Diagnosis and management of Cushing syndrome during pregnancy

INTRODUCTION — Pregnancy is uncommon in women with Cushing syndrome as hypercortisolism typically results in anovulatory infertility. However, over 260 cases of Cushing syndrome in pregnancy have been reported; these pregnancies are associated with high rates of fetal morbidity and mortality. The diagnosis of Cushing syndrome during pregnancy can be challenging, due to the physiologic changes of the hypothalamic-pituitary-adrenal (HPA) axis during normal pregnancy. Early diagnosis and treatment are essential to reduce maternal and fetal complications.

This topic will review issues related to Cushing syndrome in pregnancy. The clinical manifestations, diagnosis, and treatment of Cushing syndrome in nonpregnant patients are discussed separately. (See "Epidemiology and clinical manifestations of Cushing syndrome" and "Establishing the diagnosis of Cushing syndrome" and "Overview of the treatment of Cushing syndrome".)

EPIDEMIOLOGY — Cushing syndrome is associated with a high prevalence (up to 75 percent) of ovulatory disturbances induced by cortisol excess including amenorrhea and polycystic ovary syndrome [1-4]. As a result, it is uncommon for women with untreated Cushing syndrome to become pregnant [2]. In a systematic review of 168 studies and case reports, 263 pregnancies in 220 women with Cushing syndrome were described; most cases were associated with important maternal and fetal complications [5]. (See 'Maternal and fetal complications' below.)

The majority of cases of Cushing syndrome during pregnancy are in women who are newly diagnosed during the pregnancy, although in the series described [5], 49 of the 263 pregnancies (19 percent), were in women with a diagnosis and cure prior to pregnancy.

The remaining 214 had active Cushing syndrome while pregnant. Among the 214 women with active Cushing syndrome, 30 (14 percent) were diagnosed prior to pregnancy, 138 (65 percent) during the pregnancy, and 46 (22 percent) soon after pregnancy (the women diagnosed postpartum presumably had hypercortisolism while pregnant).

CAUSES — In nonpregnant women, most cases of Cushing syndrome are due to a corticotropin (ACTH)-secreting pituitary adenoma (Cushing disease), with adrenal adenomas accounting for approximately 15 percent of cases [6]. In contrast, the frequency of ACTH-independent cases is considerably higher in pregnant women with Cushing syndrome. In the systematic review of Cushing syndrome described above, 116 of 213 women (54 percent) with active disease had ACTH-independent Cushing syndrome, primarily adrenal adenomas [5].

Similar results have been seen in other studies [1,2,6-10]. The reason for this unusual pattern is not known. Pure cortisol-secreting adrenal adenomas may interfere with reproductive function less than the mixed cortisol and androgen excess produced by adrenal carcinomas or ACTH-dependent adrenal hyperplasia. This pattern also may reflect the milder hypercortisolism generally evident with adrenal adenomas, which may lead to less suppression of the hypothalamic-pituitary-ovarian axis.

Seven pregnant women with the ectopic ACTH syndrome have also been reported [1,5,7,11-13]. (See "Causes and pathophysiology of Cushing syndrome", section on 'Epidemiology'.)

In occasional cases, pregnancy appears to induce Cushing syndrome. In a small number of patients, Cushing syndrome remits completely postpartum [2,7,14] and, in a few reported cases, recurs during subsequent pregnancies [15,16]. Possible mechanisms include placental corticotropin-releasing hormone (CRH) production [17], ectopic expression in the adrenal cortex of receptors for luteinizing hormone (LH)/human chorionic gonadotropin (hCG; leading to adrenal adenoma or massive adrenal hyperplasia) [18-21], and estrogen-dependent nodular adrenal hyperplasia [8]. (See "Cushing syndrome due to primary pigmented nodular adrenocortical disease".)

In addition, exacerbation of Cushing syndrome during pregnancy can result from placental-derived ACTH stimulation of ACTH receptors in the adrenal adenoma. In one patient with an adrenal adenoma and modest cortisol secretion, Cushing syndrome became overt during pregnancy and regressed partially after delivery. ACTH receptors, but not aberrant LH/hCG receptors, were identified in the adenoma, suggesting that activation of ACTH receptors in the adrenal adenoma was due to ACTH of placental origin [22].

HPA CHANGES IN NORMAL PREGNANCY — Normal pregnancy is associated with changes in the maternal hypothalamic-pituitary-adrenal (HPA) axis, including increased production of corticosteroid-binding globulin (CBG); increased concentrations of serum, salivary, and urinary free cortisol (UFC); and lack of suppression of serum cortisol after dexamethasone. In addition, the placenta produces corticotropin (ACTH) and corticotropin-releasing hormone (CRH) with the latter's placental expression displaying paradoxical enhancement by glucocorticoid. A further complication is the increase in CRH-binding protein, which may reduce CRH bioavailability [23]. Thus, the biochemical diagnosis of Cushing syndrome during pregnancy is more challenging than in the nonpregnant state.

●Serum cortisol – Serum total and free cortisol concentrations are increased in pregnant women [7,24]. One study in healthy women reported that plasma cortisol values in pregnancy increased from 14.9±3.4 mcg/dL (411±94 nmol/L) at 12 weeks to 35.2±9.0 mcg/dL (971±248 nmol/L) at 26 weeks gestation and changed minimally thereafter [25].

●Salivary cortisol – While the bedtime nadir of cortisol is maintained [7], salivary cortisol levels increase during pregnancy. In one study using an enzyme-linked immunosorbent assay (ELISA), the upper limit of normal values were 0.13 mcg/dL (3.6 nmol/L) in nonpregnant control women, 0.25 mcg/dL (6.9 nmol/L) in the first trimester, 0.26 mcg/dL (7.2 nmol/L) in the second trimester, and 0.33 mcg/dL (9.1 nmol/L) in the third trimester [26]. Another study using radioimmunoassay found elevated midnight salivary cortisol values in 18 women during the second and third trimesters (2.41±1.33 ng/mL, range 0.68 to 6.80) compared with 89 control women (1.37±0.57 ng/mL, range 0.16 to 3.0) [27].

●Urinary cortisol – Urine cortisol also increases during pregnancy. One study using radioimmunoassay found a two- to threefold increase in UFC in the second and third trimesters, compared with nonpregnant women [7]. A second study using liquid chromatography/mass spectrometry compared values in each trimester (first trimester 135±10 nmol/d, second trimester 187±13 nmol/d, third trimester 242±15 nmol/d) with those of nonpregnant women (78±12 nmol/d) [24].

●Plasma ACTH – Plasma ACTH concentrations are slightly lower than those of nonpregnant women during early pregnancy but gradually increase to slightly above normal; in one report, ACTH values using radioimmunoassay were 23±4.6 pg/mL at twelve weeks, with a progressive increase to 59±16 pg/mL at 37 weeks [25]. The increase in ACTH is due at least in part to placental ACTH secretion.

●Dexamethasone suppression testing – Suppression of cortisol by dexamethasone is blunted during pregnancy, particularly in the second and third trimesters. In one report, the percent reduction of morning serum cortisol levels after dexamethasone (1 mg administered at 11 PM) was 87 percent in nonpregnant control women, compared with pregnant women who had lower degrees of suppression: 83, 44, and 37 percent in the first, second, and third trimesters, respectively [28].

CLINICAL FEATURES

Overlap with features of pregnancy — The clinical features of Cushing syndrome during pregnancy (weight gain, fatigue, hypertension) overlap with many of the signs and symptoms that women without Cushing syndrome may experience during pregnancy [1].

Cushing syndrome is often not detected until the second trimester, possibly because the potential signs and symptoms of Cushing syndrome are attributed to pregnancy rather than Cushing syndrome [29].

Biochemical findings — In the series of 136 pregnant women with Cushing syndrome described above [6], mean corticotropin (ACTH) concentrations were considerably higher in women with Cushing disease than in women with adrenal adenomas [6]. These results are discussed below. (See 'Additional evaluation to determine cause' below.)

Imaging — In the same study, among those with Cushing disease, only 13 of 18 had adequate imaging; 6 of 13 had macroadenomas, a higher percentage than is evident in nonpregnant patients [6]. Detailed imaging data were not available from the systematic review described above [5].

Maternal and fetal complications — The fetus is partially protected from the hypercortisolemia because placental 11-beta-hydroxysteroid dehydrogenase type 2 converts 85 percent of maternal cortisol to biologically inactive cortisone [11]. (See "Placental development and physiology", section on 'Steroid hormones'.)

However, maternal and neonatal complications occur commonly [5]. In a review of 136 pregnant women with Cushing syndrome (78 with no treatment and 107 live births), maternal complications included [6]:

●Gestational diabetes or impaired glucose tolerance – 25 percent

●Hypertension – 68 percent

●Preeclampsia – 14 percent

●Heart failure – 3 percent

●Psychiatric disorders – 4 percent

Two maternal deaths have also been reported [1].

In the same series of 136 cases including 107 live births, neonatal complications included [6]:

●Intrauterine deaths/spontaneous abortions – 5 percent, n = 6 (other series report have reported higher rates [30])

●Premature delivery – 43 percent, n = 46

●Intrauterine growth retardation – 21 percent, n = 22

●Still births – 6 percent, n = 8

●Neonatal adrenal insufficiency – 2 percent, n = 2

A separate series including 214 (128 with no treatment) cases of women with active Cushing syndrome demonstrated an overall fetal loss rate of 24 percent (48 of 214 pregnancies). Other outcomes included high rates of intrauterine fetal death and spontaneous abortion (17 percent), and premature delivery (66 percent) [5]. Women who received treatment had lower rates of fetal loss as described in the next section.

DIAGNOSIS DURING PREGNANCY

Initial testing — The evaluation and diagnosis of suspected Cushing syndrome in pregnancy differs from that for nonpregnant patients [31]. Unfortunately, there are no pregnancy-specific guidelines for interpretation of these tests, but we use both the urinary free cortisol (UFC) and late-night salivary cortisol tests [26,27]. We do not use the dexamethasone suppression test to diagnose Cushing syndrome during pregnancy, because of the risk for false-positive results [10]. (See 'HPA changes in normal pregnancy' above.)

We consider the diagnosis of Cushing syndrome in pregnancy to be established when:

●UFC – UFC is more than four times the upper limit of normal (for nonpregnant women) or

●Late-night salivary cortisol – Late-night salivary cortisol is three to four times above the upper limit of normal (for nonpregnant women) [24].

The lack of well-established reference ranges for these tests in normal pregnancy renders diagnosis less clear in milder cases [24,27]. The diagnosis of Cushing syndrome in pregnancy is also challenging because Cushing syndrome can easily masquerade as gestational diabetes or hypertension during pregnancy. (See "Establishing the diagnosis of Cushing syndrome".)

For any given salivary cortisol assay, values obtained in the second to third trimester of more than three to four times the normal reference range appear to support the diagnosis of Cushing syndrome. Measurement of bedtime or midnight plasma or salivary cortisol are commonly used tests to establish hypercortisolism in nonpregnant patients. Midnight salivary cortisol is significantly increased in normal pregnant women compared with nonpregnant controls, but can be distinguished from most patients with Cushing disease [26,27].

One study that used enzyme-linked immunosorbent assay (ELISA) to set late-night salivary cortisol reference ranges in pregnancy found that median values in pregnant women without Cushing syndrome were twice those of nonpregnant healthy controls. Individual values overlapped with those seen in nonpregnant Cushing syndrome patients [26]. Another study using radioimmunoassay found that midnight salivary cortisol values in normal pregnant women (1.6 to 9.13 ng/mL) overlapped with those seen in nonpregnant Cushing syndrome patients (3.88 to 49.1 ng/mL) [27]. Additional studies are needed using other assay platforms to determine appropriate reference ranges in pregnancy.

Additional evaluation to determine cause — Initial biochemical testing to determine etiology is directed to identifying adrenal causes of Cushing syndrome, which are more common in pregnancy. This strategy also defers more risky testing (corticotropin-releasing hormone [CRH], which is no longer available, magnetic resonance imaging [MRI], inferior petrosal sinus sampling [IPSS]).

Measurement of corticotropin (ACTH) and the high-dose dexamethasone suppression test provide complementary information:

●Dexamethasone suppression tests – As noted, we do not use the 1 mg overnight dexamethasone test to make the diagnosis of hypercortisolism during pregnancy. However, a high-dose (8 mg) dexamethasone suppression test (see 'Biochemical findings' above) is sometimes performed to help determine etiology.

There are limited data on the responses of serum cortisol to dexamethasone (8 mg) in pregnant women with adrenal causes of Cushing syndrome and pituitary adenomas. In a review of 136 women with Cushing syndrome during pregnancy, 14 women underwent an overnight 8 mg or two-day, high-dose dexamethasone suppression test. (See "Establishing the cause of Cushing syndrome", section on 'High-dose dexamethasone suppression tests'.)

Using a criterion of 80 percent or greater suppression of serum or urinary cortisol to identify Cushing disease, all patients with ACTH-independent Cushing syndrome were correctly identified but three of seven with Cushing disease failed to suppress [6]. (See "Dexamethasone suppression tests", section on 'High-dose DSTs'.)

●ACTH levels – Because of the persistent elevations of ACTH in normal pregnancy, a suppressed ACTH is not always found in primary adrenal causes of Cushing syndrome. In the series of 136 pregnant women with Cushing syndrome described above [6], mean ACTH concentrations for the different categories of Cushing syndrome were:

Cushing disease (n = 18) 69.4±25.9 pg/mL (15.3±5.7 pmol/L)

Adrenal adenoma (n = 13) 32.5±17.6 pg/mL (7.2±3.9 pmol/L); 5 of 13 women had undetectable ACTH (ie, <10 pg/mL)

These values were obtained at various times in pregnancy, using older radioimmunoassay techniques, as well as more modern "sandwich" immunoassays. As a result, there may be significant overlap between the expected value in normal pregnancy, unless the values are higher (indicating Cushing disease) or lower (indicating adrenal adenoma).

Patients without suppression after high-dose dexamethasone and a borderline or low ACTH (<10 pg/mL [4.4 pmol/L]) are likely to have an adrenal etiology. Ultrasound-identified adrenal lesions were present in 73 percent of cases in the series described [6]. If the adrenal ultrasound is negative, we suggest an adrenal MRI without a contrast agent, but not in the first trimester, because of concerns about potential adverse fetal effects.

●Other testing – Higher ACTH levels (>10 pg/mL [4.4 pmol/L]), or suppression after dexamethasone suggest Cushing disease. There are few data on the utility of the CRH or desmopressin stimulation tests or pituitary MRI. MRI can eventually be performed, but not in the first trimester, because of concerns about potential adverse fetal effects; during the remainder of gestation, it may be performed, but not with gadolinium contrast, which is labeled by the US Food and Drug Administration (FDA) as a class C drug. Performing MRI during pregnancy without gadolinium contrast reduces its sensitivity (from approximately 50 to 38 percent) [32]. In one study, it detected a pituitary adenoma in five of eight patients with Cushing disease, including three macroadenomas [6].

CRH stimulation and IPSS have been reported in less than 10 pregnant patients [6]. If IPSS is done, access via the jugular vein, with abdominal and chest lead shielding, is recommended to reduce radiation exposure to the fetus during fluoroscopy. (See 'ACTH-dependent Cushing syndrome' below.)

MANAGEMENT — The goal in women with Cushing syndrome diagnosed prior to pregnancy should always be treatment and a cure before proceeding with attempts to conceive. We suggest that all pregnant women with Cushing syndrome be followed by both a maternal-fetal medicine specialist and an endocrinologist with experience in the diagnosis and treatment of Cushing syndrome.

Medical therapy and surgical intervention both reduce the risk of fetal loss significantly, although the high rates of preterm birth and low birth weight do not appear to improve [5,6]. The choice of medical or surgical therapy is individualized and depends upon the cause of the patient's Cushing syndrome. Women in whom an adrenal adenoma or pituitary adenoma have been clearly identified may undergo surgical resection in the second trimester. Medical treatment options are also available.

Timing of treatment — Women with Cushing syndrome during pregnancy should be treated as early as possible, ideally before 20 weeks gestation. It has been difficult to evaluate the true impact of therapy because it has been typically instituted relatively late in pregnancy (because most women are not diagnosed until the second trimester). (See 'Clinical features' above.)

In the series of 136 pregnant women with Cushing syndrome described above [6], the live birth rate appeared to be slightly better in women who received early treatment compared with those who were managed conservatively; 50 live births out of 56 pregnancies (89 percent) if treatment started at a mean gestational age of 20 weeks, versus 59 live births out of 74 pregnancies (76 percent) in the group receiving no therapy in the third trimester [6]. In the series of 214 patients, those who received no treatment had a higher overall fetal loss (31 percent) than those receiving medical (21 percent) or surgical (13 percent) treatment [5]. However, complications, including premature delivery and eclampsia, occurred in some women in spite of treatment.

ACTH-independent Cushing syndrome — As noted above, approximately 60 percent of pregnant women with Cushing syndrome have corticotropin (ACTH)-independent Cushing syndrome (45 percent adrenal adenoma and 15 percent adrenal carcinoma). Adrenal surgery has been performed at 16 to 21 weeks of pregnancy without complicating the pregnancy [1,7].

Adrenal surgery — For pregnant women with ACTH-independent Cushing syndrome, we suggest the following:

●Surgical tumor resection appears to be more uniformly successful than medical therapy [5]. Therefore, for most women, we suggest early second trimester adrenal surgery. (See "Adrenalectomy techniques" and "Overview of the treatment of Cushing syndrome".)

Medical therapy

●For women who do not want surgery or are diagnosed later, we suggest metyrapone, given every six to eight hours, at a dose sufficient to reduce maternal hypercortisolism into the normal pregnancy range [7,33]. It may become necessary to induce early delivery of the fetus if complications of eclampsia supervene [7,15,34]. (See "Medical therapy of hypercortisolism (Cushing's syndrome)".)

●Other therapeutic options for ACTH-independent disease are associated with potential problems (see "Medical therapy of hypercortisolism (Cushing's syndrome)"):

•Ketoconazole is teratogenic and toxic to animal embryos but has been used successfully and without harm to fetuses from as early as the seventh week of pregnancy [35,36]. Similar to metyrapone, it is given every six to eight hours, at a dose sufficient to reduce maternal hypercortisolism into the normal pregnancy range. Because of the potential for feminization of a male fetus and maternal hepatotoxicity, we suggest that its use be reserved for those who need emergency medical therapy but cannot tolerate metyrapone.

•Mitotane should not be given to pregnant women, as it crosses the placenta and is teratogenic [37]. However, it was used in one woman without complications [38].

ACTH-dependent Cushing syndrome

Surgery or medical therapy — For pregnant women with Cushing disease, transsphenoidal surgery (ideally early in the second trimester) or medical therapy (generally with metyrapone) have both been used [1,5,6,39,40].

At one institution, four pregnant women with Cushing disease underwent transsphenoidal surgery; remission was achieved in three women, but fetal/neonatal deaths occurred in two pregnancies [6]. One intrauterine death due to a tight nuchal cord, occurred at 33 weeks in a woman with persistent Cushing disease postoperatively. The second occurred after early delivery (24 weeks) for reversal of cord blood flow in a woman who had remission of her hypercortisolism postoperatively but persistent hypertension and eventual toxemia. (See "Transsphenoidal surgery for pituitary adenomas and other sellar masses".)

Management of comorbidities — In addition to specific treatment of Cushing syndrome, women should receive additional treatment related to any comorbidities, such as diabetes and hypertension, both of which are common in this population [5,6]. Patients should be counselled about the potential for premature labor. Definitive treatment should be initiated postpartum.

Glucocorticoids during labor and delivery — Women who achieve surgical remission during pregnancy should receive hydrocortisone sufficient to achieve trimester-specific levels of urine cortisol. At the time of labor and delivery, they should be treated as if they have adrenal insufficiency. After delivery, the hydrocortisone dose should be gradually decreased over three months as corticosteroid-binding globulin (CBG) decreases [24]. (See "Treatment of adrenal insufficiency in adults", section on 'Pregnancy and labor'.)

If women are receiving steroidogenesis inhibitors, they should be discontinued (or reduced) when labor begins to allow for increased cortisol production and should be restarted at a lower dose following delivery. (See "Medical therapy of hypercortisolism (Cushing's syndrome)".)

Outcomes — Treating Cushing syndrome during pregnancy reduces maternal morbidity and fetal mortality. However, it does not appear to reduce the frequency of intrauterine growth restriction (approximately 30 percent) or preterm birth [5,6,10,30,41,42].

One review reported an improvement in live birth rates with treatment (59 live births in 74 pregnant women who did not receive therapy [76 percent] and 50 live births in 56 pregnant women who did receive treatment [89 percent]) [6]. In the systematic review of 213 cases of women with active Cushing syndrome during pregnancy, fetal loss occurred in 30 of 128 women (31 percent) who were untreated, compared with 5 of 24 (21 percent) who received medical therapy (most often metyrapone) and 6 of 49 (13 percent) who underwent surgery (transsphenoidal surgery or adrenalectomy [unilateral in most cases]) [5].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Diagnosis and treatment of Cushing syndrome".)

SUMMARY AND RECOMMENDATIONS

●Clinical features – Pregnancy is uncommon in women with Cushing syndrome as hypercortisolism typically results in anovulatory infertility. Other clinical features of Cushing syndrome during pregnancy, including weight gain, fatigue, and hypertension, overlap with many of the signs and symptoms that women without Cushing syndrome may experience during pregnancy. (See 'Clinical features' above.)

●Causes – Of those who do become pregnant, approximately 55 percent have corticotropin (ACTH)-independent hypercortisolism (primarily adrenal adenomas). This pattern is different from that seen in nonpregnant individuals with Cushing syndrome, the majority of whom have ACTH-dependent hypercortisolism (ACTH-secreting pituitary tumors). (See 'Causes' above.)

●Diagnosis

•Initial testing – The evaluation of suspected Cushing syndrome in pregnancy differs from that in nonpregnant patients. We suggest combined use of urinary free cortisol (UFC) and late-night salivary cortisol, although there are no pregnancy-specific guidelines for interpretation of these tests. (See 'Initial testing' above.)

•Diagnostic criteria – We consider the diagnosis of Cushing syndrome in pregnancy to be established when UFC is more than four times the upper limit of normal (for nonpregnant women) or when salivary cortisol is three to four times above the upper limit of normal (for nonpregnant women). (See 'Diagnosis during pregnancy' above.)

•Evaluation to determine cause – We perform a high-dose (8 mg) dexamethasone suppression test and measure ACTH to help determine the etiology. (See 'Additional evaluation to determine cause' above.)

●Treatment

•ACTH independent – We suggest surgical resection of adrenal adenomas causing Cushing syndrome during pregnancy (Grade 2C). This is typically performed at 16 to 21 weeks of pregnancy without complicating the pregnancy. (See 'Management' above and "Overview of the treatment of Cushing syndrome", section on 'Unilateral adrenalectomy'.)

•ACTH dependent – For women with a well-established diagnosis of Cushing disease, we suggest transsphenoidal surgery (Grade 2C), ideally in the second trimester. (See 'ACTH-dependent Cushing syndrome' above.)

•Women who decline or are not candidates for surgery – For women who do not want surgery or are diagnosed later, we suggest metyrapone to control maternal hypercortisolism (Grade 2C). It may become necessary to induce early delivery of the fetus if complications of eclampsia supervene. (See 'Management' above and "Medical therapy of hypercortisolism (Cushing's syndrome)", section on 'Metyrapone'.)

ACKNOWLEDGMENT — The views expressed in this topic are those of the author(s) and do not reflect the official views or policy of the United States Government or its components.