Prenatal treatment of congenital adrenal

hyperplasia resulting from 21-hydroxylase

deficiency

In an attempt to prevent in utero virilization o f female fetuses with 21-hydroxylase deficiency, six
mothers at risk were treated with either hydrocortisone (n = 1) or dexamethasone (n = 5) in early
pregnancy. Treatment was continued to term in the two pregnancies in which the diagnosis o f an
affected female fetus was confirmed. In patient 1 (hydrocortisone treatment) fetal adrenal suppression
was only partial but the external genitalia were only slightly abnormal. In patient 2 (dexamethasone
treatment) fetal adrenal suppression was achieved and the external genitalia were normal at birth.
These encouraging results open a new prospect for treating congenital adrenal hyperplasia in utero.
(J PEDIATR 105:799, 1984)

Michel David, M.D., and Maguelone G. Forest, M.D., Ph.D. Lyon Cedex, France

A MAJOR COMPLICATION of congenital adrenal hyper-
plasia resulting from 21-hydroxylase deficiency is in utero
virilization of female fetuses, which is caused by overpro-
duction of adrenal androgens, probably as early as 10 to 16
weeks gestation. Recently the antenatal diagnosis of CAH
has been made by HLA typing and steroid measurements
in the amniotic fluid at mid-gestation,~ but diagnosis is
then too late to undertake any form of treatment. Because
cortisoF and dexamethasone3 cross the placenta, we
attempted prenatal treatment of fetuses with CAH by
giving the mother glucocorticoids from early pregnancy.
Because it is currently not possible to make the diagnosis of
CAH in utero before the tenth week of gestation, the
decision to start treatment must be m~de before knowing
whether the fetus of a high-risk mother has CAH. From
the review of data obtained in animals,4 we had some
concern regarding the possibility of short- or long-term
From the Pediatric Endocrinology Clinic and l N S E R M U. 34,
Hbpital Debrousse.
Supported by a grant from the University Claude-Bernard and by
INSERM.
Presented in part at the First Meeting o f the French Society for
Pediatric Research, Montpelier, France, Sept. 8, 1983.
Submitted for publication Dec. 27, 1983; accepted May 11,
1984.
Reprint requests: Maguetone G. Forest, M.D., Ph.D., 1 N S E R M
U. 34, HOpital Debrousse, 29 Rue Soeur Bouvier, 69322 Lyon
Cedex 05, France.
harmful effects on the fetus. However, there is no substan-
tial evidence of fetal effects of high doses (maternal
Cushing disease) or low doses of glucocorticoid therapy in
human pregnancy5"6 other than rare transient neonatal
adrenal insufficiency.7 A careful review of the literature
revealed that the potential risk of facial malformation in a
fetus whose mother receives glucocorticoid treatment
throughout pregnancy is small when there is no familial
history of cleft palate? We decided, therefore, to start
CAH Congenital adrenal hyperplasia
DHA Dehydroepiandrosterone
HLA Histocompatibility leukocyte antigens
OHP 17a-Hydroxyprogesterone
maternal treatment as soon as pregnancy was diagnosed, to
perform amniocentesis at mid-gestation, and to stop treat-
ment if the fetal karyotype and HLA typing indicated
either a male or a normal or heterozygote female. We have
used this protocol over the last 4 years in six high-risk
mothers. We present the history and results obtained in the
only two pregnancies in which a female fetus was predicted
to be affected and treatment was continued to term.
METHODS
Mother 1 (the first at-risk mother to be treated), a
33-year-old white woman had previously had a miscar-
riage, then a girl with CAH. She wanted a second child but

The Journal o f P E D I A T R I C S 799

800 David and Forest The Journal of Pediatrics
November 1984

Fig. i. External genitalia in newborn infants of family 1: !ndex case (A) and patient treated in utero (B).

Table I. Postnatal hormone values in two patients and their siblings

Age OHP Androstenedione Testosterone Plasma renin activity

(days) (ng/dl) (ng/dl) (ng/dl) (ng/ml/hr)
Family 1
Index case 1 4,238 469 72
3 4,022 365 44 50
Patient 88 19,070 2,685 416 48
I 21,510 4,898 380
Family 2
Index case 21 22,410 174 150
Patient 1 1,738 200 42
6 7,347 215 41
10 11,320 443 69 120
Normal girls 0 to 1 431 175 174 75 46 8.8 8.8
(mean _+ SD) 3 to 21 108 48 30 7 12 8 11.6 10.5
OHP, 17a-Hydroxyprogesterone.

was anxious because of the possibility of having another
child with ambiguous genitalla. She was aware of our
project and asked for prenatal treatment when she became
pregnant again. She was then given hydrocortisone.
Mother 2 was 21 years of age when she became pregnant
for the second time. After detailed explanation of the
Possibilities of treatment during gestation, she gave
informed consent for the study. She was given dexametha-
sone.
Routine amniocenteses were performed at mid-gesta-
tion; both were uneventful. The amniotic fluids were
immediately centrifuged and the amniotic fluid cells col-
lected for determining karyotype and H L A typing after
cultures. The supernatant was kept frozen until the steroid
analysis was made.
Blood was drawn in patient 1 at 18 weeks of pregnancy,
just prior to the increase in daily dose of hydr0cortisone to
50 rag. In patient 2 blood specimens were obtained before
treatment and 2 and 26 weeks later. These samples, as well
as those obtained postnatally in the index cases and in the
patients, were immediately centrifuged an d plasma stored
at - 2 0 ~ C until analysis.
Plasma and amniotic fluid levels of 17a-hydroxyproges-
terone, androstenedione, dehydroepiandrosterone, testos-
terone, and cortisol were measured by specific radioimmu-
noassay s after purification o n celite or Sephadex LH-20
columns. Details of procedures and normal values have
been reported? Estriol was measured on ethyl ether
extracts of plasma or amniotic:~fluid by using a specific
antibody (gift of Roussel-Uclaf), the only significant
cross-reaction being 4.2% for 17r Plasma renan
actiyity was estimate d by R I A (New England Nuclear kit)
as previously reported? ~
CASE R E P O R T S
Family 1. The index case of family 1 was the first child of
nonconsanguinous parents. She was born at 7.5 months gestation
by cesarian secti0n because of placenta pra~via. Birth Weight was
2650 gin. Because of ambiguous genitalia, consisting of enlarged
clitoris and complete fusion of the labia majora (Prader stage
IV 11) (Fig. 1, A), she was immediately referred to the endocrine
Volume 105 Prenatal treatment o f adrenal hyperplasia 80 1
Number 5

Table II. U n c o n j u g a t e d steroid values in a m n i o t i c fluid o f t r e a t e d m o t h e r s

Weeks of OHP Androstenedione Testosterone Cortisol Estriol

pregnancy (ng/dl) (ng/dl) (ng/dl) (ng/dl) (ng/dl)
Mother 1 17 981 582.0 24.5 2391 --
Term 742 212.0 30.7 4758 364
Mother 2 16 49 7.6 5.7 -- 103
Normal female fetuses5
Me,i n Mid-gestation 121 39.0 4,9 497 183"
Range 52 to 225 20 to 110 2.4 to 9.8 178 tO 1090 104 to 282

*Unpublished.

Table IlI. P l a s m a steroid c o n c e n t r a t i o n s in t r e a t e d m o t h e r s

Weeks of Weeks of OHP Androstenedione Testosterone DHA Cortisol Estriol

pregnancy treatment (ng/dl) (ng/dl) (ng/dl) (ng/dl) (#g/dl) (ng/dl)
Mother 1 18 9 314 99 36 40 7'. 1 465
Mother 2 4 0 970 385 86 1065 37.7 25
6 2 374 239 40 54 1.4 20
32 26 125 213 55 28 1.6 60
Normal 6 to 10 150 to 300 150 to 250 40 to 100 350 to 500 7.8 to 23 23 _+ 12
14 to 16 150 t0 300 150 to 250 40 to 100 350 to 500 7.8 to 23 183 _+ 74
32 to 34 250 to 500 210 _+ 84 121 -_+ 46 360 _+ 176 26.8 to 36.5 450 to 950

unit. No gonads were palpable, and the karyotype was 46,XX.
CAH was suspected and confirmed by hormonal studies (Table I).
Substitutive treatment (both glucocorticoids and mineralocori.i-
coids) was given on day 4 of life, before clinical salt loss was
apparent. The results of the plastic surgery (partial clitoridectomy
and perineoplasty), done at 1 year of age, are satisfactory, but
vaginoplasty will be needed.
Maternal treatment (hydrocortisone 40 mg/day orally) was
started during the second pregnancy at 914 weeks gestation.
Amniocentesis was performed at 16.7 weeks. Studies of amniotic
fluid cells showed a female fetus with HLA haplotypes identical to
those in the older sister. The diagnosis of CAH was confirmed by
abnormally high amniotic fluid levels of OHP, 64-androstenedione,
and testosterone (Table ll).
The mother at that time had normal O H P and cortisol values.
Maternal adrenal suppression was suggested, however, by the
extremely low levels of DHA and the lower than normal levels of
androstenedione and testosterone (Table III).
C0mplcte fetal adrcnat ~uppression was obviously not achieved
at m~d-pregnancy, as shown by the high levels of estriol in the
mother and abnormally high amniotic fluid levels of OHP,
androstenedione, and testosterone, compatible with the in utero
d!agnosis of CAH, ~ despite high amniotic fluid levels of cortisol
(Table iI). Thus the maternal dose was increased to 50 mg
hydrocortisone daily. At delivery (cesarean section) the low
amniotic fluid levels of estriol (364 ng/dl; normal 600 to 1400
ng/dl) indicated that the fetal adrenal glands were partially
suppressed.
Despite our doubts about the efficacy of the treatment, the
neonate had only moderate virilization of the genitalia: clitoral
hypertrophy and slight posterior fusion (Prader stage I!) (Fig. 1,
B). Birth weight was 4180 gm. Postnatal hormonal studies
confirmed the diagnosis of CAH (Table 1). Therapy was started at
2 days of age. Development was normal. The relative and absolute
size of the clitoris decreased wkh age. She is now 50 months of
age, and the urologist thinks no plastic surgery will be needed.
Family 2. The index case of family 2 was the female firstborn
child of nonconsanguinous parents, a full-term baby weighing
3890 gm at birth. Despite marked Sexual ambiguity, the infant
was routinely discharged from the obstetrical ward. In fact, the
child had severe masculinization of the external genitalia, with
penoscrotal hypospadias: the Phallus was 1.8 cm in length, the
glans was clearly delineated but the prepuce was not circular, the
urethral meatus was proximal to the til5 of the glans, and there
were no palpabie gonads (Fig. 2, A). At 3 weeks of age the infant
was referred for severe salt 10ss. Hormonal studies (Table I) and
karyotype (46,XX) confirmed the diagnosis of CAH. Glucocorti-
cold and mineralocorticoid treatment was immediately started.
Satisfactory perineoplasty was performed at 15 months of age, but
a secondary operation will be needed to provide a normal vaginal
outlet.
The mother became pregnant again 18 months later, and agreed
to treatment. She was given 0.5 mg dexamethasone twice a day
starting in the fifth week of pregnancy. Dexamethasone was used
rather than hydrocortisone because of th~incompiete fetal adrenal
suppression obtained in the first treated fetus.
Amniocentesis at 15 weeks gave apparently contradictory
results: fetal 46,XX karyotype, HLA identical to the affected
802 David and Forest
Fig. 2. External genitalia in newborn infants of family 2: index case (A) and Patient treate d in utero (B).
sibling, but normal and even low steroid values (Table II). In fact,
they were to be expecte d in the case of efficient adrenal suppres-
sion in a CAH affected fetus. Indeed, amniotic fluid levels of OHP
and androstenedione were lower than normal, and those of
testosterone and estriol were normal. Maintenance of adrenal
suppression throughout pregnancy in this fetus was suggested by
the lack of any significant rise in maternal estriol levels (Table
Iti). The mother's adrenal glands were completely suppressed, as
shown by the fall in OHP, DHA, and cortisol and the reduction in
androstenedione and testosterone plasma values (Table III).
Maternal treatment was continued at the same dosage until term.
After a spontaneous delivery the baby, weighing 3150 gm, had
normal external genitalia: normal clitoris, almost no posterio r
fusion of the labia majora, and a normal vaginal opening (Fig. 2,
B).
Despite the evidence of maintained adrenal suppression in
utero, the effects of transplacental steroid suppression were
rapidly reversed in the neonate. Hormonal and clinical changes
confirmed the diagnosis of a severe form of salt-losing CAH.
Plasma OHP values on the first day of life were in the upper range
of normal; those of adrostenedione and testosterone were normal.
A few days after birth there was a dramatic rise in OHP plasma
concentrations. The baby was transferred to our clinic on the
seventh day Of life. At that time, despite salt supplementation (1
gm/day) given during the time needed to collect biochemical
specimens, plasma Na concentration fell to 128 mEq/L, and on
day 10, plasma values of OHP, androstendione, and testosterone
(Table I) were clearly diagnostic of the disease. She was then
given substitutive, therapy, and has done well since.
DISCUSSION
Postnatal treatment of C A H , based on a negative
feedback mechanism, is achieved by giving physiologic
doses of hydrocortisone and mineralocorticoids. It is not
certain at what time physiologic feedback is operative in
utero, but the fact that female fetuses with C A H are
masculinized is evidence that it is active at the critical
time ~2 when testosterone and dihydrotestosterone 13 are
responsible for male differentiation (9 to 14 weeks) (see
review in Forest~4). Thus the basis for treating female
The Journal of Pediatrics
November 1984
fetuses with C A H is theoretically the same as postnata!ly,
provided that therapy is initiated before this critical period.
Placental transfer of cortisol has been demonstrated in
late gestation. 2 It also occurs in mid-gestation, as shown by
the high levels in the amniotic fluid from patient l,
increasing when the therapeutic dose was increased in the
mother. The transplacental passage of dexamethasone has
been shown in early gestation 3 and suggested in late
gestation? 5 Placental transfer and half-life are greater for
dexamethasone than for cortisol. Dexamethasone, being
also more effective in suppressing A C T H , ~6is thus theoret-
ically a better form of treatment, as shown by our results
adrenal suppression appeared partial in fetus 1 but com-
plete in fetus 2.
It remains difficult to prove the efficiency of an in utero
treatment because of the paucity of biologic markers: it is
not ethical to obtain fetal blood for this purpose, and the
mother's estriol levels are only indirect evidence. Our main
purpose was to prevent the masculinization of external
genitalia in these two fetuses at risk of CAH. It is common
experience that the degree of fetal virilization varies
somewhat in the same families and is not strictly correlated
with the degree of hormonal abnormalities at the time of
postnatal diagnosis (variable intercurrent stress). Prader
stage s I I I to IV are, however, the most frequent (85% in
our experience) in severe CAH. The two infants treated in
utero both had a severe form of 21-OH deficiencY, as
evidenced by their very high postnatal plasma O H P levels
before institution of substitutive therapy. Each infant was
considerably less virilized than her respective sibling, and
the first infant, whose adrenal suppression in utero was
partial, had moderate abnormalities of the genitalia,
whereas the second, with apparently complete adrenal
suppression in utero, had normal genitalia. Reconstructive
surgery will not be necessary in the latter infants; in our
experience, this has been observed only once in more than
50 female infants with C A H untreated in utero.
Despite the lack o f definitive proof that prenatal treat-
Volume 105
Number 5
ment was directly responsible, the results appear encourag-
ing to us. The treatment has been well tolerated by all
mothers. In addition, it is easy and inexpensive, and no
malformations were observed (facial in particular) in
either infant.
We thank Dr. P. Guibaud for genetic counseling, Dr. H. Betuel
for HLA typing, M. T. Zabot for culturing the amniotic fluid
cells, Dr. D. Germain and Dr. C. Laurent for fetal and patient
karyotypes, Dr. P. Baxter for editorial help, and Miss J. Bois for
typing the manuscript.
REFERENCES
1. Forest MG, Betuel H, Couillin P, Bou6 A: Prenatal diagnosis
of congenital adrenal hyperplasia (CAH) due to 21-hydroxy-
lase deficiency by steroid analysis in the amniotic fluid of
mid-pregnancy: Comparison with HLA typing in 17 pregnan-
cies at risk for CAH. Prenat Diagn 1:197, 1981.
2. Beitins IZ, Bayard F, Ances IG, Kowarski A, Migeon C J:
The metabolic clearance rate, blood production, interconver-
sion and transplacental passage of cortisol and cortisone in
pregnancy near term. Pediatr Res 7:509, 1973.
3. Chrousos GP, Evans MI, Mann DL, Larsen JW, Loriaux DL,
Fletcher 3C, Schulman JD: Attempted prevention of abnor-
mal genital masculinization in congenital adrenal hyperplasia
(CAH) by adrenocortical suppression in utero (abstract 441).
Pediatr Res 17:160A, 1983.
4. Taensch HW Jr: Glucocorticoid prophylaxis for respiratory
distress syndrome: A review of potential toxicity. J PEDIATR
87:617, 1975.
5. Yackel DB, Kemper RD, McConahey WM: Adrenocortico-
steroid therapy in pregnancy. Am J Obstet Gynecol 96:985,
1966.
6. Ferrier PE: Editorial: Des t6ratog~nes. Helv Paediatr Acta
30:5, 1975.
Prenatal treatment o f adrenal hyperplasia 803
7. Kreines K, de Vaux WD: Neonatal adrenal insufficiency
associated with maternal Cushing's syndrome. Pediatrics
47-516, 1971.
8. Davies JI: The fetal adrenal. In Tulchinsky D, Ryan K J,
editors: Maternal-fetal endocrinology. Philadelphia, 1980,
WB Saunders, pp 242-251.
9. Forest MG, de Peretti E, Leeoq A, Cadillon E, Zabot MT,
Thoulon JM: Concentration of 14 steroid hormones in human
amniotic fluid of mid-pregnancy. J Clin Endocrinol Metab
51:816, 1980.
10. Sassard J, Sann L, Vincent M, Francois R, Cier JF: Plasma
renin activity in normal subjects from infancy to puberty.
J Clin Endocrinol Metab 40:524, 1975.
11. Prader A, Gurtner HP: Das Syndrorn des Pseudohermaph-
roditismus masculinus bei Kongenitaler Nebennierenrinden-
Hyperplasie ohne Androgen-Uber produktion (Adrenaler
Pseudohermaphroditismus masculinus). Helv Paediatr Acta
10:397, 1955.
12. Grumbach MA, Ducharme JR: The effects of androgens on
sexual development: Androgen-induced female pseudoher-
maphroditism. Fertil Steril 11:157, 1960.
13. Siiteri PK, Wilson JD: Testosterone formation and metabo-
lism during male sexual differentiation in the human embryo.
J Clin Endocrinol Metab 34:74, 1973.
14. Forest MG: Development of the male reproductive tract. In
White RD, editor: Aspects of male infertility. Int Perspect
Urol 4:1, 1982.
15. Osathanondh R, Tulchinsky D, Kamali H, Fencl M, Taeusch
HW: Dexamethasone levels in treated pregnant women and
newborn infants. J PEDIATR90:617, 1977.
16. Khalid BAK, Burke CW, Hurley DM, Funder JW, Stockigt
JR: Steroid replacement in Addison's disease and in subjects
adrenalectomized for Cushing's disease: Comparison of vari-
ous glucocorticoids. J Clin Endocrinol Metab 55:551, 1982.