The majority of pituitary tumors in pregnant women are microprolactinomas (diameter less than 10 mm). Macroprolactinomas (diameter 10 mm or greater) are less common, affecting approximately 24% of pregnant women with prolactinoma,1 and the precise incidence of these tumors in pregnancy is unknown. Other pituitary tumors occur more rarely in pregnant women and may be functioning (producing growth hormone, adrenocorticotrophic hormone, or thyroid-stimulating hormone) or nonfunctioning pituitary adenomas.
Pregnant women with microprolactinomas typically have uncomplicated pregnancies with no symptoms of tumor enlargement.1 Among all prolactinomas, approximately 46% enlarge during pregnancy2; symptomatic enlargement occurs less frequently. Approximately 30% of macroprolactinomas undergo symptomatic enlargement, whereas only 3% of the more common microprolactinomas do. If women with macroprolactinomas are treated prenatally with surgery or radiation, their risk of tumor enlargement is markedly reduced.3
Symptoms consistent with enlargement are reported in five of nine cases of nonfunctioning pituitary adenoma reported in the literature.4–7 We identified three reported women with thyrotrophinomas,8–10 two of whom had gestational tumor enlargement. Three retrospective series described 72 pregnancies in 58 women with acromegaly. Maternal complications included type 2 diabetes, gestational diabetes mellitus, pregnancy-induced hypertension, and preeclampsia and 8% had symptomatic enlargement.11,12 Combined data from 136 pregnancies in 122 women with Cushing's disease demonstrated a high rate of adverse outcomes, including preterm labor (43%), intrauterine growth restriction (21%), stillbirth (6%), preeclampsia (14%), diabetes, or impaired glucose tolerance (25%).13
The aim of this study was to describe the incidence, characteristics, management, and outcomes of a U.K. national case series of pregnant women with pituitary tumors.
MATERIALS AND METHODS
A national, prospective, observational, population-based case series study was undertaken over 3 years (March 2010 to February 2013) using the U.K. Obstetric Surveillance System. The U.K. Obstetric Surveillance System is a research platform with reporting clinicians in all U.K. consultant-led maternity units designed to study rare complications of pregnancy on a national basis. The U.K. Obstetric Surveillance System utilizes a prospective monthly case collection scheme that includes all 202 consultant-led obstetric units in the United Kingdom. Women are not contacted directly and no personally identifiable information is collected. Because all women with pituitary tumors should have consultant-led care, we anticipate that the study is likely to have covered all women with pituitary tumors within the entire U.K. birth cohort. The study was approved by the Riverside Ethics Committee, London (reference number 09/H0706/78).
Exposed women were defined as those with macroprolactinomas (10 mm or greater), nonfunctioning pituitary adenomas, Cushing's disease, acromegaly, or thyrotropinomas diagnosed before or during pregnancy. Because the U.K. Obstetric Surveillance System is limited to the study of rare diseases (incidence of less than 1 in 2,000 deliveries), women with microprolactinoma (less than 10 mm in diameter) were excluded from the study.
Women who met the criteria for the study were identified by the obstetrician or endocrinologist responsible for their care. On reporting a woman with a pituitary tumor, the clinician was asked to complete a data collection form. Anonymized data were collected about pituitary tumor diagnosis, monitoring and management before pregnancy and antenatally as well as maternal demographics, obstetric and medical history, delivery, and perinatal outcomes. Thirty-eight women were excluded (Fig. 1) because they did not meet the entry criteria (n=16) or a data collection form was reported in error, duplicated, or notes could not be found (n=22). Two comparison groups were used. The first comparison group included women in an established U.K. Obstetric Surveillance System database with an uncomplicated singleton (n=2,205) or twin (n=27) pregnancy. The U.K. Obstetric Surveillance System comparison group comprised pregnant women from whom data had previously been collected between February 2005 and February 2006. They included the two women delivering in the same hospital before U.K. Obstetric Surveillance System cases with the conditions under study at that time (antenatal pulmonary embolism, eclampsia, or peripartum hysterectomy). They did not have the conditions being studied for the U.K. Obstetric Surveillance System and were not matched for gestational week. Data were collected retrospectively by the U.K. Obstetric Surveillance System investigators from their medical charts on any complications of pregnancy they experienced. These data were collected in an identical manner to the data on exposed women through specific data collection forms completed by clinical staff; when forms were returned with invalid or out-of-range responses, clinicians were contacted and asked to correct the information. The control women did not have any known pituitary pathology at the time they were identified. These data were collected 5 years before commencement of the U.K. Obstetric Surveillance System Pituitary Study. Information from the Office of National Statistics in England for the years of the study was used for a second comparison group to calculate incidence rates of each tumor in pregnancy. The Office of National Statistics database is based on statutory birth and death registrations and includes the information recorded on birth and death certificates; no further validation of these data takes place. For the Office of National Statistics data, we could not separate singleton and multiple pregnancies and this should be taken into account in pregnancy outcome comparisons.
The study was advertised at the U.K. Society for Endocrinology Annual Conference and all U.K. endocrinologists were contacted by email and by post to ask them to inform their local U.K. Obstetric Surveillance System team of any pregnant women with pituitary tumors during the study period. No additional cases were identified. The main outcome measures were the incidence, management, and frequency of adverse maternal and offspring outcomes of pituitary tumors in pregnancy.
Statistical analysis was performed using Stata 12.1. P values were considered significant if ≤.05. Women with multiple pregnancies were included in comparisons of sociodemographic data and to calculate the incidence of pituitary tumors in the United Kingdom, but women with multiple pregnancies were excluded from the U.K. Obstetric Surveillance System control group for pregnancy outcome comparisons, because this group has an increased risk of both maternal and neonatal adverse outcomes. For Office of National Statistics control data, it was not possible to remove multiple pregnancies from the data set. Incidence rates were estimated with 95% confidence intervals (CI). Adjustment for variables such as maternal age or occupation was not possible for comparison of adverse pregnancy outcomes as a result of small numbers (less than eight for all outcomes). Presence or absence of tumor expansion was compared between different disease groups using Fisher exact test. For categorical data, unadjusted risk ratios (RRs)14 were used with Fisher exact P values (more appropriate for small sample sizes) and risk differences. For continuous data, mean differences and t tests were calculated.15 No woman included in the analysis had data included from repeat pregnancies.
During the 3-year study period, there were 71 confirmed cases of the pituitary tumors included in this study in pregnant women (Fig. 1). There were 2,703,102 maternities (defined as women giving birth to one or more liveborn or stillborn neonates after 24 weeks of gestation) in the United Kingdom in the same study period (from March 2010 to February 2013). The estimated incidence of each of the tumors in pregnant women is shown in Table 1. Because the majority of women identified had macroprolactinomas and nonfunctioning pituitary tumors, the principal focus of the study relates to these cases. There were three cases of Cushing's disease and three of acromegaly that were not included in subsequent analysis. Only one of these six women had a normal delivery at term; the other five pregnancies resulted in surgical termination, ectopic pregnancy, or unexplained stillbirth.
Affected women were more likely to be older than women in the U.K. Obstetric Surveillance System control group (mean difference 4.06, CI 2.76–5.36, P<.001) (Table 2). There were no differences in ethnic group or body mass index.
The presenting symptoms differed between women with macroprolactinoma and nonfunctioning pituitary adenoma (Fig. 2). Ninety-eight percent (48/49) of macroprolactinomas and 75% (12/16) of nonfunctioning pituitary adenomas were identified before pregnancy. Women with a macroprolactinoma were more likely to present with symptoms of hormone excess such as amenorrhea (RR 5.06, CI 1.36–18.8, P<.001) and galactorrhea (RR 2.45, CI 0.63–9.6, P=.20). In contrast, women with nonfunctioning pituitary adenoma were more likely to present with clinical features consistent with tumor mass expansion such as visual symptoms (RR 4.59, CI 1.48–14.3, P=.011).
There were 10 patients with tumor expansion in pregnancy; six had a macroprolactinoma and four a nonfunctioning pituitary adenoma (Table 3). Apart from previous treatment with surgery, radiotherapy, or both, there were no significant factors in the demographic or previous obstetric features to predict tumor expansion. Of the 60 women in whom a macroprolactinoma or nonfunctioning pituitary adenoma was diagnosed before pregnancy, nine had undergone surgery, radiotherapy, or both before pregnancy. None of this group had tumor expansion during pregnancy. Of the remaining 51 patients who did not have surgery or radiotherapy before pregnancy, seven (13.7%) had tumor expansion, six of whom had symptoms (Table 3). All the patients with macroprolactinoma with symptomatic expansion were diagnosed before pregnancy; only one of the four patients with nonfunctioning pituitary adenoma with symptoms of tumor expansion was diagnosed prepregnancy.
There was no significant difference in the size of the macroprolactinoma before conception in those who did or did not have symptomatic tumor expansion (mean diameter 13.8 mm compared with 13.0 mm, respectively). The symptoms reported in association with tumor expansion included headache, visual impairment, and neurologic impairment (Table 3).
In the entire case series, 17 women had symptoms (11/49 macroprolactinomas and 6/16 nonfunctioning pituitary adenomas) suggestive of tumor expansion (headache, visual field changes, or specific neurologic symptoms as described in Table 3). The majority (80%, CI 51.9–96) of the patients with symptoms had further investigation with formal visual field testing. A larger proportion of women with macroprolactinomas (75.5%, CI 61.1–87) had formal visual field tests performed during pregnancy than women with nonfunctioning pituitary adenoma (56.3%, CI 29.9–80) (RR 1.34, CI 0.85–2.13, P=.21). There was no difference in the proportion of women with macroprolactinoma or nonfunctioning pituitary adenoma that had impaired visual fields.
Of the 60 women diagnosed before pregnancy, 46 of 48 (95.9%, CI 86.0–99.5) with a macroprolactinoma and none of the women with a nonfunctioning pituitary adenoma received medical treatment before they conceived (Table 4). Cabergoline was the most likely dopaminergic drug to be prescribed to those with macroprolactinoma before pregnancy. It was prescribed to 32 women (65.3%, CI 50.4–78.3) compared with bromocriptine (10 [20.4%], CI 10.2–34.3) or quinagolide (4 [8.2%], CI 2.3–19.6). Nineteen women stopped taking cabergoline before the end of the first trimester as did five women treated with bromocriptine. Eighteen women with a macroprolactinoma continued medication during the first trimester of pregnancy.
A larger proportion of women with macroprolactinoma had dopamine agonists prescribed during pregnancy (47%, CI 32–62) compared with only 12.5% of those with nonfunctioning pituitary adenoma (CI 1.6–38; RR 3.76, CI 0.99–14.2, P=.018). Only 6 of the 49 women with a macroprolactinoma took dopamine agonists throughout the entire pregnancy; one was treated with bromocriptine and five took cabergoline. Seven women (five with macroprolactinoma and two with nonfunctioning pituitary adenoma) started taking dopamine agonists de novo in pregnancy. Of the two women with nonfunctioning pituitary adenoma, one started taking cabergoline in the second trimester and one in the third trimester.
Of the 15 women with symptoms suggestive of tumor enlargement, six (40%, CI 16.3–67.7) were prescribed dopamine agonists in the first trimester.
The pregnancy outcome data are only presented for singleton pregnancies as a result of the known increased risk of adverse outcomes with multiple pregnancies.16 Overall women with macroprolactinoma or nonfunctioning pituitary adenoma in pregnancy did not have increased rates of adverse outcomes (preterm birth, pregnancy-induced hypertension, preeclampsia, or stillbirth) compared with women in the control group (Table 5).
Women with nonfunctioning pituitary adenoma had significantly more cesarean deliveries (50%) than those in the U.K. Obstetric Surveillance System (23%) (RR 2.14, CI 1.31–3.52, P=.032) and Office of National Statistics control groups, that is, those from the national comparison group (24%) (RR 2.06, CI 1.26–3.36, P=.035). In contrast, women with macroprolactinoma were no more likely to have cesarean deliveries (27%) compared with women in the U.K. Obstetric Surveillance System control group (RR 1.14, CI 0.71–1.82, P=.61) or women in the Office of National Statistics control group (23% compared with 24%; RR 1.09, CI 0.68–1.74, P=.74). Women with macroprolactinoma or nonfunctioning pituitary adenoma were more likely to have induction of labor than women in the Office of National Statistics control group (37% and 44%, respectively, compared with 20% of women in the control group) (RR 1.95, CI 1.44–2.66, P<.001).
Of the 10 women who had tumor expansion in pregnancy, four had spontaneous vaginal deliveries, one had an induction of labor and vaginal delivery, and five had cesarean deliveries. Of these five women, only two cesarean deliveries were for symptoms associated with raised intracranial pressure: one was the result of worsening vision, diabetes insipidus, and preeclampsia, and another woman had pituitary apoplexy. The other cesarean deliveries were performed as a consequence of failed induction and maternal request after previous cesarean delivery.
The numbers were too small in the exposed group of pregnant women to be able to make a robust statement about the risk of birth defects.
This large prospective U.K. case series study of macroprolactinomas and nonfunctioning pituitary adenomas in pregnant women has provided valuable insights into the presentation, course, and management of these tumors. One advantage of a case series drawn from a national cohort of known size is that it provides a more accurate estimate of the incidence of these rare tumors in pregnancy. Nonfunctioning pituitary adenoma is the second most frequently occurring pituitary macroadenoma in the nonpregnant population and therefore it is surprising that so few cases in pregnancy have been documented. In our study women with pituitary macroadenomas in pregnancy were likely to be older than women in the control group.
The pituitary gland normally increases in size in pregnancy as a likely consequence of estrogen-stimulated hyperplasia and hypertrophy of lactotroph cells.17,18 This prospective U.K. study identified a lower rate of symptomatic macroprolactinoma expansion during pregnancy than has previously been reported.3 This may be because 94% (46/49) of women with a macroprolactinoma took a dopamine agonist before pregnancy, and 39% (18/46) continued it during the first trimester. The preferred dopamine agonist was cabergoline. The relatively high proportion of women who continued dopamine agonists is similar to the findings of a study that questioned endocrinologists about how they would treat these patients; 82% stated that they would continue treatment in women with large macroprolactinomas.19 Many endocrinologists continue dopamine agonists in women with large macroprolactinomas, consistent with guidelines of the Pituitary Society20 and the Endocrine Society21 (summarized in Box 1). However, the British Medicine and Healthcare Products Advisory Authority guidance in 2008 recommended stopping cabergoline 1 month before pregnancy as a result of concern about cardiac fibrosis.22 Thus, current practice may reflect the fact that most clinicians are likely to be more concerned about the risks of tumor enlargement than unsubstantiated theoretical risks of cardiac fibrosis. It is encouraging that this prospective study demonstrated no cases of symptomatic macroprolactinoma enlargement in women previously treated with surgery or radiotherapy, consistent with previous reports.3
The Pituitary Society guidance on management of macroprolactinomas in pregnancy recommends close surveillance if dopamine agonists have been stopped,20 and The Endocrine Society recommends that for women with a macroprolactinoma who have not undergone pituitary surgery, “it is prudent to undertake more frequent clinical examination and formal visual field testing.”21 The Society's guidance has never been prospectively evaluated. In our study, 75% of women with macroprolactinomas and 56% (9/16) with nonfunctioning pituitary adenoma had a formal visual field test performed at least once during the pregnancy. However, only 33% of women with macroprolactinoma had visual fields monitored in each trimester. It is possible that some cases of tumor expansion might have been identified before onset of symptoms if surveillance had been performed more regularly.
There were two U.K. Obstetric Surveillance System cases of pituitary apoplexy. One had a macroprolactinoma and the other nonfunctioning pituitary adenoma, and both were diagnosed as having pituitary tumors before pregnancy. Both patients were managed conservatively with good outcomes. This is a rare, serious complication of pituitary adenoma in pregnancy and to date only 15 cases have been reported in the literature (PubMed search, keywords: pregnancy, pituitary, tumor, non-functioning, macroprolactinoma, acromegaly, Cushing, TSHoma, apoplexy, cabergoline, bromocriptine, quinagolide, pegvisomont, somatostatin analogue, octreotide, lanreotide; dates: 1985–2015).6 Most patients were not diagnosed as having a pituitary tumor until they presented with apoplexy in pregnancy. Box 2 summarizes the symptoms of pituitary apoplexy and of pituitary tumor expansion in pregnant women reported in the literature and in the current study.
There was no increase in the rate of congenital malformations associated with dopamine agonist use in the first trimester. Although this is consistent with the current literature,2,23,24 the current study was not powered to detect a difference. Similarly, women with macroprolactinoma or nonfunctioning pituitary adenoma in pregnancy did not have increased rates of preterm labor, hypertensive disease, preeclampsia, or fetal loss.
Potential limitations of this study include the fact that microprolactinomas were not included. Unfortunately, this was not possible because they were too common to include in a U.K. Obstetric Surveillance System study of rare disorders of pregnancy. It would have been valuable to establish whether the findings in macroprolactinomas can be generalized to all prolactinomas. Some cases may have been missed if they were managed by neurologists or neurosurgeons, and a future study could benefit from also writing to these specialists to ensure all U.K. cases were identified.
In summary, the majority of women with macroprolactinoma and nonfunctioning pituitary adenoma have good pregnancy outcomes, although pituitary apoplexy occurred in one woman with each type of tumor. This study demonstrates that nonfunctioning pituitary adenoma occurs more commonly in pregnancy than previously thought and that this group of tumors can present de novo with symptoms of expansion. Ergot-containing dopamine agonists, particularly cabergoline, were the treatment of choice most frequently used for symptomatic expansion of pituitary tumors, and the majority of women with symptoms suggestive of tumor expansion can be successfully treated without surgery. A significant number of clinicians elect to continue or restart dopaminergic therapy during pregnancy in women with macroprolactinomas, consistent with national guidelines.
1. Molitch ME. Endocrinology in pregnancy: management of the pregnant patient with a prolactinoma. Eur J Endocrinol 2015;172:R205–13.
2. Lebbe M, Hubinont C, Bernard P, Maiter D. Outcome of 100 pregnancies initiated under treatment with cabergoline in hyperprolactinaemic women. Clin Endocrinol (Oxf) 2010;73:236–42.
3. Molitch ME. Pituitary tumors and pregnancy. Growth Horm IGF Res 2003;13(suppl A):S38–44.
4. Kupersmith MJ, Rosenberg C, Kleinberg D. Visual loss in pregnant women with pituitary adenomas. Ann Intern Med 1994;121:473–7.
5. Masding MG, Lees PD, Gawne-Cain ML, Sandeman DD. Visual field compression by a non-secreting pituitary tumour during pregnancy. J R Soc Med 2003;96:27–8.
6. Kita D, Hayashi Y, Sano H, Takamura T, Hayashi Y, Tachibana O, et al. Postoperative diabetes insipidus associated with pituitary apoplexy during pregnancy. Neuro Endocrinol Lett 2012;33:107–12.
7. de Heide LJ, van Tol KM, Doorenbos B. Pituitary apoplexy presenting during pregnancy. Neth J Med 2004;62:393–6.
8. Chaiamnuay S, Moster M, Katz MR, Kim YN. Successful management of a pregnant woman with a TSH secreting pituitary adenoma with surgical and medical therapy. Pituitary 2003;6:109–13.
9. Blackhurst G, Strachan MW, Collie D, Gregor A, Statham PF, Seckl JE. The treatment of a thyrotropin-secreting pituitary macroadenoma with octreotide in twin pregnancy. Clin Endocrinol (Oxf) 2002;57:401–4.
10. Caron P, Gerbeau C, Pradayrol L, Simonetta C, Bayard F. Successful pregnancy in an infertile woman with a thyrotropin-secreting macroadenoma treated with somatostatin analog (octreotide). J Clin Endocrinol Metab 1996;81:1164–8.
11. Cheng S, Grasso L, Martinez-Orozco JA, Al-Agha R, Pivonello R, Colao A, et al. Pregnancy in acromegaly: experience from two referral centers and systematic review of the literature. Clin Endocrinol (Oxf) 2012;76:264–71.
12. Caron P, Broussaud S, Bertherat J, Borson-Chazot F, Brue T, Cortet-Rudelli C, et al. Acromegaly and pregnancy: a retrospective multicenter study of 59 pregnancies in 46 women. J Clin Endocrinol Metab 2010;95:4680–7.
13. Lindsay JR, Jonklaas J, Oldfield EH, Nieman LK. Cushing's syndrome during pregnancy: personal experience and review of the literature. J Clin Endocrinol Metab 2005;90:3077–83.
14. Rotherman KJ. Modern epidemiology. Boston (MA): Little Brown; 1986.
15. Satterwaite FE. An approximate distribution of estimates of variance components. Biometrics 1946;2:110–4.
16. Doyle P. The outcome of multiple pregnancy. Hum Reprod 1996;11(suppl 4):110–7.
17. Goluboff LG, Ezrin C. Effect of pregnancy on the somatotroph and the prolactin cell of the human adenohypophysis. J Clin Endocrinol Metab 1969;29:1533–8.
18. Scheithauer BW, Sano T, Kovacs KT, Young WF Jr, Ryan N, Randall RV. The pituitary gland in pregnancy: a clinicopathologic and immunohistochemical study of 69 cases. Mayo Clin Proc 1990;65:461–74.
19. Almalki MH, Ur E, Johnson M, Clarke DB, Imran SA. Management of prolactinomas during pregnancy—a survey of four Canadian provinces. Clin Invest Med 2012;35:E96–104.
20. Casanueva FF, Molitch ME, Schlechte JA, Abs R, Bonert V, Bronstein MD, et al. Guidelines of the Pituitary Society for the diagnosis and management of prolactinomas. Clin Endocrinol (Oxf) 2006;65:265–73.
21. Melmed S, Casanueva FF, Hoffman AR, Kleinberg DL, Montori VM, Schlechte JA, et al. Diagnosis and treatment of hyperprolactinemia: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2011;96:273–88.
22. Medicines & Healthcare Products Regulatory Agency. Drug safety update. London (UK): MHRA; 2008;2.
23. Colao A, Abs R, Bárcena DG, Chanson P, Paulus W, Kleinberg DL. Pregnancy outcomes following cabergoline treatment: extended results from a 12-year observational study. Clin Endocrinol (Oxf) 2008;68:66–71.
24. Robert E, Musatti L, Piscitelli G, Ferrari CI. Pregnancy outcome after treatment with the ergot derivative, cabergoline. Reprod Toxicol 1996;10:333–7.
25. Duley L. The global impact of pre-eclampsia and eclampsia. Semin Perinatol 2009;33:130–137.