1. General Considerations and Recommendations
Hydrocortisone, prednisone, and prednisolone are inactivated by the placental type 2 11β-HSD; dexamethasone is not inactivated and should therefore be used in pregnant patients only if a fetal effect is required (i.e., in the case a prenatal therapy has been chosen, see below); in all other cases, dexamethasone should be discontinued and switching to hydrocortisone or other corticosteroids is recommended, with appropriate equivalent dosage, without modifications of the usual maintenance dose during the first two trimesters. In patients with adrenal insufficiency, glucocorticoid replacement therapy is usually increased by 20–40% during the third trimester; similar recommendations exist for congenital adrenal hyperplasia (CAH)
[1]. It is of note that 17-OH-P tends to physiologically increase throughout the pregnancy, whereas androstenedione increases, reaching a plateau at the 12th gestational week. Dose titration therefore requires trimester-appropriate reference ranges. The daily schedule of glucocorticoid administration is still debated. A “reverse circadian rhythm” administration, with a larger dose in the evening, could in theory obtain a better reduction in ACTH and androgen activity, though losing the possibility of mimicking the physiological circadian rhythm of cortisol production. However, a clear benefit of one timing schedule versus the other has not yet been observed
[2][3].
Mineralocorticoid requirements tend to increase during pregnancy too, given the anti-mineralocorticoid effects of the increased progesterone levels, and blood pressure and serum potassium should be used to titrate fludrocortisone doses instead of renin, which is unreliable during pregnancy
[4][5][6].
As for other stressing events, labor and delivery require an increase in glucocorticoid administration. Current recommendations consist of 100 mg i.m./i.v. hydrocortisone at the onset of active labor, followed by 200 mg/24 h in fractioned doses (both orally or iv). The dose may be increased in case of complications
[1][4]. After delivery, an orally administrated double dose should be maintained for 2–4 days
[1]; hydrocortisone should usually be reduced to 100 mg/24 h in four daily doses during the first day post-partum and 50 mg in three daily doses during the second day post-partum. No specific protocols exist for following further reduction, but pre-pregnancy doses can be restored if there are no clinical complications
[1][4].
Breastfeeding is recommended
[1], with some authors suggesting breastfeeding before taking hydrocortisone to reduce the already minimal concentration excreted in breast milk
[4].
2. Prenatal Therapy
Prenatal therapy regimens, since their first proposal in 1984
[7], have not been validated by solid studies. If prenatal treatment is chosen, as presented below, it is recommended that it is carried out in the context of experimental therapies in referral centers, with appropriate long-term follow-up registries, including prenatally treated children and adults
[8]. Based on embryology considerations, such therapy should be started by the ninth gestational week to be effective in avoiding hyperandrogenism and subsequent genital development abnormalities in female fetuses. Dexamethasone is usually administered for this prenatal treatment at the dosage of 20 μg/kg/day (based on pre-pregnancy body weight), fractioned in one, two, or usually three daily doses, up to 1.5 mg/day
[9][10]. Recently, Stachanow et al. proposed a markedly reduced dose of 7.5 μg/kg/day
[11]. Such therapy should be started by the 6–8th and continued up to the 16th gestational week minimum in order to be effective.
The benefit/risk ratio of prenatal glucocorticoid treatment is still controversial because of safety concerns
[12].
The expected benefits of prenatal therapy depend on the suppression of fetal ACTH, which avoids fetal hyperandrogenism and female genital virilization in at least 80–85% of cases
[13]. The alternatives are genital feminization surgery or no intervention at all, but both carry a risk of sexual, psychological, and reproductive adverse outcomes
[12][14]; an appropriate prenatal therapy would therefore lower this risk, especially in the more severe 21-OHD-null genotype group, which show worse genital surgery and psychological outcomes
[10]. The inappropriate treatment of most fetuses, with their potential exposure to the following risks, must be weighed against the benefits of the minority of appropriately treated female fetuses carrying biallelic classical CAH mutations.
Animal studies have suggested potential risks of teratogenicity and alterations in brain structure and cognitive development, behavior, metabolic profile, and HPA axis; however, rodents are not a solid model in this case because of the different glucocorticoid sensitivity. Studies including non-rodent animals have focused on late pregnancy exposure to high doses of glucocorticoids for preterm births
[10]. Given the hypothesized different outcomes of early vs. late exposure to glucocorticoids during pregnancy
[15][16], these results are not applicable to evaluate prenatal treatment in CAH
[10]. The focus will therefore on studies involving humans.
Regarding perinatal clinical outcomes, some evidence suggests that there may be an increased risk of cleft palate and other median-line alterations in patients prenatally treated with dexamethasone
[8][17]. Some authors have reported an increased risk for normal to low birth weight and for cerebral palsy, albeit non-statistically significant
[18]. Grunt et al. reported two cases of acute encephalopathy, one of which with permanent sequelae
[19].
As for the long-term risks of prenatal therapy, they need to be further analyzed
[8][10]. Several authors, mostly based in Sweden, where appropriate follow-up registries have been implemented, have reported alterations in children and adults prenatally treated with dexamethasone, mainly regarding cognitive and behavioral functions, especially in females
[10][20]: increased social anxiety
[21] and reduced sociability
[22], impaired verbal intelligence and working memory
[21][23][24], reduced cognitive abilities
[25], and effects on gender role behavior
[26]. One paper reported that prenatal therapy reduced cognitive abilities in non-CAH patients but improved them in CAH patients
[27].
Conversely, no general differences were reported by other studies
[22][28], and specifically no increase in anxiety, better sociability, and no differences in behavioral problems
[27][29][30][31] and cognitive functions
[21][28]. The same cohorts with reported cognitive alterations in children showed no such differences at an older age, suggesting the possibility of improvement/regression of these observations over time
[32][33].
These concerns regarding mental and cognitive functions are supported by studies reporting different brain morphology in dexamethasone-exposed non-CAH fetuses
[34] and by the adverse cognitive and behavioral effects observed for betamethasone, even if in different clinical settings
[10][35][36].
As for metabolic and cardiovascular health, there have also been reports of a worse insulin-secreting capacity (based on insulin level evaluation) and lipid profile
[37][38], with unknown long-term effects on metabolic and cardiovascular risk
[12]. In theory, dexamethasone may be associated with altered renal, pulmonary, and pancreatic development, with a subsequent potential increased risk of hypertension, metabolic alterations, and allergic disorders, but none of these adverse outcomes have been clearly demonstrated
[8]. A recent study showed no altered blood pressure profile in prenatally treated patients, including adults
[39].
A meta-analysis of eight observational studies has not corroborated the reported findings on brain function nor metabolic profile
[40]. Moreover, some studies with larger cohorts have strongly advocated prenatal therapy to be safe and effective
[41].
There is a strong need for prospective research on short- and especially long-term effects of prenatal dexamethasone treatment
[8][10]. Most of the cited works on this issue have been published by a Swedish research group, carrying out extensive research on these long-term effects
[9]; however, the studied population is often of a small size. Follow-up registries must be implemented in other centers and multicentric studies must be carried out, as internationally advocated. Moreover, research on the hypothalamus–pituitary–adrenal axis function and metabolic and cardiovascular health outcomes should be implemented
[10].
As for maternal safety, the risk of body weight increase, appearance of cushingoid features, edema, sleep, and mood disturbances should be taken into account. No increased risk of hypertension, gestational diabetes mellitus, or miscarriages has been observed in 21-OHD pregnancies
[14][42][43][44][45][46].
Given the uncertainty regarding prenatal therapy fetal safety, some authors strongly oppose it
[14][47], with national and international guidelines recommending it to be carried out only as experimental therapy in referral centers with appropriate informed consent and long-term follow-up registries
[8][48][49]. Nowotny et al. performed a survey of 36 centers from 14 European countries to evaluate current practice of prenatal treatment for 21-OHD: 13 of these centers carry out prenatal therapy, with great variability in terms of prenatal diagnostics and therapy starting points, dose fractioning, and the absence of follow-up registries in more than half of the responding centers
[9]. The authors underline how European cooperation may increase scientific data, potentially leading to more conclusive results.
It is of note that most of the above cited follow-up studies and reports include fetuses treated with dexamethasone at 20 μg/kg/day. As already mentioned, a reduced dose of 7.5 μg/kg/day has recently been suggested to lower the adverse effects associated with dexamethasone, but data are limited
[11]. Moreover, new early prenatal diagnostic tests (already available yet not widespread) may reduce the number of inappropriately treated fetuses (i.e., cell-free fetal DNA testing). From this perspective, the risk/benefit ratio of prenatal therapy may improve.
The following paragraphs will present all possible scenarios based on the biological parents’ genotype.
3. CAH-Affected Father and Mother
If both parents have classical CAH, there is a 100% chance of a classical-CAH-affected fetus (50% female; 50% male). In this case, only sex determination is needed to guide the decision of considering prenatal therapy because all females would be appropriately treated and all males should be excluded. If early fetal sex testing is available, waiting for its results is recommended. If early testing is not feasible, dexamethasone should be considered; if started, karyotype analysis from chorionic villus sampling should be performed as soon as possible, and in the case of a male fetus, switching back to hydrocortisone should not be delayed
[41].
Conversely, if one of the two parents is affected by classical CAH and the other by non-classical CAH, the risk of having a child with classical CAH depends on their specific genotype: non-classical CAH patients may be compound heterozygous carriers of one classical-CAH mutation in 2/3 cases
[50], bringing the risk of having a child with classical CAH to 50% (25% females); the remaining 50% of fetuses would be affected by compound heterozygous non-classical CAH. In this case, decisions on prenatal therapy should rely on available prenatal diagnostic tests:
-
If early (i.e., results available by the eighth gestational week maximum) free fetal DNA testing is available, the decision of whether to start prenatal therapy can be made after fetal genetic testing. It is recommended to switch the patient’s therapy to hydrocortisone (if not already the chosen drug) to avoid inappropriate glucocorticoid delivery to the embryo/fetus until a prenatal diagnosis can be made on fetal DNA. In the case of a male fetus or a compound heterozygous non-classical CAH female fetus, hydrocortisone may be continued until delivery. Conversely, if a female fetus carries a biallelic classical CAH mutation, prenatal therapy should be considered.
-
If free fetal DNA testing is available for fetal sex determination only, the same recommendation of switching to hydrocortisone applies. In the case of a male fetus, continuing with hydrocortisone is recommended; in the case of a female fetus, there is a 50% chance of biallelic mutation so that prenatal therapy may be considered.
-
If no free fetal DNA testing is available, a prenatal diagnosis of biallelic CAH mutations could only be made after the 10th gestational week. In this case, a prenatal glucocorticoid treatment with dexamethasone before fetal genetic testing should be considered
[43]; if karyotype analysis shows a male fetus, such therapy should be interrupted, with careful perinatal care to avoid acute adrenal insufficiency in the case of classical CAH
[43]. The 75% risk of exposing a non-female and/or non-classical-CAH fetus to steroid excess must be taken into account
[8].
4. Classical-CAH Mother + Heterozygous Father
In these scenarios, the pregnant patient will already be under replacement therapy with glucocorticoids (i.e., hydrocortisone, prednisolone, or dexamethasone). Two scenarios may present:
- 1.
-
If the biological father carries a non-classical-CAH heterozygous mutation, the fetus will have a 50% chance of inheriting the non-mutated allele from the father and be a healthy carrier, and a 50% chance of being a compound heterozygous of classical + non-classical CAH mutations, clinically translating in non-classical CAH. In both cases, no prenatal therapy would be useful and avoiding dexamethasone for the pregnant patient is mandatory.
- 2.
-
If the father carries a classical-CAH heterozygous mutation, the risk of passing classical CAH to the child is 50%. Prenatal therapy may be considered, with prenatal diagnostic considerations.
5. Classical-CAH Father + Heterozygous Mother
In this scenario, the fetal genetic risk evaluation is the same as in other section. However, being the mother a healthy carrier, prenatal therapy would carry more risks of glucocorticoid excess to her. Appropriate counselling and careful risk/benefit evaluation is crucial in this setting.
6. Non-Classical-CAH-Affected Father and Mother
In this scenario, the fetal genetic risk of having classical CAH depends on the parents’ genotype. If both parents carry one classical-CAH allele as compound heterozygosity, the chance of classical CAH for the fetus is 25%. This would mean that if prenatal therapy is chosen, one in eight fetuses would be appropriately treated biallelic-mutated females, with seven inappropriately exposed to glucocorticoid excess. In this scenario, prenatal diagnosis appears to be especially important.
7. Non-Classical-CAH-Affected Parent and Classical-CAH-Carrier Parent
In this case, if the parent with non-classical CAH carries one classical-CAH allele as compound heterozygosity, the chance of classical CAH for the fetus is 25%.
8. Classical-CAH-Carrier Father and Mother
In this case, the fetal genetic risk of classical CAH is 25%.
9. Unknown-Status Parent and CAH-Affected/-Carrier Parent
If one parent has not been genotyped, the possibility of them being a healthy carrier must be considered. The estimated prevalence of carrier status in the general population is around 1:60 but varies based on ethnic background (i.e., up to 1:3 in Ashkenazi Jews, but mostly varying between 1:55 and 70 among the Caucasian and non-Caucasian white population; not enough literature data was found for other ethnic groups)
[50]. If the affected parent has classical CAH, the chance of having a child with classical CAH is therefore 1 in 120. Conversely, if the affected parent has non-classical CAH, it depends on their genotype and specific genetic counselling is advised. Carrying one classical-CAH mutation as compound heterozygosity (or as a healthy carrier) brings the risk to 1 in 240. If the non-classical CAH parent-specific genotype is not known, given the 2/3 risk of being a compound heterozygous carrier of one classical-CAH mutation, the chance of having a child affected by classical CAH is 1 in 360. In these scenarios, the risk/benefit ratio appears too high to recommend prenatal dexamethasone therapy unless early prenatal diagnosis is available.