Ovulation Induction and Controlled Ovarian Stimulation, 2st ed.

3. Diagnosis and Causes of Anovulation

Roy Homburg1


Homerton Fertility Centre, Homerton University Hospital, London, UK


The history and examination alone will often point toward the diagnosis and dictate the order in which examinations should be performed. Using this approach and good common sense, laboratory examinations, expense and time can be limited to a minimum. There are four major categories of causes of anovulation: hypothalamic-pituitary failure, hypothalamic-pituitary dysfunction, ovarian failure and hyperprolactinaemia. This classification has the advantage of being treatment orientated, i.e. once the diagnosis of anovulation has been made and its cause determined, the treatment for induction of ovulation in that particular condition will be clear. Hypothalamic-Pituitary Failure is a situation in which gonadotrophin concentrations are so low as to be completely unable to stimulate follicle development and oestrogen production from the ovaries – hypogonadotrophic-hypogonadism. Hypothalamic-pituitary dysfunction is characterized by normal oestradiol and FSH concentrations and usually presents as oligo- or amenorrhea. Almost 90 % of ovulatory disorders are due to this type of dysfunction and a large majority of these are due to polycystic ovary syndrome (PCOS). Ovarian failure is characterized by amenorrhea, hypo-oestrogenism and high concentrations of FSH. Hyperprolactinaemia may often, but not always, present with galactorrhea. Anovulation caused by hyperprolactinaemia is usually associated with prolactin concentrations more than twice the upper limit of normal. Using the simple diagnostic scheme presented, not only can the cause of anovulation be found with a minimum of fuss but it will dictate the starting treatment required.


AnovulationOligo-ovulationPlasma progesteroneProgesteroneOligomenorrheaBasal body temperatureVaginal ultrasoundOvulationCorpus luteumPolycystic ovary syndromeAmenorrheaOestrogenHypogonadotrophicHypergonadotrophic hypogonadismHyperandrogenismSemenIntra-uterine inseminationICSIIUIIVFHystero-salpingographyLaporoscopyHypothalamic-Pituitary FailureGonadotrophinWeight-related amenorrheaKallmann’s syndromeGnRHHypothalamic-Pituitary DysfunctionOestradiolFSHPCOSOvarian drillingOvarian failureOocytesAnti-Mullerian hormoneAMHHyperprolactinaemia

3.1 Prevalence

Infertility is thought to affect about 15–16 % of couples after 1 year of unprotected regular intercourse. After 2 years, with no treatment, about half of these will still not have conceived and after a further year, about 7 % in all will remain infertile. Most couples will turn for help after 1 year depending on their particular culture. That means that one in seven couples will look for advice after 1 year.

It has proved very difficult to estimate what proportion of infertility is due to anovulation. This prevalence is very much influenced by the particular specialization and indeed geographical location of the reporting centre. Thus estimates have varied between 20 and 40 % of the causes of infertility as due to anovulation or severe oligo-ovulation.

3.2 Diagnosis

The importance of a detailed gynaecological and medical history cannot be emphasized enough. The clues should be sought by listening carefully and asking the correct direct questions. This should be followed by a thorough gynaecological and general physical examination. The history and examination alone will often point toward the diagnosis and dictate the order in which examinations should be performed. Using this approach and good common sense, laboratory examinations, expense and time can be limited to a minimum. A suggested ‘check-list’ is presented in Table 3.1.

Table 3.1

A suggested checklist for history taking and physical examination of the female partner of a couple seeking help for infertility at the first clinic visit. Note that these are headings only, contain no further details and obviously, the answers to direct questioning should prompt further reflex, relevant questions, e.g. a history of amenorrhea, is it primary or secondary? Hot flushes? Sense of smell? etc.



 Female partner

 Male partner


Previous pregnancies

Duration of infertility

Past medical history

Intercurrent illnesses/medications/drugs/alcohol

Family history

Previous contraception

Previous treatment for infertility

Age at menarche

Cycle regularity

Menstrual loss/pain/last menstrual period

Complaints of hirsutism, acne, galactorrhea

Sexual activity/problems


Body build

Weight, height, body mass index

General physical examination

Distribution of hair growth/hirsutism



Gynaecological examination

 Vulva, vagina, cervix, uterus, adnexae

Any form of menstrual irregularity, not within the limits of a 24–35 day cycle, strongly suggests anovulation or oligo-ovulation. The converse is not necessarily so as the occasional woman with regular bleeding may also be anovulatory. Painful menstruation usually indicates that ovulation is occurring.

Plasma progesterone concentrations are arguably, the most accurate way to estimate whether ovulation has occurred.

·               For women with a regular cycle of say 28 days, a plasma progesterone estimation on cycle day 20 or 21 of 8 ng/ml (25 nmol/l) or more, will rule out a diagnosis of anovulation.

·               If the usual cycle is say 35 days in length, then this examination should be done around cycle day 28, i.e. about 7 days before the expected menstruation.

·               For women with mild oligomenorrhea (cycle length >35 days) progesterone can be measured on day 28 and then once a week following that until menstruation occurs.

·               If periods only occur less than once every 2 months or in cases of amenorrhea, there is little point in hunting for progesterone estimations as the diagnosis of severe oligo- or anovulation is self-apparent.

A basal body temperature (BBTchart has been used for many years to estimate whether ovulation is occurring. The principle is that the secretion of progesterone following ovulation, into the circulation, will cause a rise in body temperature of about 0.5 °C. The typical BBT chart will thus be bi-phasic i.e. the temperature following ovulation will be higher than in the first part or follicular phase. The day before the temperature rise is usually denoted the day of ovulation. Although the BBT is a simple, cheap and non-invasive screening test, it suffers from many inaccuracies, particularly false negatives, and is open to much misinterpretation. It is very doubtful whether the BBT still has a place in the routine screening for ovulatory problems. Further, it has been found to be a niggling nuisance for many women as temperature must be measured every morning, immediately on waking. It could still be used for the timing of sampling for a progesterone estimation, about 7 days following the temperature rise or as a rough guide for timing of intercourse. This latter indication is also doubtful as once the BBT chart shows a rise, ovulation has been and gone!

A vaginal ultrasound examination before and after ovulation, should record a large developing dominant follicle which disappears following ovulation. In addition, most competent ultrasonographers are able to diagnose the presence of a corpus luteum if ovulation has occurred. This will be accompanied by a small amount of fluid in the pouch of Douglas which can also be spotted on ultrasound examination.

Physical examination can give many clues as to the cause of anovulation. Most obvious at first glance, is the weight of the patient. Weight and height should always be recorded and the body mass index (BMI) calculated. This is done with the following formula:

 $$ \mathrm{BMI}=\frac{\mathrm{Weight}\;\left(\mathrm{kg}\right)}{\mathrm{Height}\;\mathrm{in}\;{\mathrm{metres}}^2} $$

A normal BMI is 20–25.

·               <20 is underweight,

·               25.1–30 is overweight

·               >30 is frank obesity.

Some geographical variations in these diagnoses exist. For example, in most South-East Asian communities, any BMI > 25 is regarded as obesity.

Overweight and obesity is often associated with polycystic ovary syndrome (PCOS) and in turn PCOS is often characterized by hirsutism and/or acne, both of which are easily discernable on examination. In cases of suspected PCOS who are obese, acanthosis nigricans, dark discolouration of the skin in the axillary or nuchal regions, is a tell-tail sign of insulin resistance. Waist circumference should be measured at the level of the iliac crests in all overweight women as this again may be a good reflection of insulin resistance when >88 cm.

Women whose BMI is <20 may have irregular or absent ovulation due to so-called, weight-related amenorrhea. This may be due to loss of weight due to dieting and to anorexia nervosa in its extreme. Direct questioning regarding diet, alcohol or drug abuse are mandatory.

Physical examination can also reveal signs of oestrogen deficiency such as poor breast development, lack of development of the vulva, vaginal dryness, lack of additional secondary sexual characteristics. These signs indicating oestrogen deprivation could be due to either hypo- or hypergonadotrophic hypogonadism, when either is associated with primary amenorrhea. Although Turners syndrome is rare as a cause of amenorrhea, it can often be easily diagnosed by the typical body habitus; short stature, webbed neck, cubitus valgus and often a systolic cardiac murmur.

Distribution of hair growth should be noted. A male distribution would indicate hyperandrogenism and a lack of body hair could be a sign of androgen insensitivity. Clitoral enlargement or lack of development would be in parallel to these respective conditions in their extreme.

3.2.1 Following the Diagnosis of Anovulation

Needless to say, having made a diagnosis of anovulation and before embarking on treatment, some basic examinations, notably a semen analysis, must be performed. It is not unusual to encounter a multifactorial cause for infertility. One normal semen examination should be enough to lay a possible male factor aside before starting ovulation induction. There is some controversy regarding the usefulness of a post coital test (PCT) but when a reasonable quantity of progressively motile sperm are seen in good quality cervical mucus, this can be very reassuring and also confirms that coitus is being practiced and coital technique is normal. The results of at least two abnormal semen examinations, the second usually in a washed specimen, will usually dictate whether intra-uterine insemination is indicated with ovulation induction or whether there is an indication to forget an ovulation induction regimen and proceed directly to IVF-ET or ICSI.

If there is a previous history in the female partner of sexually transmitted disease, a complicated delivery, Caesarean section, pelvic inflammatory disease, endometriosis or surgical interventions in the pelvic region, including appendicectomy, a screening test, usually hystero-salpingography (HSG), should be performed. If this confirms tubal patency and a normal uterine cavity, then treatment can be commenced. Abnormal findings in the HSG will dictate what steps are to be taken further. These may include a diagnostic laparoscopy and hysteroscopy which may be diagnostic or operative or gross tubal damage demonstrated on the HSG may indicate direct progress to IVF. Some centers use laparoscopy as a screening test if the history is suggestive of a possible mechanical factor but I have found that the HSG serves this purpose well and is certainly a less invasive technique. If the HSG is suggestive of a tubal lesion or peri-tubal adhesions, then resort to a laporoscopy is justified.

3.3 Causes of Anovulation

There are numerous conditions that can cause anovulation. Here I will give a brief description of each before classifying them into a scheme which can greatly simplify the individual diagnosis and has the advantage of being treatment orientated, i.e. once the diagnosis of anovulation has been made and its cause determined, the starting treatment for induction of ovulation in that particular condition will be clear. The classification of the causes of anovulation described here are based on a long-standing World Health Organisation (WHO) classification scheme which I have adapted slightly.

3.3.1 Hypothalamic-Pituitary Failure (WHO Group I)

This is a situation in which gonadotrophin concentrations are so low as to be completely unable to stimulate follicle development and oestrogen production from the ovaries – hypogonadotrophic-hypogonadism. Anovulation and amenorrhea are the consequences.

Weight-related amenorrhea is the commonest ‘hypothalamic’ cause of this condition – loss of weight due to a crash diet or frank anorexia nervosa.

Stress, in its extreme, including very strenuous exercise, e.g. marathon running, is not an uncommon cause.

Kallmanns syndrome is a hypothalamic amenorrhea associated with anosmia.

Other causes include craniopharyngioma and debilitating systemic diseases. Not a few cases are idiopathic.

The commonest ‘pituitary’ causes are hypophysectomy, radiotherapy for pituitary tumours and severe post-partum haemorrhage (Sheehan’s syndrome).

If the pituitary is intact and the hypothalamus is failing to function, ‘replacement therapy’ with pulsatile GnRH is highly effective. Direct stimulation of the ovaries with gonadotrophins also serves the purpose for ovulation induction whether the cause is of hypothalamic or pituitary origin. See Chap. 6 for a detailed description of ovulation induction for women with hypogonadotrophic-hypogonadism.

3.3.2 Hypothalamic-Pituitary Dysfunction (WHO Group II)

Hypothalamic-pituitary dysfunction is characterized by normal oestradiol and FSH concentrations and usually presents as oligo- or amenorrhea. Almost 90 % of ovulatory disorders are due to this type of dysfunction and a large majority of these are due to PCOS. Although oestrogens are constantly being produced by the ovaries in this type of dysfunction, there is a loss of the ebb and flow of FSH and LH levels that are characteristic of the normal ovulatory cycle. Polycystic ovary syndrome may present in many ways. The heterogeneity of its presentation and, indeed, of its laboratory findings, have made the definition and diagnosis of PCOS a contentious issue over the years. Following a meeting in Rotterdam in 2003, a consensus was reached and published and has since been widely adopted. This will hopefully now enable a much needed unification of the definition and the possibility to compare data for scientific and clinical purposes.

Polycystic ovary syndrome can be diagnosed when at least two of the three following criteria are present:




Precise definitions of each of the above criteria are shown in Table 3.2.

Table 3.2

The diagnosis of polycystic ovary syndrome. Any two of the following three criteria serve to make the diagnosis

Oligoor anovulation

Oligo-ovulation is the occurrence of ovulation no more than once every 35–180 days

Anovulation is no ovulation for at least six consecutive months

Hyperandrogenism – clinical and/or biochemical

Clinical signs include hirsutism, acne, alopecia (male pattern balding) and frank virilisation

Biochemical indicators include raised total testosterone, androstendione and free androgen index

Polycystic ovaries

The presence of 12 or more follicles in either ovary measuring 2–9 mm in diameter and/or increased ovarian volume (>10 mm)

Note that other similarly presenting aetiologies, e.g. congenital adrenal hyperplasia, androgen secreting tumours and Cushing’s syndrome should be excluded

Polycystic ovary syndrome (PCOS) is a very heterogeneous syndrome, often first diagnosed when the patient presents complaining of infertility. The syndrome is associated with approximately 75 % of the women who suffer from infertility due to anovulation. The majority of women with anovulation or oligo-ovulation due to PCOS have menstrual irregularities, usually oligo/amenorrhea, associated with clinical and/or biochemical evidence of hyperandrogenism. Almost all these women will have a typical ultrasonic appearance of the ovaries. Making the diagnosis of PCOS is important as this will dictate the treatment plan, the prognosis and will serve in the avoidance of possible complications of treatment.

Practically, the diagnosis of PCOS can be made in almost every case without blood sampling. Although not essential for initial diagnostic or therapeutic decisions, for screening I usually take a blood sample for LH, FSH, total testosterone and fasting glucose and insulin concentrations. The ratio of fasting glucose to insulin levels gives only a fair indication of insulin sensitivity but as hyperinsulinaemia is present in about 80 % of obese women and 30–40 % of women of normal weight with PCOS and is strongly associated with anovulation, then it is certainly useful to know for possible therapeutic intervention. Except for research purposes, the attempted and often inaccurate estimation of insulin sensitivity is best ignored in routine practice. The LH value may be expected to be high in about half of the women with PCOS and when high (>10 IU/L) is thought to be detrimental to successful ovulation induction and to the incidence of miscarriage.

When suggested by the history and physical examination, further laboratory examinations may be needed to exclude similarly presenting syndromes. Total testosterone concentrations, together with a history of rapid progress of hyperandrogenic symptoms are useful for the screening of androgen producing tumours and 17-hydroxy progesterone when highly elevated is pathognomonic for 21-hydroxylase deficiency, the commonest form of late onset congenital adrenal hyperplasia. If Cushing’s syndrome is suspected following the physical examination, it should be investigated using the accepted adrenal function tests. Many other, more sophisticated, examinations may be performed for research purposes; measuring sex hormone-binding globulin can be used in the calculation of the free androgen index, an oral or intravenous glucose tolerance test or even insulin clamp will give more accurate information on insulin metabolism and for more information regarding the metabolic status of the patient, lipids, homocysteine and plasminogen activator inhibitor-1 concentrations may be the subject of the particular investigation.

Women with oligomenorrhea obviously have endogenous oestrogen production and therefore an intact hypothalamus and pituitary, an adequate secretion of GnRH and FSH and ovaries that are capable of responding to FSH. The dysfunction lies in the coordination and synchronization of these elements and the aim of treatment, whether it be with anti-oestrogenic agents, life-style changes, insulin lowering medications, laparoscopic ovarian drilling etc. is to rearrange correct synchronization in the timing and amount of hormonal discharge. Another approach is to bypass these steps and stimulate the ovary directly with gonadotrophins.

See Chap. 7 and following chapters for a full account of the management and treatment of PCOS.

When amenorrhea is encountered, it should first be established whether endogenous oestrogen production is present. This can be done simply and cheaply by administering an oral progestin for about 5 days. If a withdrawal bleeding follows, endogenous oestrogen production is present and a WHO Group II type of anovulation is established.

3.3.3 Ovarian Failure (WHO Group III)

This group of anovulatory disorders is characterized by amenorrhea, hypo-oestrogenism and high concentrations of FSH. It is often accompanied at its onset by hot flushes.

The underlying cause is the inability of the ovaries to respond to FSH and may be due to:

·               The onset of a ‘natural’ menopause (>40 years of age)

·               Premature menopause (<40 years old)

·               A chromosomal abnormality (e.g. Turner’s syndrome).

In each case the ovaries are either completely devoid of oocytes or have a severely depleted number of oocytes. A premature menopause may be familial and therefore suspected from a family history or it may be the result of an auto-immune disease, chemotherapy or direct radiation of the ovaries. Very often the underlying cause remains obscure.

The definitive diagnosis of ovarian failure should be established simply by an estimation of FSH which is >25 IU/L. Simple as it may sound, there are pitfalls to making this definitive diagnosis. Concentrations of FSH have been known to fluctuate considerably and not once I have been embarrassed by making a firm diagnosis, preaching doom and gloom to the unfortunate patient and then witnessing a return of normal menstruation, ovulation and, rarely, spontaneous pregnancy. Although this may be anticipated in some cases following recovery from chemotherapy, in others, rarely, this may also happen. An intermediate or transitional stage of this process may be termed impending ovarian failure. At this stage, menstruation and even ovulation may be occurring regularly, often with a short follicular phase and cycle length, but they are accompanied by infertility, poor or absent response to exogenous gonadotrophins and a raised FSH concentration on day 2–3 of the cycle. The finding of a very low or immeasurable values of serum anti-Mullerian hormone (AMH) may be a more reliable marker than FSH for the diagnosis of ovarian failure (see Chap. 5).

Except for the preservation of the cancer patient’s own ovarian tissue or oocytes before treatment or for the older patient who had the good sense to freeze eggs at an earlier age, oocyte donation is the only feasible treatment for infertility in these patients. Other treatments employing oestrogens, GnRH analogues or cortisone have proved worthless. Replacement therapy with oestrogen and progesterone should be prescribed in this situation. This is especially important for the younger patients to avoid the long-term consequences of oestrogen deprivation.

3.3.4 Hyperprolactinaemia (WHO Group IV)

Hyperprolactinaemia is a not uncommon condition, which may often, but not always, present with galactorrhea. Anovulation caused by hyperprolactinaemia is usually associated with prolactin concentrations more than twice the upper limit of normal. It is not unusual to find mildly raised prolactin concentrations which are not associated with anovulation and these should not be treated. It should also be remembered that mildly raised prolactin concentrations may be seen in about 30 % of women with PCOS but again, there is no point in treating them specifically with prolactin lowering medications.

Major causes of hyperprolactinaemia are pituitary adenomas, hypothyroidism, medications such as the phenothiazines, some sedatives and hypotensive agents. It follows that when hyperprolactinaemia is found to be associated with anovulation, an examination of the pituitary gland by MRI or CT scan should be performed. In about 50 % of these cases a prolactin secreting adenoma will be found. If a macroadenoma (>10 mm) is visualized it is probably wise to extend the investigation to include visual fields due to the close proximity of the optic chiasma on to which a large tumour may impinge. A microadenoma (<10 mm) is a commoner finding. In either case, prolactin lowering drugs are available today which are very effective in restoring ovulation and even in reducing the size of the adenoma. Thyroid stimulating hormone (TSH) concentrations should be estimated as a screening test for thyroid disorders and if raised should be followed up with more specific thyroid function tests. Stopping the offending medication, treating hypothyroidism or administering prolactin lowering drugs will decrease prolactin concentrations and restore ovulation in the vast majority of these women.

Unlike all the rest of the hormones secreted by the anterior pituitary which have releasing hormones secreted by the hypothalamus, prolactin has an inhibiting hormone (probably dopamine). This means that a hypothalamic lesion or hypothalamic suppression by drugs (e.g. phenothiazines), if sufficiently serious, will lower gonadotrophin secretion from the anterior pituitary but may well raise prolactin concentrations. Very often then, a hyperprolactinaemia sufficient to cause anovulation is associated with low LH concentrations and this is the probable connection between the two conditions.

See Chap. 12 for a detailed description of the management of hyperprolactinaemia.

3.4 Diagnostic Schemes

Having classified the possible causes of anovulation into four groups, diagnostic schemes will now be presented. These schemes point a very direct way to arrive at an accurate diagnosis of the cause of the anovulation in the shortest possible time and with the minimum of investigations.

Assuming anovulation or severe oligo-ovulation has been proven, Fig. 3.1 illustrates a rapid diagnostic method to classify the cause of the ovulatory disturbance into one of the four groups described above. Minimal laboratory examinations are required in this scheme as endogenous oestrogen production can be estimated by a progestin withdrawal test in the case of amenorrhea/anovulation. This is unnecessary if oligo- rather than amenorrhea is the presenting complaint. This leaves only prolactin to be measured and, in the case of a negative progestin withdrawal, FSH concentrations are measured to find out if the problem is hypogonadotrophic or hypergonadotrophic hypogonadism.


Figure 3.1

A rapid diagnostic method for the cause of an ovulatory disturbance

If amenorrhea (no ovulation in 6 months) is the presenting symptom, then the scheme in Fig. 3.2 can be easily applied to arrive at the diagnosis and indicate the way forward. For the sake of completeness, the diagnostic scheme for the causes of amenorrhea should include Outflow tract defects (WHO Group V) which, while not usually associated with anovulation, are a cause of amenorrhea which must be taken into possible consideration. This situation can be diagnosed if both progestin and oestrogen/progestin withdrawal do not produce bleeding and FSH levels are in the normal range. When the amenorrhea is primary, possible diagnoses in this group include imperforate hymen and congenital absence of the uterus. When secondary, severe intrauterine adhesions (Ascherman’s syndrome) may be the cause.


Figure 3.2

Scheme for the diagnosis of the cause of amenorrhea. TSH thyroid stimulating hormone, MRI magnetic resonance imaging, PCOS polycystic ovary syndrome

If oligo-ovulation (one ovulation every 35–180 days) or regular anovulatory cycles is the presenting symptom, the scheme illustrated in Fig. 3.3 will be helpful. In any of these situations, the aim is to arrive at a correct diagnosis for the cause of the anovulation in the minimum amount of time and with a minimum of investigations. As this classification is very much treatment orientated, once the diagnosis is made it will indicate what is the correct treatment suitable for that specific diagnosis.


Figure 3.3

Scheme for the diagnosis of the cause of oligomenorrhea. PCOS polycystic ovary syndrome, TT total testosterone, DHEAS dihydroepiandrosterone sulphate

3.5 Conclusions