogenous LH in these women. However, human menopausal gonadotrophin and pure FSH preparations have both been successfully used for ovulation induction in women with PCOS. Anovulatory women with PCOS have an increased number of FSH-sensitive small antral follicles and are therefore at increased risk of multifollicular development when receiving exogenous FSH. In view of this hypersensitivity of the polycystic ovaries, the conventional high-dose (150 IU) stepwise protocol, which has been in use since the 1970s, has now largely been replaced by low-dose treatment regimens, which can be applied in a step-up, step-down or sequential fashion. The aim of this approach is to allow FSH to rise slowly just above the FSH threshold while avoiding an excessive ovarian response, thus minimising the risks of OHSS and multiple pregnancy. The chronic low-dose, step-up gonadotrophin regimen has been the preferred method for ovulation induction in
women with PCOS in many centres since 1990 (Fig. 3). This approach is characterised by small initial daily FSH doses (typically 50–75 IU), which are continued for a long period (usually 14 days) without any change in dose and then increased gradually, if necessary, by small amounts (e.g., 25–37.5 IU) every 5–7 days until a dominant follicle emerges. This approach seems to give the best result, with high rates of monofollicular ovulation (70%) and low rates of multiple pregnancy (5%) and OHSS (o1%), while maintaining good pregnancy rates (20% per cycle and 40% per patient). The main problem of this regimen is, however, the long duration of treatment. An attempt by some researchers to reduce the 14-day initial period to 7 days resulted in similar success rates, although there was a slightly higher rate of multiple pregnancy in the 7-day starters. Others have attempted to shorten the 14-day interval to 5 days, but this resulted in an overstimulation rate of 3% and a 26% rate of multiple pregnancies. The step-down approach applies a high starting dose that continues until an ovarian response has been established. The FSH dose is then reduced in small decremental amounts in two steps (Fig. 4). The aim of this approach is to mimic the physiological FSH threshold/window concept of the normal ovulatory cycle, which is characterised by an intercycle elevation in FSH above the threshold for about 5 days (the window) to allow the selection of one leading follicle, followed by a gradual decline in FSH with the subsequent decreasing dependence of the selected (dominant) follicle on FSH. Therapy usually commences with 150 IU FSH per day, reduced by one decrement (37.5 IU) when one follicle has been selected (10 mm), with a further reduction 3 days later by another decrement to 75 IU per day; this is then continued until the day of injection of human chorionic gonadotrophin. The advocates of this approach claim that it is as successful as the chronic low-dose, step-up protocol in achieving monofollicular development and has the advantage of reducing the duration of the treatment. However, the step-down approach requires more stringent follicular monitoring than the step-up regimen. An alternative method is the sequential protocol, which starts with an initial step-up administration of gonadotrophin (similar to the low dose step-up protocol) followed by a step-down regimen once a dominant follicle (X14 mm) has been selected. Two step decrements 37.5 IU/day
Two step increments 25–37.5 IU/day 150 IU/day Until a follicle ≥10 mm
50–75 IU/day 14 days
7 days
7 days
Figure 3 The chronic low-dose, step-up follicle-stimulating hormone stimulation protocol, characterised by a low starting dose with no change for the first 14 days, and then small increments if necessary at 7-day intervals until the injection of human chorionic gonadotrophin or up to 28 days.
112.5 IU/day 3 days
75 IU/day Until hCG injection
Figure 4 The step-down follicle-stimulating hormone (FSH) stimulation protocol, which is based on the FSH threshold/ window concept. It is characterised by a high starting dose (to reach the threshold), which is continued until a dominant follicle (X10 mm) emerges. The dose is then decreased by small decrements in two steps (the end of the window). hCG, human chorionic gonadotrophin.
ARTICLE IN PRESS Reproductive consequences of polycystic ovarian syndrome
Pregnancy loss Pathogenesis Women with PCOS have a high rate of early miscarriage (30–50%) compared with the general population. The mechanism of pregnancy loss in PCOS remains largely unknown. Hypersecretion of LH, hyperandrogenaemia and/ or hyperinsulinaemia have been postulated as the possible underlying mechanisms of early pregnancy loss in women with PCOS. High LH concentrations The literature on the influence of LH hypersecretion on pregnancy outcome is conflicting, although there is considerable evidence to implicate LH hypersecretion in the pathogenesis of early pregnancy loss in women with PCOS. It has been postulated that an elevated concentration of LH results in premature oocyte maturation through an inhibition of oocyte maturation inhibitor. By the time of ovulation, the released egg is physiologically aged and has a diminished capacity to be fertilised, or if fertilised it has a high chance of undergoing miscarriage. Normally, oocytes are maintained by oocyte maturation inhibitor in the first meiotic division until just before ovulation, when oocyte maturation is completed. Although the precise nature of oocyte maturation inhibitor is uncertain, there is evidence to suggest that it is either cyclic adenosine monophosphate (cAMP) or an unknown factor that is activated by cAMP. In women with normal ovulation, stimulation by the preovulatory rise in LH leads to a fall in cAMP and a resumption of meiosis. In PCOS, the granulosa cells acquire LH receptors prematurely, possibly due to hyperinsulinaemia. As a result, inhibition of oocyte maturation inhibitor occurs early, resulting in premature oocyte maturation. Hyperandrogenaemia Data on the influence of high androgen concentrations on pregnancy outcome are scarce and conflicting. In vitro studies have confirmed an inhibitory effect of androgens on endometrial cell growth and activity. The expression of HOXA-10, which is a gene essential for endometrial receptivity, has been shown in in vitro studies to be suppressed by testosterone. It has therefore been hypothesised that androgens might have a detrimental effect on endometrial function, preventing effective endometrial development in the both follicular and the luteal phase.
279 that suppresses LH and/or androgen production (such as pituitary downregulation and LOD) or improves insulin sensitivity (such as dietary modifications and metformin therapy). Evidence from randomised clinical trials has shown that a pituitary suppression of LH with GnRH analogues in women with PCOS with recurrent miscarriage does not improve the live-birth rate. On the other hand, LOD and metformin have been shown to reduce the miscarriage rate in women with PCOS. Current evidence is, however, not strong enough to allow the recommendation of these measures for improving the pregnancy outcome in this group.
Conclusion Women with PCOS often present with reproductive failure as either infertility or early pregnancy loss. The main underlying pathogenesis of these reproductive disorders lies in the disordered folliculogenesis and steroidogenesis, which are the result of an interaction between hyperinsulinaemia, hypersecretion of LH and hyperandrogenaemia. Anovulatory women with PCOS who are obese or overweight should first undergo a weight-reduction programme before embarking on medical treatment. Clomifene citrate remains the standard first-line medical treatment for ovulation induction in infertile PCOS women. Women who fail to conceive on clomifene citrate can be offered LOD, gonadotrophin therapy or metformin. For recurrent miscarriage associated with PCOS, no effective treatment is available, although LOD and metformin present promising treatment modalities for this condition.
Practice points
A weight reduction of 5–10% in obese/overweight
anovulatory women with PCOS restores ovulation in almost all cases. Clomifene citrate is the standard first-line medical treatment for ovulation induction in anovulatory women with PCOS and can achieve high pregnancy rates in properly selected cases. LOD and gonadotrophin therapy are equally effective in inducing ovulation in clomifene-resistant women with PCOS, but LOD offers several advantages, such as the very low multiple pregnancy rate. The chronic low-dose step-up protocol for FSH stimulation is the preferred approach as it induces mono-ovulation in most cases, thus minimising the risks of OHSS and multiple pregnancy.
Hyperinsulinaemia and obesity It has been reported that insulin resistance, which is common in PCOS, is associated with recurrent miscarriage through a mechanism yet to be determined. It is possible that hyperinsulinaemia might affect pregnancy outcome through the interaction with LH and androgen synthesis.
Research directions
Management
The effectiveness of weight-reducing measures such
In the absence of a good understanding of the pathogenesis of early pregnancy loss in women with PCOS, an effective treatment to improve pregnancy outcome cannot be established. It seems logical to apply a therapeutic method
as the lipase inhibitor orlistat and stomach surgery in anovulatory women with PCOS who are obese. The role of metformin in the management of anovulatory infertility and recurrent miscarriage in women with PCOS.
ARTICLE IN PRESS 280
S. Amer
The role of LOD in the management the of recurrent
miscarriage in women with PCOS. Investigation into the possible role of the endometrium in the pathogenesis of early pregnancy loss in women with PCOS. The value of GnRH antagonists in ovulation induction in women with PCOS.
Further reading 1. Amer SA, Li TC. Laparoscopic ovarian surgery for polycystic ovarian syndrome. In: Tarlatzis B, editor. European practice in gynaecology and obstetrics—ovulation induction. Paris: Elsevier; 2002. p. 137–53. 2. Amer S, Li TC, Gopalan V, Ledger WL, Cooke ID. Long term follow up of patients with polycystic ovarian syndrome after laparoscopic ovarian drilling: clinical outcome. Hum Reprod 2002;17:2035–42. 3. Amer S, Li TC, Ledger WL. Ovulation induction using laparoscopic ovarian drilling in women with polycystic ovarian syndrome: predictors of success. Hum Reprod 2004;19:1719–24.
4. Balasch J. The role of FSH and LH in ovulation induction: current concepts and the contribution of recombinant gonadotropins. In: Gardner DK, Weissman A, Howeles C, Shoham Z, editors. Textbook of assisted reproductive techniques. London: Martin Dunitz; 2001. p. 546–65. 5. Balen AH, Tan SL, Jacobs HS. Hypersecretion of luteinising hormone: a significant cause of infertility and miscarriage. Br J Obstet Gynaecol 1993;100:1082–9. 6. Franks S, White D. Low-dose gonadotrophin treatment in polycystic ovary syndrome: the step-up protocol. In: Tarlatzis B, editor. European practice in gynaecology and obstetrics— ovulation induction. Paris: Elsevier; 2002. p. 101–7. 7. Macklon NS, Fauser BCJM. The step-down protocol. In: Tarlatzis B, editor. European practice in gynaecology and obstetrics— ovulation induction. Paris: Elsevier; 2002. p. 111–8. 8. Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group. Revised 2003 consensus on diagnostic criteria and longterm health risks related to polycystic ovary syndrome (PCOS). Hum Reprod 2004;19:41–7. 9. Speroff L, Fritz A. The ovary—embryology and development. In: Clinical gynaecologic endocrinology and infertility, 7th ed. Baltimore: Williams & Wilkins; p. 98–111. 10. Van der Spuy SM. The pathogenesis of infertility and early pregnancy loss in polycystic ovary syndrome. Best Pract Res Clin Obstet Gynaecol 2004;18:755–71.
Available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/curobgyn
Reproductive consequences of polycystic ovarian syndrome Saad Amer Department of Obstetrics and Gynaecology, University of Nottingham, The Medical School, Derby City General Hospital, Uttoxeter Road, Derby DE22 3DT, UK
KEYWORDS Polycystic ovarian syndrome; Anovulation; Ovulation induction; Recurrent miscarriage; Laparoscopic ovarian diathermy; Hyperandrogeneaemia; Insulin resistance
Summary Polycystic ovarian syndrome (PCOS) is a common endocrine disorder affecting women in their reproductive years. It is frequently associated with reproductive dysfunction, including anovulatory infertility and early pregnancy loss. The underlying pathophysiology of PCOS is not fully understood, although there is considerable evidence to suggest that an excess of ovarian androgen production, either genetically determined or due to hyperinsulinaemia or hypersecretion of luteinising hormone (LH), remains central in the pathogenesis of PCOS. Chronic anovulation seems to be the result of abnormal folliculogenesis characterised by follicular arrest at the small antral phase with escape from atresia. Hypersecretion of LH, hyperandrogenaemia and/or hyperinsulinaemia has been postulated as the possible underlying mechanism of early pregnancy loss in women with PCOS. Anovulatory infertility in PCOS women can be treated with insulin-sensitising measures (such as weight reduction and metformin), clomifene citrate, laparoscopic ovarian diathermy (LOD) and ovarian stimulation with follicle-stimulating hormone. LOD and metformin may help to reduce the risk of miscarriage in women with PCOS, although the effectiveness of these measures remains to be established. & 2006 Elsevier Ltd. All rights reserved.
Introduction Polycystic ovarian syndrome (PCOS) is one of the most common endocrinopathies, affecting 5–10% of women of reproductive age, and is a major cause of anovulatory infertility, accounting for about 75% of cases. According to the 2003 Rotterdam Consensus, PCOS is defined as a syndrome of ovarian dysfunction along with the cardinal features of hyperandrogenism and polycystic ovary morphology. The Corresponding author. Tel.: +44 01332 724 612.
E-mail address: [email protected]. 0957-5847/$ - see front matter & 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.curobgyn.2006.06.002
exact pathophysiology of PCOS remains uncertain. An increasing body of evidence suggests, however, that an excess of ovarian androgen production, either genetically determined or due to extra-ovarian factors such as hyperinsulinaemia or disturbances of the hypothalamic– pituitary–ovarian axis, remains central in the pathogenesis of PCOS. Clinically, PCOS is associated with short- and long-term consequences. The short-term consequences include three groups of disorder: hyperandrogenic, reproductive and metabolic. These manifestations co-exist in variable combinations in different women with PCOS. The longterm sequelae include diabetes mellitus, dyslipidaemia,
ARTICLE IN PRESS 274 hypertension, cardiovascular disease and endometrial carcinoma. The reproductive problems associated with PCOS relate to anovulatory subfertility and early pregnancy loss. The purpose of this review is to give an overview of the pathogenesis of these two common reproductive disorders and to present the different treatment options available.
Anovulatory infertility Chronic anovulation is very common in women with PCOS, affecting more than 80% of them, and is often associated with menstrual irregularities, mainly oligomenorrhoea or amenorrhoea (about 66%), which characteristically date from the time of the menarche. Polymenorrhoea, menorrhagia and dysfunctional uterine bleeding are relatively less common, affecting 4–14% of these women.
Pathogenesis Chronic anovulation in PCOS seems to be the result of two main underlying ovarian disorders: abnormal folliculogenesis and steroidogenesis. Although these two disorders are interlinked, it is difficult to determine the initiating disorder. Abnormal folliculogenesis Follicular development normally starts before birth with the daily recruitment of a cohort of primordial follicles. Under an unknown stimulus, these follicles are transformed into primary, secondary and then small antral follicles of 2–5 mm diameter. This initial development requires low levels of follicle-stimulating hormone (FSH) and takes about 70–80 days. Once the follicles reach that stage, they become FSHdependent. In the absence of an adequate FSH stimulus, these follicles will undergo atresia by default. At puberty and with the maturation of the hypothalamo-pituitary system, FSH rises to levels which intiate ovulatory cycles. In the late luteal phase of normally cycling women and with the intercyclical elevation of FSH above a certain threshold, several of these small antral follicles are recruited (i.e., rescued from atresia) and undergo further growth. Once a leading follicle reaches a diameter of 9–10 mm, the granulosa cells acquire luteinising hormone (LH) receptors, and further follicular development becomes LH-dependent. The rising oestrogen secretion by the leading follicle will result in a negative-feedback decline of FSH and a positive-feedback increase in LH. As a result, the dominant follicle continues to mature owing to the rising level of LH, while all the other follicles undergo atresia owing to the fall in FSH. In PCOS, despite a normal stock number of primordial follicles and a normal early FSH-independent follicular development, follicular growth becomes arrested at the small antral phase, with failure of dominance and escape from the natural process of atresia. This results in an increased number of primary, secondary and small antral follicles (2–8 mm in diameter). The mechanism of this disturbed folliculogenesis in PCOS remains largely unknown. Several theories have been postulated to explain the maturation arrest and the escape from atresia of the antral follicles in PCOS. Theories explaining follicular arrest
S. Amer include relative FSH deficiency, abnormal LH stimulus, a deficiency of certain local growth factors and abnormal ovarian steroidogenesis. The hypothesised relative FSH deficiency may be due to an abnormally increased inhibin B secretion by the increased number of small antral follicles and/or increased ovarian and/or peripheral oestrogen production as a result of hyperandrogenaemia. An abnormal LH stimulus has also been postulated to explain the premature follicular arrest in PCOS. There is evidence that the granulosa cells of small antral follicles in anovulatory women with PCOS acquire LH receptors prematurely (at follicular diameter of 4 mm), possibly due to hyperinsulinaemia. LH receptor acquisition of the granulosa cells seems to switch the follicle from proliferation to differentiation, resulting in a suppression of granulosa cell growth and ultimately inducing an arrest of follicular development and a failure of dominance. Some growth factors have also been implicated in the arrest of follicular growth in PCOS, for example a deficiency of growth differentiation factor 9 and increased activity of epidermal growth factor and transforming growth factor a. Increased secretion of the cytokine tumour necrosis factor a in women with PCOS has also been associated with disordered folliculogenesis. Attenuated apoptosis due to disregulation of the antiapoptotic and proapoptotic factors has been postulated as a possible mechanism of escape of the antral follicles from atresia in women with PCOS. Although the exact mechanism of disordered apoptosis is not well understood, it seems to be due to the abnormal expression of certain growth factors.
Abnormal steroidogenesis Normally, the secondary follicle acquires a theca layer characterised by LH receptors and steroidogenic capacity, whereas the granulosa cells contain receptors for FSH. According to the two-cell, two-gonadotrophin model, LH stimulates the theca cells to produce androgens, which are the precursors for oestrogen synthesis. The androgens then diffuse to the granulosa cells, where FSH stimulates the expression of cytochrome P450 aromatase, which converts the androgens to oestrogens. The rising intraovarian oestrogen and inhibin B concentrations result in a negative feedback on FSH secretion and a positive feedback on LH secretion. The resulting increase in LH, together with the rising inhibin B level, leads to an increase in thecal androgen production. The granulosa cells of the dominant follicle gradually acquire LH receptors, and most of the physiological actions of FSH on granulosa cells can be exerted by LH. In the presence of increasing levels of androgen precursors, the granulosa cells of the dominant follicle, stimulated by the rising LH, continue to produce increasing levels of oestrogens despite decreasing FSH levels. LH results in an increase in steroidogenesis, early progesterone production and luteinisation. Through a positive-feedback mechanism, progesterone, together with the high oestrogen levels, induces the mid-cycle LH surge, which results in ovulation. In PCOS, excess ovarian androgen production appears to be central in the pathogenesis of PCOS. Whether hyperandrogenaemia is the cause or the result of disordered folliculogenesis remains to be elucidated. Although it is
ARTICLE IN PRESS Reproductive consequences of polycystic ovarian syndrome
275
possible that the increased number of small antral follicles produces excess androgens, it is also possible that a genetically determined hypersecretion of androgens is responsible for the disordered folliculogenesis. The excess androgen secretion by the theca cells results in an increased availability of precursors for oestrogen production in the granulosa cells. The granulosa cells in anovulatory women with PCOS show LH-induced aromatase activity in the small antral follicles (secondary to hyperinsulinaemia), resulting in enhanced oestrogen production. The increased levels of circulating oestrogens result in an increased positive feedback on LH and a negative feedback on FSH secretion, thus causing disordered folliculogenesis, abnormal steroidogenesis and abnormal gonadotrophin secretion.
6 months before commencing clomifene citrate treatment. A weight loss of just 5–10% has been shown to reverse the deleterious effects of obesity on ovarian function and can restore reproductive function in a majority of patients within 6 months of weight reduction. Although it is effective, cheap and safe, weight loss represents a major challenge to clinicians as only a small minority of obese women manage to achieve significant weight reduction. Metformin has recently been increasingly used in many centres to help obese women with PCOS to reduce their weight, although controversy remains over its effectiveness in achieving weight reduction. When lifestyle measures fail, consideration should be given to other weight-reducing measures such as the lipase inhibitor orlistat or minimally invasive stomach surgery, especially in morbidly obese women, although the effectiveness and safety of these methods remain to be established.
Ovulation induction Several methods have been successfully used to restore ovulation and thereby fertility in anovulatory women with PCOS, including weight reduction, clomifene citrate, laparoscopic ovarian diathermy (LOD), metformin and gonadotrophin therapy (Fig. 1). Weight reduction Central obesity, which is very common in women with PCOS, accentuates insulin resistance, hyperinsulinaemia and hyperandrogenaemia. As a result, obesity not only exaggerates the disordered ovarian function in PCOS women, but also increases ovarian resistance to various methods of ovulation induction. It is therefore universally agreed that women with an elevated body mass index (BMI; 430 kg/m2) should first be advised to undergo a weight-loss programme for
Weight reduction
Clomifene citrate Clomifene citrate, which is an antioestrogenic drug, is widely used globally as the first-line medical method for ovulation induction in women with PCOS owing to its simplicity of use, low cost and relative safety and efficacy. It is a non-steroidal synthetic oestrogen that is related to the synthetic oestrogen diethylstilboestrol and is known for its antioestrogenic and weak oestrogenic properties. By binding to the oestrogen receptors of the hypothalamic– pituitary system, clomifene blocks the negative-feedback effect of oestradiol on gonadotrophin-releasing hormone (GnRH) secretion, resulting in an increase in the GnRH pulse amplitude that leads to increased gonadotrophin secretion from the pituitary. The resultant increase in the FSH stimulus to the ovary triggers an ovulatory cycle.
All overweight/obese women with PCOS should first be encouraged to lose weight before medical ovulation induction
First line Clomifene citrate
Clomifene citrate is the standard firstline method of medical ovulation induction in anovulatory women with PCOS
Second line LOD
FSH
Metformin
The second-line treatment after clomifene citrate resistance or failure is debatable, with competition between laparoscopic ovarian diathermy (LOD), gonadotrophin ovarian stimulation and metformin to be the preferred choice
Figure 1 Management options for anovulatory infertility associated with polycystic ovarian syndrome (PCOS). FSH, folliclestimulating hormone.
ARTICLE IN PRESS 276
S. Amer
Clomifene has also been shown to exert antioestrogenic activity in other oestrogen-dependent tissues, including the ovary, uterus and vaginal mucosa, via the same mechanism. This unwanted effect of clomifene citrate may explain the discrepancy between the high ovulation and low pregnancy rates achieved with clomifene. It may also explain why the pregnancy rate decreases with higher doses of clomifene, owing to the increased antioestrogenic activity. The prolonged antioestrogenic effects of clomifene citrate on the uterus result in delayed endometrial maturation, poor-quality cervical mucus and an alteration in uterine blood flow. Clomifene citrate is given for 5 days in the early follicular phase (days 2–6) of the menstrual cycle, initially at a daily dose of 50 mg, which is increased in a stepwise progression if ovulation does not occur up to a maximum daily dose of 150 mg. This achieves an ovulation rate of 60–85% and a pregnancy rate of 30–40%. Once ovulation has been achieved on a certain dose, treatment is continued with that dose for 6–12 months. Provided that all other subfertility factors have been excluded, the cumulative conception rate with clomifene citrate continues to increase until it reaches a plateau at treatment cycle 12. Prolonging clomifene treatment beyond 12 cycles has been linked with an increased risk of borderline or invasive ovarian tumour and should therefore be discouraged. Clomifene should be discontinued if it fails to induce ovulation on the maximum dose (clomifene citrate resistance) or if it fails to achieve pregnancy after 6–12 cycles despite ovulation (clomifene citrate failure). It is recommended that women receiving clomifene citrate should be monitored for follicular development with serial pelvic ultrasound scans, at least in their first cycle of treatment, to exclude multifollicular development and to minimise the risk of multiple pregnancies. If mono-ovulation is confirmed, further clomifene citrate cycles could just be monitored with a mid-luteal phase serum progesterone concentration. A progesterone level of 25 nmol/l or more is indicative of ovulation. Women with PCOS who are likely to be clomifeneresistant are those with an elevated BMI, amenorrhoea, marked hyperandrogenaemia and insulin resistance. Age and duration of infertility are the most important determinants of the probability of conception. The younger the woman and the shorter the duration of her infertility, the higher are her chances of conceiving. Other important factors determining the chances of pregnancy are the dose and duration of clomifene citrate. Most pregnancies (50%) occur with
Table 1
clomifene citrate 50 mg, and only about 10% occur with 150 mg. Drawbacks to clomifene treatment include a miscarriage rate of up to 40%, an increased risk of multiple pregnancies and a risk of ovarian hyperstimulation syndrome (OHSS). A possible increase in the risk of ovarian cancer has been suggested if more than 12 cycles are used. In addition, there are common but less serious side effects such as hot flushes, headaches and nausea. Laparoscopic ovarian diathermy Women with PCOS who experience clomifene citrate resistance or failure can be offered LOD, gonadotrophin ovarian stimulation or metformin. LOD, which is as effective as gonadotrophin therapy, offers several advantages over gonadotrophins (Table 1). Importantly, LOD results in monoovulation, with an incidence of multiple pregnancies no higher than the background rate. LOD corrects the endocrine abnormalities associated with PCOS, such as the excess secretion of LH and androgens. These favourable endocrine effects of LOD may reduce the risk of miscarriage known to be increased in women with PCOS. A recent longitudinal long-term follow-up study in our centre has shown that about two-thirds of women with PCOS undergoing LOD respond to the treatment, as evidenced by the resumption of a regular menstrual cycle. In about 50% of the initial responders, the effects are only transient (up to 1 year), whereas the remaining 50% (i.e., a third of the total number of women) continue to have regular ovulatory cycles for many years. As far as conception is concerned, women can expect an approximately 1 in 2 chance of conception within the first year, with a reduction in the incidence of miscarriage to a rate comparable to that of the general population. In about a third of cases, the improvement in reproductive performance seems to last for many years. Several techniques of LOD have been described, of which the most widely used technique is described here (Table 2). A specially designed monopolar electrocautery needle (Fig. 2) is used to penetrate the ovarian capsule at a number of points. The site of application should be away from the ovarian hilum (to minimise the risk of ovarian atrophy) and the fallopian tube (to reduce the risk of mechanical infertility). As the needle is pushed into the ovarian capsule, electricity is activated for 5 s using a monopolar coagulating current set at 30 W. The amount of energy applied per puncture is 5 s 30 W ¼ 150 J. The optimum number of punctures has been shown to be four per ovary.
Advantages of laparoscopic ovarian diathermy over gonadotrophin therapy.
It corrects the endocrine abnormalities of polycystic ovarian syndrome, such as increased levels of luteinising hormone and
androgens. It is less costly. It avoids complex monitoring. A single treatment produces ovulatory cycles for several months or years. It avoids ovarian hyperstimulation syndrome. There is no increase in multiple pregnancy. The incidence of miscarriage may be reduced. It allows concomitant testing of tubal patency and pelvic anatomy.
ARTICLE IN PRESS Reproductive consequences of polycystic ovarian syndrome
Table 2 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
277
Summary of the technique of laparoscopic ovarian diathermy.
Three-puncture laparoscopy is employed. The utero-ovarian ligament is grasped with non-toothed grasping forceps. The ovary is lifted up away from the bowel and stabilized. A specially designed monopolar electrocautery needle is used (Fig. 2). The needle is applied at a right angle to the surface of the ovary. The needle should situated be away from ovarian hilum and the fallopian tube. The power is set at 30 W (coagulating). As the needle penetrates the capsule, the electricity is activated for 5 s. The capsule of the ovary is penetrated to a depth of 6–8 mm. Four punctures are made in each ovary. The ovary is cooled with Hartmann’s solution after drilling. A crystalloid solution is instilled at the end of the procedure.
Figure 2
Laparoscopic ovarian drilling.
Women with PCOS and marked obesity (BMIX35 kg/m2), marked hyperandrogenism (testosteroneX4.5 nmol/l or free androgen indexX15) and/or a long duration of infertility (43 years) appear to be resistant to LOD. It is therefore recommended that alternative methods of treatment should be considered for this group of patients. On the other hand, high pretreatment serum LH concentrations in LOD responders appear to predict a higher probability of pregnancy. The main complications of LOD include adhesion formation and a theoretical risk of excessive ovarian damage leading to ovarian failure. These complications are, however, rare and appear to be of little clinical significance.
Metformin The link between insulin resistance and PCOS led many authors to consider insulin-sensitising agents for the management of this syndrome. These agents, which have been used for many years in type 2 diabetes, have recently been increasingly used worldwide in women with PCOS. The most commonly used agent is metformin, which is the only currently available biguanide drug. Despite its therapeutic benefits in PCOS, the mechanism of action of metformin in women with this syndrome remains uncertain. It improves insulin sensitivity by increasing peripheral glucose uptake in response to insulin at post-
receptor level. This in turn results in correction of the associated hyperinsulinaemia, which is responsible for the hypersecretion of ovarian androgens. In theory, the resulting decrease in androgen production improves the intraovarian microenvironment, which leads to a normalisation of ovarian follicular development. Metformin does not cause hyperinsulinaemia and is therefore not associated with hypoglycaemia. Hypoglycaemia could, however, occur when caloric intake is deficient or when strenuous exercise is not compensated by caloric supplementation. The exact role of metformin in PCOS has still to be established. The most common indication for metformin in PCOS is to induce ovulation in women seeking fertility treatment. Most gynaecologists reserve metformin use for women with PCOS who are resistant to clomifene citrate, although an increasing number of reproductive medicine specialists use it as a first-line treatment for ovulation induction in overweight/obese women with PCOS. If ovulation is not achieved after 3 months of metformin therapy, clomifene citrate could be added. The effective dose range of metformin in women with PCOS has yet to be determined. The drug should be started at a low dose, with gradual dose escalation, both to reduce gastrointestinal side effects and to permit identification of the minimum dose required to achieve regular ovulatory cycles. The usual starting dose is 500 mg twice a day or 850 mg once a day, given with meals. Dosage increases should be made in increments of 500 mg weekly or 850 mg every 2 weeks, up to a total of 2550 mg per day, given in divided doses. Patients should be warned against excessive alcohol intake while receiving metformin as this could precipitate lactic acidosis. Women with PCOS receiving metformin should be monitored for gastrointestinal side effects and for achievement of ovulation. If, after 3 months of treatment, ovulation has not been achieved, consideration should be given to increasing the dose of metformin, adding clomifene citrate or changing to an alternative therapy. If metformin is to be given for a long duration (41 year), initial and periodic monitoring of renal and hepatic functions should also be performed annually. The exact success rate of metformin in infertile women with PCOS is still uncertain. Most studies reported ovulation rates of 30–45% using metformin alone, although more recent reports revealed higher success rates.
ARTICLE IN PRESS 278
S. Amer
Metformin should be temporarily discontinued in patients undergoing radiological studies involving the intravascular administration of iodinated contrast materials, which can lead to an acute alteration of renal function and have been associated with lactic acidosis in patients receiving metformin. Metformin therapy should also be temporarily suspended for any surgical procedure associated with a restricted intake of food and fluids. Other contraindications to the use of metformin include renal dysfunction, hepatic impairment and congestive heart failure. The most common adverse reactions (45%) reported during metformin therapy include diarrhoea (50%), nausea/vomiting (25%), flatulence (12%), asthenia (9%), indigestion (7%), abdominal discomfort (6%) and headache (5%). Lactic acidosis is a very rare (0.003%), but serious, metabolic complication that can occur due to metformin accumulation; when it occurs, it is fatal in approximately 50% of cases. Patients at risk include those with renal insufficiency, hepatic impairment or congestive heart failure. Excessive alcohol intake is also known to potentiate the effects of metformin hydrochloride on lactate metabolism. Gonadotrophin therapy In many centres, gonadotrophin therapy is the preferred second-line treatment for induction of ovulation after clomifene citrate failure in anovulatory women with PCOS. Theoretically, the preferred gonadotrophin preparation for induction of ovulation in women with PCOS is one that does not contain LH, in view of the high levels of endogenous LH in these women. However, human menopausal gonadotrophin and pure FSH preparations have both been successfully used for ovulation induction in women with PCOS. Anovulatory women with PCOS have an increased number of FSH-sensitive small antral follicles and are therefore at increased risk of multifollicular development when receiving exogenous FSH. In view of this hypersensitivity of the polycystic ovaries, the conventional high-dose (150 IU) stepwise protocol, which has been in use since the 1970s, has now largely been replaced by low-dose treatment regimens, which can be applied in a step-up, step-down or sequential fashion. The aim of this approach is to allow FSH to rise slowly just above the FSH threshold while avoiding an excessive ovarian response, thus minimising the risks of OHSS and multiple pregnancy. The chronic low-dose, step-up gonadotrophin regimen has been the preferred method for ovulation induction in
women with PCOS in many centres since 1990 (Fig. 3). This approach is characterised by small initial daily FSH doses (typically 50–75 IU), which are continued for a long period (usually 14 days) without any change in dose and then increased gradually, if necessary, by small amounts (e.g., 25–37.5 IU) every 5–7 days until a dominant follicle emerges. This approach seems to give the best result, with high rates of monofollicular ovulation (70%) and low rates of multiple pregnancy (5%) and OHSS (o1%), while maintaining good pregnancy rates (20% per cycle and 40% per patient). The main problem of this regimen is, however, the long duration of treatment. An attempt by some researchers to reduce the 14-day initial period to 7 days resulted in similar success rates, although there was a slightly higher rate of multiple pregnancy in the 7-day starters. Others have attempted to shorten the 14-day interval to 5 days, but this resulted in an overstimulation rate of 3% and a 26% rate of multiple pregnancies. The step-down approach applies a high starting dose that continues until an ovarian response has been established. The FSH dose is then reduced in small decremental amounts in two steps (Fig. 4). The aim of this approach is to mimic the physiological FSH threshold/window concept of the normal ovulatory cycle, which is characterised by an intercycle elevation in FSH above the threshold for about 5 days (the window) to allow the selection of one leading follicle, followed by a gradual decline in FSH with the subsequent decreasing dependence of the selected (dominant) follicle on FSH. Therapy usually commences with 150 IU FSH per day, reduced by one decrement (37.5 IU) when one follicle has been selected (10 mm), with a further reduction 3 days later by another decrement to 75 IU per day; this is then continued until the day of injection of human chorionic gonadotrophin. The advocates of this approach claim that it is as successful as the chronic low-dose, step-up protocol in achieving monofollicular development and has the advantage of reducing the duration of the treatment. However, the step-down approach requires more stringent follicular monitoring than the step-up regimen. An alternative method is the sequential protocol, which starts with an initial step-up administration of gonadotrophin (similar to the low dose step-up protocol) followed by a step-down regimen once a dominant follicle (X14 mm) has been selected. Two step decrements 37.5 IU/day
Two step increments 25–37.5 IU/day 150 IU/day Until a follicle ≥10 mm
50–75 IU/day 14 days
7 days
7 days
Figure 3 The chronic low-dose, step-up follicle-stimulating hormone stimulation protocol, characterised by a low starting dose with no change for the first 14 days, and then small increments if necessary at 7-day intervals until the injection of human chorionic gonadotrophin or up to 28 days.
112.5 IU/day 3 days
75 IU/day Until hCG injection
Figure 4 The step-down follicle-stimulating hormone (FSH) stimulation protocol, which is based on the FSH threshold/ window concept. It is characterised by a high starting dose (to reach the threshold), which is continued until a dominant follicle (X10 mm) emerges. The dose is then decreased by small decrements in two steps (the end of the window). hCG, human chorionic gonadotrophin.
ARTICLE IN PRESS Reproductive consequences of polycystic ovarian syndrome
Pregnancy loss Pathogenesis Women with PCOS have a high rate of early miscarriage (30–50%) compared with the general population. The mechanism of pregnancy loss in PCOS remains largely unknown. Hypersecretion of LH, hyperandrogenaemia and/ or hyperinsulinaemia have been postulated as the possible underlying mechanisms of early pregnancy loss in women with PCOS. High LH concentrations The literature on the influence of LH hypersecretion on pregnancy outcome is conflicting, although there is considerable evidence to implicate LH hypersecretion in the pathogenesis of early pregnancy loss in women with PCOS. It has been postulated that an elevated concentration of LH results in premature oocyte maturation through an inhibition of oocyte maturation inhibitor. By the time of ovulation, the released egg is physiologically aged and has a diminished capacity to be fertilised, or if fertilised it has a high chance of undergoing miscarriage. Normally, oocytes are maintained by oocyte maturation inhibitor in the first meiotic division until just before ovulation, when oocyte maturation is completed. Although the precise nature of oocyte maturation inhibitor is uncertain, there is evidence to suggest that it is either cyclic adenosine monophosphate (cAMP) or an unknown factor that is activated by cAMP. In women with normal ovulation, stimulation by the preovulatory rise in LH leads to a fall in cAMP and a resumption of meiosis. In PCOS, the granulosa cells acquire LH receptors prematurely, possibly due to hyperinsulinaemia. As a result, inhibition of oocyte maturation inhibitor occurs early, resulting in premature oocyte maturation. Hyperandrogenaemia Data on the influence of high androgen concentrations on pregnancy outcome are scarce and conflicting. In vitro studies have confirmed an inhibitory effect of androgens on endometrial cell growth and activity. The expression of HOXA-10, which is a gene essential for endometrial receptivity, has been shown in in vitro studies to be suppressed by testosterone. It has therefore been hypothesised that androgens might have a detrimental effect on endometrial function, preventing effective endometrial development in the both follicular and the luteal phase.
279 that suppresses LH and/or androgen production (such as pituitary downregulation and LOD) or improves insulin sensitivity (such as dietary modifications and metformin therapy). Evidence from randomised clinical trials has shown that a pituitary suppression of LH with GnRH analogues in women with PCOS with recurrent miscarriage does not improve the live-birth rate. On the other hand, LOD and metformin have been shown to reduce the miscarriage rate in women with PCOS. Current evidence is, however, not strong enough to allow the recommendation of these measures for improving the pregnancy outcome in this group.
Conclusion Women with PCOS often present with reproductive failure as either infertility or early pregnancy loss. The main underlying pathogenesis of these reproductive disorders lies in the disordered folliculogenesis and steroidogenesis, which are the result of an interaction between hyperinsulinaemia, hypersecretion of LH and hyperandrogenaemia. Anovulatory women with PCOS who are obese or overweight should first undergo a weight-reduction programme before embarking on medical treatment. Clomifene citrate remains the standard first-line medical treatment for ovulation induction in infertile PCOS women. Women who fail to conceive on clomifene citrate can be offered LOD, gonadotrophin therapy or metformin. For recurrent miscarriage associated with PCOS, no effective treatment is available, although LOD and metformin present promising treatment modalities for this condition.
Practice points
A weight reduction of 5–10% in obese/overweight
anovulatory women with PCOS restores ovulation in almost all cases. Clomifene citrate is the standard first-line medical treatment for ovulation induction in anovulatory women with PCOS and can achieve high pregnancy rates in properly selected cases. LOD and gonadotrophin therapy are equally effective in inducing ovulation in clomifene-resistant women with PCOS, but LOD offers several advantages, such as the very low multiple pregnancy rate. The chronic low-dose step-up protocol for FSH stimulation is the preferred approach as it induces mono-ovulation in most cases, thus minimising the risks of OHSS and multiple pregnancy.
Hyperinsulinaemia and obesity It has been reported that insulin resistance, which is common in PCOS, is associated with recurrent miscarriage through a mechanism yet to be determined. It is possible that hyperinsulinaemia might affect pregnancy outcome through the interaction with LH and androgen synthesis.
Research directions
Management
The effectiveness of weight-reducing measures such
In the absence of a good understanding of the pathogenesis of early pregnancy loss in women with PCOS, an effective treatment to improve pregnancy outcome cannot be established. It seems logical to apply a therapeutic method
as the lipase inhibitor orlistat and stomach surgery in anovulatory women with PCOS who are obese. The role of metformin in the management of anovulatory infertility and recurrent miscarriage in women with PCOS.
ARTICLE IN PRESS 280
S. Amer
The role of LOD in the management the of recurrent
miscarriage in women with PCOS. Investigation into the possible role of the endometrium in the pathogenesis of early pregnancy loss in women with PCOS. The value of GnRH antagonists in ovulation induction in women with PCOS.
Further reading 1. Amer SA, Li TC. Laparoscopic ovarian surgery for polycystic ovarian syndrome. In: Tarlatzis B, editor. European practice in gynaecology and obstetrics—ovulation induction. Paris: Elsevier; 2002. p. 137–53. 2. Amer S, Li TC, Gopalan V, Ledger WL, Cooke ID. Long term follow up of patients with polycystic ovarian syndrome after laparoscopic ovarian drilling: clinical outcome. Hum Reprod 2002;17:2035–42. 3. Amer S, Li TC, Ledger WL. Ovulation induction using laparoscopic ovarian drilling in women with polycystic ovarian syndrome: predictors of success. Hum Reprod 2004;19:1719–24.
4. Balasch J. The role of FSH and LH in ovulation induction: current concepts and the contribution of recombinant gonadotropins. In: Gardner DK, Weissman A, Howeles C, Shoham Z, editors. Textbook of assisted reproductive techniques. London: Martin Dunitz; 2001. p. 546–65. 5. Balen AH, Tan SL, Jacobs HS. Hypersecretion of luteinising hormone: a significant cause of infertility and miscarriage. Br J Obstet Gynaecol 1993;100:1082–9. 6. Franks S, White D. Low-dose gonadotrophin treatment in polycystic ovary syndrome: the step-up protocol. In: Tarlatzis B, editor. European practice in gynaecology and obstetrics— ovulation induction. Paris: Elsevier; 2002. p. 101–7. 7. Macklon NS, Fauser BCJM. The step-down protocol. In: Tarlatzis B, editor. European practice in gynaecology and obstetrics— ovulation induction. Paris: Elsevier; 2002. p. 111–8. 8. Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group. Revised 2003 consensus on diagnostic criteria and longterm health risks related to polycystic ovary syndrome (PCOS). Hum Reprod 2004;19:41–7. 9. Speroff L, Fritz A. The ovary—embryology and development. In: Clinical gynaecologic endocrinology and infertility, 7th ed. Baltimore: Williams & Wilkins; p. 98–111. 10. Van der Spuy SM. The pathogenesis of infertility and early pregnancy loss in polycystic ovary syndrome. Best Pract Res Clin Obstet Gynaecol 2004;18:755–71.