Patients with pituitary tumors who become pregnant may present with symptoms that suggest pituitary enlargement such as headaches, nausea, or visual disturbances. However, the physical signs that are typically present with large pituitary tumors such as bitemporal hemianopsia, diplopia, or often signs of III, IV, VI nerve palsies, are not present.
Not all pituitary enlargement during pregnancy indicates the presence of a tumor. During pregnancy, the normal pituitary gland enlarges considerably, due to the estrogen-stimulated lactotroph hyperplasia; on magnetic resonance imaging (MRI) scans, the peak size of up to 12 mm occurs in the first few days postpartum.
Macroadenomas, both hormone-producing and nonfunctioning, often cause varying degrees of hypopituitarism. Loss of normal luteinizing hormone (LH) and follicle-stimulating hormone (FSH) secretion is common. Therefore, to achieve ovulation, these hormones may need to be given; in vitro fertilization is often employed.
Women with prolactinomas are usually infertile due to inhibition of pulsatile gonadotropin releasing hormone secretion. Normalization of prolactin (PRL) levels with dopamine agonists or surgery is required before ovulation and conception can occur. The high estrogen levels of pregnancy can stimulate prolactinoma growth which may cause problems.
About two-thirds of women with acromegaly are infertile due to coexistent hyperprolactinemia in about 20%, hypopituitarism in about 10%, growth hormone (GH) excess in about 15%, and multiple causes in the remainder. These hormonal abnormalities need to be corrected in many cases before ovulation can occur. Less than 150 women with active acromegaly and pregnancy have been reported.
Most women with Cushing’s disease are anovulatory and infertile and their adrenocorticotropic hormone (ACTH) and cortisol hypersecretion usually needs to be corrected before conception can occur. Fewer than 150 women with active Cushing’s disease and pregnancy have been reported.
Diagnosing Cushing’s syndrome during pregnancy may be difficult. Both conditions may be associated with weight gain in a central distribution, fatigue, edema, emotional upset, glucose intolerance, and hypertension. The striae associated with normal pregnancy, however, are usually pale and red or purple in Cushing’s syndrome. Hirsutism and acne indicating excessive androgen production, proximal myopathy, and bone fractures all point to Cushing’s syndrome.
Only 3 cases of pregnancy in women with thyroid stimulating hormone (TSH)-secreting adenomas have been reported. Issues during pregnancy include the hyperthyroidism as well as tumor size, as most TSH-secreting adenomas are large.
Women with clinically nonfunctioning adenomas (CNFAs) are anovulatory only if the tumor is large enough (usually >1 cm in diameter) to cause hypopituitarism. Few women with CNFAs and pregnancy have been reported.
Key Laboratory and Imaging Tests
Patients with macroadenomas need to be evaluated for hypopituitarism in the usual fashion. Usually this will have been done prior to pregnancy and appropriate treatment begun; only monitoring and adjustments of hormone doses are then needed during pregnancy. During pregnancy, analog free thyroxine (T4) assays are not reliable, and total T4 levels should be monitored and kept near the upper limit of normal. Because the hypopituitarism is the cause of the hypothyroidism, measurement of TSH levels is not useful.
As with other forms of hypocortisolism, measurement of urine cortisol levels during pregnancy is not useful. Serum bound and free cortisol levels increase during pregnancy, and appropriate normal values for each trimester have not been well established, nor have the results of various stimulation tests. As GH treatment is generally not continued during pregnancy, its assessment is not performed.
Not all women with hyperprolactinemia have prolactinomas and other causes need to be excluded (see Table I). PRL levels rise gradually throughout gestation in normal individuals with levels peaking at over 200 ng/ml. This increase in PRL prepares the breast for lactation. In women with prolactinomas, PRL levels may rise during gestation with no change in tumor size and conversely, may remain stable despite an increase in tumor size. Therefore, routine measurement of PRL levels during pregnancy is inadvisable. Only 2.4% of microadenomas enlarge during pregnancy, so women harboring them can be followed symptomatically.
Visual field testing and MRI scans are performed only in women with visual symptoms or progressive headaches. In contrast, 21% of macroadenomas enlarge during pregnancy; in such women formal visual field testing is warranted each trimester and MRI scans done when clinically indicated. Usually gadolinium is not given because of caution by the neuroradiologists, albeit there is no evidence of gadolinium-induced teratogenicity or other adverse effects. With significant tumor growth, a re-evaluation for hypopituitarism may be necessary.
The diagnosis of acromegaly during pregnancy is complicated by the fact that beginning in the second half of pregnancy, circulating levels of a GH variant (GHv) made by the syncytiotrophoblastic epithelium of the placenta increase, and pituitary GH secretion decreases due to negative feedback effects of insulin-like growth factor 1 (IGF-1). In patients with acromegaly who have autonomous GH secretion and become pregnant, both forms of GH persist in the blood but conventional assays cannot distinguish between them.
If it is critical to make a diagnosis during pregnancy, it may be possible by demonstrating GH pulsatility with frequent sampling, as GH secretion in acromegaly is highly pulsatile, but that of the variant is not. Neither the GHv nor pituitary GH from a GH-secreting tumor are suppressible by an exogenous glucose load.
Only four patients with tumors secreting GH have been reported to have enlargement of their tumors with a resultant visual field defect in one during pregnancy. Therefore, patients with acromegaly with macroadenomas should be monitored for symptoms of tumor enlargement, but visual field testing and MRI’s are done only when clinically indicated.
The diagnosis of Cushing’s syndrome during pregnancy is complicated by the finding that cortisol levels rise progressively over the course of gestation in normal individuals, due both to the estrogen-induced increase in cortisol binding globulin (CBG) levels, and an increase in cortisol production. Therefore, the bioactive “free” fraction as assessed by urinary free cortisol levels are increased 2-3 fold. ACTH levels coming from placental production may be elevated in normal pregnancy and may not be suppressed in Cushing’s syndrome due to adrenal causes.
The overnight dexamethasone test usually demonstrates inadequate suppression during normal pregnancy. A persistent circadian variation in the elevated levels of total and free serum cortisol during normal pregnancy may be most helpful in distinguishing Cushing’s syndrome from the hypercortisolism of pregnancy, as this finding is characteristically absent in all forms of Cushing’s syndrome. Midnight levels of salivary cortisol during pregnancy have not been standardized.
Most patients with Cushing’s disease have microadenomas and MRI scanning of the pituitary is of little use because many ACTH-producing adenomas may be too small to see, and because of the high frequency of pituitary incidentalomas. Little experience has been reported with CRH stimulation testing or petrosal venous sinus sampling during pregnancy.
It is very important to localize whether there is an adrenal source of the hypercortisolism vs. an ACTH-producing pituitary tumor, in order to initiate the correct therapy. This differential diagnosis may be exceedingly difficult due to the nonsuppressibility of ACTH in all forms of Cushing’s syndrome in pregnancy, and usually multiple tests may need to be done, including adrenal ultrasound.
TSH-secreting tumors have as their distinguishing biochemical characteristic normal-to-elevated TSH levels in the face of overtly elevated thyroid hormone levels. Total thyroid hormone levels rather than the free fractions should be assessed during pregnancy. Because these tumors are usually quite large, assessment by formal visual fields in each trimester is warranted, with MRI reserved for visual field or clinical symptomatology indicating tumor growth.
CNFAs by definition have no tumor hypersecretion that needs monitoring. Only 2 cases have been reported with tumor enlargement during pregnancy. Visual field testing is indicated for patients with macroadenomas with suprasellar extension, but MRI is reserved for those who develop visual field abnormalities or progressive headaches.
Management and Treatment of the Disease
Hyperprolactinemia causes anovulation and PRL levels must be brought to normal or near-normal levels to allow ovulation to occur. The choice of therapy has important consequences for decisions regarding pregnancy. The dopamine agonists, bromocriptine and cabergoline, are usually considered the primary mode of therapy, restoring ovulatory menses in about 80% and 90% respectively.
In addition, macroadenoma size can be reduced by >50% in 50-75% of patients with bromocriptine, and in >90% of patients with cabergoline. Usually the dopamine agonist is stopped once a woman has missed her menstrual period and pregnancy is diagnosed in order to limit fetal exposure.
Transsphenoidal surgery can normalize PRL levels in 80-90% of patients with microadenomas when performed by an experienced pituitary neurosurgeon. However, there is a recurrence rate of hyperprolactinemia of about 20% so that the true long-term remission rate is only 50-60%. Hypopituitarism is very rare when it is performed on women with microadenomas (tumors <10 mm in diameter). For patients with macroadenomas (tumors >10 mm in diameter), surgical cure rates are considerably lower and there is a considerably greater risk of causing hypopituitarism.
About 20-30% of patients with acromegaly have hyperprolactinemia and treatment with a dopamine agonist may be necessary to permit ovulation and conception in these patients. Most patients with acromegaly are treated with surgery as primary therapy and those not cured by surgery are usually treated medically with the somatostatin analogs octreotide and lanreotide. Cabergoline and pegvisomant may also be helpful in some cases.
The considerations regarding the use of bromocriptine and cabergoline in women with prolactinomas also apply to those with acromegaly. Only 36 patients have been reported who were treated during pregnancy with somatostatin analogs and no malformations were found in their children. However, somatostatin analogs bind to somatostatin receptors in the placenta and cross the placenta, and therefore can affect developing fetal tissues where somatostatin receptors are widespread, especially in the brain.
Because of the limited data documenting safety, octreotide and other somatostatin analogs usually should be discontinued if pregnancy is considered and contraception should be used when these drugs are administered. Considering the prolonged nature of the course of most patients with acromegaly, interruption of medical therapy for 9-12 months should not have a particularly adverse effect on the long-term outcome. On the other hand, these drugs can control tumor growth, and for enlarging tumors, their reintroduction during pregnancy may be warranted vs. operating. Pegvisomant, a GH receptor antagonist, has been given to only 2 patients with acromegaly during pregnancy and the safety of this approach has not been established.
It is clear from several analyses of pregnant patients with Cushing’s syndrome that the frequency of live births is increased and the frequency of prematurity is decreased when active treatment is instituted by a gestational age of 20 weeks. Although any surgery poses risks for the mother and fetus, it appears that with Cushing’s syndrome, the risks of not operating are considerably higher than those of proceeding with surgery.
Transsphenoidal resection of pituitary ACTH-secreting adenomas has been carried out successfully in several patients during the second trimester. Medical therapy for Cushing’s syndrome during pregnancy with metyrapone and ketoconazole is not very effective and intrauterine growth retardation has been reported with ketoconazole. Furthermore, hepatotoxicity with ketoconazole limits its use. Aminoglutethimide and mitotane should be avoided because of teratogenicity and mifepristone is an abortifacient.
There are no safety data for the use of pasireotide during pregnancy, and the cautions discussed above for acromegaly regarding the use of somatostatin analogs during pregnancy also applies to patients with Cushing’s disease. Thus, transsphenoidal surgery is the preferred treatment. In patients not cured by surgery, ketoconazole probably is the best treatment with care being taken to monitor liver function tests.
For women with TSH-secreting tumors who become pregnant, the most pressing issue is the need to control hyperthyroidism and that can usually be done with standard antithyroid drugs. Generally, propylthiouracil is preferred during the first trimester to avoid teratogenicity, and methimazole in the second and last trimester to avoid liver toxicity. However, with growing macroadenomas, somatostatin analogs may be necessary for tumor size control and may be necessary to control the hyperthyroidism if thionamides are ineffective.
Pregnancy would not be expected to influence tumor size in patients with CNFAs, and only 2 cases have been reported in which tumor enlargement during pregnancy resulted in a visual field defect. In 1 case, the patient responded rapidly to bromocriptine treatment, probably due to shrinkage of the lactotroph hyperplasia with decompression of the chiasm, and probably with little or no direct effect on the tumor itself. Thus dopamine agonists can be tried if there is evidence of clinically significant tumor enlargement, and surgery performed, if there is no response to the dopamine agonist.
Hypopituitarism in patients with large tumors may be due to the tumor itself or to therapeutic modalities such as surgery or irradiation. The hypopituitarism may be partial or complete and loss of gonadotropin secretion is common. Induction of ovulation may be difficult and a variety of techniques have been used including administration of human chorionic gonadotropin (hCG) and FSH, pulsatile gonadotropin releasing hormone (GnRH), and in vitro fertilization.
Once pregnancy is established, of course, there is no need for further use of gonadotropins or estrogen. Because of increased thyroxine turnover and volume of distribution in pregnancy, T4 levels usually fall and TSH levels rise with a fixed thyroxine dose over the course of gestation. The average increase in thyroxine need in these patients is about 0.05 mg/day. Because patients with hypothalamic/pituitary dysfunction may not elevate their TSH levels normally in the face of increased need for thyroxine, it is reasonable to increase the thyroxine supplementation by 0.025 mg during the first trimester and by additional 0.025 mg in the second trimester, also following total, rather than free T4 levels, which are less reliable during pregnancy.
The dose of chronic glucocorticoid replacement does not usually need to be increased during pregnancy. Hydrocortisone is metabolized by the placental enzyme type 2 11β-hydroxysteroid dehydrogenase, so the fetus is generally protected from any overdose of hydrocortisone; the usual dose is in the range of 15-20 mg/day given in 2 or 3 divided doses. Additional glucocorticoids are needed for the stress of labor and delivery, such as 75 mg of hydrocortisone IV every 8 hours with rapid tapering postpartum. Prednisolone does not cross the placenta and prednisone crosses only minimally.
There is little data on the use of GH during pregnancy in hypopituitary individuals and in most series, GH therapy has been stopped at conception. As the GHv, which is biologically active, is produced by the placenta in substantial amounts beginning in the second half of pregnancy and can access the maternal circulation, then at most the mother would be GH-deficient only in the first half of pregnancy.
Patients with prolactinomas who are treated with dopamine agonists generally do well. When the drugs are stopped after pregnancy is diagnosed, no increase in spontaneous abortions, ectopic pregnancies, trophoblastic disease, multiple pregnancies, or malformations were found in over 6,000 pregnancies in which bromocriptine was used, and almost 800 pregnancies in which cabergoline was used. Although the safety database of continuous dopamine agonist therapy during pregnancy is very small, the data suggests that it probably is not harmful.
The stimulatory effect of the hormonal milieu of pregnancy and the withdrawal of the dopamine agonist may result in significant prolactinoma enlargement requiring intervention in 2.4% of women with microadenomas, 21% of those with macroadenomas that had not undergone prior surgery or radiotherapy, and 4.7% of those with macroadenomas that had undergone prior surgery or radiotherapy. In almost all cases, such enlargement can be successfully treated with reinstitution of a dopamine agonist. If the pregnancy is sufficiently advanced, another approach is to deliver the baby. Surgical decompression is only undertaken if the other approaches fail.
In patients with acromegaly, because of the GH-induced insulin resistance, the risk of gestational diabetes is increased. Gestational hypertension is also increased, but cardiac disease has not proved to be an issue, probably because of the relatively young age of women who became pregnant. Standard treatments for gestational diabetes and hypertension can be used without any special considerations because of the acromegaly.
Cushing’s syndrome is associated with a pregnancy loss rate of 25% due to spontaneous abortion, stillbirth, and early neonatal death because of extreme prematurity. The passage of cortisol across the placenta may rarely result in suppression of the fetal adrenals. Hypertension develops in most mothers with Cushing’s, and diabetes and myopathy are frequent. Postoperative wound infection and dehiscence are common after cesarean section. Because of all the adverse effects to mother and baby, treatment during pregnancy is advised. Standard treatments for gestational diabetes and hypertension can be used without any special considerations because of the Cushing’s syndrome.
Hyperthyroidism due to TSH-secreting tumors can be associated with prematurity and lack of maternal weight gain. However, too few cases have been reported to determine the frequency of such complications.
Most CNFAs are actually gonadotroph adenomas. Usually, tumors that result in symptoms and signs are large, causing mass effects of headaches, visual disturbance, and hypopituitarism. Two patients have been reported who had gonadotroph adenomas secreting intact FSH with a resultant ovarian hyperstimulation syndrome; both became pregnant, 1 after having the FSH hypersecretion controlled by bromocriptine, and the second following surgical removal of the tumor.
Although the malformation rate is normal in pregnancies complicated by hypopituitarism, there seems to be an increased frequency of cesarean sections, miscarriages, and small for gestational age babies.
Most patients with pituitary tumors who become pregnant are under the chronic care of an endocrinologist. This physician is responsible for directing the overall care of the patient and the one to decide on which approach is necessary for patients who wish to become pregnant. However, when a decision has been made to try to achieve pregnancy, consultation with an obstetrician is important. Depending upon the experience of the endocrinologist in dealing with this combination of events, referral to a high-risk obstetrician may be helpful. Decisions regarding surgery are made in conjunction with an experienced pituitary neurosurgeon.
Rarely, patients are first diagnosed with a pituitary tumor during pregnancy where the obstetrician is generally in charge of care and seeks referral from the endocrinologist. It is only after endocrinology consultation that a neurosurgeon might be consulted.
What’s the Evidence?/References
Casanueva, FF, Molitch, ME, Schlechte, JA, Abs, R. “Guidelines of the Pituitary Society for the diagnosis and management of prolactinomas”. Clin Endocrinol (Oxf). vol. 65. 2006 Aug. pp. 265-73.
Melmed, S, Casanueva, FF, Hoffman, AR, Kleinberg, DL. “Diagnosis and treatment of hyperprolactinemia: an Endocrine Society clinical practice guideline”. J Clin Endocrinol Metab. vol. 96. 2011. pp. 273-88.
Katznelson, L, Laws, ER, Melmed, S, Molitch, ME. “Endocrine Society. Acromegaly: an endocrine society clinical practice guideline”. J Clin Endocrinol Metab. vol. 99. 2014. pp. 3933-3951.
Nieman, LK, Biller, BM, Findling, JW, Newell-Price, J. “The diagnosis of Cushing's syndrome: an Endocrine Society Clinical Practice Guideline”. J Clin Endocrinol Metab. vol. 93. 2008. pp. 1526-40.
Stagnaro-Green, A, Abalovich, M, Alexander, E, Azizi, F. “The American Thyroid Association Taskforce on Thyroid Disease During Pregnancy and Postpartum. Guidelines of the American Thyroid Association for the Diagnosis and Management of Thyroid Disease During Pregnancy and Postpartum”. Thyroid. vol. 21. 2011. pp. 1081-1125.
Karaca, Z, Kelestimur, F. “Pregnancy and other pituitary disorders (including GH deficiency)”. Best Pract Res Clin Endocrinol Metab. vol. 25. 2011. pp. 897-910.
Molitch, ME. “Pituitary disorders during pregnancy”. Endocrinol Metab Clin N Amer. vol. 34. 2006. pp. 99-116. (Two overall reviews of this subject that go into more depth on each condition.)
Molitch, ME. “Endocrinology in pregnancy: management of the pregnant patient with prolactinoma”. Eur J Endocrinol. vol. 172. 2015. pp. R205-R213. (Up-to-date review of the most common pituitary tumor causing problems during pregnancy.)
Cheng, V, Faiman, C, Kennedy, L, Khoury, F. “Pregnancy and acromegaly: a review”. Pituitary. vol. 15. 2012. pp. 59-63.
Cheng, S, Grasso, L, Martinez-Orozco, JA, Al-Agha, R. “Pregnancy in acromegaly: experience from two referral centers and systematic review of the literature”. Clin Endocrinol (Oxf). vol. 76. 2012. pp. 264-271. (Two reviews of acromegaly and pregnancy.)
Caron, P, Broussaud, S, Bertherat, J, Borson-Chazot, F. “Acromegaly and pregnancy: a retrospective multicenter study of 59 pregnancies in 46 women”. J Clin Endocrinol Metab. vol. 95. 2010. pp. 4680-7. (The largest series reported to date of patients with acromegaly who get pregnant.)
Maffei, P, Tamagno, G, Nardelli, GB, Videau, C. “Effects of octreotide exposure during pregnancy in acromegaly”. Clin Endocrinol (Oxf). vol. 72. 2010. pp. 668-77. (A discussion of the effects of octreotide during pregnancy.)
Lindsay, JR, Nieman, LK. “The hypothalamic-pituitary-adrenal axis in pregnancy: challenges in disease detection and treatment”. Endocr Rev. vol. 26. 2005. pp. 775-99. (An overall review of the changes in the HPA axis during pregnancy and the problems faced in making diagnoses.)
Lindsay, JH, Jonklaas, J, Oldfield, EH, Nieman, LK. “Cushing’s syndrome during pregnancy: personal experience and review of the literature”. J Clin Endocrinol Metab. vol. 90. 2005. pp. 3077-83. (A presentation of several cases and a detailed review of the literature on Cushing’s syndrome during pregnancy.)
Vilar, L, Freitas Mda, C, Lima, LH, Lyra, R. “Cushing’s syndrome in pregnancy: an overview”. Arq Bras Endocrinol Metabol. vol. 51. 2007. pp. 1293-1302. (A nice review of the literature on Cushing’s syndrome in pregnancy.)
Negro, R, Mestman, JH. “Thyroid disease in pregnancy”. Best Pract Res Clin Endocrinol Metab. vol. 25. 2011. pp. 927-43. (These authors review the changes occurring in the thyroid during pregnancy and the management of thyroid dysfunction during pregnancy.)
Kübler, K, Klingmüller, D, Gembruch, U, Merz, WM. “High-risk pregnancy management in women with hypopituitarism”. J Perinatol. vol. 29. 2009. pp. 89-95. (This paper reviews aspects of the management of hypopituitarism during pregnancy.)
Canfield, MA, Honein, MA, Yuskiv, N, Xing, J. “National estimates and race/ethnic-specific variation of selected birth defects in the United States, 1999-2001”. Birth Defects Res A Clin Mol Teratol. vol. 76. 2006. pp. 747-56. (This paper provides data on adverse outcomes in the general population, for comparison with data from patients with various conditions.)
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.