Marc A. Fritz
Amenorrhea is the absence or cessation of menses and may result from a wide variety of pathologic conditions. It is a common symptom of an underlying abnormality in the reproductive system that may be anatomic (developmental or acquired), organic, or endocrinologic in nature. This chapter outlines the differential diagnosis of amenorrhea, discusses the indications and methods for evaluation, and describes options for treatment once a diagnosis and the patient's goals are clearly defined. However, the most common cause of amenorrhea in women of reproductive age is pregnancy and that should be ruled out before considering other etiologies.
THE NORMAL MENSTRUAL CYCLE
Normal menstrual function involves a complex multilevel integration of endocrine signals, local autocrine and paracrine mechanisms, and target cell receptors, all operating at four distinct levels: the genital tract, the ovary, the pituitary gland, and the hypothalamus. First, normal menstrual function requires a normal genital outflow tract. The uterus must have a functional endometrium capable of response to estrogen and progesterone, and be continuous with the cervix, vagina, and introitus. Second, the ovaries must contain follicles responsive to pituitary follicle-stimulating hormone (FSH) and luteinizing hormone (LH) stimulation. Progressive follicular development and ovulation further require the normal operation of local intraovarian regulatory mechanisms that are sensitive to changes in the endocrine milieu. Third, pituitary gonadotrophs must have the capacity to synthesize and secrete gonadotropins in response to hypothalamic gonadotropin-releasing hormone (GnRH) stimulation. The relative amounts of FSH and LH released reflect changes in the pulsatile pattern of GnRH secretion and the feedback modulation of ovarian steroid and peptide hormones. Finally, specialized neurosecretory cells located in the medial basal hypothalamus (arcuate nucleus) must have functional communication with the pituitary gland and be able to synthesize and release GnRH in a pulsatile pattern that varies in response to stimuli from the environment and feedback signals from the periphery.
Amenorrhea may result from congenital or acquired disease or dysfunction at the level of the genital tract, the ovary, the pituitary, or the hypothalamus. In fact, the single most common cause of amenorrhea—chronic hyperandrogenic anovulation (polycystic ovary syndrome; PCOS)—involves a number of interrelated pathophysiologic mechanisms that operate at the ovarian, pituitary, and hypothalamic levels, and does not fall neatly into any one specific category. The wide array of disorders that may be responsible suggests that amenorrhea may present a daunting diagnostic challenge but, in truth, evaluation is relatively straightforward, logical, and requires only tests and procedures with which all gynecologists should be quite familiar. With few exceptions, an accurate diagnosis can be confidently established in very little time and without great expense.
Differential Diagnosis of Amenorrhea
Although the list of potential causes of amenorrhea is long, the majority of cases relate to one of five conditions: pregnancy, PCOS, hypothalamic amenorrhea, hyperprolactinemia, and ovarian failure (Table 34.1). All of the remaining causes are relatively uncommon and only occasionally encountered in a lifetime of clinical practice.
|
|
|
TABLE 34.1. Causes of amenorrhea |
Genital Tract Abnormalities
The embryology of the female genital tract involves the medial migration and midline fusion of the paired müllerian (paramesonephric) ducts to form the uterus, cervix, and upper vagina, and the vertical fusion of that developing ductal system with the invaginating urogenital sinus to form the lower vagina and the introitus. Outflow tract abnormalities that result from failure of müllerian duct development include vaginal/müllerian agenesis and androgen insensitivity syndrome (AIS), where the uterus is altogether absent. Abnormalities caused by failure of vertical fusion include imperforate hymen, transverse vaginal septum, and cervical atresia.These conditions result in an accumulation of menstrual effluent above the level of obstruction (cryptomenorrhea). With two notable exceptions, all outflow tract abnormalities are developmental in origin and therefore cause primary amenorrhea. Asherman syndrome and cervical stenosis/obstruction are acquired conditions and therefore are causes of secondary, rather than primary, amenorrhea. Asherman syndrome results from intrauterine adhesions that obstruct or obliterate the uterine cavity as a consequence of inflammation (postpartum endometritis, retained products of conception) or trauma (curettage). Severe cervical stenosis with complete outflow obstruction is a rare complication of cervical conization procedures or other surgical treatments for cervical intraepithelial neoplasia.
Ovarian Disorders
Ovarian failure occurs when few or no follicles remain that are capable of producing estradiol in response to pituitary gonadotropin stimulation. Follicular depletion may occur during embryonic life with no follicles remaining by infancy or early childhood, after puberty has begun but before menarche, or at some later time before menopause would normally be expected. Therefore, depending on when the available supply of ovarian follicles is functionally depleted, puberty may not occur, it may begin normally but stop before the first menses, or it may progress normally to and beyond menarche with secondary amenorrhea having onset at some later point in time.
Gonadal dysgenesis is among the most common of all causes of primary amenorrhea (approximately 30%–40%) and results from an absence of ovarian follicles or accelerated follicular depletion during embryogenesis or the first few years of life. The gonads of affected individuals contain only stroma and appear as fibrous streaks. The most common form of gonadal dysgenesis is Turner syndrome, classically associated with a 45,X karyotype, but also with an assortment of other structural X chromosome abnormalities (deletions, ring, and iso-chromosomes). These X chromosomal anomalies may be present in all or only in some of the cells of the body (mosaicism), depending on the stage of embryonic development at the time that they arise. Although less common, individuals with gonadal dysgenesis also may have a normal 46,XX or a 46,XY karyotype (Swyer syndrome) in all cells or in one or more cell lines in mosaic individuals (e.g., 45,X/46,XX; 45,X/46,XY). Most, in the absence of ovarian follicles, have no significant secondary sexual development. A few may have transient normal ovarian function, the extent depending on when the limited supply of ovarian follicles is depleted. Approximately 15% begin but do not complete pubertal development and approximately 5% have sufficient follicles to complete puberty and begin spontaneous menstruation. Spontaneous pregnancies rarely occur and are associated with a relatively high risk for sex chromosome aneuploidy and spontaneous abortion.
Premature ovarian failure (POF) results in secondary amenorrhea at some time after puberty has been completed. It is distinguished from gonadal dysgenesis on the basis of ovarian morphology and histology; instead of streak gonads, the ovaries in POF more closely resemble those of postmenopausal women. Approximately 1% to 5% of women will develop POF before the age of 40 years. The karyotype in individuals with POF is most often normal (46,XX), but also may reveal mosaicism (e.g., 45,X/46,XX). A specific cause for early follicular depletion in POF frequently cannot be determined and is presumed to result from inadequate germ cell migration during embryogenesis or accelerated atresia. POF is frequently associated with autoimmune disorders and in some cases (e.g., Addison disease) appears to result from an autoimmune lymphocytic oophoritis. Radiation and chemotherapy are two other important causes of ovarian failure. The effects of both are dependent upon dose and the age at time of treatment. Galactosemia is an autosomal recessive disorder of galactose metabolism caused by a deficiency of the enzyme galactose 1-phosphate uridyltransferase and another, albeit very rare, cause of POF. Affected women have fewer primordial follicles presumably due to the cumulative toxicity of galactose metabolites on germ cell migration and survival.
Other, very rare, ovarian disorders that may cause amenorrhea include 17α-hydroxylase deficiency, aromatase deficiency, and the gonadotropin-resistant ovary syndrome. Unlike in ovarian failure, the ovaries of individuals with these disorders contain follicles and oocytes, but cannot produce estrogen. The enzyme 17 µ-hydroxlase mediates an early step in steroid hormone synthesis, without which progesterones cannot be converted to androgens and subsequently, estrogens. The enzyme aromatase mediates the conversion of androgenic precursors to estrogens; individuals with aromatase deficiency generally exhibit sexual ambiguity at birth, virilization at puberty, and multicystic ovaries. The gonadotropin-resistant ovary syndrome results from genetic mutations in the FSH or LH receptor or post-receptor signaling defects that prevent the ovaries from responding normally to gonadotropin stimulation; although present, ovarian follicles fail to develop beyond the early antral stage and therefore produce little estrogen.
Pituitary Disorders
Pituitary tumors may cause amenorrhea by directly compressing pituitary gonadotrophs or distorting the portal venous network that delivers hypothalamic GnRH stimulation, resulting in decreased FSH and LH secretion. They also may cause inadequate or excessive production of other pituitary hormones, via compression of pituitary tissue, interference with the portal venous circulation, or autonomous secretion (functional pituitary adenomas). Directly or indirectly, these endocrinopathies may disrupt normal ovarian function, thereby causing amenorrhea. Although metastatic lesions may occasionally be seen, malignant pituitary tumors are extremely rare. Virtually all pituitary tumors are benign adenomas that may be functional or nonfunctional; functional tumors may secrete prolactin, growth hormone (GH), thyroid-stimulating hormone (TSH), or adrenocorticotropic hormone (ACTH).
Other uncommon pituitary disorders that may cause amenorrhea include the empty sella syndrome and Sheehan syndrome. The empty sella syndrome results from herniation of the subarachnoid space containing cerebrospinal fluid into the sella turcica with compression of the pituitary gland against the sellar floor, giving the sella an “empty” appearance when viewed by computed tomography (CT) or magnetic resonance imaging (MRI). Sheehan syndrome results from acute infarction and necrosis of the pituitary gland as a rare complication of shock due to obstetric hemorrhage. Depending on the extent of pituitary damage, clinical consequences may be limited to disorders of reproductive function (failed lactation, amenorrhea), or more general, with multisystem failure due to panhypopituitarism.
Hypothalamic Disorders
Absent or abnormal patterns of pulsatile hypothalamic GnRH secretion that fail to stimulate normal levels or patterns of pituitary gonadotropin secretion are the most common cause of amenorrhea. Both PCOS and hypothalamic amenorrhea that may result from emotional, nutritional, or physical stress are basically hypothalamic disorders, but their pathophysiology and clinical presentations differ considerably. Women with PCOS exhibit an increased frequency of pulsatile GnRH secretion that results in increased LH synthesis, hyperandrogenism, and impaired follicular maturation. In contrast, the inconsistent and generally low frequency pattern of pulsatile GnRH secretion in women with hypothalamic amenorrhea results in low levels of pituitary gonadotropin release that fail to stimulate or sustain progressive follicular development.
Occasionally, a hypothalamic tumor (craniopharyngioma, meningioma, hamartoma, chordoma) may distort the tuberoinfundibular tract or portal venous network, thereby interfering with effective delivery of GnRH stimulation, and result in decreased pituitary FSH and LH secretion. Alternatively, interference with hypothalamic dopamine delivery to pituitary lactotrophs may cause hyperprolactinemia. In either case, hypothalamic tumors may result in a secondary hypogonadotropic hypogonadism and amenorrhea. In other rare instances, GnRH deficiency is congenital and associated with midline craniofacial defects or with anosmia due to a failure of olfactory axonal and GnRH neuronal migration during embryogenesis (Kallmann syndrome), resulting in primary amenorrhea and sexual infantilism.
EVALUATION OF AMENORRHEA
A detailed medical history and physical examination are always important. In the patient with amenorrhea, elements of particular interest include growth and secondary sexual development (breast and pubic hair), menstrual history (if any), previous surgery or trauma to the pelvis or central nervous system (CNS), family history of hereditary disorders, evidence of physical, psychological or emotional stress, symptoms and signs of hirsutism or galactorrhea, as well as reproductive tract anatomy.
Medical History
The age at which menarche should be expected varies but, in general, the first menses should occur within 2 to 3 years after the initiation of pubertal development. In most young girls (approximately 80%), the first sign of puberty is an acceleration of growth, followed by breast budding (thelarche), and the appearance of pubic hair (adrenarche). In the remainder, adrenarche precedes thelarche by a brief interval, but the two events typically are closely linked. Consequently, menarche should be expected as early as age 10 (when puberty begins at age 8), and rarely later than age 16 (when puberty begins at age 13). On average, in the United States, the mean ages for thelarche, adrenarche, and menarche in black girls are 6 to 12 months earlier than in white girls. When secondary sexual development fails to begin by age 14, or begins but fails to progress at the normally expected pace, evaluation is indicated. Once menstrual cycles have been established, amenorrhea for an interval equivalent to three previous cycles, or 6 months, warrants evaluation.
Questions relating to past medical history, general health, and lifestyle may identify a severe or chronic illness (diabetes, renal failure, inflammatory bowel disease), head trauma, or evidence of physical, psychological, or emotional stress. Weight loss or gain and the frequency and intensity of exercise may be revealing. Headaches, seizures, vomiting, behavioral changes, or visual symptoms may suggest a CNS disorder. Vaginal dryness or hot flushes are evidence of estrogen deficiency and suggest ovarian failure. Progressive hirsutism or virilization is evidence of hyperandrogenism that may result from PCOS, nonclassic (late-onset) congenital adrenal hyperplasia (CAH), or an androgen-producing tumor of the ovary or adrenal gland. Bilateral galactorrhea suggests hyperprolactinemia. Cyclic pelvic or lower abdominal pain or urinary complaints may be caused by developmental anomalies resulting in obstructed menstrual flow, including an imperforate hymen, transverse vaginal septum, or cervical atresia. A previous inguinal hernia repair or curettage suggests the possibility of a developmental anomaly or damage to the reproductive tract. The timing and duration of any treatment with progestational agents (oral contraceptive pills [OCP], depot-medroxyprogesterone acetate), GnRH agonists (leuprolide, goserelin, nafarelin), or other medications (phenothiazines, reserpine derivatives, amphetamines, opiates, benzodiazepines, antidepressants, dopamine antagonists) or drugs (opiates) may provide important diagnostic clues.
Physical Examination
Body habitus often provides important clinical information. Height, weight, and body mass index (BMI) should be recorded. Short stature (less than 60 inches) is a hallmark of gonadal dysgenesis; sexual infantilism, webbing of the neck, low set ears and posterior hairline, widely spaced nipples, short fourth metacarpal, and a wide carrying angle of the arms (cubitus valgus) are among the classical stigmata of Turner syndrome. Low body weight is frequently associated with hypothalamic amenorrhea resulting from poor nutrition (eating disorders) or physical, psychological, or emotional stress. Obesity or an increased waist-to-hip ratio (>0.85) is often associated with insulin resistance and chronic anovulation.
Examination of the skin may reveal a soft, moist texture as seen in hyperthyroidism; a rapid pulse and classic eye signs (exophthalmos, lid lag), a fine tremor, and hyperreflexia may provide further evidence to suggest a diagnosis of Graves disease. Conversely, dry, thick skin, a slow pulse, diminished reflexes, and thinning of the hair suggest hypothyroidism. Both hypothyroidism and hyperthyroidism may be associated with amenorrhea. Orange discoloration of the skin in the absence of scleral icterus may result from hypercarotinemia associated with excessive ingestion of low-calorie, carotene-containing fruits and vegetables in dieting women. Acanthosis nigricans, velvety hyperpigmented skin most commonly observed at the nape of the neck, in the axillae, and beneath the breasts, strongly suggests severe insulin resistance and the possibility of diabetes. Acne and hirsutism are indications of hyperandrogenism that may result from chronic anovulation (PCOS), nonclassic CAH, or ingestion of androgenic anabolic steroids. When accompanied by any sign of frank virilization (deepening of the voice, frontotemporal balding, decrease in breast size, increased muscle mass, clitoromegaly), the possibility of ovarian hyperthecosis or an ovarian or adrenal neoplasm must be considered.
Breast development, as assessed by Tanner staging, is a reliable indicator of estrogen production or exposure to exogenous estrogens. Arrested breast development suggests a disruption of the hypothalamic–pituitary–ovarian (HPO) axis. When menarche has not followed adult breast development, a developmental anomaly of the reproductive tract should also be considered. The breast examination should include gentle compression, beginning at the base and moving toward the nipple. Microscopic examination of any expressed cloudy or white nipple secretions that demonstrate lipid droplets indicate true galactorrhea and suggest hyperprolactinemia.
Abdominal examination may rarely reveal a mass as may result from hematometra or an ovarian neoplasm. Growth of hair from the pubic symphysis to the infraumbilical region suggests hyperandrogenism. Abdominal striae raise the possibility of Cushing syndrome, but much more often result from progressive obesity or previous pregnancy.
As noted earlier, thelarche and adrenarche typically are closely linked events during puberty and, in general, breast development and growth of pubic hair progress in a symmetric manner. The Tanner stages of breast and pubic hair development should be consistent. Absent or scant growth of pubic hair is a classic sign of AIS when breast development is asymmetrically advanced. Attempts at office examination of the vagina in sexually infantile girls or those with a small hymeneal ring are often difficult and counterproductive, but whenever feasible, speculum examination should be performed. A patent vagina and visible cervix excludes müllerian/vaginal agenesis, AIS, and most obstructive causes of amenorrhea. In those with an absent or infantile vaginal orifice, rectal examination should be performed and may reveal a distended hematocolpos above the obstruction when the uterus is present and functional.
Diagnostic Evaluation
A careful history and physical examination will always narrow the range of diagnostic possibilities that must then be differentiated. The subsequent laboratory investigation or imaging should be focused on that differential diagnosis and, with few exceptions a diagnosis can be quickly and easily established. Sexual ambiguity and virilization should be evaluated as separate disorders, mindful that amenorrhea is an important element of their presentation.
Abnormal Genital Tract Anatomy
A history of primary amenorrhea accompanied by physical examination that reveals an absent or blind vagina indicates a developmental anomaly of the genital outflow tract. The list of diagnostic possibilities is short and includes an imperforate hymen, a transverse vaginal septum, cervical atresia, müllerian/vaginal agenesis, and androgen insensitivity syndrome (Fig. 34.1). Because each of these disorders has unique features, they generally are not difficult to distinguish.
|
|
|
FIG. 34.1. Congenital absence of the vagina. The external genitalia are entirely normal in appearance without any evidence of ambiguity but there is simply no vagina present. The normal pubic hair indicates androgen responsiveness and eliminates the possibility of androgen insensitivity syndrome indicating that this is simply congenital absence of the müllerian duct system. |
Patients with an imperforate hymen or transverse vaginal septum/cervical atresia typically present at the expected time of menarche with complaint of cyclic perineal, pelvic, or abdominal pressure or pain and exhibit normal secondary sexual development. In those with an imperforate hymen, genital examination reveals no obvious vaginal orifice, but a thin, often bulging, blue perineal membrane and a fluctuant mass, resulting from the accumulation of mucus and blood in the vagina. In those with a transverse vaginal septum or cervical atresia, examination typically reveals a normal vaginal orifice, a short blind vagina of varying length, and a fluctuant pelvic mass well above to the level of obstruction (hematocolpos, hematometra, hematosalpinx). Differentiation of imperforate hymen and transverse vaginal septum/cervical atresia generally requires no laboratory investigation. Whereas a transabdominal or transperineal ultrasound examination will reveal the level and volume of sequestered menses, an abdominalpelvic MRI provides greater anatomic detail and helps to define the nature of an anomaly (Fig. 34.2). In some instances, laparoscopy may be required to clearly identify the anatomy of a developmental anomaly.
|
|
|
FIG. 34.2. Congenital absence of the lower one third of the vagina. Magnetic resonance imaging demonstrates that a uterus is present (A) as well as an upper vaginal pouch (B). The upper vagina has formed a hematocolpos which began soon after menarche and presented as a pelvic mass. |
In contrast, patients with amenorrhea resulting from müllerian agenesis generally are otherwise entirely asymptomatic. They exhibit normal breast and pubic hair development, an absent vagina, and have no symptoms or signs of cryptomenorrhea because the uterus is altogether absent. The diagnosis is usually self-evident from physical examination alone. Further evaluation to exclude skeletal and urinary tract anomalies is indicated because approximately 12% to 15% of women with müllerian agenesis have skeletal abnormalities (vertebral anomalies are most common) and one third or more have urinary tract anomalies (ectopic kidney, renal agenesis, horseshoe kidney, abnormal collecting system).
In girls who have not yet reached the age when menarche (and cryptomenorrhea) would be expected if a uterus was present, imaging must be interpreted cautiously because even abdominal/pelvic MRI can be misleading when the reproductive organs are immature. Remaining alert to the diagnostic possibilities, careful observation over time is preferable to invasive investigations that are otherwise unnecessary.
Normal breast development, absent or sparse growth of pubic hair, and a short blind vagina clearly suggest AIS. Although the chromosomal sex is male (46,XY), the phenotype is female. The testes are undescended, often palpable in the inguinal canals (most commonly at the level of the external inguinal ring), and produce normal male levels of testosterone and müllerian inhibitory hormone (MIH). Whereas end-organ insensitivity to androgen action (due to abnormalities of the androgen receptor) prevents normal masculinization of the genitalia, MIH inhibits müllerian development in a normal fashion. Consequently, the external genitalia are those of a female (absent androgen action), the uterus is absent (normal MIH action), and the vagina is short and ends blindly (derived from the invaginating urogenital sinus, absent androgen action). The diagnosis may be suspected when other family members (e.g., aunt, sister) are affected with this X-linked disorder. Incomplete penetrance may result in growth of more pubic hair than might be expected and sometimes can be misleading. However, a serum testosterone concentration easily distinguishes androgen insensitivity syndrome (AIS; normal or modestly elevated above the range observed in normal males) from müllerian agenesis (normal range for a female). A karyotype firmly establishes the diagnosis.
Cervical obstruction/stenosis and Asherman syndrome are abnormalities of genital tract anatomy, but physical examination of the genital tract most often is normal. When cervical stenosis causes symptoms, worsening dysmenorrhea or prolonged light staining or spotting after menses are the most common complaints; amenorrhea is a rare occurrence. In women with a history of previous conization or other cervical surgery or ablative therapy, uterine soundingmay help to establish a diagnosis. Similarly, most women in whom previous infection or surgical trauma has resulted in intrauterine synechiae present with dysmenorrhea, hypomenorrhea, subfertility, or recurrent early pregnancy loss, rather than amenorrhea. In women whose history clearly suggests the possibility of intrauterine adhesions, ultrasound and hysterosalpingography can reveal their location and extent, but hysteroscopy is the definitive method for diagnosis.
Normal Genital Tract Anatomy
When physical examination reveals normal genital tract anatomy, further evaluation is required to determine the cause of amenorrhea. The possibility of pregnancy should always be considered and excluded.
When breast growth is absent or inconsistent with age and associated with primary amenorrhea, the cause of delayed puberty should be determined. The vast majority of these cases have no pathology. In the remainder, evaluation may reveal thyroid disease, chronic illness (malabsorption, renal disease, eating disorders, inflammatory bowel disease), ovarian failure (e.g., gonadal dysgenesis), a pituitary disorder (tumor, empty sella syndrome, hyperprolactinemia), or a hypothalamic cause (Kallmann syndrome; physical, emotional, or psychological stress; tumor). Wrist x-rays for bone age and a GnRH stimulation test are important components of the evaluation for delayed puberty in children and adolescents, but the diagnostic possibilities and scope of evaluation in adolescents and adults with either primary or secondary amenorrhea are otherwise very much the same.
Thyroid Function Tests
Initial evaluation should include a measurement of serum TSH. The newest generation of ultrasensitive TSH assays in common use provides the means to detect both primary hypothyroidism (elevated TSH) and primary hyperthyroidism (low TSH). Either may result in chronic anovulation and amenorrhea. Any abnormal value should be confirmed and accompanied by measurement of serum thyroxine (tetraiodothyronine; T4) to better define the nature and extent of the thyroid disorder.
When TSH is elevated and the T4 concentration is normal, the diagnosis is subclinical hypothyroidism, best viewed as a compensated state wherein normal levels of T4 are maintained, but only under increased levels of pituitary stimulation. On rare occasions, both TSH and T4 levels may be low, suggesting hypothyroidism of pituitary origin that will require additional evaluation to include hypothalamic/pituitary imaging (MRI) and careful assessment to determine whether other pituitary functions also are affected.
Prolactin
A serum prolactin determination is another component of the initial evaluation of amenorrhea. Hyperprolactinemia may be associated with a variety of menstrual disturbances—oligomenorrhea and amenorrhea being the most common. In general, prolactin concentrations are higher in amenorrheic than in oligomenorrheic hyperprolactinemic women. Hyperprolactinemia may occasionally result in delayed puberty and primary amenorrhea, if it arises before menarche, and is among the most common causes of secondary amenorrhea. Hyperprolactinemia inhibits pulsatile hypothalamic GnRH secretion, resulting in depressed levels of pituitary FSH and LH secretion. The end result is anovulation or more profound hypogonadotropic hypogonadism, depending on the extent to which gonadotropin secretion is suppressed. One cannot rely on the symptom or finding of galactorrhea to identify individuals whose amenorrhea may result from hyperprolactinemia. Only approximately one third of hyperprolactinemic women will exhibit galactorrhea, probably because breast milk production requires several other hormones, including GH, T4, cortisol, insulin, and most important, estrogen and progesterone. Serum prolactin determinations therefore should be obtained in all amenorrheic women.
Hyperprolactinemia has many causes. Among these are:
· prolactin-secreting pituitary adenomas
· other pituitary or hypothalamic tumors that may distort the portal circulation and thereby prevent effective delivery of hypothalamic dopamine (the putative prolactin inhibitory factor or hormone)
· a variety of drugs that lower dopamine levels or inhibit dopamine action (amphetamines, benzodiazepines, butyrophenones, metoclopramide, methyldopa, opiates, phenothiazines, reserpine, and tricyclic antidepressants)
· breast or chest wall surgery, cervical spine lesions, or herpes zoster (activation of the afferent sensory neural pathway that stimulates prolactin secretion, in a manner similar to suckling)
· hypothyroidism (increased hypothalamic thyrotropin-releasing hormone stimulates pituitary prolactin secretion directly)
· pharmacologic estrogens (OCP)
· other rare, nonpituitary sources (lung and renal tumors) or causes of decreased prolactin clearance (renal failure).
All causes must be considered and systematically excluded; a careful history will eliminate many of the possibilities. When medications are the cause, prolactin concentrations are usually only moderately elevated and levels greater than 100 ng/mL generally are uncommon. Although medications may offer an obvious explanation, one cannot confidently assume they are the cause of hyperprolactinemia. If possible, a trial discontinuation or use of an alternative medication should be considered. When that is not possible and hypothyroidism has been excluded, further evaluation to exclude hypothalamic and pituitary tumors and other causes of hyperprolactinemia is appropriate.
Imaging of the hypothalamic and pituitary regions to exclude mass lesions in hyperprolactinemic patients can be accomplished with head CT or MRI. MRI generally is regarded as the superior method because it is more accurate for identification of very small lesions or an empty sella and better defines tumor margins and relationships to surrounding structures. The indications for CT or MRI in the evaluation of hyperprolactinemia remain controversial. Those who advocate liberal use of imaging correctly emphasize that the likelihood of a pituitary tumor does not correlate with the prolactin concentration. Pituitary microadenomas (≤10 mm) are very common (10%–30% prevalence in autopsy studies) and even macroadenomas (>10 mm) may be associated with only modest elevations (25–100 ng/mL) of prolactin because they are nonfunctional tumors or may have undergone necrosis. Moreover, imaging may reveal evidence of other hypothalamic disease that may be important (tumor, tuberculosis, sarcoidosis, aqueductal stenosis) and amenable to specific treatment. Those who prefer a more selective approach stress that imaging is costly and has a relatively low yield when performed routinely. They correctly emphasize that pituitary tumors rarely grow (even in pregnancy), are not a contraindication to hormone replacement or OCP, and that their natural course is unaffected by treatment with dopamine agonists (bromocriptine, pergolide, cabergoline). In this view, because diagnosis of a pituitary microadenoma generally has little or no impact on clinical management decisions, MRI should be limited to those with grossly elevated prolactin levels (>100 ng/mL) who are hypoestrogenic, in whom a macroadenoma of greater clinical significance is more likely, and to those with suspicious symptoms (visual disturbances, headaches) or findings (visual field defects, abnormal optic fundi). For asymptomatic patients with moderate hyperprolactinemia (20–100 ng/mL), some advocate a less costly coned down lateral view of the sella turcica and reserve MRI for those with an enlarged or abnormal sella (erosion of the sellar floor or clinoid processes) or hypothalamic calcifications that suggest a tumor (e.g., craniopharyngioma). The most prudent approach is to obtain an MRI whenever persistent hyperprolactinemia cannot be confidently attributed to medication or hypothyroidism.
FSH
Initial evaluation of amenorrheic women with normal genital tract anatomy should include measurement of serum FSH to distinguish ovarian failure (elevated FSH) from hypothalamic/pituitary disease or dysfunction that yields inadequate or ineffective patterns of gonadotropin secretion (low or normal FSH).
An elevated FSH level generally is a reliable indicator of ovarian failure, but must be interpreted in the context of the clinical presentation. During the perimenopause, regardless whether it occurs prematurely or at the usual age, FSH levels may rise well before menses have ceased entirely. Those follicles that remain also are relatively insensitive to FSH, but many can and will respond to rising levels of stimulation when the requisite threshold FSH concentration is achieved. Once follicular growth begins and estrogen levels rise, FSH concentrations decline, albeit transiently, before rising again to the levels necessary to stimulate new follicular growth. FSH levels therefore are dynamic, often fluctuate widely during the perimenopause, and must be interpreted cautiously. The FSH level is high in the enigmatic gonadotropin-resistant ovary syndrome that may result from inactivating mutations in the FSH or LH receptor. However, specific efforts to diagnose these rare conditions (ovarian biopsy, genotyping) are academic and have no practical clinical value because the prognosis for future fertility is extremely poor and treatment options are no different from those for women with true ovarian failure. Serum FSH concentrations also are elevated in individuals with 17α-hydroxylase or aromatase deficiencies and galactosemia, but these conditions are extremely rare and do not enter into clinical consideration.
With few exceptions, a high serum FSH level is an indication of ovarian failure. A history of previous radiation or chemotherapy may provide an obvious explanation. Doses of radiation under approximately 100 rads generally have no significant effect, but risk of ovarian damage rises progressively with higher doses. Individuals treated when young may have only transient amenorrhea with a return of menstrual cycles months or years later, but also are more likely to develop POF. Those treated as adults are at greater risk for immediate and irreversible ovarian failure. Alkylating agents (e.g., cyclophosphamide), used in the treatment of malignancies and other diseases (systemic lupus erythematosus), are extremely toxic to gonadal tissues. As with radiation, the dose required to induce ovarian failure is inversely related to age at the time of treatment. Other chemotherapeutic agents have the potential for ovarian damage, but their effects are less clear; risk increases with the number of agents involved in combination therapies.
When ovarian failure occurs before age 30 and cannot be confidently explained, a karyotype should be obtained. An abnormal karyotype may be observed in up to one-half of all women with primary amenorrhea. Classic Turner syndrome (45,X), structural abnormalities of the X chromosome (deletion, ring, iso-chromosome), and mosaicism (e.g., 45,X/46,XX) are the most common abnormalities found. A karyotype also will detect the presence of a Y chromosome that might not otherwise be suspected. The phenotype in Swyer syndrome (46,XY gonadal dysgenesis) and some with Turner mosaicism (45,X/46,XY) is female because the dysgenetic (streak) gonads fail to produce both MIH and androgens. Consequently, the uterus, fallopian tubes, cervix, and vagina develop normally but the genitalia do not masculinize. However, a peripheral leukocyte karyotype alone cannot exclude the presence of occult Y chromosomal material. Further analysis with fluorescence in situ hybridization (FISH) using one or more probes that are specific for segments of the Y chromosome is required and should be performed in any individual with a 45,X karyotype or 45,X mosaic cell line. An occult Y chromosome must be identified because affected individuals are at significant risk (approximately 25%) for developing a unique type of germ cell tumor (gonadoblastoma) that may contain malignant elements (dysgerminoma, embryonal cell carcinoma, choriocarcinoma). Virtually all such tumors arise early in life. Over the age of 30, therefore, karyotype is unnecessary and ovarian failure can be confidently regarded as premature menopause. Approximately 25% of patients with gonadal dysgenesis have a normal karyotype (46,XX). As gonadal dysgenesis with a normal karyotype is associated with neurosensory deafness, audiometry should be considered.
In women with secondary amenorrhea and unexplained POF, further evaluation to exclude autoimmune disease is appropriate as up to 40% may have autoimmune disorders. POF develops in 10% to 60% of women with Addison disease (adrenal insufficiency) and also is more common in women with diabetes mellitus (type 1), myasthenia gravis, and parathyroid disease than in healthy women. However, only those with Addison disease are likely to have a demonstrable autoimmune lymphocytic oophoritis. Ovarian biopsy is not indicated in clinical practice, but because POF may be a component of a polyglandular syndrome, general screening for autoimmune disorders is reasonable. Thyroid abnormalities are the most common and can be identified by measuring TSH, T4, and thyroid autoantibodies (antiperoxidase, antithyroglobulin). Screening may also include 24-hour urinary free cortisol, fasting blood glucose, serum calcium and phosphorus, and antinuclear antibody. More extensive or specific testing for autoimmune disorders is unnecessary in the absence of other clinical signs and symptoms of disease. Although autoimmune screening is logical and practical, no one or combination of serum markers can confirm a diagnosis of autoimmune ovarian failure. A low normal serum FSH concentration is an indication of hypothalamic or pituitary dysfunction and the most common result observed in clinical practice. PCOS and hypothalamic amenorrhea are the two main diagnostic possibilities and generally are easily distinguished by their clinical presentations.
POLYCYSTIC OVARY SYNDROME
Although there is no universally accepted definition of PCOS, diagnosis generally is based on three criteria—ovulatory dysfunction, clinical evidence of hyperandrogenism (hirsutism, acne, androgenic alopecia) or hyperandrogenemia, and exclusion of other disorders (hyperprolactinemia, thyroid abnormalities, nonclassic CAH). Women with PCOS more commonly exhibit oligomenorrhea (75%) than amenorrhea (25%); irregular and infrequent menses typically begin soon after menarche, but may emerge later, often in association with progressive weight gain. Signs of androgen excess generally do not become evident until years later and progress gradually. Transvaginal ultrasound examination typically reveals ovaries that are modestly enlarged and contain numerous small follicles aligned in the periphery (“string of pearls”), but such findings are not useful for diagnosis because in up to one-third of normal women between the ages of 18 and 25 years, the ovaries have a similar polycystic appearance. Women with PCOS are frequently insulin-resistant (insulin sensitivity is reduced by 30%–40%), exhibit compensatory hyperinsulinemia (up to 80%), and are predisposed to glucose intolerance. Approximately 30% have demonstrable impaired glucose intolerance by glucose tolerance testing; fasting glucose is elevated in less than 10%. A fasting glucose/insulin ratio less than 4.5 is a fairly specific but somewhat insensitive diagnostic criterion for insulin resistance. Obesity is a common feature of women with PCOS (50%–75%) and exacerbates insulin resistance and the hyperinsulinemia that is associated with elevated androgen levels. Prolactin levels are mildly elevated in 10% to 25% of women with PCOS. Although the ratio of serum LH/FSH is frequently increased (>2.0), measurement of the serum LH level generally is not useful or necessary. The diagnosis of PCOS is not based on findings of ovarian or hormonal abnormalities, but on a history of chronic anovulation and clinical findings of androgen excess or obesity.
When hirsutism is severe or perimenarchial in onset, the possibility of nonclassic CAH also must be considered. Most commonly, nonclassic CAH results from a deficiency of the enzyme 21-hydroxylase which mediates an essential step in cortisol synthesis. Affected individuals cannot efficiently convert 17-hydroxyprogesterone (17-OHP) to 11-deoxycorticosterone (DOC; an intermediate step in cortisol synthesis). A follicular phase 17-OHP level greater than 2 ng/mL merits further evaluation with an ACTH stimulation test (serum 17-OHP before and 30–60 minutes after intravenous injection of 250 g ACTH) to confirm the diagnosis (post-stimulation serum 17-OHP concentration >1,000 ng/dL).
Determination of Estrogen Status
Evaluation of estrogen levels would seem logical for differentiating PCOS from hypothalamic amenorrhea and other hypoestrogenic disorders. Unfortunately, available methods cannot easily and reliably define the level of ovarian estrogen production. One cannot rely on symptoms and signs of estrogen deficiency to identify hypogonadal women. Genitourinary atrophy develops only gradually and is uncommonly observed in young women, even when estrogen levels are extremely low, and vasomotor symptoms typically are absent in women with hypothalamic dysfunction. Other methods for assessing the level of ovarian estrogen production include immunoassay of the serum estrogen concentration and “bioassays” based on clinical observation of the amount and character of cervical mucus (“estrogenic” mucus being clear, watery, and relatively abundant) or results of a “progestin challenge test” (presence or absence of withdrawal bleeding after administration of an exogenous progestin). Each of these methods may be useful, but each clearly also has pitfalls.
A serum estradiol measurement is easy to perform and relatively inexpensive. One might reasonably expect low estrogen levels in women with hypothalamic amenorrhea and normal levels in women with PCOS. Unfortunately, estradiol concentrations fluctuate erratically and may be normal or low on any given day, and therefore can be misleading. Whereas observations of estrogenic cervical mucus clearly suggest a normal level of ovarian estrogen production, the absence of such findings cannot be confidently interpreted because many women exhibit such mucus only in the late follicular phase of the cycle when estrogen levels are relatively high, or not at all.
The progestin challenge is based on the observation that progestin treatment (e.g., medroxyprogesterone acetate 10 mg daily for 5–7 days or progesterone in oil 100 mg i.m.) will induce menses only in those with normal circulating estrogen concentrations. For this purpose, a pure progestational agent must be used; endogenous estrogen status cannot be inferred from the response to an OCP that contains both estrogen and progestin. A positive test (bleeding after completion of progestin treatment) implies normal levels of estrogen production and a negative test (no withdrawal menses) suggests frank hypogonadism. However, withdrawal bleeding correlates poorly with estrogen status; both false-positive (withdrawal bleeding despite generally low levels of estrogen production) and false-negative (absent bleeding despite significant estrogen production) results are common. Up to 20% of women with oligomenorrhea or amenorrhea in whom substantial estrogen is present do not exhibit withdrawal bleeding. Conversely, up to 40% of women whose amenorrhea relates to stress, exercise, weight loss, or hyperprolactinemia, in whom estrogen levels are generally low, exhibit withdrawal bleeding. A false-positive progestin challenge also has been frequently observed in women with POF.
Hypothalamic Amenorrhea
In the absence of obesity or evidence of hyperandrogenism characteristic of PCOS, the most likely cause of amenorrhea in women with a normal or low serum FSH level is a functional disorder of the hypothalamus or higher CNS centers. Women with such “hypothalamic amenorrhea” generally present with secondary amenorrhea that is frequently accompanied by history of emotional stress, weight loss (dieting), poor nutrition (eating disorders, chronic illness), or regular strenuous exercise (endurance training). In contrast to women with PCOS, they typically have normal or low body weight and are poorly estrogenized. In the context of low levels of estrogen production, a low or normal serum FSH has the same implication—a dysfunctional HPO axis—because if that axis were intact and fully functional, the classic negative feedback relationship between estrogen and FSH would stimulate a compensatory increase in FSH secretion and result in an elevated serum FSH concentration. A low or normal serum FSH therefore clearly suggests hypothalamic or pituitary disease or dysfunction. Most women with hypothalamic amenorrhea have a normal FSH concentration; extremely low or undetectable FSH levels are seldom seen except in women with large pituitary tumors or anorexia nervosa.
Amenorrhea associated with weight loss due to dieting is common; anorexia nervosa is fortunately much less common (15/100,000 women/year). In athletic women, the risk of amenorrhea is increased approximately three-fold over that in nonathletic women with the highest prevalence observed in endurance athletes (long-distance running). Chronic debilitating diseases (end-stage renal disease, malignancy, acquired immune deficiency syndrome, malabsorption) also may result in anovulation and amenorrhea. In such cases, hypothalamic amenorrhea represents a functional suppression of the reproductive system that may be viewed as a psychobiologic response to psychological, physical, or nutritional stress. One proposed unifying hypothesis emphasizes the concept of energy balance: When available energy is excessively diverted (exercise), or insufficient (dieting, malnutrition), reproduction is suspended in order to support essential metabolism for survival. The mechanism responsible may involve a stress-induced increase in hypothalamic corticotropin-releasing hormone (CRH) and endogenous opioid secretion that inhibits pituitary gonadotropin release directly, or inhibition of pulsatile hypothalamic GnRH secretion by increased dopamine or opioids.
Hypothalamic–Pituitary Imaging
The diagnosis of hypothalamic amenorrhea cannot be established until organic disease of the CNS, hypothalamus, or pituitary gland is excluded with a CNS MRI. Imaging is prudent, even when emotional stress, weight loss, poor nutrition, or regular strenuous exercise appear to offer an explanation for hypothalamic dysfunction and amenorrhea. Imaging may reveal a pituitary tumor, an empty sella, or evidence of a hypothalamic tumor, anomaly, or other disease.
The vast majority of pituitary tumors are prolactinomas or nonfunctioning adenomas, but other varieties are rarely encountered. Additional evaluation can help to define the nature of a tumor and the extent to which other pituitary functions may be compromised. Grossly elevated prolactin levels clearly suggest a prolactinoma; more modest prolactin elevations may be observed when large tumors infarct or a nonfunctioning adenoma distorts sellar anatomy and disrupts normal dopamine delivery to pituitary lactotrophs.
When the clinical presentation suggests Cushing syndrome, screening (24-hour urinary free cortisol, overnight dexamethasone suppression test) is indicated; additional evaluation (serum ACTH, adrenal CT or MRI) is required when results are abnormal. Physical findings that suggest acromegaly are an indication to measure serum insulin-like growth factor-1 (IGF-1; somatomedin-C); if elevated, an oral glucose tolerance test with GH levels should be performed (lack of suppression is diagnostic). Pituitary function testing (TSH, T4, prolactin, 24-hour urinary free cortisol, IGF-1) also is indicated when imaging reveals a macroadenoma or an empty sella to insure that the other trophic pituitary hormones are normal. Dynamic testing is unnecessary in otherwise asymptomatic women with a microadenoma or normal sella. The empty sella syndrome generally is a benign condition and not progressive. However, because of the possibility of an unrecognized coexisting tumor, periodic surveillance with a prolactin determination and MRI are indicated.
Anatomic abnormalities of the hypothalamus are distinctly uncommon. Hypothalamic tumors (craniopharyngioma, hamartoma, meningioma) generally are rare and other mass lesions (tuberculosis, sarcoidosis) are even more so. Congenital hypogonadotropic hypogonadism associated with anosmia or hyposmia (Kallmann syndrome) is another rare inheritable disorder associated with a specific anatomic defect: Hypoplastic or absent olfactory sulci. The condition results from a failure of both olfactory axonal and GnRH neuronal migration during development and is genetic in origin (X-linked, autosomal dominant, or autosomal recessive). The most common form (X-linked) derives from mutations in a single gene (KAL) on the short arm of the X chromosome that encodes a protein (anosmin-1) necessary for normal neuronal migration. The clinical presentation (primary amenorrhea, sexual infantilism, hypogonadotropic hypogonadism, and anosmia) and common association with other anatomic (cleft lip and palate) and neurologic (hearing loss, cerebellar ataxia, color blindness) abnormalities is not difficult to recognize. Very often, imaging fails to reveal any anatomic abnormality of the pituitary or hypothalamus and in otherwise asymptomatic individuals, the diagnosis is, by exclusion, hypothalamic dysfunction.
TREATMENT OF AMENORRHEA
Treatment of amenorrhea is obviously focused on the specific etiology, when known, and always is tailored to the goals of the patient.
Genital Tract Abnormalities
In women with vaginal/müllerian agenesis, the primary goal of treatment—creation of a functional vagina—can be accomplished with a variety of methods when the time is appropriate. In most cases, progressive vaginal dilation will be successful in the motivated patient. The technique involves application of pressure to the point of moderate discomfort (approximately 20–30 min/d) using commercially available vaginal dilators, first in a posterior direction (to create a pouch), and then in the usual line of the vaginal axis (after about 2 weeks). After the desired depth is achieved, increasingly larger dilators will expand the vaginal diameter and create a functional vagina in approximately 3 to 6 months.
Operative treatment of women with vaginal/müllerian agenesis generally should be reserved for those who refuse or poorly tolerate vaginal dilation. Traditionally, a neovagina has been created by dissection of the rectovaginal space and placement of a skin graft, held in place with a soft mold until the graft was established (McIndoe procedure). Subsequent regular intercourse or vaginal dilation must be maintained to avoid risk of fibrosis and loss of function. More recently, an operation that employs a transabdominal traction device has been described that can create a functional vagina within 7 to 9 days. Where available and applicable, the technique would appear to offer significant advantages over the traditional vaginoplasty procedure. In the rare woman with vaginal/cervical agenesis with a well-formed uterine body, it may be technically possible to create a neovagina continuous with the uterus with the goal of preserving fertility, but the associated long-term infectious morbidity has led most to conclude that the uterus should be removed in most circumstances.
Reassurance and support are important elements of the management of vaginal/müllerian agenesis. Affected women should be counseled that although they are infertile, normal sexual function can be expected and that genetic offspring can be achieved by in vitro fertilization using oocytes retrieved from the normal ovaries and the sperm of the partner, with subsequent transfer of embryos to a gestational surrogate. A growing experience with these techniques has revealed no evidence to indicate that congenital absence of the vagina and uterus is heritable in a dominant fashion.
AIS presents both similar and different treatment challenges. A functional vagina again can be created by any of the techniques described previously for women with vaginal/müllerian agenesis, but surgical treatment much less often is required as the vagina generally is short but otherwise normal rather than altogether absent. Genetic offspring, of course, are not possible because women with AIS do not have oocytes. They have testes, most commonly located in the inguinal canals at the level of the external inguinal ring but sometimes found within the abdomen. Like women with gonadal dysgenesis whose karyotype contains a Y chromosome (46,XY, 45,X/46,XY), individuals with AIS are at increased risk for development of tumors in their undescended testes. Therefore, the testes should be removed. However, the risk of neoplasia is lower (approximately 5%–10% vs. 25%), tumors are rarely encountered before puberty, and the secondary sexual development that accompanies puberty in women with AIS from the aromatization of testosterone is more natural than can be accomplished pharmacologically. Consequently, removal of the testes and hormone replacement are best delayed until pubertal maturation is complete, generally by age 16 to 18 years.
Initial treatment of women with an imperforate hymen, transverse vaginal septum, or cervical atresia centers on relief of symptoms related to accumulated menstrual fluid and debris. Surgical correction of imperforate hymen is straightforward, requiring only a cruciate incision in the hymen to the base of the hymeneal ring and excision of its central portion to allow drainage of sequestered menstrual fluid and subsequent normal menstruation. Surgical management of a transverse vaginal septum can be very challenging. In its simplest form, the procedure involves excision of the septum or dissection through the atretic segment to connect the margins of the lower and upper vaginal canals with split-thickness skin grafting occasionally required if the distance between the two precludes a primary junction without tension. In rare women with cervical atresia, as in those with vaginal agenesis who have a functional uterine body, the generally poor outcome and high postoperative morbidity associated with heroic efforts to preserve the uterus and fertility suggest that removal of the müllerian remnants is the safest management strategy.
Uterine sounding and gentle cervical dilation often are all that is required to correct a symptomatic cervical obstruction/stenosis. Operative hysteroscopy is the preferred method for treatment of intrauterine synechiae that may be lysed by blunt dissection, scissors, electrodissection, or with a laser; any of these methods achieves results superior to blind curettage. When adhesions are severe or extensive, an intrauterine balloon catheter (left in place for approximately 7–10 days) may be useful to maintain separation between the walls of the uterine cavity. Treatment with a broad-spectrum antibiotic and an inhibitor of prostaglandin synthesis (nonsteroidal antiinflammatory drugs) helps to minimize risk of infection and uterine cramping. High-dose exogenous estrogen treatment (2.5 mg conjugated equine estrogens/day or its equivalent for approximately 4 weeks) also generally is recommended to encourage rapid endometrial reepithelialization and proliferation. Repeated procedures may be required to restore a normal uterine cavity. Approximately 50% to 75% of affected women may be expected to achieve a successful pregnancy, although risks of preterm labor, placenta accreta, placenta previa, and postpartum hemorrhage are increased.
Ovarian Disorders
Women with gonadal dysgenesis and sexual infantilism should be offered growth and sex hormone replacement therapy to promote growth to maximum potential, normal bone density development, and secondary sexual maturation. Treatment with exogenous recombinant human GH (50g/kg/d) stimulates an acceleration of growth that generally can be sustained for 6 years or more and results in an adult height as much as 10 cm over initial predicted height. In those with short stature, sex hormone replacement should be postponed until the bone age reaches 12 or greater to avoid premature epiphyseal closure and allow a longer interval of time for long bone growth. Sex hormone replacement therapy should follow the normal sequence of sex hormone production observed during adolescence and begin with low doses of estrogen alone (0.3 mg conjugated equine estrogens or 0.5 mg micronized estradiol). Dosage should be increased gradually and after approximately 6 to 12 months or first evidence of vaginal bleeding, cyclic treatment with a progestin should be added. To achieve maximum bone and breast development, higher doses (1.25–2.5 mg conjugated equine estrogens or 2–3 mg micronized estradiol) often are required. Once secondary sexual development is completed, longer-term sex hormone replacement should continue to avoid the otherwise inevitable emergence of estrogen deficiency symptoms and bone mineral depletion. Alternatively, a low-dose OCP may be offered, for convenience.
Pregnancy may be achieved in women with gonadal dysgenesis through in vitro fertilization using oocytes provided by a genetically normal oocyte donor. However, pregnancy carries a unique risk for women with 45,X gonadal dysgenesis (Turner syndrome) as numerous cases of aortic dissection, aneurysm, and spontaneous rupture during pregnancy now have been reported. Preconceptional echocardiography may identify patients at risk for whom pregnancy is contraindicated, but catastrophic events may occur despite the absence of abnormal findings. Should they conceive spontaneously or by using donor oocytes, echocardiography should be performed at least once in each trimester, and the practitioner should remain alert to any evidence of progressive dilation of the ascending aorta. Women with 45,XY or 45,X/46,XY gonadal dysgenesis should have their gonads removed soon after diagnosis to avoid the substantial risk of gonadal tumors. Unlike those with AIS whose functional testes produce sex steroids that will promote normal secondary sexual maturation, women with gonadal dysgenesis have nonfunctional streak gonads. Consequently, there is no advantage to delay their removal.
The primary goals of treatment for women with POF that develops after secondary sexual maturation is complete generally are relief from symptoms of estrogen deficiency and prevention of premature bone mineral depletion. Standard regimens of cyclic or combined continuous estrogen/progestin hormone replacement therapy or a low-dose OCP will meet the need, and either is appropriate. In vitro fertilization with use of donor oocytes offers the possibility of pregnancy and the prognosis for success in most such women is excellent. Women in whom POF is diagnosed before age 30 and in whom karyotype reveals a mosaic 45,X cell line must be carefully evaluated and counseled on the potential risks of aortic dissection and rupture during pregnancy, as described earlier. Monitoring or treatment should be offered for any associated autoimmune disorders.
Pituitary Disorders
The overwhelming majority of pituitary tumors are prolactin-secreting adenomas or nonfunctioning tumors. TSH-, ACTH-, or GH-secreting adenomas generally are rare. A pituitary tumor often will be discovered when imaging is performed in women with hyperprolactinemia and occasionally in euprolactinemic women with hypogonadotropic hypogonadism that cannot be confidently attributed to weight loss, eating disorders, or regular strenuous exercise (Fig. 34.3). It may be difficult to differentiate a prolactin-secreting adenoma from a nonfunctioning tumor that disrupts normal hypothalamic dopamine delivery, except perhaps when prolactin levels are markedly elevated. However, because treatment options in hyperprolactinemic amenorrheic women with small intrasellar tumors are the same in either case, a specific diagnosis is not critical. The nature of a tumor may become clear only after treatment with a dopamine agonist restores normal prolactin levels; the majority of prolactin-secreting adenomas shrink in size during treatment and those that do not are more likely to be nonfunctioning tumors. In euprolactinemic amenorrheic women, a pituitary tumor also may be only an incidental finding, having no clinical significance.
|
|
|
FIG. 34.3. A pituitary tumor. Magnetic resonance imaging of a prolactin-secreting pituitary tumor (three large arrows). Note the lateral location and elevation of the diaphragm sella. However, the tumor does not compress the optic chiasm (single arrow) and thus no visual disturbances would be anticipated. |
Whereas transsphenoidal surgery was commonly performed in the past to remove pituitary tumors, it is now generally reserved for women with rare TSH-, ACTH-, or GH-secreting adenomas, those with large nonfunctioning tumors associated with complaints of headache or visual disturbances, and those in whom medical treatment (described subsequently) fails or is poorly tolerated. Surgery achieves immediate reduction of prolactin levels and restores cyclic menses in approximately 30% of women with prolactin-secreting macroadenomas and in up to 70% with microadenomas. However, residual or recurrent tumors and hyperprolactinemia are common and surgery may be complicated by cerebrospinal fluid leakage, meningitis, diabetes insipidus, or other trophic pituitary hormone deficiencies that require further treatment. Postoperative monitoring involves periodic serum prolactin determinations and repeated imaging. Radiation therapy for pituitary tumors is even less attractive because the response to radiation is slow and such treatment risks development of panhypopituitarism over time. As a result, radiation generally is reserved for postoperative recurrence of large tumors.
Dopamine-agonist therapy is the mainstay of treatment for hyperprolactinemia, independent of whether a pituitary tumor is identified. Bromocriptine is highly efficacious but side effects (nausea, headache, orthostatic hypotension, dizziness, nasal congestion) are common. The dose should be titrated incrementally to normalize serum prolactin levels, beginning with a small dose (1.25–2.5 mg) at bedtime, with a second dose taken with breakfast or lunch when required; most patients will require 5.0 mg per day or less.
Cabergoline may be an effective alternative when bromocriptine treatment cannot be tolerated, although its lower frequency of side effects and less frequent dosing (0.25–3.0 mg every 3–7 days) also make it an attractive initial therapeutic choice. Vaginal administration of bromocriptine or cabergoline is effective and may help to reduce side effects when oral treatment is poorly tolerated. Dopamine-agonist treatment restores menses and ovulatory function in up to 80% of amenorrheic hyperprolactinemic women, generally within approximately 6 to 8 weeks; reduction or cessation of galactorrhea typically requires somewhat longer-term treatment.
Medical treatment with a dopamine agonist (bromocriptine, cabergoline) promotes shrinkage of pituitary macroadenomas in the majority of cases, usually within 6 to 12 weeks. Large tumors that fail to shrink despite effective suppression of excess prolactin secretion generally are nonfunctioning adenomas that cause hyperprolactinemia by interfering with delivery of dopamine from the hypothalamus and require surgery.
Regardless of the treatment they receive, women with macroadenomas should be monitored with serum prolactin determinations and imaging every 6 to 12 months for at least 2 years, less often thereafter if tumor size and prolactin levels remain stable, and sooner if prolactin levels rise significantly or symptoms of headache or visual disturbances emerge or recur. Similar monitoring with serial prolactin determinations (every 6 to 12 months) and periodic imaging (annually for 2 years) is recommended for those with microadenomas. Dopamine-agonist therapy is the best initial choice for persistent or recurrent tumor or hyperprolactinemia. For the few patients who may require radiation therapy, ongoing surveillance must be implemented to detect any evidence of developing panhypopituitarism.
Pituitary tumors generally grow slowly or not at all, even during pregnancy. No more than approximately 5% of women with pituitary microadenomas will experience tumor growth during pregnancy, even fewer develop signs or symptoms as a result, and only a rare patient will require surgical intervention. The risk is greater but generally still modest for those with macroadenomas (approximately 15%). Routine serial visual field examinations and serum prolactin determinations therefore are unnecessary, but the patient and her physician must be alert to symptoms that may emerge in any trimester. Headaches generally precede visual disturbances and if either appears, monitoring with serum prolactin measurements, visual field examinations, and imaging should begin. With rare exceptions, dopamine-agonist therapy will arrest and reverse tumor expansion and eliminate associated symptoms and poses no significant risk to mother or fetus.
Dopamine-agonist therapy clearly is the treatment of choice for anovulatory hyperprolactinemic women who desire pregnancy or have significant breast tenderness or troublesome galactorrhea, with or without a pituitary adenoma. In the absence of other coexisting causes of infertility, approximately 80% of anovulatory hyperprolactinemic women treated with dopamine agonists may be expected to conceive. In the few who cannot tolerate medical treatment with bromocriptine or cabergoline but desire pregnancy, ovulation induction with exogenous gonadotropins is an effective alternative as prolactin does not alter the effect of the gonadotropins on the ovarian follicles. Women not seeking pregnancy may be offered hormone replacement therapy or even a low-dose OCP that will restore menses and effectively eliminate or prevent the consequences of any associated estrogen deficiency.
The uncommonly encountered empty sella syndrome generally is entirely benign and does not progress to pituitary failure. Treatment and subsequent surveillance are the same as in women with a pituitary adenoma. In the rare patient whose amenorrhea results from Sheehan syndrome and its related pituitary insufficiency, more than simple sex steroid replacement therapy may be required. Depending on the extent and scope of pituitary damage, such women also may require glucocorticoid and thyroid hormone replacement. For those affected women who desire a subsequent pregnancy, ovulation induction may be achieved, but only with exogenous gonadotropins.
Hypothalamic Disorders
Chronic anovulation resulting from abnormal patterns of pulsatile hypothalamic GnRH secretion is the cause of amenorrhea most commonly encountered in practice. A hypothalamic etiology can be established only after first excluding other peripheral, ovarian, and pituitary disorders (normal serum TSH and prolactin levels, low or normal serum FSH, negative imaging, where indicated). Most such women are estrogenized and can also be obese or androgenized (PCOS). Some have a more profound hypogonadism with grossly low levels of estrogen production (hypothalamic amenorrhea), often in association with history of emotional, nutritional, or physical stress.
Treatment of women with PCOS is based on whether pregnancy is an immediate goal. Those who desire pregnancy are candidates for ovulation induction. For those not seeking pregnancy, the goals of treatment are to establish regular menses and prevent the consequences of chronic unopposed estrogen stimulation on the endometrium (dysfunctional uterine bleeding, endometrial hyperplasia, endometrial adenocarcinoma), to prevent the emergence or progression of hirsutism, and to reduce the longer-term risks of diabetes and cardiovascular disease associated with the disorder.
The wide range of therapeutic options for ovulation induction in anovulatory women with PCOS who desire pregnancy is only summarized here. When obese, weight loss should always be encouraged and may be all that is required to restore menstrual cyclicity. Even if weight loss does not restore ovulatory function, it may be expected to improve the response to ovulation-inducing agents. The substantial costs, risks, and logistic demands of more aggressive forms of treatment might thus be avoided, but weight loss should never be a prerequisite for treatment.
Clomiphene citrate is the initial treatment of choice as it is safe, inexpensive, and has few serious side effects. Clomiphene is a competitive estrogen receptor antagonist that acts centrally to deplete hypothalamic estrogen receptors, thereby interfering with estrogen-negative feedback and resulting in a compensatory alteration in pulsatile GnRH secretion that stimulates increased pituitary gonadotropin release and, in turn, drives ovarian follicular activity. Treatment should begin with a low dose (50 mg/d cycle days 3–7 or 5–9 after a spontaneous or progestin-induced menses) and subsequently increase in increments (100 mg/d, 150 mg/d) as needed to achieve ovulation. Most women with PCOS will respond to clomiphene, but many prove resistant and ultimately require alternative treatment. Among clomiphene-resistant anovulatory women with PCOS, a significant majority have insulin resistance. Metformin (1,000–2,000 mg/d in divided doses) is an insulin-sensitizing agent that can restore spontaneous menses and cyclic ovulation in many amenorrheic women with PCOS. Consequently, metformin can be used as the first treatment option for ovulation induction, adding clomiphene in those who fail to respond or, as is more commonly done, metformin can be reserved for those who first prove resistant to clomiphene. In either case, many who fail to ovulate in response to either alone will respond when the two are used in combination. Although the safety of metformin treatment in pregnancy has not been established, preliminary evidence suggests that it may reduce the incidence of spontaneous abortion and gestational diabetes in the subset of women with PCOS and insulin resistance.
Those who prove resistant to clomiphene and metformin are candidates for treatment with exogenous gonadotropins. Because such women often are very sensitive to low doses of gonadotropins, unifollicular ovulation can be difficult to achieve and the risk for both multiple pregnancy and ovarian hyperstimulation syndrome is increased. Consequently, treatment must be carefully monitored and is best provided only by clinicians having the necessary training or experience.
Ovarian drilling is a contemporary version of the classic ovarian wedge resection and another treatment option for ovulation induction in clomiphene-resistant, hyperandrogenic, anovulatory women with PCOS (Fig. 34.4). The technique involves laparoscopic cautery, diathermy, or laser vaporization of the ovaries at multiple sites, the objective being to decrease circulating and intraovarian androgen levels by reducing the volume of ovarian stroma. However, the effects of treatment are temporary and postoperative adhesions that may compromise fertility are a distinct possibility. Drilling also has the potential to destroy oocytes and shorten the reproductive lifespan and therefore is best limited to those for whom all other alternatives have been exhausted.
|
|
|
FIG. 34.4. The ovaries in polycystic ovary syndrome shown in a woman with polycystic ovary syndrome at laparotomy. Note the large size of the ovaries, relative to the uterus, with a smooth ovarian capsule without evidence of ovulatory events. |
OCPs have long been the mainstay of treatment for amenorrheic women with PCOS who do not desire pregnancy because the combined actions of OCP offer a number of important benefits. Programmed cyclic OCP treatment restores regular, predictable menses, protects the endometrium from the adverse effects of otherwise unopposed estrogen stimulation, and provides effective contraception for the occasional spontaneous ovulation. OCPs decrease circulating androgens by suppressing typically increased serum LH levels that stimulate excess ovarian androgen production and further reduce levels of circulating free androgen by stimulating hepatic production of sex hormone binding globulin. Consequently, within 3 to 6 months after initiation of treatment hair typically grows more slowly, becomes both finer and lighter in color, and therefore is less noticeable and more easily managed.
For those with more severe hirsutism in whom treatment with OCP alone fails to achieve the desired effect, spironolactone can be added (100–200 mg/d in divided doses) to provide more effective control. In the dose range indicated, spironolactone acts as a competitive androgen receptor antagonist and blocks androgen action on the pilosebaceous unit (a hair follicle, sebaceous glands, and arrector pili muscle). The multiple actions of combined treatment with OCP and spironolactone (suppression of excess androgen production, increase in androgen binding, and blockade of androgen action) are complementary and generally quite effective for the management of hirsutism. In extreme cases, a more profound suppression of excess androgen production can be achieved by treatment with a GnRH agonist (e.g., leuprolide acetate) with superimposed cyclic or combined continuous hormone replacement with standard doses of estrogen and progestin to prevent the otherwise inevitable emergence of estrogen deficiency symptoms. Suppression of adrenal androgen production by chronic glucocorticoid treatment risks suppression of the adrenal response to stress, immunosuppression, and progressive bone mineral depletion, should therefore be reserved only for those with documented nonclassic CAH, and must be carefully monitored when offered.
Insulin resistance certainly contributes to the pathophysiology of PCOS by local amplification of ovarian androgen production in response to LH (directly or via IGF-1) and by reducing hepatic synthesis of sex hormone binding globulin (directly). Moreover, hyperandrogenism clearly is associated with an atherogenic lipid profile. There also is evidence that in individuals with glucose intolerance, metformin treatment can reduce the risk of progression to frank diabetes and, presumably, the increased risk of cardiovascular disease associated with that disease. However, these longer-term potential benefits of primary metformin therapy remain unproven. Moreover, the drug is relatively costly and often poorly tolerated (nausea, vomiting, diarrhea). Under such circumstances, whether primary metformin treatment offers sufficient benefits to justify its use as an alternative or complement to other already proven treatment strategies (weight loss, exercise, OCP, statin therapy) remains controversial. Based on the evidence available to date, primary metformin treatment cannot be recommended for all women with PCOS, but may merit serious consideration in selected individuals.
In the absence of obesity, hirsutism, or other evidence of hyperandrogenism that is characteristic of PCOS, the most likely cause of amenorrhea in women with a normal or low-serum FSH level is a functional disorder of the hypothalamus or higher centers. As for women with PCOS, treatment for women with hypothalamic amenorrhea depends on whether pregnancy is an immediate goal. Those who desire pregnancy clearly are candidates for ovulation induction. For those not seeking pregnancy, the goals of treatment are to establish regular menses, if desired, and to prevent the consequences of chronic estrogen deficiency (genitourinary atrophy, progressive bone demineralization) as most, but not all women with hypothalamic amenorrhea exhibit frankly low levels of endogenous estrogen production.
The range of therapeutic options for ovulation induction for women with hypothalamic amenorrhea (Fig. 34.5) is somewhat narrower than for those with PCOS and again is only summarized here. In women with a low BMI, weight gain should be encouraged but weight gain, like weight loss, can be very difficult to achieve. Women with a low BMI should be educated on the relationship between weight and menstrual function and advised that weight gain may be all that is needed to restore menstrual cyclicity and ovulation. At the least, weight gain may be expected to increase the likelihood of response to more conservative ovulation induction strategies (clomiphene citrate) involving less cost and risk than otherwise may be required. Those in whom symptoms (fasting, purging) or physical findings suggest a potentially serious underlying eating disorder (bulimia, anorexia) should be offered psychiatric evaluation and care before attempting pregnancy as such disorders pose unique risks to both mother and fetus.
|
|
|
FIG. 34.5. The ovary in hypothalamic amenorrhea. Laparoscopy of the pelvis demonstrates a normal-appearing uterus with the left ovary having a smooth capsule without evidence of follicular activity. This is consistent with the lack of gonadotropin stimulation with a normal but unstimulated ovary. |
Similarly, women whose hypothalamic amenorrhea is associated with a high level of regular strenuous exercise should be educated on the pathophysiology involved. Reduced physical activity may restore menstrual cyclicity and ovulation and, at the least, less exercise may improve the probability that conservative methods of ovulation induction will succeed. However, once again, reduced levels of exercise should not be a prerequisite for treatment. For many, high levels of exercise are intrinsic to lifestyle or represent the preferred method for relieving stress and may not be desirable or even feasible.
Ovulation induction should be offered to those who desire pregnancy. Because of its relatively low cost and risk, clomiphene citrate is again the initial drug of choice. However, unlike women with PCOS, those with hypothalamic amenorrhea often fail to respond to clomiphene treatment. Clomiphene response requires an intact and functional HPO axis, something that most women with hypothalamic amenorrhea do not have (as reflected by low or normal serum FSH levels despite frankly low estrogen production). Consequently, one of two alternative approaches to ovulation induction frequently is required. Exogenous synthetic GnRH, administered in a pulsatile fashion via a portable and programmable infusion pump can effectively restore normal levels of pituitary gonadotropin secretion and spontaneous ovulatory function. GnRH pump therapy requires relatively little monitoring once the effective dose has been established, allows ovulation to occur spontaneously, and is associated with only modest risk of a multiple pregnancy. However, many women object to having an indwelling intravenous catheter over an extended interval of time (approximately 3 weeks) and reject this option.
Most often, treatment with exogenous gonadotropins that directly stimulate ovarian follicular development is the preferred method for ovulation induction. Exogenous human chorionic gonadotropin is administered to trigger ovulation when follicular maturation is complete. Treatment requires careful monitoring with serial determination of serum estrogen concentrations and transvaginal ultrasound examinations to determine the size, number, and maturity of developing follicles and to minimize the risks of excessive follicular development, the ovarian hyperstimulation syndrome, and multifetal gestation.
Hormone replacement therapy is indicated for those women with hypothalamic amenorrhea who do not desire pregnancy and in whom weight gain or decreased exercise fails to restore menstrual function. In those who are simply anovulatory and not frankly hypoestrogenic or at risk for unwanted pregnancy, cyclic progestational therapy may be all that is required to restore predictable menses and to prevent the endometrial consequences of chronic unopposed estrogen stimulation. However, the level of estrogen production in such women often fluctuates unpredictably and is difficult to judge confidently on clinical or laboratory criteria. Consequently, treatment with cyclic or combined continuous hormone replacement or OCP is prudent to prevent any prolonged interval of hypoestrogenism and the depletion of bone mineral that may result. Given that spontaneous ovulation may occasionally occur, sexually active women with hypothalamic amenorrhea who do not desire pregnancy are best managed with OCP.
SUMMARY POINTS
· Amenorrhea is a common symptom resulting from an abnormality in the reproductive system that may be anatomic (developmental or acquired), organic, or endocrinologic in nature.
· Congenital anomalies of the müllerian duct system typically present with normal puberty and may be associated with abdominal pain if there is functional endometrium present.
· Gonadal dysgenesis represents the lack of any ovarian function prior to puberty leading to sexual infantilism.
· Premature ovarian failure presents as amenorrhea associated with an elevated FSH level and, if occurring prior to age 30, may be associated with the presence of a Y chromosome.
· Pituitary tumors are very common, with the majority secreting prolactin and associated with hypoestrogenic amenorrhea and galactorrhea.
· Hypothalamic amenorrhea is the general term used to describe hypoestrogenic amenorrhea when no anatomic abnormality is identified. It is commonly associated with weight loss or athletic stress.
· Polycystic ovarian syndrome is a very common clinical problem which may present with amenorrhea but is associated with excess androgen production. A significant majority of these women have insulin resistance.
· The most common cause of amenorrhea in women of reproductive age is pregnancy.
SUGGESTED READINGS
Differential Diagnosis of Amenorrhea
1. Koutras DA. Disturbances of menstruation in thyroid disease. Ann NY Acad Sci 1997;816:280–284.
2. Laufer MR, Floor AE, Parsons KE, et al. Hormone testing in women with adult-onset amenorrhea. Gynecol Obstet Invest 1995;40:200–203.
3. Reindollar RH, Byrd JR, McDonough PG. Delayed sexual development: a study of 252 patients. Am J Obstet Gynecol 1981;140:371–380.
4. Reindollar RH, Novak M, Tho SPT, et al. Adult-onset amenorrhea: a study of 262 patients. Am J Obstet Gynecol 1986;155:531–543.
5. Styne DM. New aspects in the diagnosis and treatment of pubertal disorders. Pediatr Clin North Am 1997;44:505–529.
6. Warren MP. Evaluation of secondary amenorrhea. J Clin Endocrinol Metab 1996;81:437–442.
Genital Tract Problems
7. American College of Obstetricians and Gynecologists Committee on Adolescent Health Care. Nonsurgical diagnosis and management of vaginal agenesis. ACOG Committee Opinion No. 274, July 2002. Obstet Gynecol 2002;100:213–216.
8. Blask AR, Sanders RC, Rock JA. Obstructed uterovaginal anomalies: demonstration with sonography. Part II. Teenagers. Radiology 1991;179:84–88.
9. Burgis J. Obstructive müllerian anomalies: case report, diagnosis, and management. Am J Obstet Gynecol 2001;185:338–344.
10. Fujimoto VY, Miller JH, Klein NA, et al. Congenital cervical atresia: report of seven cases and review of the literature. Am J Obstet Gynecol1997;177:1419–1425.
11. Kim HH, Laufer MR. Developmental abnormalities of the female reproductive tract. Curr Opin Obstet Gynecol 1994;6:518–525.
12. March CM. Intrauterine adhesions. Obstet Gynecol Clin North Am 1995;22:491–505.
13. Quigley CA, De Bellis A, Marschke KB, et al. Androgen receptor defects: historical, clinical, and molecular perspectives. Endocr Rev 1995;16:271–321.
14. Reinhold C, Hricak H, Forstner R, et al. Primary amenorrhea: evaluation with MR imaging. Radiology 1997;203:383–390.
15. Williams DM, Patterson MN, Hughes IA. Androgen insensitivity syndrome. Arch Dis Child 1993;68:343–344.
Gonadal Dysgenesis
16. Guidozzi F, Ball J, Spurdle A. 46,XY pure gonadal dysgenesis (Swyer-James syndrome)—Y or Y not?: a review. Obstet Gynecol Surv 1994;49:138–146.
17. Ranke MB, Saenger P. Turner's syndrome. Lancet 2001;358:309–314.
18. Rosenfield RL. Diagnosis and management of delayed puberty. J Clin Endocrinol Metab 1990;70:559–562.
19. Simpson JL, Rajkovic A. Ovarian differentiation and gonadal failure. Am J Med Genet 1999;89:186–200.
20. Verp M, Simpson JL. Abnormal sexual differentiation and neoplasia. Cancer Genet Cytogenet 1987;25:191–218.
Premature Ovarian Failure
21. Aittomaki K, Lucena JL, Pakarinen P, et al. Mutation in the follicle-stimulating hormone receptor gene causes hereditary hypergonadotropic ovarian failure. Cell 1995;82:959–968.
22. Byrne J, Mulvihill JJ, Myers MH, et al. Effects of treatment on fertility in long-term survivors of childhood cancer. N Engl J Med 1987;317:1315–1321.
23. Clark ST, Radford JA, Crowther D, et al. Gonadal function following chemotherapy for Hodgkin's disease: a comparative study of MVPP and a seven-drug hybrid regimen. J Clin Oncol 1995;13:134–139.
24. Conway GS, Kaltsas G, Patel A, et al. Characterization of idiopathic premature ovarian failure. Fertil Steril 1996;65:337–341.
25. Gradishar WJ, Schilsky RL. Ovarian function following radiation and chemotherapy. Semin Oncol 1989;16:425–436.
26. Guerrero NV, Singh RH, Manatunga A, et al. Risk factors for premature ovarian failure in females with galactosemia. J Pediatr 2000;137:833–841.
27. Hoek A, Schoemaker J, Drexhage HA. Premature ovarian failure and ovarian autoimmunity. Endocr Rev 1997;18:107–134.
28. Howell S, Shalet S. Gonadal damage from chemotherapy and radiotherapy. Endocrinol Metab Clin North Am 1998;27:927–943.
29. Laml T, Schulz-Lobmeyr I, Obruca A, et al. Premature ovarian failure: etiology and prospects. Gynecol Endocrinol 2000;14:292–302.
30. McDonough PG. The search for the gonadotropin-resistant ovary—a lost planet or a dinosaur? Fertil Steril 1999;71:1169–1170.
31. Morishima A, Grumbach MM, Simpson ER, et al. Aromatase deficiency in male and female siblings caused by a novel mutation and the physiological role of estrogens. J Clin Endocrinol Metab 1995;80:3689–3698.
32. Saenger P, Wikland KA, Conway GS, et al. Fifth International Symposium on Turner Syndrome. Recommendations for the diagnosis and management of Turner syndrome. J Clin Endocrinol Metab 2001;86:3061–3069.
33. van Kasteren YM, Schoemaker J. Premature ovarian failure: a systematic review on therapeutic interventions to restore ovarian function and achieve pregnancy. Hum Reprod Update 1999;5:483–492.
Pituitary Disorders (Hyperprolactinemia)
34. Colao A, Loche S, Cappa M, et al. Prolactinomas in children and adolescents. Clinical presentation and long-term follow-up. J Clin Endocrinol Metab1998;83:2777–2780.
35. Conner P. Fried G. Hyperprolactinemia; etiology, diagnosis and treatment alternatives. Acta Obstet Gynecol Scand 1998;77:249–262.
36. Hodgson SF, Randall RV, Holman CB, et al. Empty sella syndrome. Report of 10 cases. Med Clin North Am 1972;56:897–907.
37. Maguire GA. Prolactin elevation with antipsychotic medications: mechanisms of action and clinical consequences. J Clin Psychiatry 2002;63:56–62.
38. Molitch ME. Disorders of prolactin secretion. Endocr Metab Clin North Am 2001;30:585–610.
39. Tan SL. Hyperprolactinemia and the management of related tumors. Curr Opin Obstet Gynecol 1990;2:378–385.
40. Touraine P, Plu-Bureau G, Beji C, et al. Long-term follow-up of 246 hyperprolactinemic patients. Acta Obstet Gynecol Scand 2001;80:162–168.
Hypothalamic Amenorrhea
41. Berga SL. Behaviorally induced reproductive compromise in women and men. Semin Reprod Endocrinol 1997;15:47–53.
42. Hall JE. Physiologic and genetic insights into the pathophysiology and management of hypogonadotropic hypogonadism. Ann Endocrinol 1999;60:93–101.
43. Hurst BS. Ovulation induction in the nonestrogenized patient. Semin Reprod Endocrinol 1996;14:299–308.
44. Laughlin GA, Dominguez CE, Yen SSC. Nutritional and endocrine-metabolic aberrations in women with functional hypothalamic amenorrhea. J Clin Endocrinol Metab 1998;83:25–32.
45. Marshall JC, Eagleson CA, McCartney CR. Hypothalamic dysfunction. Mol Cell Endocrinol 2001;183:29–32.
46. Miller KK, Klibanski A. Amenorrheic bone loss. J Clin Endocrinol Metab 1999;84:1775–1783.
47. Warren MP. Stiehl AL. Exercise and female adolescents: effects on the reproductive and skeletal systems. J Am Med Wom Assoc 1999;54:115–120.
Polycystic Ovarian Syndrome
48. Azziz R, Dewailly D, Owerbach D. Non-classic adrenal hyperplasia: current concepts. J Clin Endocrinol Metab 1994;78:810–815.
49. Bili H, Laven J, Imani B, et al. Age-related differences in features associated with polycystic ovary syndrome in normogonadotropic oligo-amenorrhoeic infertile women of reproductive years. Eur J Endocrinol 2001;145:749–755.
50. Dunaif A, Thomas A. Current concepts in the polycystic ovary syndrome. Annu Rev Med 2001;52:401–419.
51. Glueck CJ, Phillips H, Cameron D, et al. Continuing metformin throughout pregnancy in women with polycystic ovary syndrome appears to safely reduce first-trimester spontaneous abortion: a pilot study. Fertil Steril 2001;75:46–52.
52. Glueck CJ, Wang P, Kobayashi S, et al. Metformin therapy throughout pregnancy reduces the development of gestational diabetes in women with polycystic ovary syndrome. Fertil Steril 2002;77:520–525.
53. Hughes E, Collins J, Vandekerckhove P. Clomiphene citrate for ovulation induction in women with oligo-amenorrhoea. In: Cochrane Database of Systemic Reviews. The Cochrane Library 2000; Issue 2 (CD000056). Oxford: Update Software.
54. Hughes E, Collins J, Vandekerckhove P. Gonadotrophin-releasing hormone analogue as an adjunct to gonadotropin therapy for clomiphene-resistant polycystic ovarian syndrome. In: Cochrane Database of Systemic Reviews. The Cochrane Library 2000; Issue 2 (CD000097). Oxford: Update Software.
55. Hughes E, Collins J, Vandekerckhove P. Ovulation induction with urinary follicle stimulating hormone versus human menopausal gonadotropin for clomiphene-resistant polycystic ovary syndrome. Cochrane Database of Systemic Reviews. The Cochrane Library 2000; Issue 2 (CD000087). Oxford: Update Software.
56. Legro RS. Polycystic ovary syndrome: the new millennium. Mol Cell Endocrinol 2002;186:219–225.
57. Nestler JE, Jakubowicz DJ, Evans WS, et al. Effects of metformin on spontaneous and clomiphene-induced ovulation in the polycystic ovary syndrome. N Engl J Med 1998;338:1876–1880.
