Relevant BNF sections: 6.5.1, 6.7.1, 22.214.171.124
Around one in six couples wishing to have children have problems conceiving, and in a fifth of these, the cause is identified as an ovulatory disorder.1 Here we describe the management of women with infertility who present in general practice with amenorrhoea or oligomenorrhoea, often the first indication of possible anovulatory infertility.
Statistics from Altmetric.com
Relevant BNF sections: 6.5.1, 6.7.1, 126.96.36.199
Is the woman ovulating normally?
The physiological processes controlling menstrual cycles are complex. Cycles are initiated by pulsatile release of gonadotrophin-releasing hormone (GnRH) from the hypothalamus. GnRH stimulates the pituitary to release follicle-stimulating hormone (FSH) and luteinising hormone (LH), which then stimulate follicle growth in the ovary. Levels of these hormones vary throughout the month and at around day 14 of the menstrual cycle a rapid increase in serum LH concentration triggers ovulation.
'Normal' ovulation is likely to be occurring in any woman who has regular menstrual cycles (i.e. in whom the cycle length does not vary by more than 2 days either side of a mean with an inter-cycle length of 23-35 days). A woman with amenorrhoea is unlikely to be ovulating, and in a woman with oligomenorrhoea (i.e. an interval of 36 days to 6 months between menstruation) ovulation is probably infrequent, if occurring at all.
Causes of anovulation
Polycystic ovary syndrome
Polycystic ovary syndrome (PCOS) is the commonest cause of anovulatory infertility, accounting for about 80% of all cases.2 The diagnosis of PCOS requires the presence of at least two out of the following three criteria: oligomenorrhoea and/or anovulation; hyperandrogenism (clinical and/or biochemical); polycystic ovaries (defined as the presence of 12 or more follicles in each ovary measuring 2-9mm in diameter and/or ovarian volume above 10cm).3* Oligomenorrhoea or amenorrhoea occurs in around two-thirds of women with PCOS, a finding even more common in women with the syndrome who are obese (i.e. have a BMI of 30kg/m or above).5 The cause of PCOS is not known, but seems to involve insulin resistance and impaired glucose tolerance. The resulting raised serum insulin concentrations stimulate androgen production by the ovaries, which in turn inhibits normal follicular development.
* Note that the definition of PCOS has changed since our article on the condition in 2001.4
Premature ovarian failure
Menopause in women aged below 40 years is an explanation for anovulation and accounts for some 4% of cases of anovulatory infertility.2 In some of these women, ovulation stops prematurely because of the development of antiovarian antibodies, or as a result of cancer chemotherapy or radiotherapy. Another cause of ovarian failure is gonadal dysgenesis, commonly due to an abnormal X chromosome complement, as in Turner's syndrome. Women with Turner's syndrome usually present with primary amenorrhoea because the ovaries have failed to develop or have degenerated before puberty. However, women with mosaic chromosomal abnormality may start menstruation and then undergo a premature menopause.
In premature menopause, amenorrhoea is associated with raised serum concentrations of both FSH (above 20units/L) and LH (above 40units/L). Women aged under 40 years who have been diagnosed as having ovarian failure on the basis of raised concentrations of gonadotrophins should be referred for chromosomal analysis. Their ovaries are unlikely to respond to treatment to induce ovulation and treatment of infertility is likely to require egg donation.
Hyperprolactinaemia, the features of which may include amenorrhoea, infertility and galactorrhoea, accounts for about 5% of those who are diagnosed with anovulatory infertility.2 The most common causes of hyperprolactinaemia are medicines that block dopamine receptors (e.g. antipsychotics and the anti-emetics metoclopramide and domperidone) or inhibit dopamine synthesis (e.g. benserazide). It can also be caused by tricyclic antidepressants, SSRIs and histamine (H2)-receptor antagonists. Non-drug causes of hyperprolactinaemia include a prolactin-secreting pituitary tumour (prolactinoma), most commonly a microprolactinoma (i.e. tumour less than 10mm in diameter); with such a tumour, prolactin concentrations are usually raised to above 1,000milliunits/L. Rarer causes include a non-prolactin-secreting hypothalamic tumour that compresses the pituitary stalk, so blocking the usual dopa-minergic inhibition of prolactin. With such a tumour, prolactin levels may be raised only modestly (e.g. remaining below 1,000milliunits/L). As secretion of prolactin is often stimulated by stress, serum concentrations of the hormone can be raised after a routine physical examination, and so it is usually worthwhile repeating the measurement if a high initial concentration is found.
Hypothalamic amenorrhoea, which is associated with a reduced secretion of GnRH and subsequently gonadotrophin deficiency and amenorrhoea, is caused by anorexia nervosa, bulimia, intense exercise regimens, severe stress, or being underweight (i.e. BMI below 19kg/m whether or not associated with an eating disorder). Hypothalamic amenorrhoea is the cause of around 5% of cases of anovulatory infertility.2 It may be difficult to distinguish hypothalamic amenorrhoea from pituitary dysfunction and so where the diagnosis is suspected, specialist advice should be sought.
Other possible causes of infertility
Anabolic steroids, such as those used as performance-enhancing drugs, can cause amenorrhoea or oligomenorrhoea.6 There is some evidence to suggest that NSAIDs can cause reversible infertility,7 but this is by no means proven.
Assessment and advice
It is important to establish the woman's menstrual history, to measure her weight and height for estimation of BMI, and to get a detailed drug history, including any non-prescribed drugs. In women with long menstrual cycles (more than 35 days) that are irregular, a blood test to measure mid-luteal serum progesterone concentration (i.e. measured 7 days before the expected onset of menstruation) may be a helpful indicator of whether the woman is ovulating, with concentrations below 16nmol/L suggesting that ovulation is unlikely to have occurred. Timing of the test is likely to be difficult in women with irregular cycles, and so it may need to be repeated (e.g. weekly until menstruation starts).8 Serum testosterone concentrations should be measured in women with symptoms of hyperandrogenism. Measurement of serum concentrations of FSH and LH can help confirm a diagnosis of ovarian failure. Measurement of serum prolactin concentration and of thyroid function may help identify anovulation due to drug-induced hyperprolactinaemia or to pituitary disorders.
Women suspected of having anovulation and whose serum prolactin concentrations are raised should first stop any suspected drug treatment, if possible. The woman should be referred to a specialist if the high concentrations are not due to a drug; the drug treatment cannot be stopped; or there is evidence of pituitary disease (e.g. visual symptoms, low thyroid hormone levels).
As the woman will be attempting to conceive, she should be offered a general health check in preparation for a future pregnancy, including checking rubella immunity status (and, if necessary, offering rubella vaccination), advising the woman to take folic acid while trying to conceive and during the first 12 weeks of a pregnancy in order to help prevent a neural tube defect, and reviewing any medicines that could damage the baby. The woman should also be given lifestyle advice (e.g. on stopping smoking and minimising alcohol consumption).
The recently published National Institute for Clinical Excellence (NICE) guideline on assessment and treatment of fertility problems recommends that people should be treated by a specialist team because this is likely to improve effectiveness and efficiency of treatment and is known to improve satisfaction among patients.8 The guideline does not set out where initial investigations should take place (e.g. whether in primary or secondary care). However, there may be established local protocols which could help to avoid unnecessary duplication of tests, such as measurement of FSH and LH. Before starting any ovulation-induction therapy in a specialist unit, the woman may also undergo tests of Fallopian tube patency and for chlamydial infection.
Treating infertility in PCOS
Treatment of women with PCOS should begin with lifestyle measures. For those who are obese, a combination of weight loss and exercise may stimulate ovulation. In one study, 18 women (60% with PCOS) who had a BMI of at least 30kg/m, who had been infertile for at least 2 years and not responded to therapy with clomifene* (an anti-oestrogen stimulator of ovulation; see below), underwent weekly 2-hour group sessions for 6 months that involved exercising and advice on diet.9 This was not a randomised trial, and the 5 women who dropped out of the programme were used as 'controls'. By the end of 6 months, the 13 women who continued to attend the group sessions had lost an average of 6.3kg and after 6 months 12 had started ovulating. A year after the end of the programme, 11 had become pregnant (5 spontaneously without any specific treatment). By contrast, those who dropped out lost an average of 1.4kg and none conceived spontaneously during or up to 12 months after the trial.
* This new spelling reflects the use of recommended International Non-proprietary Names (rINNs) instead of former British Approved Names (BANs), as required by European law. For more information, see the British National Formulary, or http://medicines.mhra.gov.uk/inforesources/productinfo/banslist.pdf
If the woman is not overweight or weight loss is ineffective, an anti-oestrogen is usually the first-line ovulation-induction therapy. Where this fails or is inappropriate, second-line treatment options are gonadotrophins, metformin or surgery (laparoscopic ovarian diathermy).
As weight loss can be difficult to achieve, there should not be undue delay in referring a woman who is overweight for specific treatment to induce ovulation if she has failed to lose weight, particularly if she is aged over 35 years.
Oral anti-oestrogen therapy
Clomifene acts by blocking oestrogen receptors in the hypothalamus, so reducing the negative feedback effect of circulating oestrogen. This in turn leads to increased FSH secretion, which then results in follicular development. The summary of product characteristics (SPC) for clomifene recommends a dose of 50mg daily for 5 days, starting within 5 days of onset of progestogen-induced or spontaneous menstrual bleeding. Ovulation usually occurs within 5-10 days after taking the last clomifene tablet, with around two-thirds of women ovulating in the first cycle and most by the third cycle. Ultrasound monitoring should be used to identify whether the woman has produced a dominant follicle, an indicator of likely ovulation. Some specialists also monitor serum progesterone concentrations to help confirm ovulation, and also LH concentrations, because clomifene can cause an exaggerated secretion of LH in the mid-follicular phase, an effect associated with a reduced chance of conception (because it changes the consistency of cervical mucus). If the woman responds to clomifene by producing more than two follicles, usual practice is to advise her against becoming pregnant and the daily dose is typically reduced to 25mg (an unlicensed dose). If ovulation has not occurred, the dose is typically increased to 100mg or 150mg (both unlicensed doses). Women who are overweight are less likely to respond to clomifene than those of normal weight.10
There are relatively few published randomised double-blind trials of clomifene. A systematic review identified three such studies, involving a total of 217 women. A meta-analysis of data from the trials showed that, compared with placebo, clomifene (50-250mg daily) during 1-5 treatment cycles increased the likelihood of ovulation (odds ratio 6.82, 95% CI 3.92-11.85) and pregnancy (odds ratio for conception per treatment cycle 3.41, 95% CI 1.23-9.48).11 Clomifene is reported to induce ovulation in around 60-85% of women with anovulatory infertility, with 30-50% becoming pregnant.10,12 The NICE guideline recommends that women who are ovulating with clomifene therapy, but who do not become pregnant after 6 months, should be offered continued therapy with clomifene together with intrauterine insemination.
Clomifene can cause hot flushes, nausea, depression, and breast tenderness and there have been reports of convulsions. It can also cause blurred vision (due to increased pressure in the anterior chamber of the eye), in which case, the drug should be stopped immediately. Clomifene therapy increases the likelihood of multiple pregnancy (usually two embryos, but occasionally more) to an estimated rate of 8-13% of induced pregnancies13 (compared with a natural multiple birth rate of 1-2%8). Ovarian hyperstimulation syndrome is a rare, and potentially life-threatening effect characterised by enlarged ovaries (filled with enlarging follicles), causing abdominal pain, distension, ascites and haemoconcentration. It requires management in hospital.
There have been concerns that multiple courses of clomifene therapy (12 or more treatment cycles) may increase the risk of ovarian cancer. The SPC recommends that women receive no more than around 6 treatment cycles. By contrast, NICE recommends that there may be clinical benefit in using up to 12 cycles in some women.8 This is because cumulative pregnancy rate continues to increase with use of up to 10-12 cycles of clomifene treatment.
Who should start clomifene?
The need for monitoring with ultrasound (at least in the first cycle of treatment) and serum endocrine tests means that, in general, clomifene should not be started in general practice. However, it is possible for GPs to continue to prescribe clomifene after the appropriate dose has been established; this should be done as part of a formal shared-care agreement between hospital and primary care.
Tamoxifen has a similar structure and mode of action to clomifene and one of its licensed uses is for the treatment of anovulatory infertility. However, it is rarely used in this way, with its place being limited to treating women who experience troublesome unwanted effects with clomifene.
Injectable preparations of FSH and LH, given either alone or in combination, are available for inducing ovulation in women with PCOS who are resistant to clomifene or those in whom clomifene induces high serum concentrations of LH and so alters cervical mucus. In published case series, cumulative live birth rates from the use of gonadotrophin therapy for PCOS were 65% after 6 cycles of treatment and 75% after 12 cycles.15 However, the use of such therapy to induce ovulation carries a significantly higher risk of multiple pregnancy than does clomifene, and is more likely to cause ovarian hyperstimulation syndrome. Therapy using parenteral gonadotrophins must therefore be initiated and supervised by a hospital specialist.
The role of insulin resistance as a principal factor in the pathogenesis of PCOS has led to the use of metformin to treat women with the condition,4 an unlicensed use in the UK. The drug's mode of action in this setting is not clear, but it is thought to reduce androgen levels by increasing sensitivity to insulin. Evidence from a systematic review of 13 randomised controlled trials (involving a total of 310 women) suggests that when used to treat oligomenorrhoeic or amenorrhoeic women with PCOS, metformin increases the likelihood of ovulation compared with placebo (46% of women vs. 24%, p<.0001).16 It did not affect pregnancy rates, although most of the trials reporting pregnancy rates were of less than 4 months' duration. However, a meta-analysis of data from three trials suggests that adding metformin to clomifene increases the likelihood of ovulation (odds ratio 4.41, 95% CI 2.37-8.22, p<.0001) and of pregnancy (odds ratio 4.40, 95% CI 1.96-9.85, p=0.0003) compared with clomifene alone.16 Two of the trials involved women known to be resistant to clomifene, while in the trial in which women were not selected on the basis of being resistant to clomifene, the addition of metformin did not offer any benefit. The NICE guideline recommends that women with PCOS who have not responded to clomifene and who have a BMI greater than 25kg/m should be offered metformin plus clomifene.8
Metformin may cause nausea, vomiting, flatulence and diarrhoea although starting at a low dose and titrating up at 2-weekly intervals, and taking it with food can reduce the likelihood of these symptoms. The drug is contraindicated in pregnancy and in patients with any degree of renal impairment, because it can cause lactic acidosis.
Laparoscopic ovarian diathermy
The aim of this surgical procedure is to cauterise the surface of each ovary at several points. How this might bring about ovulation is not known. However, in women who do not ovulate with clomifene therapy, the method appears as effective as gonadotrophin therapy in increasing pregnancy rates and is less likely to result in a multiple pregnancy.17 It can be the treatment of choice after clomifene in a woman who in any case needs a laparoscopy for assessment of her Fallopian tubes.
Ovulatory abnormalities secondary to prolactinomas often respond well to treatment with dopamine receptor agonists (usually bromocriptine or cabergoline). Such treatment should be carried out under specialist supervision. Ideally, the dopamine agonist should be taken until conception but sometimes it is necessary for the drug to be continued throughout pregnancy (e.g. in the treatment of a macroprolactinoma, prolactin concentrations may rise rapidly when the dopamine agonist is stopped). Safety of the drugs in pregnancy is therefore an important consideration. Compared with bromocriptine, cabergoline appears more effective in increasing pregnancy rates, and causes fewer and less severe gastrointestinal symptoms,18 but there is less experience of its use in pregnancy. The SPC for cabergoline advises discontinuing the drug 1 month prior to intended conception. By contrast, the SPC for bromocriptine advises discontinuing bromocriptine after the first missed menstrual period. Therefore, in reality, bromocriptine is probably more commonly used. Data from two small retrospective studies suggest no increased rate of congenital abnormalities in the offspring of women taking cabergoline at conception.19,20 For women intolerant of, or unresponsive to, bromocriptine, many specialists recommend continuing cabergoline until pregnancy is achieved and then stopping treatment.21
Treating hypothalamic amenorrhoea
Women whose amenorrhoea appears related to factors such as being underweight or undertaking excessive exercise need appropriate lifestyle advice. Specialist referral should be considered so that treatment with pulsatile gonadotrophins or gonadotrophin-releasing hormones administered by subcutaneous infusion pump can be started. Such treatment is often effective in inducing ovulation (in a case series, cumulative pregnancy rate was 82% after 12 cycles of therapy).15 However, restoration of weight is mandatory, because while it may be possible to stimulate ovulation with drugs, women who are underweight are more likely than those of normal weight to have problems during pregnancy, including miscarriage, intrauterine growth retardation, intrauterine death and stillbirth. Women with eating disorders, such as anorexia nervosa, need specialist management before starting fertility treatment.
Failure to ovulate is a common cause of infertility, the underlying causes of which include polycystic ovary syndrome, hyperprolactinaemia and hypothalamic amenorrhoea. Much can be done in general practice to diagnose ovulatory disorders and to ensure that the woman receives appropriate management with minimal delay. Ideally, there should be local protocols on initial assessment and referral to avoid duplication of investigations. Where appropriate, management should be based on lifestyle changes (including appropriate weight gain or loss). Otherwise, referral to a specialist for drug or surgical treatments to bring about ovulation may be needed. Continued care can be part of a formal shared-care agreement between hospital and primary care.
[M=meta-analysis; R=randomised controlled trial]
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.