Anovulatory subfertility is a heterogeneous condition with various underlying causes, which should be identified with appropriate history taking, physical examination and relevant investigations. Optimisation of body weight is essential in either underweight, overweight or obese individuals. Women with hypogonadotrophic anovulation can be treated with pulsatile gonadotrophin-releasing hormone therapy or a gonadotrophin preparation containing both follicle-stimulating hormone or luteinising hormone activities. For normogonadotrophic anovulation, clomiphene citrate should be used as first-line medical treatment. Metformin co-treatment with clomiphene citrate may be considered in a subgroup of women with polycystic ovary syndrome who are obese or clomiphene-resistant. Ovulation induction with gonadotrophin or laparoscopic ovarian drilling is the next option. Dopamine agonist is indicated for anovulation as a result of hyperprolactinaemia.
Introduction
Ovulation disorders is the cause of infertility in about 15–20% of women seeking treatment for subfertility. Subfertility caused by ovulation disorders can be classified according to the site of deficiency in the hypothalamic–pituitary–ovarian axis ( Table 1 ) as in the World Health Organization (WHO) classification.
| Hypogonadotrophic hypogonadism (World Health Organization group I) |
| Idiopathic hypogonadotrophic hypogonadism |
| Kallmann’s syndrome (isolated gonadotrophin deficiency and anosmia) |
| Functional hypothalamic dysfunction (e.g. excessive weight loss such as in anorexia nervosa, exercise, stress, drugs, iatrogenic) |
| Pituitary tumour, pituitary infarct (e.g. Sheehan’s syndrome) |
| Normogonadotrophic normogonadic ovarian dysfunction (World Health Organization group II) |
| Polycystic ovary syndrome |
| Hypergonadotrophic hypogonadism (ovarian failure) (WHO group III) |
| Genetic (e.g. Turner’s syndrome) |
| Autoimmune causes |
| Infection (e.g. mumps oophoritis) |
| Iatrogenic (e.g. surgical menopause, post-radiotherapy or chemotherapy) |
| Idiopathic |
| Other endocrinopathies, such as hyperprolactinaemia, thyroid dysfunction, other conditions of androgen excess such as congenital adrenal hyperplasia and androgen-secreting adrenal and ovarian tumours. |
Initial assessment and investigations
Fig. 1 is a flowchart of diagnosis and treatment for anovulatory infertility.
Initial assessment and investigations
Fig. 1 is a flowchart of diagnosis and treatment for anovulatory infertility.
History and physical examination
Anovulatory women typically present with oligomenorrhoea or amenorrhoea, although about 10% of women with regular menstrual cycles could be anovulatory. A detailed menstrual history needs to be taken, including the cycle length, regularity and the last menstrual period. Symptoms associated with hyperprolactinaemia (e.g. galactorrhoea, headache, visual disturbance), thyroid dysfunction and climacteric symptoms should be enquired as well, in addition to any related drug or medical history that might have precipitated the menstrual cycle disturbance. Hyperandrogenic symptoms, such as hirsutism, acne, greasy skin, and male-pattern alopecia, may be evident in polycystic ovarian syndrome (PCOS) or other disorders of androgen excess. Other relevant history would include any recent weight changes, diet, stress and exercise pattern, as well as history of secondary sexual development.
In examination, stigmata of chromosomal abnormalities, secondary sexual development, body mass index, signs of hyperandrogenism, galactorrhoea and goitre, as well as abnormal genital development, should all be noted.
Investigations
A mid-luteal phase serum progesterone test is the recommended investigation for assessment of ovulation status, and a level of 30 nmol/l or above confirms presence of ovulation. It is preferably taken about 7 days before the next menstrual period, and the correctness of the timing can be assessed by noting the menstrual date after the blood test. Borderline levels (15–30 nmol/l) are usually caused by mis-timed blood taking, and a repeat test may help confirmation. Basal body temperature charts and urine luteinising hormone assays are less robust investigations and subject to greater variations; these are not recommended for determination of ovulation.
A blood test for serum follicle-stimulating hormone (FSH), thyroid-stimulating hormone (TSH) and prolactin are the basic work-up for oligoamenorrhoea who are presumed to be anovulatory. A progestogen challenge test would offer a functional test of the oestrogenic status. Androgen profile may be checked, although universal consensus for defining biochemical hyperandrogenism is lacking, and reference ranges vary between laboratories. In case of severe hyperandrogenaemia (e.g. total testosterone > 5 nmol/l) or signs of severe hirsutism or virilisation, it needs to be followed by checking 17-hydroxyprogesterone to exclude late-onset congenital adrenal hyperplasia, an overnight dexamethasone suppression test to exclude Cushing’s syndrome, as well as appropriate imaging (ultrasound scan of the pelvis and computed scan of the adrenals) to exclude an androgen-secreting tumour.
In the Rotterdam criteria, ultrasonographic features of polycystic ovaries is one of the diagnostic criteria for PCOS. Polycystic ovary morphology is defined as either 12 or more follicles measuring 2–9 mm in diameter, increased ovarian volume (> 10 ml), or both; only one ovary fitting this definition is sufficient for the definition. If a dominant follicle (> 10 mm) or a corpus luteum is present, a repeat scan in the next cycle should be carried out. Luteinising hormone to FSH ratio is no longer regarded as a diagnostic criteria for PCOS, owing to the large variations.
Before managing anovulatory subfertility, semen analysis of the male partner should be carried out to exclude severe male factors that would alter the choice of treatment. Unless tubo-peritoneal factor is suggested by history, simple oral medical treatment for ovulation induction can be commenced without tubal assessment.
Treatment principles for the common causes
Hypogonadotrophic hypogonadism (World Health Organization group I)
Hypogonadotrophic hypogonadism may be manifested as primary or secondary oligo-amenorrhoea. Serum FSH, luteinising hormone and oestradiol levels are typically low, although the degree may vary and is sometimes not clearly distinguishable from WHO group II anovulation. If the cause is treatable, treatment should be directed to the cause. Surgery may be indicated in cases of intracranial tumours. Women with anorexia nervosa may benefit from psychotherapy and weight gain after extensive counselling. Ovulation can be induced by pulsatile gonadotrophin-releasing hormone therapy (GnRH) administration (for hypothalamic but not pituitary causes) or gonadotrophins (containing both FSH and luteinising hormone) if anovulation persists despite optimisation of body weight.
Normogonadotrophic anovulation (World Health Organization group II)
With normogonadotrophic anovulation, PCOS accounts for most of these women, although the typical full-blown picture may not be present in some cases. Other causes of androgen excess may also contribute to this picture. Weight reduction should be the first-line treatment in obese women with PCOS, and this may result in resumption of spontaneous ovulation and also improve their response to ovulation induction if indicated. Ovulation induction can be achieved with clomiphene citrate or aromatase inhibitors. Those not responsive to, or failing, oral treatment, may be offered gonadotrophins or ovarian drilling. Other causes of androgen excess should be managed accordingly.
Hypergonadotrophic hypogonadism (World Health Organization group III)
Women with hypergonadotrophic hypogonadism may present with primary or secondary amenorrhea with elevated FSH and low oestradiol levels. Ovarian biopsy for detecting the presence of follicles in case of resistant ovary syndrome is not recommended because of the invasive nature and doubtful value of the procedure. About one-half of young women with ovarian failure may have intermittent and unpredictable ovulation, and spontaneous pregnancies have been reported in about 5–10% of cases subsequent to the diagnosis. Any form of ovulation-induction treatment, however, is not advisable in these women. The only realistic option is assisted reproduction using donor eggs.
Hyperprolactinaemia
Hyperprolactinaemia interferes with the pulsatile secretion of GnRH and impairs normal ovarian function. Causes of hyperprolactinaemia include prolactin-producing adenoma, other pituitary tumours that block the inhibitory signal of the hypothalamus, primary hypothyroidism, chronic renal failure, and drugs such as neuroleptics and calcium channel blockers.
Conditions of falsely high prolactin levels should be noted so as to avoid unnecessary interventions. Transiently raised prolactin level can be triggered by breast examination and stress, for instance as a result of venepuncture. Macroprolactinaemia can also be a cause of pseudohyperprolactinaemia. Macroprolactin are high molecular weight polymers of prolactin molecules that are biologically inactive. It can be differentiated by polyethylene glycol study.
Asymptomatic women with hyperprolactinaemia can be observed without treatment. In anovulatory women with hyperprolactinaemia, dopamine agonist is the first-line treatment to lower the prolactin level and shrink the prolactinoma if present. The risk of tumour expansion with neurological sequelae is rare with microadenoma (< 10 mm). Women with macroprolactinoma (> 10 mm) may be managed with the neurosurgeon’s input, and they should conceive after normalisation of serum prolactin and significant reduction of tumour volume so as to reduce the neurological risk of optic chiasm compression during pregnancy. Surgical treatment by transphenoidal pituitary adenectomy and rarely radiotherapy may be required if medical treatment fails to shrink a macroadenoma.
Weight management as a therapeutic option
Weight gain in underweight women
Undernutrition and underweight can exert an inhibitory effect on the hypothalamo–pituitary–ovarian axis, thereby suppressing ovulation. Although underweight does not seem to adversely affect the pregnancy rate in fertility treatment, underweight or malnourished women who conceive have higher risks of obstetric complications, such as hyperemesis gravidarum, anaemia, fetal growth restriction and premature delivery. Therefore, women with eating disorders should be advised to postpone conception until remission and normalisation of body weight. Nutritional counselling should be offered.
Weight reduction in overweight and obese women
Overweight and obese women are associated with a higher incidence of menstrual disturbance, ovulation disorders and subfertility in women. They have poorer response to ovulation induction and a higher risk of pregnancy complications, such as miscarriage, gestational diabetes, hypertension, fetal macrosomia and intrapartum problems.
Weight loss should be advised before fertility treatment in overweight and obese women. Even a modest weight loss of 5% may restore spontaneous ovulation and possibly improve pregnancy rate. Data are limited on its effect on pregnancy complications. Weight loss should be achieved with lifestyle modification by caloric restriction and increased physical exercise. Reduced caloric intake by 500–1000 kcal/day has been suggested to be effective, aiming at reducing the body weight by 7–10% over a period of 6–12 months. Structured moderate exercise lasting for 30 mins or more per day is advisable. In individuals who experience difficulty in reducing significant weight with lifestyle intervention alone, the use of anti-obesity drugs can be an adjunct. Orlistat and metformin are the options currently, and their use is probably safe for women planning for pregnancy. Bariatric surgery can be an option for refractory cases.
The effects of caloric restriction, excessive physical exertion or pharmacological intervention in the periconceptional period are not yet known, and hence those interventions should precede any planned pregnancy or fertility treatment.
Medical options for ovulation induction
Dopamine agonists
The dopamine agonists (bromocriptine, carbergoline and quinagolide) inhibit prolactin secretion from the pituitary lactotrophs, leading to restoration of gonadal function and shrinkage of prolactinoma. Bromocriptine is given at a daily dosage of 2.5–20 mg in divided doses two to three times a day. Cabergoline and quinagolide have longer biological half-lives than bromocriptine. Cabergoline is given once or twice weekly and quinagolide once daily. Response can be monitored by menstrual pattern and serum prolactin levels.
Dopamine agonist therapy restores ovulation in about 90% of women with anovulation related to hyperprolactinaemia. Cabergoline and quinagolide are more effective than bromocriptine in restoring normal prolactin concentrations and ovulatory cycles. In women who do not ovulate even when prolactin concentrations are within normal range, dopamine agonists can be combined with anti-oestrogen or gonadotrophin as appropriate.
Common side-effects with bromocriptine include nausea, vomiting, abdominal cramps, vertigo, postural hypotension, headaches and drowsiness. Around 12% of women discontinue the treatment for this reason. This can be minimised by gradual step up from a low starting dose and taking the drug at bedtime, or by giving vaginal bromocriptine. Significantly fewer side-effects were reported in women taking cabergoline and quinagolide compared with bromocriptine.
Dopamine agonists have not been associated with any adverse effect on pregnancy or fetal development, although we commonly recommend women with microprolactinomas or idiopathic hyperprolactinaemia to stop treatment once pregnancy is confirmed in order to avoid any potential harm. Treatment may be continued during pregnancy in cases of macroprolactinoma or where tumour expansion is evident.
The European Medicines Agency has issued warnings on the risk of cardiac valvular fibrosis associated with cabergoline, which were reported mostly with long-term use in high doses. Bromocriptine and quinagolide have weaker affinity for the 5HT-2B receptor and are hence thought to be less valvulopathic, although few data are available, especially with the low-dose use for treatment of hyperprolactinaemia.
Anti-oestrogens
Clomiphene citrate
Clomiphene citrate is commonly used as the first-line drug in treatment of WHO group II anovulation. It is an orally active non-steroidal compound that acts primarily by its anti-oestrogenic property. It displaces endogenous oestrogen from oestrogen receptors in the hypothalamic–pituitary axis, diminishing its negative feedback and hence increasing the secretion of endogenous GnRH and gonadotrophins, which subsequently induce ovulation.
It should be started at 50 mg daily for 5 days after a spontaneous or progestogen-induced withdrawal bleeding. The recommended maximum dose is 150 mg per day, as no clear evidence of efficacy has been shown at higher doses. Commencement from any day between days 2 to 5 produced the same results. Ovulation usually occurs within 5–10 days after the last tablet. If no ovulation occurs, the dose can be stepped up at increments of 50 mg daily until ovulation occurs, or a maximum dose of 150 mg daily is reached.
It is recommended that the response is monitored, at least during the first treatment cycle. Follicular tracking by transvaginal pelvic ultrasound helps to identify non-response, excessive response or reduced endometrial thickness. Ovulation can be confirmed by checking serum progesterone in the mid-luteal phase.
Treatment with clomiphene citrate can achieve ovulation, pregnancy and live birth rates of 73%, 36% and 29% per woman, respectively. Treatment should generally be limited to six ovulatory cycles, but further treatment up to 12 cycles may be considered on an individual basis. Woman who are resistant to maximum dose of clomiphene citrate or who fail to conceive after six ovulatory cycles of clomiphene citrate treatment should be offered second-line treatment.
Two randomised-controlled trials showed no significant difference in pregnancy rate with human chorionic gonadotropin (hCG) administration or not, nor in the incidence of miscarriage or multiple pregnancy. Therefore, the use of hCG trigger is not routinely needed except in women where failure of follicular rupture is evident.
Clomiphene citrate is generally well tolerated, although side-effects, including hot flushes, breast discomfort, abdominal distension, nausea, vomiting, nervousness, mood swings, dizziness, hair loss and disturbed vision may be occasionally encountered. About 7–10% of pregnancies induced by clomiphene citrate are twins, and 0.5–1% are triplets. Although mild ovarian enlargement is relatively common, severe ovarian hyperstimulation syndrome (OHSS) is rare. No increase in spontaneous abortion or congenital abnormalities in pregnancies induced by clomiphene citrate occur.
Tamoxifen
Tamoxifen is a triphenylethylene derivative that is structurally similar to clomiphene citrate. The suggested dose in ovulation induction is 20–40 mg daily for 5 days after a spontaneous period or withdrawal bleed. A meta-analysis reported that tamoxifen and clomiphene citrate had similar ovulation rates and pregnancy rate per cycle. The use of tamoxifen for ovulation induction, however, is an off-label use, despite the available evidence on its efficacy and safety for such indication.
Insulin-sensitising agents
In women with PCOS, insulin resistance is one recognised metabolic disturbance. Insulin-sensitising agents can increase the insulin responsiveness in target tissues, reduce the compensatory hyperinsulinaemia, hence ameliorate its adverse effect on ovulatory function. Metformin, a biguanide, is one of the most commonly used insulin sensitiser, and can be given at 500 mg once daily or 850 mg twice daily with meals. Gastrointestinal upset, including nausea, vomiting and diarrhoea, are the most common side-effects. Lactic acidosis is a rare though serious complication, and hence it is contraindicated in women with renal, hepatic or major cardiovascular disease or hypoxia.
As reported in a Cochrane review, metformin used alone improves the ovulation rate (OR 2.12; 95% CI 1.5–3.0) and clinical pregnancy rate (OR 3.86; 95% CI 2.18 to 6.84) compared with placebo or no treatment, but not the livebirth rate (OR 1.0; 95% CI 0.16 to 6.39). When compared with clomiphene citrate, metformin leads to lower ovulation rate (OR 0.48; 95% CI 0.41 to 0.57) and clinical pregnancy rate (OR 0.63; 95% CI 0.43 to 0.92), and a non-significant trend of lower livebirth rate (OR 0.67; 95% CI 0.44–1.02). Co-treatment with metformin and clomiphene improves the ovulation rate (OR 1.76; 95% CI 1.51 to 2.06) and clinical pregnancy rate (OR 1.48; 95% CI 1.12 to 1.95), but not the livebirth rate (OR 1.05; 95% CI 0.75 to 1.47) compared with clomiphene nitrate alone.
Previous subgroup meta-analyses showed a higher clinical pregnancy rate after metformin plus clomiphene citrate co-treatment compared with clomiphene citrate alone in obese only, but not in non-obese women, and in women who are resistant to clomiphene citrates only. Another systematic review also suggested that metformin plus clomiphene citrate produced higher live birth rates than clomiphene citrate alone only in women resistant to clomiphene citrate but not in clomiphene-citrate-naïve women.
No good evidence has supported the safe and effective use of other insulin sensitisers in fertility treatment.
Aromatase inhibitors
Aromatase catalyses the conversion of androstenedione and testosterone to oestrone and oestradiol, respectively. Aromatase inhibitors block the production of oestrogen, reducing the negative feedback to the hypothalamic–pituitary axis, hence increasing endogenous FSH secretion. They have been used for many years as an adjunctive treatment for breast cancer and are gaining in popularity for ovulation induction in women with PCOS.
Letrozole is the third-generation aromatase inhibitor most commonly used for ovulation induction. When compared with clomiphene citrate, letrozole has a much shorter half-life and hence minimal suppressive effect on the endometrium. The recommended regimen is 2.5–5 mg per day for 5 days commencing in the first 5 days of spontaneous or induced bleeding, or as a single dose of 20 mg on day 3 of the period. The monitoring is similar to that of clomiphene citrate. The use in ovulation induction, however, is an off-label use.
Letrozole gave an ovulation rate of 70–84% and a pregnancy rate of 20–27% per cycle in women with PCOS resistant to clomiphene citrate. A meta-analysis showed no significant difference between letrozole and clomiphene citrate in the ovulation rate, pregnancy rate per cycle or per patient.
Letrozole is generally well tolerated. Earlier reports suggested that letrozole results in more monofollicular development and significantly lower multiple pregnancy rate compared with clomiphene citrate. In a recent randomised-controlled trial, it had comparable chance of twin pregnancies as clomiphene citrate (8.3% v 9.1%). A case report of a triplet pregnancy after ovulation induction with letrozole has also been published.
Teratogenic effects of letrozole have been described in animal studies. An abstract report suggested that the use of letrozole for subfertility treatment might be associated with a higher risk of congenital cardiac and bone malformations in newborns. A retrospective study with a much larger sample size could not show any difference in the overall rates of major and minor congenital malformations among newborns from mothers who conceived after letrozole or clomiphene citrate treatments.
Gonadotrophin-releasing hormone
Gonadotrophin-releasing hormone given in a pulsatile fashion restores the normal pattern of gonadotrophin secretion as in spontaneous menstrual cycles, leading to the development of a single dominant follicle. It is indicated for the treatment of hypogonadotrophic anovulation caused by hypothalamic dysfunction but not pituitary problem. It is given subcutaneously or intravenously through a small butterfly cannula using a small battery-operated pump.
Cumulative pregnancy rates of 80% and 93% have been reported after six and 12 cycles of treatment, respectively. Multiple pregnancy rates ranged between 3.8 and 13.5%. Other drawbacks include the inconvenience of having the needle in situ for a long period, needle site reaction and infection, displacement and pump failure, and its relatively high cost.
Gonadotrophin
The use of exogenous gonadotrophins is to overcome the FSH threshold required for follicular development. Follicle-stimulating hormone is the key gonadotrophic hormone for follicular development. In women with hypogonadotrophic hypogonadism, a preparation containing both FSH and luteinising hormone gives better outcome than purely FSH because luteinising hormone is required for ovarian steroidogenesis to achieve optimal endometrial proliferation. In these women, luteal phase support is necessary.
Chronic low-dose, step-up protocol
Chronic low-dose, step-up protocol is currently the recommended protocol in many centres worldwide. The principle is to determine the FSH threshold gradually, avoiding excessive stimulation and multifollicular development. Follicle stimulating hormone is commenced at a low starting dose (37.5–75 IU/day) for at least 10–14 days and stepped up at weekly intervals by increments of 37.5 IU up to a maximum of 225 IU/day if there is no response. The same dose is maintained once follicular growth is observed. Once one to two dominant follicles reach 18 mm in mean diameter, hCG is given to trigger ovulation. It may take up to several weeks to achieve an ovarian response in the first treatment cycle in women with a high FSH threshold. In subsequent cycles, FSH can be started at a dose that gives rise to ovarian response in the first cycle.
Step-down protocol
The step-down protocol mimics the physiological hormonal cycle. Gonadotrophin injection is commenced at a daily dose of 150 IU starting from days 2 or 3 of the cycle, and the ovarian response is monitored by ultrasound every 2–3 days. The same dose is continued until a dominant follicle reaches 10 mm, which is then reduced to 112.5 IU/day for 3 days and then further down to 75 IU/day until hCG is administered to trigger ovulation. It requires more intense monitoring than the step-up protocol. The step-down protocol has a shorter duration of stimulation compared with the step-up protocol, but a higher rate of multifollicular development and ovarian hyperstimulation syndrome, as well as a lower ovulation rate. The pregnancy rate is comparable between the two regimens.
Ovulation induction with low-dose gonadotrophin regimens has monofollicular ovulation rate and pregnancy rates of about 70% and 20%, respectively, and a cumulative pregnancy rate of 55–70%. The rates of ovarian hyperstimulation syndrome and multiple pregnancy are as low as less than 1% and 6%, respectively. Such results are superior to conventional dose gonadotrophin regimens, which gave significantly higher risk of multiple pregnancy and severe ovarian hyperstimulation syndrome.
Adverse effects include ovarian hyperstimulation syndrome and multiple pregnancy. It is mandatory to exercise vigilant monitoring with adjustment of the gonadotrophin dosage as appropriate. Cycles with more than two dominant follicles should be cancelled and the starting dose should be reduced in subsequent cycles.
A recent randomised-controlled trial suggested that when low-dose FSH was used as the first-line treatment in treatment-naïve women with PCOS, the reproductive outcome was significantly better than with clomiphene citrate, for live birth rate per first cycle, cumulative live birth rate over three cycles and time to pregnancy. Further studies on the cost-effectiveness of such an approach are warranted, and the choice between the two modalities of treatment would be subject to the individual clinic setting and resources and the patient’s preference.
Risk of ovarian cancer with ovulation- inducing agents
Concerns have been raised over the risk of ovarian malignancy after ovulation induction. A cohort study indicated a relative risk of 11.1 (95% CI 1.5 to 82) with long-term use of clomiphene citrate over 12 months. A collaborative analysis also showed an increased risk (OR 2.8; 95% CI 1.3 to 6.1) of invasive ovarian cancer in subfertile women who had used fertility drugs compared with fertile women. A meta-analysis showed a significantly higher risk of ovarian cancer in women exposed to fertility drugs compared with general population controls (OR 1.52; 95% CI 1.18 to 1.97), but not when compared with subfertile controls not exposed to fertility drugs. Indeed, treated infertile women had a tendency towards a lower incidence of ovarian cancer (OR 0.67; 95% CI 0.32 to 1.41) compared with untreated subfertile women. It suggested that subfertility itself rather than the use of fertility drugs is the risk factor for developing ovarian cancer. Another recent large cohort study in the Danish population reported a 46% higher risk of ovarian cancer in women attending the subfertility clinic compared with the general population after adjustment for parity. The overall risk of ovarian cancer was not significantly affected by the use of fertility drugs, including clomiphene citrate, GnRH and gonadotrophins. Overall, findings to date on ovarian cancer risk associated with fertility drug treatment seemed reassuring though not definitive.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
