Hi Jo,

 

I have not been in this position personally, and do not expect to be, but, like you, I think I'd want to know ALL the potential risks of prenatal treatment, before weighing whether it is worse risking having a virilized little girl or a baby unnecessarily exposed to steroids.  The article I posted to Aimee below, by Dr. New--- which, as Julia S. pointed out, has a lot of info about prenatal dex treatment---contains mostly the GOOD news.  Here is a more balanced view.  In this same article, Dr. Speiser also offers a great discussion about the genetic factors to consider, when weighing the risks and benefits of prenatal treatment.  I'll skim back through the article and post that part, too, when I find it. 

 

Endocrine Reviews 21 (3): 245-291
Copyright © 2000 by The Endocrine Society

Congenital Adrenal Hyperplasia due to 21-Hydroxylase Deficiency¹

Perrin C. White and Phyllis W. Speiser

G. Prenatal therapy

.......Prenatal therapy is usually coupled with prenatal diagnosis (see Section VI.H). Since dexamethasone treatment suppresses amniotic fluid adrenocortical hormones, genetic diagnosis must be performed. Although the incidence of fetal deaths in treated pregnancies does not appear to exceed that for the general population [9% spontaneous abortion rate in treated pregnancies, compared with 14% in untreated pregnancies (317)], a high rate of spontaneous abortions has been observed after chorionic villus sampling (CVS) performed to obtain tissue for genetic diagnosis (318). Either amniocentesis or CVS may be done for diagnostic purposes, but the latter should be carried out only in experienced centers at 10–12 weeks gestation (319). If the sex is male, or CYP21 genotype indicates the fetus is unaffected, dexamethasone should promptly be discontinued to minimize potential risks of glucocorticoid toxicity (Fig. 9).

2. Therapeutic risks to the fetus. Most published literature dealing with fetal exposure to glucocorticoids deals with late second or third trimester treatment. Prenatal treatment of CAH is different in that it must begin in the early first trimester to be effective in preventing female genital ambiguity. Since only one of eight fetuses is likely to be an affected female, seven of eight pregnancies will be unnecessarily exposed to at least several weeks of dexamethasone treatment before the sex and disease status can be determined. Because the unaffected offspring will not derive any benefit from glucocorticoid treatment, the issue of potential risk is all the more significant. Normal fetal cortisol production is considerably lower than that of children or adults; samples obtained during fetoscopy, show umbilical vein cortisol at 16–20 weeks gestation to average approximately 20 nmol/liter (0.72 µg/dl) (320). Thus, providing pharmacological doses of a potent, long-acting glucocorticoid such as dexamethasone might disrupt fetal physiology. Potential risks include congenital malformations such as cardiac septal hypertrophy (321), hydrometrocolpos (322), and hydrocephalus (318). Moreover, the effects of abrupt dexamethasone withdrawal on fetal development are unknown. Intrauterine growth retardation and unexplained fetal death have been observed in 2% or less of treated pregnancies (317); these statistics are not significantly different from those found in the population at large (323). The risk of overt human fetal defects appears to be low compared with complications observed in a rodent model of in utero exposure to high-dose glucocorticoids (324), which features frequent cleft palate in addition to fetal growth retardation and/or demise. Pregnant rats treated with dexamethasone, 20 µg/kg/day (the same weight-based dose used in human prenatal treatment for CAH), produced litters with average birth weights 14% below those of controls; the offspring were also hypertensive at 6 months, i.e., young adulthood (325). Mid-to-late gestation administration of dexamethasone to fetal rats alters levels of transcription factors important in brain development (326). It is conceivable that more subtle effects of glucocorticoids on the developing human brain may go unnoticed during early life. A pilot study using maternal surveys suggested that children not affected with CAH who were subjected to prenatal dexamethasone treatment were more shy than untreated children (327). However, prenatally treated children have not undergone thorough neuropsychological testing. Recent editorials called attention to these issues, citing numerous studies in animal models showing the dangers of prenatal exposure to glucocorticoids with respect to impairment of somatic growth, brain development, and blood pressure regulation (328, 329, 330). An international registry might facilitate long-term studies that could answer many of these questions.

3. Therapeutic risks to the mother. The incidence of maternal complications has varied among investigators; overall, it is about 10% (310). In evaluating such data, it is important to correlate the types of adverse side effects observed and the duration of therapy. Both American (331) and European (311) investigators have found a higher incidence of side effects in women treated from the first through third trimesters. Cushingoid features, excessive weight gain, severe striae, hypertension, and hyperglycemia are seen in this setting. These side effects most often resolve with discontinuation of treatment. Weight, blood pressure, and glucose tolerance should be closely monitored in all treated women treated to term. Serial maternal serum dexamethasone levels, if available, might prevent over- and undertreatment, but they have not been followed routinely. Maternal urinary estriol measurements have also been suggested as a guide to adjusting maternal treatment (304). A gradual decrease in the dose of dexamethasone later in gestation might decrease the incidence of maternal side effects, but there are as yet no data concerning the efficacy of such a regimen. More common maternal side effects in those treated for a shorter duration include edema, gastrointestinal upset, mood fluctuations, acne, and hirsutism; one or more of these symptoms are seen in 10–20% of women treated in early pregnancy.

Because of these concerns, caution should be exercised in recommending prenatal therapy with dexamethasone, and women must be fully informed of potential fetal and maternal risks, some of which may be as yet unrecognized. Additionally, the possibility of lack of therapeutic benefit should be disclosed when obtaining informed consent......