Bulletin Vol4, Issue 3, Winter 1998

Prolactinomas and Pregnancy

Pituitary Home

Anne Klibanski
by Anne Klibanski, M.D.

NEPTCC Newsletter MGH Neuroendocrine Center Bulletin Vol 4, Issue 3, Winter 1998

Prolactin and Prolactinomas

Prolactin, a pro-lactational hormone, is secreted by the pituitary gland in a pulsatile fashion and is kept within the normal range predominantly under the influence of dopaminergic inhibition throughout adult life. During pregnancy, prolactin secretion in normal women markedly increases from the beginning of pregnancy until delivery. The marked rise in serum prolactin concentrations, often into a range of 100 to 200 ng/ml or higher is the result of an actual increase in lactotroph number and capacity because of stimulation by estradiol. Estrogen both stimulates lactotroph number and prolactin secretion by a direct effect on prolactin transcription and also cell proliferation, and, by an effect of estrogen to decrease central dopaminergic tone. Hyperprolactinemia from a number of pathologic conditions including a pituitary tumor, primary hypothyroidism, renal failure and secondary to a number of medications including psychotropic medications, antidepressants and a number of antihypertensives, can elevate serum prolactin levels and interfere with normal ovulatory function and fertility. Pathologic hyperprolactinemia has long been known to be associated with amenorrhea, anovulation and infertility. Of importance also, is the syndrome of intermittent or transient hyperprolactinemia found in a number of women with regular menstrual periods but luteal phase insufficiency. In a paper by Huang et al1 serum prolactin levels measured in 151 patients with luteal phase deficiency showed that 21.9% of patients had mild hyperprolactinemia most commonly seen around the time of ovulation. In such patients, successful pregnancy can typically be achieved with the initiation of dopamine agonist therapy if this is found to be the only underlying abnormality. Therefore, there exists a large spectrum of abnormalities associated with hyperprolactinemia that can interfere with normal ovulatory function and conception in women.

Although prolactin secretion in patients with prolactin secreting adenomas, both microadenomas (less than 10 mm) and macroadenomas ( greater than 10 mm) is typically autonomous and poorly responsive to normal physiologic stimulants to prolactin release, patients with underlying prolactin secreting tumors are at risk for the development of tumor enlargement causing symptoms during pregnancy. Pituitary tumors themselves may enlarge under the influence of high endogenous levels of gonadal steroids, specifically estradiol. In addition, there is marked hyperplasia of normal lactotrophs. Enlargement of a prolactinoma during pregnancy may be associated with symptoms due to tumor mass such as headaches, and specifically neuroophthamalogic abnormalities including visual field loss. The likelihood of symptomatic enlargement during pregnancy is directly related to the size of the underlying lesion, and in part, the extent and anatomic configuration of the tumor itself. A number of studies have investigated the outcome of patients with pregnancy and prolactinomas2. In patients with microprolactinomas, the risk has been reported to be between 2 to 5%. In the MGH experience of over 100 pregnancies in patients with microprolactinomas, progressive symptomatic enlargement causing neurologic deficits or visual field loss is extremely rare. However, in patients with macroprolactinomas, the likelihood of symptomatic tumor enlargement is much higher and can affect up to a quarter or more of patients. Patients who have pituitary macroadenomas with significant suprasellar extension may be at higher risk for the development of visual field abnormalities during pregnancy than patients with pituitary tumors which extend laterally or inferiorly into the cavernous sinus. Overall, innumerable studies indicate that the majority of patients with prolactinomas, regardless of size, can be safely managed during pregnancy without dopamine agonist therapy.

How should patients with known prolactinomas be monitored throughout pregnancy? Because the majority of patients with prolactinomas do not develop symptomatic enlargement, it is our policy to discontinue dopamine agonist therapy when the pregnancy test becomes positive. This approach is followed uniformly in almost all patients. If a patient is beginning a pregnancy with a known lesion with previous significant chiasmal compression and visual field loss, and now has a lesion not compressing but nearly abutting the optic chiasm, the decision as to whether to continue dopamine agonist therapy during pregnancy must be made on an individual basis. In the vast majority of cases, however, patients do not continue to receive dopamine agonist therapy during pregnancy. Although there are no data to suggest that dopamine agonists, such as Parlodel, are teratogenic, these medications are not approved for use during pregnancy. Should prolactin levels be monitored during pregnancy in a patient with an underlying prolactinoma? We do not routinely monitor prolactin levels during pregnancy for the following two reasons. First, prolactin levels in a normal pregnancy can increase to levels in the hundreds range or higher and may be indistinguishable from prolactin levels during pregnancy in a patient with a prolactinoma. Second, a patient would not be treated during pregnancy if the prolactin elevation were not associated with clear-cut clinical symptoms such as neurologic or neuroophthamalogic signs. Therefore, the best way to monitor patients during pregnancy with an underlying prolactinoma is to make the patient aware of symptoms that may be suggestive of mass effect such as headache, visual abnormalities or systemic signs consistent with pituitary insufficiency. We monitor patients routinely throughout pregnancy. In patients with macroprolactinomas, visual fields should be done during pregnancy on a monthly basis. In patients with microadenomas, the frequency of follow-up is somewhat individual dependent upon the patient. Given the low prevalence of tumor enlargement in patients with microadenomas, formal visual fields testing once each trimester is probably sufficient but should certainly be done sooner should the patient develop any symptoms consistent with mass effect.

Postpartum patients with idiopathic hyperprolactinemia or microprolactinomas can certainly be allowed to nurse. Prolactin should be rechecked at three to six months following pregnancy and if prolactin continues to be elevated and is associated with menstrual abnormalities and other symptoms, dopamine agonist therapy should be reinstituted unless the patient wants to continue nursing. In patients with macroadenomas, resumption of dopamine agonist therapy postpartum should be individually made depending upon the patient's symptoms and need for medical control of the underlying tumor.

The recent introduction of Dostinex (cabergoline), a long-acting D2 specific dopamine agonist, has changed the routine management of prolactinomas. Because of the improved efficacy and lower incidence of side effects with this medication, patients are often started on cabergoline as an initial mode of therapy given on a once-a-week or twice-a-week basis. In patients who are specifically seeking pregnancy, the decision as to what dopamine agonist to use becomes more problematic. Parlodel has been in clinical use for over twenty years and there is extensive experience in inducing ovulation in patients with Parlodel. Although cabergoline has been in use for several years in Europe, the experience in inducing pregnancy in terms of safety is still relatively new. In a series published in 1997 by Ciccarelli et al, 47 women were treated with cabergoline for 1 to 82 months. There were nine patients reported who became pregnant. In two patients, one patient with a microadenoma and one patient with idiopathic hyperprolactinemia, prolactin levels remained in the normal range for one to three years after delivery. This series is too small to determine whether the spontaneous resolution which is sometimes reported after pregnancy in patients with underlying hyperprolactinemia is at all associated with the use of cabergoline. The experience to assess the possibility of teratogenicity of cabergoline is limited to a 1996 report by Musatti et al published in the European literature.4 This series reports 226 pregnancies in 205 women with follow-up data available in 204 women and there was no increase in the miscarriage rate, distribution of birth weights, or rate of congenital malformations reported among the babies of mothers who became pregnant while receiving cabergoline. Although these data are reassuring, the long-term safety of this medication in inducing pregnancy is in no way comparable to the data available in patients who have become pregnant on Parlodel. Therefore, our recommendation is to use Parlodel in patients who are actively trying to achieve a pregnancy. If a patient is receiving Dostinex for therapy of hyperprolactinemia and becomes pregnant, these preliminary data are reassuring and it will be anticipated that a large number of other such pregnancies will be reported in the next few years.

The spectrum of reproductive abnormalities associated with hyperprolactinemia ranges from frank amenorrhea and galactorrhea to very subtle luteal phase deficiencies manifested only by periovulatory hyperprolactinemia. Reproductive abnormalities associated with hyperprolactinemia in women typically respond very well to dopamine agonist therapy. Pregnancies can be achieved in patients with underlying normal gonadotroph function by inhibiting prolactin with a dopamine agonist. There is no evidence that radiation therapy will prevent tumor enlargement during pregnancy in a patient with an underlying prolactinoma and the radiation therapy itself may be deleterious to gonadotroph function in patients who still have gonadotroph capacity. Transsphenoidal surgery can be used in patients to control hyperprolactinemia and tumor size in patients who are unresponsive or poorly responsive to dopamine agonist therapy. Symptomatic tumor enlargement during pregnancy rarely occurs and is best managed with the re-institution of Parlodel therapy. Finally, transsphenoidal surgery can be considered in those patients who have symptoms during pregnancy that are refractory to dopamine agonist administration or who have rapidly developing neurologic symptoms.

  1. Huang KE, Bonfiglio TA, Muechler EK. Transient hyperprolactinemia in infertile women with luteal phase deficiency. Obstet Gynecol. 1991; 78 (4): 651-655.
  2. Molitch ME. Management of prolactinomas. Ann Rev Med. 1989; 40: 225-232.
  3. Ciccarelli E, Grottoli S, Razzore P, et al. Long-term treatment with cabergoline, a new long-lasting ergoline derivate, in idiopathic or tumorous hyperprolactinaemia and outcome of drug-induced pregnancy. J Endocrinol Invest. 1997; 20(9): 547-551.
  4. Robert E, Musatti L, Piscitelli G, Ferrari CI. Pregnancy outcome after treatment with the ergot derivative, cabergoline. Reprod Toxicol. 1996; 10(4): 333-337.

Central Hypothyroidism due to Pituitary / Hypothalamic Dysfunction

Pituitary Home

Karen K. Miller
by Karen K. Miller, M.D.

NEPTCC Newsletter MGH Neuroendocrine Center Bulletin Vol 4, Issue 3, Winter 1998

Central Hypothyroidism
Central hypothyroidism is an important complication of pituitary disease and, because TSH levels are not useful, the diagnosis and therapeutic considerations are difficult.

Central hypothyroidism is defined as a reduction in circulating thyroid hormone as a result of inadequate stimulation of a normal thyroid gland by TSH and may be secondary, due to pituitary disease, or tertiary, due to hypothalamic dysfunction. Causes include all pathologic processes that affect the hypothalamus or pituitary including tumors, Sheehan's syndrome, idiopathic hypopituitarism and infiltrative diseases, such as sarcoidosis, histiocytosis and lymphocytic hypophysitis. Radiation-induced central hypothyroidism is common in patients irradiated for pituitary tumors. Tsang et al. retrospectively examined records of 160 patients who had received radiation for non-functioning pituitary adenomas 8 to 22 years before and found that 65% required thyroid replacement therapy, with 23% of patients' hypothyroidism directly attributable to the radiation therapy received [1]. In addition, hypothyroidism is common in patients receiving radiation for nasopharyngeal or paranasal sinus tumors and brain tumors. Constine et al. evaluated the endocrine function of 32 patients who had received radiation for brain tumors, including gliomas, medulloblastomas and ependymomas, from 2 to 13 years before, and found that 65% had low free T4 levels. The probability of hypothyroidism depended on the amount of radiation received, with doses of more than 5000 rads (50 Gy) to the pituitary and hypothalamus necessary for the development of hypothyroidism. Moreover, the longer the interval since irradiation, the more likely a patient was to have developed hypothyroidism [2]. Therefore, the percentage of patients developing hypothyroidism may have been even higher had the follow-up been longer. In addition, because the onset of hypothyroidism may occur years after the administration of the radiation, constant vigilance for the signs and symptoms of hypothyroidism in this population is imperative and yearly monitoring of thyroid hormone levels obligatory.

As in primary hypothyroidism, the symptoms of central hypothyroidism include fatigue, apathy, weight gain, dry skin, constipation and cold intolerance. Signs include periorbital edema, cool extremities, delayed relaxation of the deep tendon reflexes and bradycardia. The signs and symptoms of central hypothyroidism mimic those of several other common conditions, and this disorder is therefore difficult to diagnose. A low free T4 or free T4 index and a low TSH in the setting of pituitary disease and signs and symptoms of hypothyroidism point in a straightforward manner to the diagnosis of central hypothyroidism. Unfortunately, however, the TSH is most commonly in the normal range in cases of central hypothyroidism, creating a confusing picture. Research has shown that, in some of these cases, a bio-inactive TSH resulting from abnormal glycosylation of the TSH molecule [3-6] explains the higher than expected TSH levels. Therefore, although the serum TSH is measured as normal in routine assays, performed by immunoradiometric assay (IRMA) or immunochemiluminometric assay (ICMA), only a small proportion of the TSH molecules function normally. Although these "normal" TSH values can confound the diagnosis of central hypothyroidism, it should be noted that a "normal" or slightly high TSH is inappropriate when circulating thyroid hormone levels are low, and that in cases of primary hypothyroidism, the TSH would be expected to be much higher. Therefore, TSH is not a useful screen for the diagnosis of this disorder.

The management of central hypothyroidism is further complicated by the fact that the TSH cannot be used to monitor therapeutic response to L-thyroxine therapy. When pituitary pathology is not present, the TSH provides an accurate method of assessing the appropriateness of circulating thyroid hormone levels for each particular patient. However, pituitary or hypothalamic pathology often interrupts the feedback mechanism by preventing normal release of TSH and/or TRH. The consequences of this are two-fold. First, patients with pituitary pathology or who have been irradiated cannot be screened for hypothyroidism with TSH levels alone, as a normal TSH often belies central hypothyroidism. Moreover, the usually routine monitoring of thyroid replacement becomes more problematic. Inadequate replacement doses of l-thyroxine often result in markedly subnormal TSH values. Therefore, TSH values are not reliable as an accurate reflection of thyroid status, and a free T4 or free T4 index must be used to adjust the replacement dose. However, as in primary hypothyroidism, when appropriately diagnosed and treated, management of central hypothyoidism can result in prompt resolution of symptoms.

  1. Tsang R, Brierley J, Panzarella T, et al., 1994 Radiation therapy for pituitary adenoma: treatment outcome and prognostic factors. Int J Radiat Oncol Biol Phys. 30:557-565.
  2. Constine L, Woolf P, Cann D, et al. 1993 Hypothalamic-pituitary dysfunction after radiation for brain tumors. N Engl J Med. 328:87-94.
  3. Papandreou M, Persani L, Asteria C, Ronin C, and Beck-Peccoz P. 1993 Variable carbohydrate structures of circulating thyrotropin as studied by lectin affinity chromatography in different clinical conditions. J Clin Endocrinol Metab. 77:393.
  4. Miura Y, Perkel V, Papenberg K, Johnson M, and Magner J. 1989 Concanavalin-A, lentil, and ricin lectin affinity binding characteristic of human thyrotropin: differences in the sialylation of thyrotropin in sera of euthyroid, primary and central hypothyroid patients. J Clin Endocrinol Metab. 69:985-995.
  5. Taylor T and Weintraub B. 1989 Altered thyrotropin (TSH) carbohydrate structures in hypothalamic hypothyroidism created by paraventricular nuclear lesions are corrected by in vivo TSH-releasing hormone administration. Endocrinology. 125:2189-2203.
  6. Magner J, Klibanski A, Fein H, et al. 1992 Ricin and lentil lectin affinity chromatography reveals oligosaccharide heterogeneity of thyrotropin secreted by 12 human pituitary tumors. Metabolism. 41:1009-1015.

Androgen Deficiency and It's Therapy in HIV-Infected Men

Pituitary Home

Steven Grinspoon, M.D.
by Steven Grinspoon, M.D.

NEPTCC Newsletter MGH Neuroendocrine Center Bulletin Vol 4, Issue 3, Winter 1998

Androgen Deficiency

Among men with HIV disease, gonadal dysfunction is highly prevalent. In early studies, over half of all men with AIDS demonstrated low testosterone levels. The prevalence of hypogonadism increases with severity of illness, and is most often associated with normal gonadotropin levels in 75% of cases. Potential mechanisms of secondary hypogonadism in HIV-infected men include undernutrition and chronic illness which may reduce GnRH pulsatility, inducing a state of hypogonadotropic hypogonadism. In addition, medications are a potential cause of secondary hypogonadism in this population. Megestrol acetate, a synthetic progestational agent, may result in profound hypogonadism in treated patients. More rarely, mass lesions of the pituitary and hypothalamus are seen and may represent opportunistic infections, such as toxoplasmosis or HIV-related malignancies, including lymphoma. Primary hypogonadism is often idiopathic, but may also be related to opportunistic infection or malignancy. More recent data suggest a lower prevalence of 25% for hypogonadism among HIV-infected men with advanced HIV disease. However, hypogonadism remains prevalent among HIV-infected men, even in the era of highly active antiviral therapy or HAART.

The diagnosis of hypogonadism is best made by determination of the serum free or bioavailable testosterone levels, because of increased SHBG levels in HIV disease. Measurement of gonadotropin levels is useful to differentiate primary from secondary hypogonadism, and in secondary hypogonadism, gonadotropins will be inappropriately low or normal in the setting of low testosterone levels. Imaging of the hypothalamus and pituitary glands is recommended in the setting of headache, visual changes or in association with other signs and symptoms of pituitary disease.

The sequelae of hypogonadism among HIV-infected patients include decreased muscle mass and functional capacity, fatigue and reduced quality of life. Decreased lean body mass is a significant predictor of increased mortality and reduced survival among HIV-infected patients and is therefore an important clinical endpoint in this population. Among men with AIDS wasting, testosterone levels are highly correlated with lean body mass. Furthermore, such patients demonstrate a disproportionate loss of muscle mass, in comparison to weight. Importantly, significant loss of muscle mass or sarcopenia is seen even among stable, protease inhibitor-treated patients, in whom there is a direct correlation between muscle mass and functional status.

The effects of testosterone administration in hypogonadal HIV-infected men were recently investigated. In a randomized, placebo-controlled trial, testosterone was administered at 300 mg intramuscularly every 3 weeks for 6 months. Muscle and lean body mass increased significantly in the testosterone-treated relative to the placebo-treated patients by approximately 3 kg (Figure 1). Importantly, patients reported a subjective benefit with respect to improved overall quality of life, appearance and well being. The testosterone was well tolerated and without adverse effects. The relative change in lean body mass in response to testosterone is equivalent to or greater than that of other anabolic agents used in the AIDS wasting syndrome, including growth hormone. The use of physiologic testosterone administration (200-300 mg IM q 2-3 weeks) for hypogonadal HIV-infected men is now routine, and all such patients, particularly men with the wasting syndrome, should be screened and treated for hypogonadism when appropriate

Transdermal delivery of testosterone is now an alternative to intramuscular dosing for HIV-infected patients. Two such transdermal products, Androderm (r) and Testoderm (r) are now commercially available, each with a recommended dose of 5 mg/day. Bhasin et al. recently showed a beneficial effect of transdermal testosterone administration (5 mg/day) to significantly increase lean body mass by 1.4 kg over 3 months in hypogonadal men with HIV infection. Potential advantages of transdermal dosing include more stable, steady state testosterone levels. However, further studies are necessary to insure that mean testosterone levels are sufficient in response to transdermal dosing. At the current time, either transdermal or IM therapy is recommended for hypogonadal men with HIV-infection. However, it is recommended that serum testosterone levels be checked at least once after initiation of transdermal therapy to insure adequate levels within the normal range.

A commonly asked question regarding testosterone administration in HIV-infected men is the appropriate duration of therapy. A sustained anabolic effect of testosterone administration on lean body mass of 3.7 kg or 7.6% over 12 months was previously demonstrated. No adverse effects on the prostate or PSA were seen and hematocrit increased 3.5% over this time period. These data suggest that continuation of testosterone for at least one year is beneficial and results in sustained increases in lean body mass. Serial monitoring of the prostate is important with long-term testosterone administration. Among patients who have achieved stable weight and are less ill, discontinuation of testosterone and reassessment of gonadal function is appropriate, as androgen levels may improve with nutritional and immunologic recovery.

An important question is the efficacy of androgen therapy in eugonadal men with the wasting syndrome. At the current time, such supraphysiologic dosing cannot be endorsed because of the potential risks related to prostate, mood and acne. However, studies performed under carefully controlled conditions and with appropriate monitoring are now underway to determine the efficacy of larger dose of testosterone in HIV-infected patients. A final question concerns the role of exercise therapy, in combination with anabolic therapy, to increase functional status in HIV-infected men. Data in non HIV-infected men suggest an additive effect of combined androgen and exercise therapy, but combined anabolic strategies have not been previously studied among men with AIDS wasting.

In summary, recent data suggest that hypogonadism remains an important clinical problem among HIV-infected men, even in the setting of potent antiviral agents. Hypogonadism in HIV-infected men is most often associated with low or inappropriately normal gonadotropin levels but is not often associated with a sellar mass lesion. Recent studies suggest a potent and sustained benefit of androgen therapy to reverse significant underlying muscle loss in hypogonadal HIV-infected men. At the current time, HIV infected patients with evidence of weight loss or muscle weakness and/or clinical symptoms of hypogonadism should be screened and treated for hypogonadism. The use of testosterone to increase lean body mass in eugonadal patients, must be considered experimental until further data are obtained.

[Bar Graph]

Figure Legend: Mean changeñSEM for fat-free mass assessed by dual energy x-ray absorptiometry, lean body mass determined from potassium-40 isotope analysis, and muscle mass from urinary creatinine excretion in patients who received testosterone (left) and placebo (right) over 6 months. * P less than 0.05 and **P less than 0.01 for the change from baseline between the testosterone group and the placebo group by analysis of covariance. n="number" of patients for whom paired end of study data are available. Reprinted from Reference 5 (below) with permission of the American College of Physicians.

  1. Bhasin S, Storer TW, Asbel-Sethi N, et al. 1998 Effects of testosterone replacement with a nongenital, transdermal system, Androderm, in human immunodeficiency-virus infected men with low testosterone levels. J Clin Endocrinol Metab. 83:3155-3162.
  2. Dobs AS, Dempsey MA, Ladenson PW, and Polk BF. 1988 Endocrine disorders in men infected with human immunodeficiency virus. Am J Med. 84:611-6.
  3. Grinspoon SK and Bilezikian JB. 1992 HIV disease and the endocrine system. N Engl J Med. 327:1360-1365.
  4. Grinspoon S, Corcoran C, Lee K, et al. 1996 Loss of lean body mass and muscle mass correlates with androgen levels in hypogonadal men with acquired immunodeficiency syndrome and wasting. J Clin Endocrinol Metab. 81:4051-4058.
  5. Grinspoon S, Corcoran C, Askari H, et al. 1998 Effects of androgen administration in men with aids wasting: a randomized, placebo-controlled, trial. Ann Int Med. 129:18-26.
  6. Grinspoon S, Corcoran C, Anderson E, Hubbard J, Basgoz N, and Klibanski A. 1998 Sustained anabolic effects of long-term androgen administration in men with AIDS wasting. Clin Inf Dis. In press.