CHAPTER 13 PROLACTIN AND ITS DISORDERS

CHAPTER 13 PROLACTIN AND ITS DISORDERS
Principles and Practice of Endocrinology and Metabolism

CHAPTER 13 PROLACTIN AND ITS DISORDERS

LAURENCE KATZNELSON AND ANNE KLIBANSKI

Normal Control of Prolactin Secretion
Clinical Aspects of Prolactin Physiology

Diurnal and Menstrual Cycle Variation

Eating

Stress

Age

Normal States
Clinical Manifestations of Hyperprolactinemia
Differential Diagnosis and Clinical Approach to Hyperprolactinemia
Treatment for Prolactinomas

Medical Therapy

Surgery

Radiation Therapy
Management of Microprolactinomas
Management of Macroprolactinomas
Pregnancy and Prolactinomas
Chapter References

NORMAL CONTROL OF PROLACTIN SECRETION
Prolactin secretion is controlled by dual inhibitory and stimula-tory factors (Fig. 13-1). This hormone is unique among anterior pituitary hormones, because it is primarily regulated through tonic inhibition. Two decades of investigation have demonstrated the presence of one or more prolactin-inhibiting factors (PIFs).1

FIGURE 13-1. Regulation of prolactin secretion. Prolactin release is under tonic inhibition by prolactin inhibiting factors, predominantly dopamine. Prolactin release is stimulated by a number of factors, including vasoactive intestinal peptide (VIP), thyroid-releasing hormone (TRH), and gonadotropin-releasing hormone (GnRH). Estrogens, pregnancy, and breast suckling stimulate prolactin release. Within the hypothalamus, serotinergic and dopaminergic pathways are stimulatory and inhibitory, respectively, to prolactin release. (GABA, g-aminobutyric acid.) (Modified from Molitch ME. Pathologic hyperprolactinemia. Endocrinol Metab Clin North Am 1992; 21:877.)

Dopamine is the most important PIF described. Multiple studies support the hypothesis that dopamine acts as an inhibiting factor. In vitro studies reveal that high-affinity dopamine receptors (D2) are present on lactotrope membranes, and after binding occurs, inhibition of adenylate cyclase is demonstrated.2,3 This results in a decrease in cyclic adenosine mono-phosphate (cAMP) production and the release of prolactin. Dopamine also directly inhibits prolactin biosynthesis at the level of RNA transcription. Dopamine is produced in higher nuclei in the brain and is secreted into the portal circulation to reach the pituitary. Infusion of dopamine in humans, resulting in serum dopamine concentrations similar to those found in portal blood, causes a reduction in prolactin secretion.4 Dopa-mine receptor blockade results in prolactin elevation.5 After dopamine is removed, as during pituitary stalk section, prolactin is rapidly released. These studies all point to a direct inhibitory effect of dopamine on pituitary lactotrope secretion. Most pharmacologic agents that cause prolactin release act either by blockade of dopamine receptors (e.g., haloperidol, phenothia-zines) or by dopamine depletion in the tuberoinfundibular neurons (e.g., reserpine, a-methyldopa).
Another potential PIF includes a 56-amino-acid PIF identified within the precursor for gonadotropin-releasing hormone (GnRH). This GnRH-associated peptide (GAP) inhibits prolactin secretion and reduces prolactin secretion in rats, but the role of GAP in modulating prolactin secretion in humans remains unconfirmed.6 Data also support the role of g-aminobutyric acid (GABA) as a PIF. GABA is secreted into the portal circulation, with a resulting inhibitory effect on prolactin secretion, and GABA receptors have been detected on lactotropes.7 However, the physiologic importance of GABA in prolactin regulation is unclear.
Stimulatory factors also regulate prolactin secretion. These substances may act directly on the pituitary or may act indirectly by means of dopaminergic blockade or depletion at the level of the hypothalamus. Estrogens are important physiologic stimulators of prolactin release.8 In vitro studies show that estradiol increases prolactin biosynthesis, consistent with a direct stimulatory effect of estrogen on lactotrope function.9 Chronic exposure to estrogens increases lactotrope number and size (i.e., “pregnancy cells”), and acute administration increases prolactin secretion within hours.10 Estrogens may also indirectly increase prolactin levels by altering dopaminergic tone and by increasing responsiveness to other neuromodulators.
Thyrotropin-releasing hormone (TRH) stimulates the synthesis and release of prolactin in vivo and in vitro from normal and neoplastic lactotropes. Although pharmacologic doses of TRH result in a rapid release in prolactin after intravenous administration in humans, the physiologic role of TRH in modulating prolactin secretion is not established.11 For example, suckling leads to release of prolactin without an accompanying heightened release of TRH. Hypothyroidism results in an increase in TSH and the prolactin response to TRH, and elevations in basal prolactin levels may be seen in primary hypothyroidism. The suggestion has been made that decreased hypothalamic dopamine may play a role in the hyperprolactinemia associated with hypothyroidism.
Vasoactive intestinal peptide (VIP) may selectively stimulate or potentiate the TRH effect on prolactin release.12 Evidence exists for VIP receptors on lactotropes and VIP may stimulate prolactin release in vitro. Immunoneutralization of VIP effects through administration of anti-VIP antisera diminishes the prolactin response to stimuli, including suckling, a finding that again suggests a role of VIP as a stimulatory factor.13 Data suggest that VIP may be produced in the pituitary and may stimulate prolactin release through a paracrine or autocrine mechanism.14 The clinical significance of VIP in prolactin regulation in humans is unknown.
GnRH may also have stimulatory properties. The administration of GnRH induces the acute release of prolactin in normally cycling women and hypogonadal patients.15 Moreover, incubation of human lactotropes with GnRH in vitro results in prolactin secretion.16 These investigations suggest that GnRH, directly or through a paracrine mechanism involving gonadotropes, may be important in evoking prolactin release.
Investigations have suggested a role for galanin as a potent stimulator of prolactin release. Galanin is a 29-amino-acid peptide widely distributed in the central and peripheral nervous system. In the rat, intracerebroventricular injections of galanin may increase prolactin levels.17 However, intravenous administration of galanin in humans does not raise serum prolactin levels.18 The physiologic role of galanin in human prolactin regulation remains controversial.
Serotonin is another factor that may stimulate prolactin secretion.19 Administration of serotonin antagonists decreases prolactin levels. Serotonin agonists appear to enhance prolactin secretion through specific serotonin receptors, perhaps explaining why only specific serotonin antagonists are capable of lowering prolactin secretion.20
Other factors that may have stimulatory roles include bombesin, angiotensin II, histamine (H2) antagonists, and opiates.
Human prolactin structure has partial homology with growth hormone, perhaps accounting for the lactotropic activity of growth hormone. Heterogeneous forms of prolactin are found in the circulation. Eighty-five percent of prolactin (23 kDa) detected in the pituitary and secreted into serum is non-glycosylated, but glycosylated forms have been detected.21 Approximately 8% of prolactin extractable from the pituitary is dimeric, and an additional 1% to 5% is polymeric, linked by disulfide bonds.22 These forms include “big,” “big-big,” and “little” or “native” prolactins. The significance of these forms is unknown. The larger forms of prolactin may have decreased rates of binding to the prolactin receptor and possess diminished bioactivity relative to monomeric, nonglycosylated prolactin. These forms may represent nonspecific hormonal aggregates or binding of prolactin to serum proteins. Some patients have normal reproductive function but elevated serum prolactin values; the prolactin in these patients is composed of a relatively increased component of polymeric prolactin.23 In such patients, the elevated prolactin levels may reflect increased levels of polymeric prolactin with decreased bioactivity. The suggestion has been made that certain isotypes of prolactin, specifically iso-B prolactin, may be elevated in the sera of patients with infertility and pregnancy wastage.24 This isotype may be more resistant to bromocriptine therapy than native prolactin. However, these reports need to be confirmed. The significance of the remaining fraction of glycosylated prolactin is unknown.
CLINICAL ASPECTS OF PROLACTIN PHYSIOLOGY
The physiologic causes of hyperprolactinemia are summarized in Table 13-1. The following sections describe clinical aspects of prolactin physiology.

TABLE 13-1. Physiologic Causes of Hyperprolactinemia

DIURNAL AND MENSTRUAL CYCLE VARIATION
Prolactin is secreted in a pulsatile fashion with 4 to 14 pulses per day (60% occur during sleep).25 Prolactin secretory pulses begin 60 to 90 minutes after the onset of sleep.26 The amplitude of pulses varies greatly among individuals, with peak levels occurring during the late hours of sleep. Such rises are not clearly associated with any specific stage of sleep. Although some studies have suggested that prolactin varies during the menstrual cycle, the precise nature of this relationship remains unclear. Several investigators have shown that prolactin levels are significantly higher during the ovulatory and luteal phases, particularly at midcycle.27 This midcycle rise may be the result of increased circulating periovulatory estradiol levels. However, other studies have not confirmed this finding. Prolactin is probably not necessary for ovulation, because ovulatory periods may occur in women taking bromocriptine, a medication that suppresses prolactin.
EATING
Abrupt rises in serum prolactin levels occur within an hour of eating in normal and pregnant hyperprolactinemic individuals, but they do not rise in those with prolactinomas. Amino acids metabolized from protein components of meals appear to be the main stimulants to prolactin secretion.28
STRESS
Prolactin rises during stress, including physical exertion, surgery, sexual intercourse, insulin hypoglycemia, and seizures. The nature and teleologic significance of these changes is unknown. In women, nipple stimulation, chest wall trauma or surgery, and herpes zoster infection of the breast may result in increases in prolactin levels, in part through afferent neural pathways.29 In contrast, nipple stimulation in men does not cause increased prolactin levels.
AGE
Mean levels of prolactin are slightly higher in premenopausal women than in men, probably because of a direct effect of estrogen on pituitary prolactin secretion or estrogen-induced alterations in dopaminergic tone.30 Some studies suggest that prolactin levels progressively decline in women with age, particularly after the menopause.31 The responsiveness of prolactin to various pharmacologic agents (e.g., TRH) declines with age in women, probably because of postmenopausal estrogen deficiency.
NORMAL STATES
The only established role of prolactin is to initiate and maintain lactation. Prolactin levels increase progressively with pregnancy. Estrogens play a major role in stimulation of prolactin levels, which peak at term (at 100–300 ng/mL).32 Lactation begins when estradiol levels fall at parturition. During the first 4 to 6 weeks after delivery, prolactin levels increase in the circulation to 60 times higher than baseline levels within 20 to 30 minutes of nursing.33 This elevation is associated with enhanced prolactin pulse amplitude without alteration of pulse frequency.34 The nursing stimulus effectively promotes acute prolactin release through afferent spinal neural pathways. With continued nursing, the nipple stimulation itself elicits progressively less prolactin release, and in the weeks after initiation of lactation, basal and nursing-induced prolactin pulses decrease, although lactation continues.33 Within 4 to 6 months after delivery, basal pro-lactin levels are normal, without a nursing-induced rise. The full explanation for the attenuation in basal and nursing-induced prolactin levels with continued nursing is unknown.
CLINICAL MANIFESTATIONS OF HYPERPROLACTINEMIA
The amenorrhea-galactorrhea syndrome is the classic description of the clinical manifestation of hyperprolactinemia. However, a spectrum of reproductive disorders may be seen. Prolactin elevations are found in approximately 20% of patients with secondary amenorrhea.35 Women with hyperprolactinemia may have more subtle abnormalities in gonadal function, including oligomenorrhea or alterations in luteal phase function. A subset of infertile women has been described with mild hyperprolactinemia in whom fertility was restored with bromocriptine therapy. Galactorrhea affects only ~30% of female patients with hyperprolactinemia, but the presence of galactor-rhea in a woman with an ovulatory disorder greatly increases the chance that hyperprolactinemia is the underlying cause of the amenorrhea.36 Patients with primary amenorrhea and delayed puberty may have hyperprolactinemia.37
Galactorrhea occurs in as many as 25% of women with normal serum prolactin levels. However, patients with idiopathic galactorrhea may demonstrate intermittent hyperprolactinemia. In a study of nine normoprolactinemic women with galactorrhea, eight patients had elevated levels of prolactin during sleep.38 Several studies have shown that infertile, normopro-lactinemic women with luteal phase defects may show improved luteal function or fertility after administration of bromocriptine therapy.38 Unrecognized hyperprolactinemia may occur in a subset of patients with presumed normoprolactinemic galactorrhea and luteal phase defects.
Hypogonadism is frequently found in patients with hyperprolactinemia. In women, hypogonadism includes abnormal menstrual function, dry vaginal mucosa, and dyspareunia, and in men and women, the features include fatigue and diminished libido. Multiple potential mechanisms have been hypothesized for the induction of hypogonadism by prolactin, and the antigo-nadotropic actions of prolactin may occur at multiple levels. Frequently, the hypogonadism is associated with decreased or inappropriately normal levels of luteinizing hormone (LH) and follicle-stimulating hormone (FSH) relative to the state of estrogen deficiency. Multiple investigations suggest that prolactin may suppress spontaneous LH release through decreases in endogenous GnRH levels. In castrated rats, administration of graded doses of prolactin suppress LH levels, and prolactin appears to exert a negative feedback effect on its own secretion by means of a short-loop negative feedback at the level of the hypothalamus.39,40 This feedback may be mediated through an increase in dopamine inhibitory tone. This increased hypothalamic dopamine tone, along with opiates and other factors, may suppress GnRH with a resultant decrease in LH pulses. The restoration of ovulatory menstrual periods in hyperprolactinemic women by pulsatile exogenous GnRH administration confirms the importance of endogenous GnRH abnormalities as the key mechanism of hypogonadism in these women.41 Prolactin may modulate androgen secretion at the level of the adrenal gland and ovary, resulting in increased secretion of dehydroepiandrosterone sulfate and testosterone.42 Altered ratios of estrogens and androgens may result in further abnormal gonadal function, with evidence of clinical hyperandrogenism.
If the underlying cause of the increased prolactin is a pituitary macroadenoma, the adenoma could cause compression of the normal, adjacent pituitary gland with a resultant decrease in gonadotrope function.
Men with hyperprolactinemia may have clinical manifestations of hypogonadism, such as decreased libido, impotence, infertility due to oligospermia, and gynecomastia. Galactorrhea is rare in hyperprolactinemic men because of a lack of estrogen priming of the breast.
DIFFERENTIAL DIAGNOSIS AND CLINICAL APPROACH TO HYPERPROLACTINEMIA
As shown in Table 13-1 and Table 13-2, hyperprolactinemia has multiple causes. The measurement of prolactin level should be repeated with the patient in a nonstimulated state, and, if possible, after an overnight fast in a nonstressed state. Because prolactin may be secreted to a modest degree after a breast examination, a subsequent mild increase in prolactin levels would warrant a repeat determination. Although Table 13-2 demonstrates the existence of several pathologic causes of prolactin elevation, pituitary tumors are clinically the most important. Prolactin-secreting pituitary adenomas are the most common type of pituitary tumors and may account for as many as 40% to 50% of all pituitary tumors.43 Hyperprolactinemia may be detected in as many as 40% of patients with acromegaly and has been reported in patients with Cushing disease. Hyper-prolactinemic patients with suggestive clinical manifestations should be evaluated for acromegaly and Cushing disease.

TABLE 13-2. Pathologic and Pharmacologic Causes of Hyperprolactinemia

Substantial elevation in prolactin (>150 ng/mL) in a nonpuerperal state usually indicates a pituitary tumor. Good correlation exists between radiographic estimates of tumor size and prolactin levels, and very high levels of prolactin are associated with larger tumors. Prolactinomas are classified as microade-nomas (<10 mm) and macroadenomas (>10 mm). The finding of a substantial elevation in serum prolactin in association with a pituitary lesion larger than 10 mm by radiographic analysis supports the diagnosis of a macroprolactinoma.
Modest levels of prolactin elevation (25–100 ng/mL) may be associated with several diagnoses. All other causes of hyper-prolactinemia should be excluded before a tumor is considered. Primary hypothyroidism and pregnancy should be excluded. Chronic renal disease is associated with elevations in prolactin, probably because of altered metabolism or clearance of prolac-tin or decreases in dopaminergic tone.44 Hemodialysis does not usually reverse the hyperprolactinemia.
Multiple pharmacologic causes of hyperprolactinemia are found. Ingestion of phenothiazines and other neuroleptics is a common cause for elevations in serum prolactin.44a One diagnostic problem is the evaluation of patients with psychiatric disease who are receiving phenothiazines and are found to have an elevated prolactin level. A magnetic resonance image (MRI) should be obtained for patients whose prolactin levels are above 100 ng/mL. Levels lower than 100 ng/mL are consistent with neuroleptic administration, and a scan is unnecessary unless other symptoms suggest a pituitary tumor. This strategy is based on the finding that most patients receiving neuroleptics with modest prolactin elevations have no evidence of a pituitary abnormality on MRI. Other pharmacologic agents associated with hyperprolactinemia include reserpine, a-methyldopa, cimetidine, and opiates.
Estrogen may increase prolactin levels, as is seen in pregnancy. However, the estrogen concentrations in typical oral contraceptives (e.g., 35 µg of ethinyl estradiol) are not associated with hyperprolactinemia, and no evidence exists that post-menopausal replacement estrogen causes elevations in serum prolactin. Any intrasuprasellar mass may lead to modest prolactin elevations through stalk compression, and the evaluation should include an MRI. These masses include primary pituitary tumors or meningiomas and craniopharyngiomas. Hypothalamic disorders, including destructive lesions such as tumors and granulomatous diseases, may lead to hyperprolactinemia by interfering with normal dopaminergic tone.
If an elevated serum prolactin level is not associated with primary hypothyroidism, pregnancy, or pharmacologic agents, a pituitary radiographic scan should be performed to rule out the presence of a prolactin-secreting pituitary tumor or other lesions. Microprolactinomas should be differentiated from macroprolac-tinomas. An MRI is the most sensitive tool for evaluating the sellar and suprasellar areas. If the scan shows normal sellar and extrasellar contents and no clear secondary cause of the elevated prolactin is present, the diagnosis of idiopathic hyperprolactine-mia is made. This syndrome may be the result of a small tumor that is beyond the sensitivity of the scanning technique.
The evaluation should include assessment of gonadal status, such as the presence of oligomenorrhea or amenorrhea in women and of sexual dysfunction in men. This impacts therapy, as is described later. No stimulatory or suppressive endocrine tests are available that aid in the evaluation of elevated prolactin levels. For example, a TRH test cannot be used to diagnose a pituitary tumor; although tumors typically have blunted responses after TRH stimulation, this response can be seen with other disorders.
TREATMENT FOR PROLACTINOMAS
Treatment depends on whether the patient has hyperpro-lactinemia due to an underlying cause such as drugs or hypothyroidism, or due to a prolactinoma. If the evaluation suggests the presence of a microprolactinoma, three treatment options are available: medical therapy with a dopamine agonist, careful follow-up without treatment, and, rarely, surgery. All patients with macroadenomas should be treated.
MEDICAL THERAPY
Almost all patients with hyperprolactinemia due to pituitary disease can be effectively treated medically with the dopamine agonist bromocriptine (see Chap. 21). Bromocriptine lowers serum prolactin in patients with pituitary tumors and all other causes of hyperprolactinemia. A review of early studies of bromocriptine therapy for more than 400 hyperprolactinemic patients showed that normoprolactinemia or return of ovulatory menses occurred in 80% to 90% of patients.45 Bromocriptine effectively decreases prolactin levels, normalizes reproductive function, and reverses galactorrhea. In this series, return of menstrual function was accompanied in some patients by prolactin levels that were significantly reduced but not normal. This suggests that the reduction of prolactin levels in some patients to slightly elevated levels may be sufficient for return of gonadal function. Bromocriptine is also useful in treating galactorrhea in patients with normoprolactinemic galactorrhea.
The onset of the effects of bromocriptine is rapid, usually occurring within 1 to 2 hours. The greatest decrease in prolactin levels occurs at the initiation of therapy; however, normalization may take weeks. The biologic half-life of bromocriptine is similar to its plasma half-life. Discontinuation of the drug is typically followed by a return of prolactin to elevated values. Bromocriptine decreases prolactin production and secretion, with a resultant reduction in lactotrope size and a subsequent decrease in tumor size.46
Therapy should be initiated slowly because side effects, including nausea, headache, dizziness, nasal congestion, and constipation, may occur. Gastrointestinal side effects may be minimized by starting with a very low dose (e.g., 1.25 mg, or one-half tablet) taken at night with a snack, and increasing the dose by 1.25 mg over 4- to 5-day intervals, as tolerated. This progression is continued until a dosage that normalizes prolac-tin levels is reached. The rate of dosage escalation is dictated by the clinical situation, such as the presence of mass effects. Side effects can usually be improved by continuing the medication at the same dosage or by temporarily reducing the dosage. If patients stop taking the medication for a few days, therapy should be reinstituted at a lower dosage, because side effects may return. Rarely, long-term therapy may result in side effects, including painless cold-induced digital vasospasm, alcohol intolerance, dyskinesia, and psychiatric reactions, including fatigue, depression, and anxiety.
To reduce the gastrointestinal side effects, bromocriptine has been administered intravaginally. Reductions in prolactin similar to that attained by oral bromocriptine have been achieved with the intravaginal route.47 Gastrointestinal side effects are less common, and therapy may be more effective with vaginally administered bromocriptine.48
Cabergoline is an ergoline derivative with selective, potent, and long-lasting dopaminergic properties and is highly effective in the management of hyperprolactinemia. Because of the ease of administration of cabergoline (i.e., once or twice weekly) and its improved side-effect profile relative to bromocriptine, patients show a high rate of compliance. Administration of cabergoline at doses as high as 1.0 mg twice weekly to 113 patients with micro-prolactinomas resulted in normalization of prolactin levels in 95%.49 In another study, administration of cabergoline resulted in normalization of prolactin levels in 25 of 26 patients with micro-prolactinomas.50 Cabergoline appears to be better tolerated than bromocriptine and may play an important role in the management of patients intolerant of or resistant to bromocriptine. In a multicenter, randomized, 24-week trial involving 459 women, cabergoline was more effective and better tolerated than bromocriptine.51 In a study of 27 patients with prolactinomas resistant to bromocriptine or the investigational dopamine agonist CV 205-502, including 19 subjects with macroadenomas, caber-goline administration resulted in normalization of prolactin values in 47% of patients with macroadenomas and in all patients with microadenomas.52 Therefore, use of cabergoline may be considered in all subjects with hyperprolactinemia, including those who are poorly tolerant of or resistant to bromocriptine.52a
Other dopamine agonists are available, but not in the United States; these include CV 205-502 and a long-acting preparation of bromocriptine mesylate, Parlodel LAR. CV 205-502 is a non-ergot, long-acting dopamine agonist that appears to have D2-receptor binding compared with bromocriptine. Either CV 205-502 or bromocriptine was administered in a randomized double-blind fashion to 22 patients with microprolac-tinomas.53 In this study, 91% of the patients who received CV 205-502 and 56% of those who received bromocriptine had normalization of prolactin levels. Side effects were less common in the group receiving CV 205-502, and the drug may be useful in those patients intolerant to bromocriptine.54 CV 205-502 also is effective in the management of macroprolactinomas.55
Pergolide is a dopamine agonist approved by the U.S. Food and Drug Administration for the treatment of Parkinson disease. Although not approved for use in the management of hyperprolactinemia, studies have shown that pergolide has a comparable side-effect profile to bromocriptine and may reduce prolactin levels in patients unresponsive to bromocriptine.56,56b With the availability of cabergoline, pergolide is rarely used as an alternative to bromocriptine.
SURGERY
Although surgery is not a primary mode of management for patients with prolactinomas, it may be indicated in several settings. These include patients with large tumors causing visual field deficits unresponsive to bromocriptine, those unable to tolerate dopamine agonist therapy because of its side effects, those with cystic tumors that do not respond to medical therapy, and those with tumor apoplexy. A transsphenoidal approach is used almost exclusively. When the procedure is performed by experienced surgeons, the morbidity rate is negligible. The mortality rate is less than 0.27%, and the major morbidity rate is 3%.57
A theoretic advantage of curative surgery is avoidance of long-term medication. However, clinical evidence is lacking. Among 28 patients with microprolactinomas, 24 were cured with transsphenoidal surgery based on normalization of serum prolactin levels. After approximately 4 years, 50% of these initially “cured” patients had recurrence of hyperpro-lactinemia, although none had radiographic evidence of tumor growth.58 Another study found a recurrence rate of 39% after approximately 5 years.59 In one study of patients with normal prolactin values after surgery, the overall recurrence rate was 26% at a mean follow-up of 9.2 years.60 An immediately postoperative serum prolactin value of <5 ng/mL was associated with a recurrence rate of approximately 20%. These data suggest that, although surgery may result in normalization of prolactin levels initially in patients with microprolactinomas, risk of recurrence is relatively high. In addition, achievement of a low-normal serum prolactin level is an important predictive factor for long-term cure.
Surgical cure rates for macroprolactinomas are approximately 32%, with cure defined as normal prolactin levels after surgery.45 Surgical cure is inversely proportional to serum pro-lactin levels and tumor size. Unfortunately, the recurrence rate in macroprolactinomas has been reported to be as high as 80% after curative surgery,58 although another report indicates that recurrence rates as low as 26% can be achieved.60
RADIATION THERAPY
Conventional radiotherapy (4500–5000 rad) or, rarely, proton beam therapy may be indicated in patients who are not able to tolerate medical therapy.
MANAGEMENT OF MICROPROLACTINOMAS
The decision to institute medical therapy in patients with microprolactinomas is based on the metabolic consequences of hyperprolactinemia and tumor size. Patients with hyperpro-lactinemia are usually hypogonadotropic and have accompanying menstrual irregularities. Dopamine agonist therapy can restore menstrual function in most patients with amenorrhea. Luteal phase defects associated with hyperprolactinemia can also be reversed with bromocriptine therapy. Ovulation rates greater than 90% have been reported, with induction of pregnancy in more than 80% of patients.61 Galactorrhea is not an absolute indication for dopamine agonist therapy, unless the degree of galactorrhea is significantly bothersome to the patient. The presence of amenorrhea is an indication for medical therapy because of the risk of osteoporosis associated with hyperprolactinemic amenorrhea. Some women with micropro-lactinomas may choose not to have therapy if they show no evidence of hypogonadism or amenorrhea and if they do not desire fertility. Reduction of prolactin levels frequently restores libido and increases sperm counts in hyperprolactinemic men.
Patients with microprolactinomas and those without radiographic evidence of pituitary tumors can sometimes be followed without therapy. Studies investigating the natural history of such tumors have shown that prolactin elevations usually remain stable and, in some cases, spontaneously normalize.62 In a study of 41 patients with idiopathic hyperpro-lactinemia, based on normal computed tomographic scans for 5.5 years, 67% of patients whose initial prolactin values were less than 57 ng/mL had normalization of prolactin levels.63 None of the patients with initial prolactin values above 60 ng/mL showed normalization.
These and other data suggest that the degree of prolactin elevation is a prognostic factor for spontaneous resolution. When 38 untreated patients with microprolactinomas were followed for an average of 50.5 months, 36.6% had an increase, 55.3% had a spontaneous decrease, and 13.1% had no change in prolactin levels.64 A prospective study of untreated hyperpro-lactinemic women showed that basal menstrual function is an important variable in predicting progression of the prolactin level.65 In this study, patients with normal initial menstrual function were more likely to have normalization of prolactin levels, and patients with oligomenorrhea or amenorrhea were more likely to have no change or increases in prolactin levels. Most microprolactinomas do not exhibit evidence of further growth, and prolactin levels may spontaneously normalize.
An important aspect of the natural history of microprolacti-nomas is that most tumors do not significantly increase in size. Although many of these studies used insensitive radiographic techniques, such as skull films and tomograms, they demonstrated that in patients with microprolactinomas and no radiographic evidence of a tumor, tumor size increased in 0% to 22% of patients.61,63,64,65 and 66 In a study of 43 patients with presumed microadenomas with a mean follow-up period of 5.4 years, only 2 patients showed evidence of tumor progression.66 In a prospective study, 27 women were followed for an average of 5.2 years.65 Of 14 women with normal baseline radiographic studies, 4 developed evidence of an adenoma, although none developed a macroadenoma. Of the 13 women with evidence of a tumor at baseline, only 2 showed worsening of radiographic findings. This study suggests that, although tumor growth may occur in as many as 22% of cases, it is rarely accompanied by clinical symptoms from mass effects. Follow-up of untreated patients should include serial measurement of prolactin levels and periodic MRI scans, because tumor progression may not be accompanied by an increase in prolactin levels.
The presence of osteoporosis is a key factor in the decision to institute therapy. Hyperprolactinemia is associated with trabecular and cortical osteopenia. Fourteen young hyperprolactinemic women with prolactin levels ranging from 22 to 99 ng/mL and amenorrhea for 1 to 18 years had significantly decreased cortical bone density compared with normal women.67 Additional studies have shown that hyperprolactinemic women may have trabecular osteopenia with spinal bone density 10% to 25% below normal.68,69 and 70 Spinal bone density in hyperpro-lactinemic women correlates with serum androgen levels and relative percentage of ideal body weight.68,71 and 72 Decreased bone density is thought to be due to hypogonadism and not a direct effect of prolactin, because hyperprolactinemic women with normal menstrual function do not have associated bone loss.68 Figure 13-2 shows that hyperprolactinemic patients with hypogonadism have lower bone density than eugonadal hyper-prolactinemic women. The decreased bone density in the hyperprolactinemic women with amenorrhea also correlates with the duration of amenorrhea.70

FIGURE 13-2. Spinal bone density in 13 women with hyperprolactine-mic amenorrhea (top), 12 eumenorrheic hyperprolactinemic women (middle), and 11 women with hypothalamic amenorrhea (bottom). The mean ± 1 standard deviation for 19 normal women is shown by the solid and dashed horizontal lines, respectively. (Reprinted from Klibanski A, Biller BM, Rosenthal DI, et al. Effects of prolactin and estrogen deficiency in amenorrheic bone loss. J Clin Endocrinol Metab 1988; 67:124.)

An important question is whether treatment of hyperprolac-tinemic amenorrhea may result in improvement in bone density. A prospective series in which 32 women with hyperprolactine-mic amenorrhea were randomized to medical therapy or no therapy showed that resumption of menses with medical therapy was associated with a significant increase in cortical bone density, mostly during the first 12 months of therapy.70 However, the cortical bone density achieved in this series remained below that of normal women. Of 38 women with prolactinomas examined 2 to 5 years after surgery, cortical bone mass was below normal in both cured and uncured patients, a finding which suggests that remission of hyperprolactinemia may not lead to normalization of bone density.73 Therapy for hyperprolactinemic amenorrhea with resumption of menses may lead to improvement in but not normalization of bone density.
Another important question is whether hyperprolactinemic amenorrhea is associated with progressive, accelerated bone loss. In a study of 52 hyperprolactinemic women and 41 controls over a mean period of 1.8 years, trabecular osteoporosis was marked in women with hyperprolactinemic amenorrhea, and bone loss progressed in untreated women by an average of 3.8% per year.71 Conversely, in another study of 56 hyperpro-lactinemic women, spinal bone density in women with amenorrhea did not decrease significantly over an average of 4.7 years.72 These studies suggest that a subgroup of women with hyperprolactinemic amenorrhea may be at risk for progressive osteoporosis, including those women with decreased percentages of ideal body weight and lower serum androgen levels. Because osteoporosis is common in women with hyperpro-lactinemic amenorrhea, the authors recommend treatment of amenorrheic women to improve bone density or at least prevent further deterioration.
MANAGEMENT OF MACROPROLACTINOMAS
Unlike patients with microprolactinomas, those with macropro-lactinomas always require therapy. Patients with macroade-nomas may have evidence of local mass effects due to the tumor mass, with resultant visual field abnormalities, and hypopitu-itarism due to compression of the normal pituitary gland. Aggressive management is required to prevent or reverse these complications. Bromocriptine therapy results in significant tumor shrinkage in as many as 75% of patients. Tumor size reduction may occur in weeks or over many months, and this reduction is frequently accompanied by an improvement in visual field abnormalities and pituitary function. Visual field deficits have been reported to improve within hours of institution of medical therapy. Of 27 patients treated with bromocriptine, 64% experienced a reduction in tumor size of at least 50%.74 Tumor shrinkage often occurred within 6 weeks. Sixty-six percent of patients had normalization of prolactin levels, but the fall in prolactin levels did not always correlate with reductions in tumor size.
Administration of cabergoline may lead to reduction in pro-lactin levels, tumor shrinkage, and restoration of gonadal function in patients with macroprolactinomas. When cabergoline was administered to 14 patients with macroprolactinomas, tumor shrinkage was observed in 13 (93%), and complete disappearance was documented in 2 patients.75 In one such study, cabergoline was administered to 15 patients with macroprolacti-nomas in a multicenter, 48-week trial (see Fig. 13-3).76 Normalization of prolactin levels was achieved in 73% of subjects, and mean reduction in tumor size was 31%. In this study, tumor shrinkage often occurred during the first 8 weeks; however, in some subjects, reductions in tumor volume were not seen until 24 weeks. Therefore, cabergoline may have an important role in the management of patients with macroprolactinomas.77 Dosages of 0.5 to 3.0 mg cabergoline per week are often sufficient for maximal effects. The drug may be useful as first line therapy in patients with macroprolactinomas or in those who are intolerant of or resistant to bromocriptine.52

FIGURE 13-3. Individual serum prolactin levels at baseline and 48 weeks according to final dose in macroprolactinoma patients treated with cabergoline. The dotted horizontal line indicates normal serum pro-lactin (<20ng/mL). The dotted vertical line refers to a patient who normalized PRL at week 6 but did not complete 48 weeks because of a complication (pituitary hemorrhage). (From Biller BM, Molitch ME, Vance ML, et al. Treatment of prolactin-secreting macroadenomas with the once-weekly dopamine agonist cabergoline. J Clin Endocrinol Metab 1996; 81:2338, with permission.)

The heterogeneity of prolactinoma responses to bromocriptine may be the result of variable density of the D2 receptor on tumor membranes. Good correlation exists between the density of dopaminergic binding sites and maximal inhibition of adenylate cyclase activation in bromocrip-tine-responsive and bromocriptine-resistant prolactinomas.78 This suggests that bromocriptine response depends on the density of the D2 receptor. A study of 46 prolactinomas in patients undergoing surgery because of failure to respond to bromocriptine or local compression showed no evidence of mutations in the D2 coding sequence.79 Alterations in density or function probably explain the lack of dopamine agonist effect in these patients.
Medical therapy represents the initial option for patients with macroprolactinomas with no or stable visual field deficits because of the low cure rates associated with surgery in such patients. Although transsphenoidal surgery is not used as a primary therapy for patients with macroprolactinomas, no data are available directly comparing the use of dopamine agonists with surgery as primary therapy for macroadenomas. Specific circumstances, such as the presence of a cystic prolactinoma (which often does not respond fully to dopamine agonist therapy), may predict a poor initial response to bromocriptine therapy in these patients. Cabergoline or bromocriptine is a useful adjunctive therapy in patients with large tumors, for which complete resection has not been possible.
Most men with diagnosed prolactinomas have macroade-nomas. In women, most tumors are microadenomas. In part, this difference may reflect the fact that women present earlier than men for evaluation because of complaints of menstrual disturbances. In men with hyperprolactinemia-induced hypog-onadism and normal residual pituitary function, 3 to 6 months may be required for testosterone levels to increase, normal sexual function to be restored, and sperm counts and motility to increase after prolactin levels normalize.80
PREGNANCY AND PROLACTINOMAS
Many women with hyperprolactinemia present with infertility; bromocriptine is typically used to normalize prolactin levels and allow normal ovulation to occur. After pregnancy is established, bromocriptine should be discontinued if no evidence of local tumor compression is seen. Pregnancy in normal women leads to increased pituitary size through estrogen-stimulated lactotrope hyperplasia. For patients with prolactinomas, the concern is that the high estrogen levels associated with pregnancy may lead to lactotrope stimulation and tumor growth, with resulting local complications, including visual field deficits, headaches, and diabetes insipidus. In a review of the data on pregnancy outcomes in hyperprolactinemic patients, clinically significant tumor enlargement (e.g., causing headaches, visual deficits) were found in as many as 5.5% of patients with microprolactinomas.81 Patients with microprolactinomas should be followed carefully during pregnancy, with visual field monitoring at monthly intervals.
In contrast to patients with microprolactinomas, 15.5% to 35.7% of patients with macroadenomas are at risk for clinically significant tumor enlargement during any trimester of pregnancy. Although some centers recommend transsphenoidal resection of the macroadenomas before conception, surgical resection does not prevent symptomatic enlargement during pregnancy. Monitoring of serum prolactin levels throughout pregnancy is not clinically useful, because prolactin levels increase markedly during pregnancy. The decision to reinstitute therapy depends on the development of clinical symptoms, not the serum prolactin level. If a complication due to tumor growth does occur, it is rapidly reversible with the reinstitution of bromocriptine therapy, which is then continued through term.
The outcome of bromocriptine-induced pregnancies is comparable to that of normal pregnancies. A large, international experience with bromocriptine and pregnancy suggests that bromocriptine therapy does not result in complications for the fetus82 (see Chap. 110).
Experience with cabergoline and pregnancy is much more limited. In the largest such study (226 pregnancies induced by cabergoline) no evidence was found of increased pregnancy- associated complications or birth defects in these women compared to women who did not receive cabergoline.83 However, until the experience with cabergoline is more widespread, bromocriptine is recommended as a first-choice therapy for hyper-prolactinemic women seeking pregnancy.
After delivery, breast-feeding appears to be safe in patients not receiving bromocriptine. Patients with macroadenomas should continue to be followed closely, and the decision to institute therapy should depend on tumor size and clinical symptoms.
Patients with hyperprolactinemia should undergo a complete evaluation to determine the underlying cause of the elevated prolactin level. If idiopathic hyperprolactinemia or a prolactinoma is the cause, dopamine agonist therapy is highly efficacious in lowering prolactin levels, restoring gonadal function, and reversing local tumor complications.
CHAPTER REFERENCES

1.
Blackwell RE. Hyperprolactinemia. Evaluation and management. Endocrinol Metab Clin North Am 1992; 21:105.

2.
Foord SM, Peters JR, Dieguez C, et al. Dopamine receptors on intact anterior pituitary cells in culture: functional association with the inhibition of prolactin and thyrotropin. Endocrinology 1983; 112:1567.

3.
Schettini G, Cronin MJ, MacLeod RM. Adenosine 3′,5′-monophosphate (cAMP) and calcium-calmodulin interrelation in the control of prolactin secretion: evidence for dopamine inhibition of cAMP accumulation and prolactin release after calcium mobilization. Endocrinology 1983; 112:1801.

4.
Leblanc H, Lachelin GC, Abu-Fadil S, Yen SS. Effects of dopamine infusion on pituitary hormone secretion in humans. J Clin Endocrinol Metab 1976; 43:668.

5.
Rivera JL, Lal S, Ettigi P, et al. Effect of acute and chronic neuroleptic therapy on serum prolactin levels in men and women of different age groups. Clin Endocrinol 1976; 5:273.

6.
Nikolics K, Mason AJ, Szonyi E, et al. A prolactin-inhibiting factor within the precursor for human gonadotropin-releasing hormone. Nature 1985; 316:511.

7.
Grossman A, Delitala G, Yeo T, Besser GM. GABA and muscimol inhibit the release of prolactin from dispersed rat anterior pituitary cells. Neuroen-docrinology 1981; 32:145.

8.
Raymond V, Beaulieu M, Labrie F, Boissier J. Potent antidopaminergic activity of estradiol at the pituitary level on prolactin release. Science 1978; 200:1173.

9.
Maurer RA. Estradiol regulates the transcription of the prolactin gene. J Biol Chem 1982; 257:2133.

10.
Scheithauer BW, Sano T, Kovacs KT, et al. The pituitary gland in pregnancy: a clinicopathologic and immunohistochemical study of 69 cases. Mayo Clin Proc 1990; 65:461.

11.
Bowers CY, Friesen HG, Hwang P, et al. Prolactin and thyrotropin release in man by synthetic pyroglutamyl-histidyl-prolinamide. Biochem Biophys Res Commun 1971; 45:1033.

12.
Kato Y, Iwasaki Y, Iwasaki J, et al. Prolactin release by vasoactive intestinal polypeptide in rats. Endocrinology 1978; 103:554.

13.
Abe H, Engler D, Molitch ME, et al. Vasoactive intestinal peptide is a physiological mediator of prolactin release in the rat. Endocrinology 1985; 116:1383.

14.
Hagen TC, Arnaout MA, Scherzer WJ, et al. Antisera to vasoactive intestinal polypeptide inhibit basal prolactin release from dispersed anterior pituitary cells. Neuroendocrinology 1986; 43:641.

15.
Casper RF, Yen SS. Simultaneous pulsatile release of prolactin and luteiniz-ing hormone induced by luteinizing hormone–releasing factor agonist. J Clin Endocrinol Metab 1981; 52:934.

16.
Denef C, Andries M. Evidence for paracrine interaction between gonado-trophs and lactotrophs in pituitary cell aggregates. Endocrinology 1983; 112:813.

17.
Koshiyama H, Kato Y, Inoue T, et al. Central galanin stimulates pituitary prolactin secretion in rats: possible involvement of hypothalamic vasoactive intestinal polypeptide. Neurosci Lett 1987; 75:49.

18.
Giustina A, Licini M, Schettino M, et al. Physiological role of galanin in the regulation of anterior pituitary function in humans. Am J Physiol 1994; 266:E57.

19.
Wehrenberg WB, McNicol D, Frantz AG, Ferin M. The effects of serotonin on prolactin and growth hormone concentrations in normal and pituitary stalk-sectioned monkeys. Endocrinology 1980; 107:1747.

20.
Falaschi P, Rosa M, Rocco A, et al. Effect of ritanserin, specific 5HT-2 antagonist, on PRL secretion in normal subjects and in different hyperprolactinaemic conditions. Clin Endocrinol 1991; 34:449.

21.
Lewis UJ, Singh RN, Sinha YN, Vanderlaan WP. Glycosylated human prolactin. Endocrinology 1985; 116:359.

22.
Whitaker MD, Klee GG, Kao PC, et al. Demonstration of biological activity of prolactin molecular weight variants in human sera. J Clin Endocrinol Metab 1984; 58:826.

23.
Corenblum B. Asymptomatic hyperprolactinemia resulting from macro-prolactinemia. Fertil Steril 1990; 53:165.

24.
Ben-David M, Schenker JG. Transient hyperprolactinemia: a correctable cause of idiopathic female infertility. J Clin Endocrinol Metab 1983; 57:442.

25.
Veldhuis JD, Johnson ML. Operating characteristics of the hypothalamo-pituitary-gonadal axis in men: circadian, ultradian, and pulsatile release of prolactin and its temporal coupling with luteinizing hormone. J Clin Endocrinol Metab 1988; 67:116.

26.
Sassin JF, Frantz AG, Weitzman ED, Kapen S. Human prolactin: 24-hour pattern with increased release during sleep. Science 1972; 177:1205.

27.
Franchimont P, Dourcy C, Legros JJ, et al. Prolactin levels during the menstrual cycle. Clin Endocrinol 1976; 5:643.

28.
Carlson HE. Prolactin stimulation by protein is mediated by amino acids in humans. J Clin Endocrinol Metab 1989; 69:7.

29.
Boyd AED, Spare S, Bower B, Reichlin S. Neurogenic galactorrhea-amenorrhea. J Clin Endocrinol Metab 1978; 47:1374.

30.
Katznelson L, Riskind PN, Saxe VC, Klibanski A. Prolactin pulsatile characteristics in postmenopausal women. J Clin Endocrinol Metab 1998; 83:761.

31.
Vekemans M, Robyn C. Influence of age on serum prolactin levels in women and men. BMJ 1975; 4:738.

32.
Rigg LA, Lein A, Yen SS. Pattern of increase in circulating prolactin levels during human gestation. Am J Obstet Gynecol 1977; 129:454.

33.
Noel GL, Suh HK, Frantz AG. Prolactin release during nursing and breast stimulation in postpartum and nonpostpartum subjects. J Clin Endocrinol Metab 1974; 38:413.

34.
Nunley WC, Urban RJ, Kitchin JD, et al. Dynamics of pulsatile prolactin release during the postpartum lactational period. J Clin Endocrinol Metab 1991; 72:287.

35.
Franks S, Murray MA, Jequier AM, et al. Incidence and significance of hyperprolactinaemia in women with amenorrheoa. Clin Endocrinol 1975; 4:597.

36.
Jacobs HS, Franks S, Murray MA, et al. Clinical and endocrine features of hyperprolactinaemic amenorrhoea. Clin Endocrinol 1976; 5:439.

37.
Patton ML, Woolf PD. Hyperprolactinemia and delayed puberty: a report of three cases and their response to therapy. Pediatrics 1983; 71:572.

38.
Asukai K, Uemura T, Minaguchi H. Occult hyperprolactinemia in infertile women. Fertil Steril 1993; 60:423.

39.
Bergland RM, Page RB. Can the pituitary secrete directly to the brain? (Affirmative anatomical evidence). Endocrinology 1978; 102:1325.

40.
Park SK, Selmanoff M. Hyperprolactinemia suppresses the luteinizing hormone responses to N-methyl-D-aspartate, epinephrine, and neuropeptide-Y in male rats. Endocrinology 1993; 133:2091.

41.
Polson DW, Sagle M, Mason HD, et al. Ovulation and normal luteal function during GnRH treatment of women with hyperprolactinaemic amenorrhoea. Clin Endocrinol 1986; 24:531.

42.
Lobo RA, Kletzky OA, Kaptein EM, Goebelsmann U. Prolactin modulation of dehydroepiandrosterone sulfate secretion. Am J Obstet Gynecol 1980; 138:632.

43.
Klibanski A, Zervas NT. Diagnosis and management of hormone-secreting pituitary adenomas. N Engl J Med 1991; 324:822.

44.
Cowden EA, Ratcliffe WA, Ratcliffe JG, et al. Hyperprolactinaemia in renal disease. Clin Endocrinol 1978; 9:241.

44a.
Kleinberg DL, Davis JM, de Coster R, et al. Prolactin levels and adverse events in patients treated with risperidone. J Clin Psychopharm 1999; 19:57.

45.
Molitch ME. Pathologic hyperprolactinemia. Endocrinol Metab Clin North Am 1992; 21:877.

46.
Bassetti M, Spada A, Pezzo G, Giannattasio G. Bromocriptine treatment reduces the cell size in human macroprolactinomas: a morphometric study. J Clin Endocrinol Metab 1984; 58:268.

47.
Vermesh M, Fossum GT, Kletzky OA. Vaginal bromocriptine: pharmacology and effect on serum prolactin in normal women. Obstet Gynecol 1988; 72:693.

48.
Kletzky OA, Vermesh M. Effectiveness of vaginal bromocriptine in treating women with hyperprolactinemia. Fertil Steril 1989; 51:269.

49.
Webster J, Piscitelli G, Polli A, et al. Dose-dependent suppression of serum prolactin by cabergoline in hyperprolactinaemia: a placebo controlled, double blind, multicentre study. European Multicentre Cabergoline Dose-Finding Study Group. Clin Endocrinol 1992; 37:534.

50.
Muratori M, Arosio M, Gambino G, et al. Use of cabergoline in the long-term treatment of hyperprolactinemic and acromegalic patients. J Endo-crinol Invest 1997; 20:537.

51.
Webster J, Piscitelli G, Polli A, et al. A comparison of cabergoline and bromocriptine in the treatment of hyperprolactinemic amenorrhea. N Engl J Med 1994; 331:904.

52.
Colao A, Di Sarno A, Sarnacchiaro F, et al. Prolactinomas resistant to standard dopamine agonists respond to chronic cabergoline treatment [see comments]. J Clin Endocrinol Metab 1997; 82:876.

52a.
Cabergoline in the treatment of hyperprolactinemia: a study in 455 patients. J Clin Endocrinol Metab 1999; 84:2518.

53.
Homburg R, West C, Brownell J, Jacobs HS. A double-blind study comparing a new non-ergot, long-acting dopamine agonist, CV 205-502, with bromocriptine in women with hyperprolactinaemia. Clin Endocrinol 1990; 32:565.

54.
van-der-Lely AJ, Brownell J, Lamberts SW. The efficacy and tolerability of CV 205-502 (a nonergot dopaminergic drug) in macroprolactinoma patients and in prolactinoma patients intolerant to bromocriptine. J Clin Endocrinol Metab 1991; 72:1136.

55.
Vance ML, Lipper M, Klibanski A, et al. Treatment of prolactin-secreting pituitary macroadenomas with the long-acting non-ergot dopamine agonist CV 205-502. Ann Intern Med 1990; 112:668.

56.
Kletzky OA, Borenstein R, Mileikowsky GN. Pergolide and bromocriptine for the treatment of patients with hyperprolactinemia. Am J Obstet Gynecol 1986; 154:431.

56a
Freda PU, Andreadis CI, Khandji AG, et al. Long-term treatment of prolac-tin-secreting macroadenomas with pergolide. J Clin Endocrinol Metab 2000; 85:8.

57.
Zervas NT. Surgical results in pituitary adenomas: results of an international study. In: Black PM, Zervas NT, Ridgway EC, Martin JB, eds. Secretory tumors of the pituitary gland. New York: Raven Press, 1984:377.

58.
Serri O, Rasio E, Beauregard H, et al. Recurrence of hyperprolactinemia after selective transsphenoidal adenomectomy in women with prolacti-noma. N Engl J Med 1983; 309:280.

59.
Schlechte JA, Sherman BM, Chapler FK, VanGilder J. Long term follow-up of women with surgically treated prolactin-secreting pituitary tumors. J Clin Endocrinol Metab 1986; 62:1296.

60.
Feigenbaum SL, Downey DE, Wilson CB, Jaffe RB. Transsphenoidal pituitary resection for preoperative diagnosis of prolactin-secreting pituitary adenoma in women: long term follow-up. J Clin Endocrinol Metab 1996; 81:1711.

61.
Bergh T, Nillius SJ, Wide L. Bromocriptine treatment of 42 hyperprolacti-naemic women with secondary amenorrhoea. Acta Endocrinol 1978; 88:435.

62.
Weiss MH, Teal J, Gott P, et al. Natural history of microprolactinomas: six-year follow-up. Neurosurgery 1983; 12:180.

63.
Martin TL, Kim M, Malarkey WB. The natural history of idiopathic hyper-prolactinemia. J Clin Endocrinol Metab 1985; 60:855.

64.
Sisam DA, Sheehan JP, Sheeler LR. The natural history of untreated micro-prolactinomas. Fertil Steril 1987; 48:67.

65.
Schlechte J, Dolan K, Sherman B, et al. The natural history of untreated hyperprolactinemia: a prospective analysis. J Clin Endocrinol Metab 1989; 68:412.

66.
March CM, Kletzky OA, Davajan V, et al. Longitudinal evaluation of patients with untreated prolactin-secreting pituitary adenomas. Am J Obstet Gynecol 1981; 139:835.

67.
Klibanski A, Neer RM, Beitins IZ, et al. Decreased bone density in hyper-prolactinemic women. N Engl J Med 1980; 303:1511.

68.
Klibanski A, Biller BM, Rosenthal DI, et al. Effects of prolactin and estrogen deficiency in amenorrheic bone loss. J Clin Endocrinol Metab 1988; 67:124.

69.
Koppelman MC, Kurtz DW, Morrish KA, et al. Vertebral body bone mineral content in hyperprolactinemic women. J Clin Endocrinol Metab 1984; 59:1050.

70.
Klibanski A, Greenspan SL. Increase in bone mass after treatment of hyper-prolactinemic amenorrhea. N Engl J Med 1986; 315:542.

71.
Biller BM, Baum HB, Rosenthal DI, et al. Progressive trabecular osteopenia in women with hyperprolactinemic amenorrhea [see comments]. J Clin Endocrinol Metab 1992; 75:692.

72.
Schlechte J, Walkner L, Kathol M. A longitudinal analysis of premeno-pausal bone loss in healthy women and women with hyperprolactinemia. J Clin Endocrinal Metab 1992; 75:698.

73.
Schlechte JA, Sherman B, Martin R. Bone density in amenorrheic women with and without hyperprolactinemia. J Clin Endocrinol Metab 1983; 56:1120.

74.
Molitch ME, Elton RL, Blackwell RE, et al. Bromocriptine as primary therapy for prolactin-secreting macroadenomas: results of a prospective multi-center study. J Clin Endocrinol Metab 1985; 60:698.

75.
Ferrari C, Paracchi A, Mattei AM, et al. Cabergoline in the long-term therapy of hyperprolactinemic disorders. Acta Endocrinol 1992; 126:489.

76.
Biller BM, Molitch ME, Vance ML, et al. Treatment of prolactin-secreting macroadenomas with the once-weekly dopamine agonist cabergoline. J Clin Endocrinol Metab 1996; 81:2338.

77.
Ferrari CI, Abs R, Bevan JS, et al. Treatment of macroprolactinoma with cabergoline: a study of 85 patients. Clin Endocrinol 1997; 46:409.

78.
Pellegrini I, Rasolonjanahary R, Gunz G, et al. Resistance to bromocriptine in prolactinomas. J Clin Endocrinol Metab 1989; 69:500.

79.
Friedman E, Adams EF, Hoog A, et al. Normal structural dopamine type 2 receptor gene in prolactin-secreting and other pituitary tumors. J Clin Endocrinol Metab 1994; 78:568.

80.
De Rosa M, Colao A, Di Sarno A, et al. Cabergoline treatment rapidly improves gonadal function in hyperprolactinemic males: a comparison with bromocriptine. Eur J Endocrinol 1998; 138:286.

81.
Molitch ME. Pregnancy and the hyperprolactinemic woman. N Engl J Med 1985; 312:1364.

82.
Turkalj I, Braun P, Krupp P. Surveillance of bromocriptine in pregnancy. JAMA 1982; 247:1589.

83.
Robert E, Musatti L, Piscitelli G, Ferrari CI. Pregnancy outcome after treatment with the ergot derivative, cabergoline. Reprod Toxicol 1996; 10:333.