Anovulation and Ovulatory Dysfunction

Published on 10/04/2015 by admin

Filed under Surgery

Last modified 22/04/2025

Print this page

This article have been viewed 2921 times

Chapter 19 Anovulation and Ovulatory Dysfunction

Margo Fluker, Stephanie Fisher

INTRODUCTION

Anovulation and ovulatory dysfunction are frequent reasons for gynecologic referral. The most common form of ovulatory dysfunction, polycystic ovary syndrome (PCOS), is estimated to occur in up to 6% to 10% of women.¹ Ovulatory dysfunction may result in a spectrum of clinical symptoms ranging from amenorrhea to frequent, irregular, and heavy menses. In addition to menstrual irregularities, many women will experience subfertility or other endocrine symptoms, such as hirsutism. Choice of therapy will depend on numerous factors, including the presenting symptom, age, desire for fertility, and associated conditions.

Regardless of the etiology, a spectrum of ovulatory disorders exists that may precede the eventual development of amenorrhea or occur during the recovery from amenorrhea (Fig. 19-1). Underlying abnormalities in ovulatory function may present clinically with either quantitative or qualitative changes in the menstrual cycle. A prolonged follicular phase or intermittent anovulation both result in infrequent (cycles greater than 35 days) and often heavy menstrual flow; a shortened follicular or luteal phase is associated with frequent menstrual flow and often premenstrual spotting.

Figure 19-1 Spectrum of menstrual bleeding patterns, from normal ovulatory cycles to amenorrhea.

Because ovulatory disorders encompass a spectrum of clinical presentations and a variety of underlying etiologies, this chapter necessarily overlaps with other chapters that focus on specific endocrine abnormalities. The current chapter is not intended to be an in-depth review of each of these associated topics, but rather a brief overview to help understand how they are related to the pathophysiology of anovulation.

PATHOPHYSIOLOGY

To develop an approach to the problem of anovulation, one must first have a clear understanding of the components of normal ovulation. The clinical history and functional inquiry can then be directed toward potential dysfunction in the three components of the hypothalamic-pituitary-ovarian (HPO) axis.

The Hypothalamus

The pathway ultimately responsible for ovulation is initiated by pulsatile release of gonadotropin-releasing hormone (GnRH) from neurons in the arcuate nucleus of the hypothalamus into the portal circulation. From there, GnRH induces release of the anterior pituitary hormones, luteinizing hormone (LH) and follicle-stimulating hormone (FSH). The amplitude and frequency of GnRH release are critical to normal functioning of the HPO axis.² Numerous hormones and neurotransmitters are known to modulate GnRH release, including dopamine, norepinephrine, neuropeptide Y, and endorphins. Factors affecting release of these hormones and neurotransmitters or the physical delivery of GnRH to the anterior pituitary may ultimately disrupt ovulation.

Suppression of GnRH release is associated with excessive stress (including severe chronic disease) and eating disorders, such as anorexia nervosa and bulimia. Even in the face of a normal body mass index (BMI), excessive exercise or disordered eating such as bulimia nervosa may result in ovulatory dysfunction. A spectrum of GnRH deficiency may be observed, ranging from altered frequency and amplitude of GnRH pulses to complete suppression of GnRH activity. The resulting clinical presentation ranges from oligomenorrhea to amenorrhea. Because GnRH cannot be measured directly, this is a diagnosis of exclusion. Women with hypothalamic causes of amenorrhea rarely report hypoestrogenic symptoms such as vasomotor flushing. In the presence of chronic hypoestrogenism, estrogen receptor down-regulation renders them relatively insensitive to the usual hypothalamic-mediated vasomotor symptoms triggered by low estrogen levels. This same phenomenon also renders them relatively refractory to ovulation induction with clomiphene citrate.

Excessive cortisol secretion (due to high corticotropin-releasing hormone activity), high levels of endogenous neuropeptide Y and opioids (specifically endorphins) are all thought to play a role in suppression of GnRH activity associated with psychological stress and eating disorders. Less commonly, injury, infection, or central nervous system (CNS) tumors may interfere with delivery of GnRH to the anterior pituitary via the portal circulation. Hypothalamic dysfunction or failure has been reported following severe deceleration injuries that cause pituitary stalk transection, CNS tumors such as craniopharyngiomas or hamartomas that cause stalk compression, and infiltrative diseases such as tuberculosis or sarcoidosis.

Lastly, hypogonadotropic hypogonadism may result from a GnRH receptor defect or GnRH deficiency due to failure of migration of the GnRH-producing neurons (Kallmann syndrome). These patients generally present with primary amenorrhea and lack of secondary sexual development rather than with secondary ovulatory disturbances.

More recently, the role of leptin in maintenance of BMI, regulation of appetite, and control of reproductive function has been investigated. Leptin is a 146-amino acid protein hormone expressed in numerous sites but mainly in adipose tissue. It has been shown to have a direct stimulatory effect on GnRH pulsatility, as well as LH/FSH secretion from the anterior pituitary, and is likely central to control of the HPO axis through a number of complex endocrine and paracrine actions.³ Leptin may act as a sensor of energy deficiency, limiting the high-energy cost of reproduction and growth during periods of starvation. Concentrations decrease rapidly with fasting, resulting in suppression of reproductive, thyroid, and growth hormones. A preliminary study suggests that administration of recombinant human leptin may reverse the neuroendocrine abnormalities associated with hypothalamic amenorrhea, inducing a return of hormone secretion reminiscent of puberty.⁴

Pituitary Gonadotropins

At the extremes of reproductive age, anovulation can be considered to be physiologic. With the onset of puberty, the hypothalamus is released from its prepubertal negative inhibition, and pulsatile GnRH secretion is initiated. This pulsatile GnRH release is initially nocturnal, and as puberty progresses, pulsatile secretion occurs throughout the day. In response to increasing secretion of GnRH, LH and FSH are released from the anterior pituitary gland in a similar pulsatile fashion, resulting in the production of gonadal steroids that will ultimately result in thelarche and menarche.

The secretion of LH and FSH in response to GnRH stimulation varies as puberty progresses. As the HPO axis matures, it is common to see abnormalities in ovulatory function. The average time from onset of menses to establishment of regular ovulatory cycles is 2 to 3 years.⁵

The Ovary

Depletion of follicles is a common cause of abnormalities in ovulatory function during the perimenopausal transition. Early follicular (day 3) FSH levels rise due to the relative resistance of the remaining follicles to FSH-dependent stimulation and a concomitant decrease in serum inhibin. Accelerated follicular development may occur in response to increased FSH levels, with elevated estradiol secretion early in the follicular phase, ovulation as early as day 8 to 11, and corresponding shortening of the menstrual interval (e.g., from a 28-day to a 23-day pattern).^6,⁷ As follicular number and responsiveness decline, the pattern progresses to oligo-ovulation or anovulation and eventual amenorrhea. This inevitable process of reproductive aging usually begins in the late 30s and may become clinically manifest with changing menstrual patterns throughout the 40s.⁷

A similar process, albeit occurring at an earlier age, may occur in young women who develop premature ovarian failure with loss of ovarian function before age 40.^8,⁹ Some young women with premature ovarian failure experience premature follicular depletion; others likely have a normal complement of follicles that are unresponsive to gonadotropin stimulation, possibly on an autoimmune basis. Young women with premature ovarian failure may also experience sporadic and transient return of ovarian function after diagnosis, ranging from intermittent estrogen-withdrawal bleeding to normal ovulatory cycles and the possibility of conception.¹⁰^–¹²

Lastly, the most common cause of ovulatory dysfunction is PCOS, a complex condition of uncertain etiology. PCOS is the most common endocrine disorder in women and is characterized by hyperestrogenic hyperandrogenic chronic anovulation, which may present clinically with various ovulatory and menstrual disorders, infertility, acne, hirsutism, and obesity.¹³ Numerous metabolic aberrations may also be associated with PCOS, including obesity, insulin resistance, type 2 diabetes, dyslipidemia, and cardiovascular disease. Insulin resistance appears to play a central role in the reproductive dysfunction and long-term health consequences of PCOS.¹²

When evaluating women with ovulatory dysfunction, it is important to recognize that PCOS is a syndrome with a broad spectrum of clinical phenotypes, and that no existing classification will encompass all possible phenotypes. The Rotterdam PCOS consensus workshop recently published revised criteria for the diagnosis of PCOS.¹⁴ After exclusion of other etiologies (congenital adrenal hyperplasia, androgen-secreting tumors, Cushing’s syndrome), two of three of the following criteria must be present to establish a diagnosis of PCOS:

• Oligo-ovulation or anovulation

• Clinical and/or biochemical signs of hyperandrogenism

• Polycystic ovaries (presence of ≤ 12 follicles in each ovary measuring 2 to 9 mm and/or ovarian volume ≤ 10 mL).

Although frequently associated with PCOS, obesity in itself may contribute to ovulatory dysfunction.¹⁵ Numerous studies, including the Nurses’ Health Study, have reported increased rates of ovulatory infertility with increasing BMI.^16,¹⁷ It is believed that obesity leads to a state of functional hyperandrogenism. Central (abdominal) obesity is associated with insulin resistance and high circulating insulin levels. Because the production of sex hormone-binding globulin (SHBG) in the liver is directly inhibited by insulin, it follows that women with central obesity have lower levels of SHBG and higher levels of free testosterone than age- and weight-matched controls with peripheral obesity.¹⁸ This functional hyperandrogenism, in combination with increased peripheral aromatization of estrogen in adipose tissue, leads to alteration in the balance of sex steroids and ultimately contributes to ovulatory dysfunction.

Other Endocrine Disorders

A number of other endocrine disorders may result in ovulatory dysfunction through disruption of the HPO axis. Hyperprolactinemia, for example, leads to suppression of GnRH secretion.¹⁹ Similarly, increased thyrotropin-releasing hormone (TRH) activity in women with hypothyroidism ultimately results in reduced GnRH pulsatility through the direct stimulatory effect of TRH on prolactin secretion. Hyperthyroidism may result in a range of menstrual disturbances, from frequent heavy bleeding to amenorrhea.

Congenital Adrenal Hyperplasia

Women with both classic and late onset congenital adrenal hyperplasia (CAH) may present with ovulatory disturbances similar to PCOS.²⁰ Menstrual abnormalities in these women were initially attributed to inadequate glucocorticoid suppression of adrenal androgen production, resulting in suppression of the HPO axis. Particularly in classic CAH, there is evidence of ovarian hyperandrogenism similar to PCOS despite adequate adrenal suppression with dexamethasone.

It has been postulated that perinatal exposure to hyperandrogenism in women with CAH results in hypersecretion of LH in response to GnRH stimulation at puberty. This allows the development of ovarian hyperandrogenism independent of control of adrenal androgen production.²¹ A similar abnormality has been reported but is less common in late-onset CAH.

Luteal Phase Defect

Luteal phase defect is a controversial entity thought to consist of inadequate endometrial development in the luteal phase leading to impaired endometrial receptivity, failed implantation, and early pregnancy loss. Qualitative or quantitative progesterone deficiency is felt to be central to the abnormal endometrial development. Two subgroups likely exist:

1. Short luteal phase. A short luteal phase (>10 days) is usually documented by urinary LH testing or basal body temeperature (BBT) charting. This is generally secondary to suppression of pulsatile GnRH secretion,¹⁹ leading to inadequate luteinization and reduced progesterone production. Short luteal phases occur in association with other endocrine disorders (e.g., hyperprolactinemia or hypothyroidism) or during development of or recovery from the hypothalamic-pituitary suppression that accompanies lactational amenorrhea, steroidal contraceptive use, and strenuous exercise. Correction of the underlying cause usually results in normalization of the luteal phase.

2. Disordered endometrial development. Endometrial histologic development that lags 2 or more days behind chronologic dating is often independent of luteal phase length.²² This variant of luteal phase defect is thought to be due to suboptimal progesterone secretion or inadequate endometrial response to progesterone. Although the precise etiology is uncertain, it is known to occur more commonly in the exaggerated hormonal milieu that accompanies ovarian stimulation with exogenous gonadotropins, with or without GnRH analog therapy.²³

The role of luteal phase defect in infertility remains controversial, however. Studies designed to address this issue have demonstrated a significant prevalence of luteal phase defect in the fertile population and are hampered by a lack of consensus regarding diagnostic criteria. Performance of repeated endometrial biopsies to assess histology is often impractical and the interpretation of results limited by a high degree of interobserver variability. Similarly there is no agreement on the appropriate normal values for serum progesterone. Finally, lack of a clear treatment effect with progesterone supplementation has led many to question the importance of luteal phase defect as a cause of infertility.²⁴^–²⁶

Luteinized Unruptured Follicle

Lastly, luteinized unruptured follicle syndrome has been reported as a transient or ongoing cause of ovulatory disturbance. The syndrome involves absence of sonographic signs of ovulation or lack of an ovulatory stigma at laparoscopy 48 to 72 hours after the LH surge. The phenomena may occur in conjunction with periovulatory nonsteroidal anti-inflammatory ingestion, leading to inhibition of prostaglandin-induced follicular rupture.^27,²⁸ Other cases have been reported in conjunction with suboptimal doses or improperly timed human chorionic gonadotropin injections during ovulation induction cycles.²⁹ Aside from these iatrogenic cases, there is no clear etiology, no documented treatment, and considerable controversy over the diagnostic criteria with their inherent variability and observer biases.^29,³⁰ A survey of practice patterns in the United States indicated that only 8% of university-based board-certified reproductive endocrinologists would screen for this disorder.³¹

ESTABLISHING A DIAGNOSIS

The diagnosis of anovulation or ovulatory dysfunction can be made solely on the basis of history in most cases. Menstrual bleeding occurring at regular intervals between 21 and 35 days, particularly in the presence of premenstrual molimina (i.e., breast tenderness, abdominal bloating, mood disturbance) is suggestive of ovulation; bleeding at longer or irregular intervals generally reflects anovulation.

Identification of more subtle ovulatory dysfunction may require the use of ancillary testing such as urine LH monitoring, BBT charting, or luteal phase serum progesterone measurement.

A woman with 40- to 50-day cycles may indeed be ovulatory, albeit with a prolonged and irregular follicular phase. The diagnosis will be most likely be made from biphasic BBT charts in which the progesterone-induced temperature rise occurs late (e.g., from day 26 to 36 in this case). Serum progesterone measurements are helpful if obtained in the 7 to 10 days prior to the next menses, but appropriate timing can be challenging in women with long or erratic cycle intervals. Progesterone levels may be misleading in this case if performed on the “traditional” day 21, although they could confirm ovulation if ordered on day 38 to 39, for example.

Basal body temperature charting or urine LH testing may also be useful in identification of women with ovulatory cycles accompanied by a shortened luteal phase (>10 days). This can occur in the context of a normal menstrual interval but is frequently associated with a short menstrual interval (e.g., every 21 to 24 days) and should prompt evaluation of thyroid function and prolactin secretion.

Considerable controversy exists over the criteria used to define luteal phase defects.^25,³² Histologic dating of a luteal phase endometrial biopsy specimen has been considered the gold standard, requiring a 2- to 3-day difference between histologic and chronologic dating in two separate cycles.²² However, the test is invasive and uncomfortable, and suffers from substantial intraobserver and interobserver variability.³³ Serum progesterone measurements are less invasive, although less accepted due to debate over the threshold level used for diagnosis, the number of suboptimal tests needed, and the relatively poor correlation with endometrial response.^24,^25,³⁴ Progesterone secretion is both pulsatile (in response to episodic LH release) and parabolic (with peak levels in the midluteal phase), so low levels may occur physiologically in the trough prior to each LH pulse and at the beginning and end of each luteal phase.³⁵ The threshold progesterone value used to differentiate ovulatory from anovulatory cycles (6 to 18 nmol/L or 2 to 5 ng/mL) should not be confused with that suggested for the diagnosis of defective luteal phase (21 to 32 nmol/L or 7 to 10 ng/mL).^24,³⁴

DIAGNOSIS

Various classification systems have been developed for ovulatory disorders (Table 19-1). Such systems also provide the basis for a timely and cost-effective evaluation. They revolve around three key principles that can also be used to guide the history, physical examination, investigations, and management (Table 19-2).

Table 19-1 Classification of Ovulatory Disorders

Table 19-2 Key Principles for Evaluation of Ovulatory Dysfunction

1. Identify hypoestrogenic patients who lack spontaneous or progestin-induced menstrual bleeding.

2. Differentiate hypogonadotropism, hypergonadotropism, and normo-gonadotropism.

3. Exclude underlying medical conditions that produce secondary ovulatory dysfunction.

History

History taking in women with ovulatory disorders focuses on evaluating the current and previous menstrual pattern, identifying precipitating factors or underlying medical disorders that require specific therapy, and assessing associated symptoms that may also require treatment.

The menstrual history should commence with menarche and a brief assessment of the pubertal development, particularly in the adolescent patient. This may be addressed simply by ensuring that pubertal development occurred in parallel with the patient’s peers. Bear in mind that anovulatory cycles can be a normal physiologic pattern up to 3 years from menarche.

The menstrual pattern can be described as the number of menses experienced per year (e.g., 2 or 3 menses per year) or as the shortest and longest interval between menstrual bleeding (e.g., 35 to 90 days). Cycles occurring at intervals longer than 35 days are considered oligo-ovulatory or anovulatory and may often be prolonged or heavy. In some cases, helpful information about follicular or luteal phase length and the timing of ovulation may be gleaned from BBT charts or urinary ovulation detection kits. Although no menstrual pattern is pathognomonic of the underlying condition, shortened follicular phases (e.g., 8 to 10 days) often occur early in the perimenopausal transition, shortened luteal phases (e.g., >10 days) are common with hyperprolactinemia, whereas infrequent, heavy menses are more common with PCOS.

A review of the hypothalamic-pituitary axis may reveal precipitating factors or underlying medical disorders that require specific investigation and treatment. However, many of these symptoms are nonspecific. Hypothalamic factors are common among thin women who present with oligomenorrhea or amenorrhea. Questioning should include sources of excessive life stress, chronic illness, excessive exercise, significant weight change, disordered eating, and an overall preoccupation with thinness.¹³ Recall that women with hypothalamic causes of amenorrhea rarely report hypoestrogenic symptoms such as vasomotor flushing.

Symptoms of pituitary adenoma or prolactin excess such as severe headaches, visual field change (typically bitemporal hemianopsia), or galactorrhea should be elicited. A medication history may reveal drugs that affect prolactin secretion, including dopamine antagonists (metoclopramide, domperidone) and antipsychotics (risperidone, haloperidol, phenothiazines). Symptoms of thyroid dysfunction (particularly hypothyroidism) are common and notoriously nonspecific, but should be sought in addition to a personal or family history of thyroid disease.

The most common disorder of the HPO axis that will be encountered in anovulatory patients is PCOS. Menstrual disorders, infertility, weight gain, central obesity, hirsutism, and acne are common presenting symptoms. Some affected women will also present with a history of known hypertension, dyslipidemia, or glucose intolerance. Other causes of hyperandrogenism must be excluded, including androgen-secreting tumors, CAH, and Cushing’s syndrome.¹⁴

In the assessment of hirsutism, it is important to differentiate between the slow insidious onset of hirsutism associated with PCOS or late-onset CAH and the rapidly progressive onset of hirsutism associated with virilization due to an androgen-secreting tumor. Symptoms of true virilization, such as deepening of the voice, clitorimegaly, and change in body habitus (not simply weight gain) must be excluded because their presence mandates much more extensive investigation to rule out the presence of an androgen-secreting tumor.

Specific questions regarding the severity of hirsutism will aid in assessing the need for and response to treatment. It is often helpful to document the frequency, duration, and method of hair removal and to assess a change in those parameters with treatment (i.e., electrolysis diminishing from 30 to 15 minutes weekly).

The possibility of premature ovarian failure is high in a young woman who presents with amenorrhea and vasomotor symptoms. However, these patients will frequently progress through a period of shortened menstrual cycles or oligo-ovulation before the onset of amenorrhea.⁷ Depending on the degree and duration of hypoestrogenism, progestin-withdrawal bleeding may disappear. Symptoms of hypoestrogenism may be absent if the onset was insidious; other women will present with the sudden onset of hot flashes, sleep disturbances, and mood disturbances that may be more marked than those typically seen in the natural menopausal transition. A history of precipitating factors such as chemotherapy, radiation, or extensive ovarian surgery should be sought. A family history of premature ovarian failure is present in 4% to 30%.¹⁰ A personal or family history of autoimmune disease such as diabetes, thyroid disease, Addison’s disease, lupus, or vitiligo can also be a valuable clue to potential autoimmune-mediated causes of premature ovarian failure.¹⁰

In any patient presenting with ovulatory dysfunction, the desire for fertility, the need for contraception, and the need for menstrual regulation and endometrial protection will necessarily guide the therapeutic options considered. The age of the patient and the duration of anovulatory cycles are important in establishing the potential risk of endometrial hyperplasia and the need for an endometrial biopsy and subsequent endometrial protection.

Physical Examination

The physical examination of the anovulatory patient should commence with basic observations including height, weight, calculation of BMI, and assessment of body habitus. Low BMI (>20 kg/m²) may be a clue to hypothalamic dysfunction, whereas elevated BMI (<30 kg/m²) may be suggestive of hypothyroidism (if related to recent increase in weight) or insulin resistance associated with PCOS. Other features of insulin resistance associated with PCOS include central obesity (increased waist/hip ratio) or the presence of acanthosis nigricans, a grey-brown velvety pigmentation frequently seen in the neck, axillae, and groin. Finally, consideration of PCOS should prompt an assessment of clinical hyperandrogenism. The presence of acne or hirsutism and the pattern and severity of each should be documented. An objective tool for assessment of hirsutism is the Ferriman-Gallwey score,³⁶ which is helpful, particularly in the context of research studies in which objective response to antiandrogen treatment is desirable; its utility in routine clinical assessment may be more limited. Male pattern alopecia (temporal hair loss) is also a feature of clinical hyperandrogenism.

The remainder of the physical examination should include a systematic head to toe assessment, including visual fields (bitemporal hemianopsia), a thyroid examination to exclude goiter, and breast examination for galactorrhea. The abdominal examination should pay particular attention to the presence of striae and/or male pattern hair distribution. Extensive abdominal striae may suggest cortisol excess associated with Cushing’s syndrome; a male escutcheon is further evidence of clinical hyperandrogenism. Abdominal or pelvic masses are encountered rarely.

The gynecologic examination should commence with assessment of the external genitalia for clitorimegaly. Normal estrogenization is confirmed by healthy vaginal rugae, often accompanied by copious clear cervical mucus. In contrast, low estrogen levels due to prolonged hypothalamic anovulation or premature ovarian failure results in pale, smooth vaginal mucosa with little cervical mucus. Bimanual examination should be performed to rule out adnexal masses. Finally, in women with prolonged anovulation and evidence of normal estrogenization, an endometrial biopsy should be considered to exclude pathology such as endometrial hyperplasia.

Laboratory Evaluation

The systematic and cost-effective laboratory evaluation of ovulatory dysfunction is based on the three key principles described in Table 19-2. In the initial investigation, one should always consider the possibility of pregnancy and perform a pregnancy test when clinically indicated. Once pregnancy has been excluded, the investigation should commence with measurement of thyrotropin, prolactin, and FSH (Fig. 19-2).

Figure 19-2 Guidelines for the investigation of ovulatory dysfunction.

Thyrotropin and Prolactin

Elevated thyrotropin or prolactin levels will require appropriate investigation and therapy (see Chapter 22), keeping in mind that elevation of both thyrotropin and prolactin is usually the result of hypothyroidism and that hyperprolactinemia will respond to treatment of the underlying thyroid condition.

Gonadotropins

For the purpose of assessing ovarian reserve, measurement of early follicular FSH (day 3) is ideal; however, this may not be practical in the setting of very infrequent menses.⁶

Elevated FSH concentrations raise the possibility of premature ovarian failure and should be repeated in another cycle for confirmation. Although there is considerable variation among centers, FSH concentrations above 10 to 15 IU/L are suggestive of diminished ovarian reserve, whereas concentrations above 15 to 20 IU/L are highly suggestive of impending or established premature ovarian failure.⁶ If confirmed, additional testing is prudent in women under age 40 to identify underlying karyotype abnormalities (e.g., Turner’s mosaicism), fragile X premutations, and associated autoimmune disorders.⁹

Low or normal FSH requires further assessment of estrogen status. If the patient clearly has normal estrogen levels based on physical examination (vaginal rugae, watery cervical mucus), further investigation may not be required. If this is unclear clinically, a progestin challenge test should be done. Absence of bleeding following administration of a progestin challenge suggests clinical hypoestrogenism and is in keeping with a diagnosis of hypothalamic dysfunction. A positive progestin challenge is in keeping with ovulatory dysfunction such as PCOS.

Androgens

In the absence of clinical hyperandrogenism, identification of biochemical androgen excess is rarely of value in the management of ovulatory dysfunction, because the course of management (ovulation induction or endometrial protection) would not be altered by its presence. Therefore, screening for biochemical evidence of hyperandrogenism in every patient with anovulation is generally unnecessary and not cost-effective.

In the context of clinical hyperandrogenism, it is essential to rule out the possibility of an androgen-secreting tumor. This can frequently be done on the basis of history and physical examination alone. In the absence of rapid onset of hirsutism associated with virilization, an androgen-secreting tumor is extremely unlikely, and serum androgen screening is not required. If uncertain, measurement of testosterone (adrenal or ovarian origin), androstenedione (ovarian origin), and dehydroepiandrostenedione sulfate (adrenal origin) may be helpful. The reliability of different methods of testosterone assessment is highly variable between laboratories. Depending on local assay methodology, free testosterone or free androgen index will likely represent the most accurate assessment of biologically active testosterone.

Once androgen-secreting tumors have been ruled out, the most likely diagnosis is PCOS. However, late-onset (nonclassic) CAH may be clinically indistinguishable from PCOS. For patients desiring pregnancy, CAH should be excluded by assessment of a follicular phase 17-hydroxyprogesterone level and possibly a corticotropin stimulation test. The value of screening for this rare disorder in every anovulatory patient with evidence of hyperandrogenism is controversial; in the absence of virilizing symptoms or the desire for pregnancy, management is essentially the same as for PCOS.

Finally, a diagnosis of PCOS should prompt an assessment of associated conditions, including hypercholesterolemia and insulin resistance. Although the benefit of screening for these conditions in thin women with PCOS is unclear, a fasting cholesterol panel and an assessment of insulin resistance/glucose intolerance (glucose tolerance test ± insulin levels) are certainly warranted in obese women with PCOS or those with other risk factors (e.g., family history of hypercholesterolemia or type 2 diabetes).

The etiology of ovulatory dysfunction will ultimately remain unclear (idiopathic ovulatory dysfunction) for many patients who do not strictly fit the criteria for a diagnosis of PCOS. Once diseases requiring specific therapy have been excluded, it is acceptable to manage these patients according to their presenting symptoms and desire for pregnancy.

Progestin Challenge Test

If the estrogen status of the patient is unclear from the physical examination, a progestin challenge test can be arranged at the same time as the initial blood work. Medroxyprogesterone acetate (5 to 10 mg) is administered daily for 7 to 14 days, after which a menstrual bleed should ensue in normally estrogenized patients. Any amount of spotting or bleeding in the 2 weeks after progestin withdrawal is considered a positive progestin challenge.

MANAGEMENT

Management options for women with ovulatory disorders vary according to the presenting symptoms and clinical circumstances. The principles of therapy involve the following:

1. Identify underlying disorders that require specific treatment

2. Determine if the woman desires pregnancy, contraception, or menstrual regulation

3. Offer individualized therapy based on the presenting signs and symptoms.

After ruling out underlying medical disorders that require specific therapy, the next priority is to identify predisposing or contributing factors and provide appropriate counseling or therapy. Examples include excessive stress, anxiety or depressive disorders, excessive weight gain, overt or insidious eating disorders, strenuous exercise without appropriate nutritional intake, or a preoccupation with thinness. Remember that the menstrual cycle is often a barometer of underlying health. The etiology of hypothalamic amenorrhea can often be thought of as “energy drain,” whether it be emotional, physical, or caloric. There is a metabolic cost associated with a menstrual cycle, a luteal phase thermal shift, and subsequent menstrual blood loss. Ovulatory dysfunction is unlikely to resolve until the HPO axis senses sufficient “energy” in reserve to cover this metabolic expenditure and sustain an ensuing pregnancy. The process of identifying and discussing factors that contribute to HPO dysfunction is key to initiating the necessary lifestyle changes and/or therapy.^13,³⁷

The next step is to assess the woman’s needs in terms of contraception, menstrual regulation, and the management of any associated symptoms.

Fertility

For women who desire fertility, induction of ovulation can produce the secondary benefits of cycle control and endometrial protection. Depending on clinical circumstances, it may be prudent to rule out endometrial hyperplasia prior to ovulation induction attempts.

For overweight or underweight women, achievement of a healthy body weight and healthy eating habits may be sufficient to correct the underlying ovulatory disorder. Increased activity and modest weight reduction (at least 5% of body weight) among obese oligo-ovulatory women (BMI <30 kg/m²) with PCOS leads to improvement in ovulatory function and an increase in spontaneous and treatment-associated pregnancy rates.^38,³⁹ Similarly, ovulatory function frequently improves among women with a low BMI in response to stepwise reductions in exercise frequency or intensity, increased BMI, increased caloric and protein intake, and/or modification of restricted eating patterns.¹³

After normalization of body weight and other lifestyle changes as appropriate, ovulation can be induced with clomiphene citrate.⁴⁰ Induction of ovulation with clomiphene citrate is effective in 75% to 80% of patients, although only 40% to 50% will conceive.⁴¹ Predictors of clomiphene citrate resistance include high basal FSH concentrations (impending ovarian failure), absence of progestin-induced withdrawal bleeding (hypogonadotropic amenorrhea), and PCOS associated with high BMI (<30 kg/m²) or severe hyperandrogenism.^42,⁴³ More recently, aromatase inhibitors such as letrozole have been used for ovulation induction.^44,⁴⁵ Their role in the treatment of PCOS, and specifically clomiphene-resistant PCOS, remains unclear.⁴⁶

Various approaches have been examined for the treatment of clomiphene-resistant anovulation. Extending the duration of clomiphene citrate administration to 7 to 10 days (i.e., 100 mg from days 3 to 12) is useful in some women, with a total dose that does not exceed the amount used in a conventional protocol of 200 mg for 5 days.⁴² In clomiphene-resistant women with documented insulin resistance and/or impaired glucose tolerance, addition of metformin may improve the response to clomiphene citrate.⁴⁷

Appreciating that insulin resistance is felt to be central to the pathophysiology of PCOS, particularly in obese women, there has been great interest in the use of insulin-sensitizing agents as an adjunct to ovulation induction therapy. A number of randomized, controlled trials support improved ovulation rates with metformin, either alone or in conjunction with clomiphene citrate.^47,⁴⁸ A variable improvement in cycle parameters and pregnancy rates has been reported with metformin use during in vitro fertilization treatment cycles.⁴⁷ Limited information exists regarding the outcome of pregnancies conceived on metformin or those pregnancies in which metformin was continued during pregnancy. Similarly, there is little information about the optimal duration of therapy prior to conception. It has been our practice to discontinue metformin when conception occurs, or after a 3-month trial if there has been no appreciable weight loss or improvement in clinical or endocrine parameters.

Women who do not ovulate or conceive after making appropriate lifestyle changes and receiving therapy with oral agents become candidates for injectable exogenous gonadotropins. Women with hypothalamic amenorrhea or WHO group I ovulatory disorders lack endogenous gonadotropin activity. As such, exogenous gonadotropins can be expected to restore normal ovulation and pregnancy rates.⁴⁹ A less predictable response is observed in the heterogeneous group of women with WHO group II ovulatory disorders. Predictors of lower ovulation and pregnancy rates include increasing age, obesity, hyperandrogenism, and failure to conceive despite successful ovulation with clomiphene citrate.^49,⁵⁰

Due to the cost associated with exogenous gonadotropin therapy and the increased risk of multiple pregnancy, some women with clomiphene citrate-resistant PCOS will elect to undergo laparoscopic ovarian drilling in an attempt to restore spontaneous ovulatory function. This technique has been shown to reduce circulating androgen levels and restore ovulation in up to 75% of women treated.⁵¹ Concern regarding the longevity of its effect and ultimately compromised fertility due to postoperative adhesions has limited the broad application of this technique. However, it is a reasonable alternative in selected patients.

Contraception

If contraception is needed, combination oral contraceptives (OCs) are generally the first line of therapy because of their combined contraceptive and noncontraceptive benefits. In addition to contraception, they provide menstrual regulation, protection of the endometrium from the effects of prolonged unopposed estrogen stimulation, and relief of acne and hirsutism in hyperandrogenic women. The antiandrogenic effects arise predominantly from hypothalamic-pituitary suppression that leads to diminished androgen production from the ovarian stroma. In addition, the estrogen component rapidly produces an increase in hepatic SHBG production, resulting in a decrease in free testosterone levels.⁵² Where available, cyproterone acetate, the progestational component of some OCs, has intrinsic antiandrogenic activity and is an effective choice for the treatment of menstrual irregularities associated with hyperandrogenism.

Progestin-only contraceptives (tablets, depot injections, and implants) provide contraception, cycle control, and endometrial protection and may be appropriate alternatives for some women unable to tolerate estrogen-containing OCs or in whom such OCs are contraindicated. Relief of androgenic conditions such as acne and hirsutism is more variable. Progestin-only therapy produces suppression of the HPO axis with a resulting reduction in ovarian androgen production, although possibly to a lesser degree than combination OCs. Although testosterone clearance from the circulation is increased, there is no effect on SHBG production.⁵³ In addition, the synthetic progestins possess some intrinsic androgenic activity that, for some women, will result in a paradoxical exacerbation of their androgenic symptoms.⁵²

Cycle Control and Endometrial Protection

If contraception is not required and fertility is not an issue, the need for menstrual cycle control and endometrial protection should be assessed. There are no clear guidelines for the frequency or pattern of menstrual bleeding that ensures endometrial protection in this young population of women. However, chronic anovulation is associated with an increased risk of endometrial hyperplasia and neoplasia,^54,⁵⁵ particularly in obese women (weight <90 kg).⁵⁶ Clinical experience suggests that progestin-withdrawal bleeds should be induced every 4 to 12 weeks in the face of chronic anovulation or if menses occur less frequently than every 8 weeks.⁵⁷ Endometrial biopsy should be performed prior to therapy in women with prolonged anovulation, in those with irregular or heavy bleeding, and in women with other risk factors for endometrial hyperplasia (weight <90 kg, infertility with nulliparity, or PCOS).^51,^56,⁵⁸^–⁶⁰

For women who do not wish to have cyclic menses, hypomenorrhea or amenorrhea can be induced with continuous systemic progestin therapy using a depot progestin preparation, progestin implants, or continuous local administration via a progesterone-containing intrauterine system.

Hirsutism

Women with hyperandrogenic anovulation (PCOS) frequently experience acne or hirsutism. In women desiring pregnancy, mechanical hair removal is the only appropriate option given the concern regarding antiandrogen therapy in pregnancy. However, for those who are not attempting conception, various combination therapies are available, some of which provide simultaneous contraception and cycle control. The most effective treatment for hirsutism usually involves a combination of weight loss (where appropriate), mechanical hair removal, and medical therapy. The latter usually involves androgen suppression (e.g., oral contraceptives) and/or specific antiandrogen therapy (e.g., with spironolactone, cyproterone acetate, flutamide, or finasteride).

Ovarian Failure

Fertility is often still a concern for young women with hypergonadotropism (see Chapter 20). However, women with rising FSH levels, diminished ovarian reserve, or premature ovarian failure are rarely suitable candidates for ovulation induction therapy.⁶⁰ Depending on age and other factors, there is a small chance of spontaneous conception during occasional ovulatory cycles. Approximately 5% to 10% of young women with premature ovarian failure conceive spontaneously if they enter into a remission after the diagnosis. Beyond this, oocyte donation offers the only realistic possibility of achieving a pregnancy.^10,¹¹

Women with rising FSH levels have a presumptive diagnosis of impending ovarian failure, but may still have a relatively normal menstrual pattern and few other symptoms. Although they generally require no specific therapy, they may benefit from counseling about cardiovascular and bone health, including a healthy diet, regular weight-bearing and muscle strengthening exercise, smoking cessation, modest alcohol intake, and appropriate intake of calcium and vitamin D.

As ovarian function declines and eventually ceases, women with premature ovarian failure may benefit from measures aimed at menstrual cycle control, endometrial protection, relief of hypoestrogenic symptoms, and osteoporosis prevention. Either cyclic OCs or traditional estrogen/progestin hormone replacement therapy can be used for this purpose, although OCs may achieve better cycle control in cases where some residual ovarian function persists. Young women who are still hoping for the possibility of a remission and a pregnancy may be better served by low-dose hormone replacement therapy using micronized estradiol and progesterone. Such low-dose regimens are not likely to inhibit or mask the return of spontaneous ovulation during a remission, nor do they pose safety concerns about inadvertent exposure during an early pregnancy because they involve hormones identical to those produced by the human ovary.