The management of congenital adrenal hyperplasia during preconception, pregnancy, and postpartum

Due to the possible pregnancy complications and implications on the offspring, preconception genetic testing and counseling with a high risk obstetrics specialist is recommended in women with a known diagnosis of classic CAH who are family planning [19, 20]. The carriership in the general population for classic CAH is approximately 1 in 60 (between 1 and 2 percent) [3]. Therefore, if the carriership of the partner is unknown, the risk of having a child with classic CAH is approximately 1 in 120 (Fig. 2). However, if the partner is a known carrier of classic CAH, then the risk of having an affected child is 50 percent. The majority of patients with NCCAH are compound heterozygotes with different mutations on each allele, and approximately two-thirds carry one allele that causes classic CAH (Fig. 3). [21] Theoretically and without genotyping, a patient with NCCAH has a 1 in 350 risk of having a child with classic CAH. An international multicenter study of 203 pregnancies in 101 women with NCCAH reported a much higher risk of 2.5% of conceiving a child with classic CAH and 14.8% with NCCAH [22]. Similarly, a retrospective study of children born to 190 women with NCCAH in France estimated the risk of having a child with classic CAH to be 1.5% [21]. The higher than expected rate of classic CAH in these studies may be due to the high frequency of marriages between people of similar ethnicities resulting in a higher likelihood of those that are carriers intermarrying. The estimated carrier frequency of NCCAH varies from 1:7 to 1:16, and the epidemiology of NCCAH is less well established than classic CAH because it is not detected by neonatal screening programs [4]. Therefore, CYP21A2 genotyping is recommended prior to pregnancy to better define individual risk. Routine genetic carrier screening, which includes CAH due to 21-hydroxylase deficiency, is becoming more affordable and is recommended by the American College of Obstetrics and Gynecology to be offered to all pregnant patients and those desiring pregnancy during preconception visits [23], especially if undergoing an infertility work up and in vitro fertilization (IVF).

Compared to the general population, women with classic CAH are less likely to seek motherhood and have fewer pregnancies and children [13]. In a population-based study using the national Swedish registry, only 25.4% of women with CAH had given birth compared with 45.8% of controls [24]. Interestingly, the reduction in motherhood in CAH was only observed in women with SW CAH, while those with SV and NCCAH were similar to matched controls. In the UK, the pregnancy rate of women with CAH was similar to the general UK population, however women with SW CAH were less likely to seek motherhood [12]. This is consistent with other studies that report fewer pregnancies and less desire for fertility in women with the SW phenotype [14]. There are a multitude of potential causes including possible prenatal androgen effects on the brain and behavior [25], an increased prevalence of gender identity shifts in sexual orientation and reduced likelihood to be exclusively heterosexual compared to unaffected women [26‐28], hormonal imbalances leading to anovulation, and decreased sexual satisfaction due to urogenital dysfunction [28, 29].

Women with CAH have multiple hormonal imbalances contributing to subfertility. Androgen excess suppresses late stage folliculogenesis and ovulation. Preferential secretion of LH further increases androgen levels and elevated androgens are aromatized to estrogens and both suppress gonadotropin secretion. High progesterone levels impact fertility on many levels; it disrupts GnRH pulsatility and ovulation, thickens cervical mucus and interferes with sperm motility, and impacts endometrial lining growth which impairs embryo implantation [9]. However, with glucocorticoid and mineralocorticoid treatment, menstrual cycles can become ovulatory about 40% of the time [30]. Fertility can be restored in the majority of women, with natural conception reported as high as 76% [12]. However, in a European (6 countries: Germany, France, the Netherlands, Poland, Sweden and United Kingdom) multicenter cross-sectional study of disorders of sex development, only 14.7% of 221 women with CAH had one or more children without assisted reproduction techniques [31]. Numbers of oocytes appear to not be affected in patients with CAH, as indicated by a study evaluating anti-mullerian hormone (AMH, a marker of ovarian reserve) that found no significant difference between CAH patients and controls and comparable levels between classic and NCCAH patients [32].

CAH has a spectrum of anatomic presentations due to the varied virilizing effects of androgen exposure to the external genitalia. Furthermore, surgical management may further impact anatomy and function. Anatomic factors that may interfere with fertility include a smaller vaginal introitus, decreased vaginal lubrication, dyspareunia, decreased clitoral sensitivity secondary to surgery, and decreased sexual satisfaction [33‐35]. In a meta-analysis of 29 studies, most patients were sexually active, but only 48% reported comfortable sexual intercourse [36]. These negative experiences may make intercourse less desirable and therefore decrease the chances of conception based on frequency and timing.

Women with NCCAH are at risk for spontaneous miscarriage, which has been reported in up to 25% of pregnancies and shown to be reduced with glucocorticoid treatment. [21, 22, 37] If impaired fertility occurs, prednisone or hydrocortisone can be used, and dexamethasone should be avoided since it can cross the fetoplacental barrier and may impact cognitive development [38, 39]. However, there are no clear recommendations regarding the treatment of NCCAH during pregnancy [40]. Similar to polycystic ovary syndrome, which NCCAH is sometimes misdiagnosed as, the main cause of infertility is anovulation in 30–50% of patients [22, 41]. This results in complaints of subfertility in 10–30% of women with NCCAH seeking care [21, 22]. Glucocorticoid therapy is recommended for patients with NCCAH and infertility (no conception after 12 months) or history of miscarriage. Though it is reasonable to refer to a reproductive endocrinology and infertility (REI) specialist if no conception after 6 months.

The first goal in addressing infertility in women with classic CAH is to optimize glucocorticoid treatment in an attempt to achieve regular ovulation and menstruation [5] (Fig. 4). Hydrocortisone, prednisone, or prednisolone can be given, which is preferred over dexamethasone which crosses the placenta without inactivation and exposes the fetus. Increasing glucocorticoid dose to suppress follicular phase progesterone is desired, as progesterone accumulation is likely the greatest hormonal obstacle to fertility [17, 42]. For NCCAH, infertility is an indication for glucocorticoid therapy, and prednisone 4 – 5 mg/day in two divided doses can be given and increased to 7.5 mg/day if needed. In women with classic CAH, optimal glucocorticoid management should be attempted and may include dividing doses throughout the day or using a longer acting glucocorticoid (prednisone or prednisolone) for maximal reduction of androgen and progesterone. For both classic and NCCAH, if ovulation is not achieved by glucocorticoid treatment alone, it is reasonable to consider ovulation induction[43‐49] with clomiphene citrate (only three case reports in the literature) [43, 50, 51], aromatase inhibitors, injectable gonadotropins, and adjuvant metformin [46, 47], however there is a paucity of data.

If refractory to these treatments, or in couples trying to avoid having a child with classic CAH, care with a REI specialist to utilize in vitro fertilization (IVF) can be offered, with or without preimplantation genetic testing for monogenic disorders (PGT-M). Since CAH is an autosomal recessive disorder, IVF and PGT-M is beneficial in cases where both parents are carriers of a classic CYP21A2 mutation, one parent is a classic carrier and the other is affected, or a couple has a prior classic CAH-affected child. Some studies advocate that molecular testing for CAH should be routinely considered at fertility centers because of the high frequency, potential severity of disease of having an affected child with classic CAH, and the NCCAH association with reduced fertility [52]. Typically, embryos are created by combining oocytes (after controlled ovarian hyperstimulation and egg retrieval) with sperm and growing embryos to a day 5–6 blastocyst, at which point a biopsy of the trophectoderm (5–6 cells) is performed [53]. The trophectoderm is what becomes the placenta and this avoids disrupting the inner cell mass, which becomes the fetus. PGT-M can identify which embryos are affected vs. carriers for CAH (Fig. 2 and 3). In current practice, all embryos are typically frozen while PGT-M is performed via linkage analysis, single nucleotide polymorphism (SNP) arrays or next generation sequencing (NGS) to create a case specific probe. This often requires DNA samples from first degree relatives such as a grandparent or sibling or other embryos (including aneuploid ones) and can take approximately 6–8 weeks for the embryology lab to create a custom set-up for the couple [54, 55]. Then the chosen embryo is transferred in a subsequent cycle (Fig. 4).

If resistant to medical therapy, though controversial, an alternative treatment in select patients may include laparoscopic bilateral adrenalectomy. Though the Endocrine Society Clinical Practice Guidelines for CAH recommend against it [5], infertility may be a special indication for this surgery. In a meta-analysis for 48 cases in 32 studies of bilateral adrenalectomy, three of the 35 women underwent the surgery for primary infertility. All 3 women subsequently conceived, resulting in a total of 6 healthy live born babies [56‐58]. Three additional patients underwent the surgery for obesity and virilization and all experienced spontaneous menses post-operatively [57]. However, bilateral adrenalectomy can increase the risk of future adrenal crises, which occurred in the three above mentioned patients as well as 8 of the 48 patients (17%) in another series [56, 57].

Several studies have evaluated pregnancy outcomes in women with CAH [24, 59, 60]. A study in the United States using diagnosis codes in the Health Care Cost and Utilization Project-Nationwide Inpatient Sample database from 2004–2014, compared pregnancy outcomes in the general population and those with CAH. Unfortunately, this study was unable to distinguish between classic and non-classic CAH and medication data was not available. They found increased rates of chorioamnionitis, cesarean section, maternal infection, small for gestational age infants and congenital anomalies (which were not specified) but no increase in maternal hypertension or gestational diabetes [59]. In a series of pregnancy outcomes in patients with adrenal insufficiency, 32 women had classic CAH, and they were found to have a high rate of miscarriage (43.8%), cesarean section (64.3%) and preterm labor (32.3%). In this review, 2 of the 32 women had an adrenal crisis during pregnancy [60]. The most comprehensive data to date on pregnancy in women with CAH can be found in a Swedish population-based cohort of 272 women with CAH (69 gave birth to at least 1 child) [24]. Compared to controls matched by sex, age and place of birth, women with CAH had a higher rate of gestational diabetes (4.9% versus 1.4%) and cesarean section, (51.4% versus 12.3%). No differences were found in preeclampsia, infant Apgar score, and frequency of small-for-gestational age in the neonates. This cohort included 8 women with SW, 26 women with SV and 16 women with NCCAH. Outcomes between the different cohorts of CAH were equivalent except for the cesarean section rate which was 90.9%, 65.9% and 33.3% in these groups respectively, with the rate of 90% planned cesarean section for those with salt-wasting [24]. This higher rate of planned cesarean section in women with classic SW CAH is likely associated with the desire to avoid damage to the previous genital surgery in women born with a urogenital sinus [24].

As mentioned above preimplantation genetic testing can be performed during IVF. There are several options for prenatal diagnosis during pregnancy: non-invasive prenatal testing (maternal serum cell free-DNA) for sex of the baby (10 weeks—delivery), and chorionic villus sampling (CVS) at 10–13 weeks or amniocentesis (15–20 weeks) for diagnosis of CAH. Cell-free DNA is an analysis of circulating free fetal DNA collected from maternal blood; CVS is placental villi collected by a needle or special catheter transcervically or transabdominally under ultrasound guidance; amniocentesis is a collection of fetal cells obtained via a 22-gauge spinal needle, transabdominally from a pocket of amniotic fluid under ultrasound guidance [61].

In addition to identifying fetal chromosomal sex, one study accurately identified fetal CAH status using targeted massively parallel sequencing (MPS) of cell-free DNA in maternal plasma in 14 families as early as 5 weeks and 6 days gestation [62], however this is considered experimental. Cell-free DNA is currently offered and utilized as a screening test for chromosomes 13, 18, 21, X, and Y, where CVS and amniocentesis are considered diagnostic tests and are needed to definitively confirm most genetic disorders. When a fetus is identified as 46,XX based on cell-free DNA but male or atypical genitalia are identified on a second trimester anatomy ultrasound, amniocentesis for karyotype and CYP21A2 mutation testing should be performed before assuming that the fetus has CAH, as a variety of conditions can explain these findings including testing errors [63]. It is also recommended that amniocentesis is performed for fetuses that were PGT-M tested embryos since the trophectoderm becomes the placenta, not the fetus, and CVS is a biopsy from the placenta, whereas amniocentesis analyzes fetal cells.

Based on the above limited information, we can conclude that women with CAH when they conceive have relatively successful pregnancies when compared to the general population. These pregnancies, however, are considered high risk given the complexity of medical management of CAH and comorbid conditions. It is thus important that care is coordinated between an obstetrician experienced in high-risk pregnancy and an endocrinologist experienced in the management of CAH [24]. Prenatal care should include routine monitoring for infections and preeclampsia. Although there are conflicting results regarding the risk of gestational diabetes, enhanced screening for gestational diabetes with glucose screen in the first or second trimester is prudent given that exposure to glucocorticoids may increase glucose intolerance.

Patients with classic CAH have been found to have a higher rate of chronic hypertension compared to the general population [64, 65], which along with chronic glucocorticoid use may increase the risk of intrauterine growth restriction (IUGR), which increases the risk for intrauterine fetal demise (IUFD). Often patients with CAH have comorbidities which include diabetes and obesity [65], which also might affect fetal growth [66]. Along with these theoretic risks, at least one case series of women with CAH reported increased risk of having a small for gestational age infant [59]. It is therefore reasonable to perform serial fetal growth ultrasounds starting in the third trimester. Although a higher incidence of fetal demise has not been found, studies to date comprise of small numbers, and are retrospective [24, 59, 60]. Thus, fetal wellbeing monitoring starting in the late third trimester should be performed given the high-risk nature of these pregnancies. If IUGR is detected, then standard antenatal fetal surveillance management is indicated, and timing of delivery is determined by umbilical artery doppler findings [67, 68].

Finally, women with CAH may be at increased risk of infection. Labor and delivery are associated with high rates of infection under normal circumstances, and increased risk of infection was found in pregnant women with CAH in a U.S. retrospective population-based study [59]. Thus, a high index of suspicion for chorioamnionitis should be maintained, and women with CAH should be treated with antibiotics if there is prolonged rupture of membranes, maternal temperature reaches 100.4 during labor, and prophylactically prior to incision when a cesarean section is undertaken [69, 70].

Treatment of classic CAH includes the administration of lifetime glucocorticoid and usually mineralocorticoid therapy. As stated previously, hydrocortisone and prednisolone/prednisone are preferred as they are both inactivated by 11β-hydroxysteroid dehydrogenase type 2 in the placenta and thus prevent fetal exposure. Dexamethasone, however, does cross the placenta and should be avoided during pregnancy unless the purpose is to expose the fetus (such as in the case of administration for fetal lung maturation when pre-term delivery is expected). Efficient placental aromatization of maternal androgens prevents the female fetus from virilization, thus maternal CAH is not likely to lead to in-utero fetal virilization of an unaffected female [24, 71]. If glucocorticoid therapy is initiated to treat subfertility in a woman with NCCAH, it is typically continued throughout pregnancy, but data is limited. In this cohort, the miscarriage rate can be reduced to normal in women treated with glucocorticoid prior to and during pregnancy [21].

Steroid management during pregnancy requires careful consideration. During normal pregnancy, there is a progressive increase in circulating corticotropin-releasing hormone (CRH) and adrenocorticotropic hormone (ACTH), and the estrogen-stimulated increase in sex hormone–binding globulin and corticosteroid-binding globulin cause androgen and cortisol levels to naturally increase. In addition, free cortisol increases after the second trimester [72]. It is therefore reasonable to expect that women with CAH will require an increase of their treatment dose as the pregnancy progresses. Our practice is to increase glucocorticoid dose by approximately 20% in the second or third trimester. This is in agreement with the Endocrine Society Clinical Guidelines’ suggestion that women should be maintained on their pre-pregnancy doses of glucocorticoid and mineralocorticoid, and doses of glucocorticoid should be increased by 20% to 40% from the 24th week [5, 73]. However, there are no evidence-based recommendations for steroid management during pregnancy.

Typical biomarkers such as 17-hydroxyprogesterone, androstenedione and plasma renin activity change physiologically with pregnancy. Normal ranges during pregnancy are not available and therefore these biomarkers cannot be used in the management of the pregnant woman with CAH. Free testosterone levels can be measured to target levels in the high normal range for pregnancy [74].

Care should be given to avoid cushingoid side effects from too high a dose of glucocorticoid.

Conversely, cases of adrenal crisis in pregnancy have been reported, but symptoms of adrenal insufficiency such as postural hypotension and fatigue are common during normal pregnancy [60]. It is, therefore, important to educate the patient regarding stress dosing for intercurrent illnesses, and maintain a high index of suspicion for possible adrenal insufficiency (see below). Patients should be provided an emergency hydrocortisone injection set and receive training for self-injection.

Doses of fludrocortisone may need to be increased during pregnancy as progesterone and 17-hydroxyprogesterone levels rise during pregnancy and compete with mineralocorticoid receptors. In general, hydrocortisone will exert some mineralocorticoid activity which may compensate for the need for higher fludrocortisone [49]. Plasma renin activity physiologically increases during pregnancy, [75] but treatment can be monitored through blood pressure and serum sodium and potassium [73], However, it should be noted that normative ranges may be altered during pregnancy [75]. For example, serum sodium reference ranges for pregnant women are 2–5 nmol/L lower than nonpregnant women probably due to increased glomerular filtration rate (GFR)[76]. Clinical monitoring for general wellbeing, blood pressure, and weight gain during pregnancy is essential. In principle, management of glucocorticoid and mineralocorticoid replacement is based on clinical grounds as no reliable laboratory-based assessment has been established.

Importantly, maternal dexamethasone started in the first trimester prior to testing for the chromosomal sex of the baby and administered in order to prevent virilization of a female fetus, is no longer recommended [5, 77]. Giving dexamethasone before week 9 of gestation can suppress fetal adrenal androgen production and prevent virilization of the external female genitalia during the period of urogenital organogenesis. This approach has been shown to be successful in ameliorating virilization of the external genitalia in studies in the U.S. and Europe. [62, 78‐80] However, routine administration will more often expose unaffected fetuses (seven out of eight) including unaffected and heterozygous females as well as males, who are not at risk for atypical genitalia. Given exposure of unaffected fetuses, the concern of glucocorticoid effects on fetal brain development and postnatal growth [48], the unknown long term side-effects in children, and ethical concerns, Endocrine Societies recommend that prenatal treatment is only performed as part of Institutional Review Board approved studies [5, 77]. There are no studies in the U.S. currently evaluating this.

It is unknown whether pregnancy is associated with an increased risk of adrenal crisis in women with CAH, but cases have been reported. It is important to review with the patient and family the requirements for higher doses of glucocorticoid during illnesses, similar to “sick day rules” in the nonpregnant state, [81] but which may include episodes of hyperemesis [49]. If a pregnant woman experiences an adrenal crisis, she requires urgent attention with administration of stress dose steroids and intravenous fluids in a manner similar to the nonpregnant state [82]. If the pregnancy has entered the period of fetal viability (after 24 weeks gestational age) when an adrenal crisis is encountered, it is important that fetal monitoring for wellbeing and contractions be performed during maternal treatment. In such cases the patient should be co-managed by an endocrinologist experienced in CAH care as well as a high-risk obstetrician [67, 68].

Labor is a “stress” event equivalent to major surgery and thus requires the administration of stress dose steroids. Hydrocortisone should be initiated with the onset of active labor with an intravenous bolus injection of 100 mg hydrocortisone followed by 50 mg of hydrocortisone every 6 h or a continuous infusion of 200 mg hydrocortisone/24 h throughout labor and delivery (Cesarean section or vaginal delivery). Intravenous fluids are also given, and close cardiovascular monitoring is indicated.