Download Fertility and pregnancy outcome in women with congenital adrenal

doi:10.1093/humrep/den118
Human Reproduction Vol.23, No.7 pp. 1607–1613, 2008
Advance Access publication on April 16, 2008
Fertility and pregnancy outcome in women with congenital
adrenal hyperplasia due to 21-hydroxylase deficiency
K. Hagenfeldt1, P.O. Janson2, G. Holmdahl3, H. Falhammar4,5, H. Filipsson6, L. Frisén7,
M. Thorén4,5 and A. Nordenskjöld5,8,9
1
Department of Women and Child Health, Division of Obstetrics and Gynaecology, Karolinska Institutet, Karolinska University Hospital,
171 76 Stockholm, Sweden; 2Department of Obstetrics and Gynaecology, Sahlgrenska Academy, Sahlgrenska University Hospital, 413 45
Göteborg, Sweden; 3Department of Paediatric Surgery, Sahlgrenska Academy, Sahlgrenska University Hospital, 413 45 Göteborg,
Sweden; 4Department of Endocrinology, Metabolism and Diabetes, Karolinska Institutet, Karolinska University Hospital, 171 76
Stockholm, Sweden; 5Department of Molecular Medicine and Surgery, Karolinska Institutet, 171 76 Stockholm, Sweden; 6Department of
Endocrinology, Sahlgrenska Academy, Sahlgrenska University Hospital, 41345 Göteborg, Sweden; 7Department of Clinical Sciences,
Division of Psychiatry, Karolinska Institutet, Danderyd Hospital, 182 87 Stockholm, Sweden; 8Department of Pediatric Surgery,
Karolinska Institutet, Karolinska University Hospital, 171 76 Stockholm, Sweden
9
Correspondence address. Fax: þ46 8 517 73620; E-mail: [email protected]
BACKGROUND: Low pregnancy rate has been reported in women with congenital adrenal hyperplasia (CAH) and
little information on pregnancy and children is known. METHODS: In a Swedish study, 62 adult women with CAH,
aged 18–63 years, and 62 age-matched controls were followed-up. Medical records, including those concerning pregnancies and deliveries, were examined and the 21-hydroxylase genotype of patients was noted. All women answered a
questionnaire concerning sexual and reproductive health including health of the children. RESULTS: Pregnancy and
delivery rates were significantly lower in women with CAH (P < 0.001, P < 0.0056, respectively), and the severity of the
21-hydroxylase-mutation correlated with the reduced number of children born. More women with salt-wasting CAH
were single and had not attempted pregnancy. Pregnancies were normal except for a significantly increased incidence
of gestational diabetes in CAH patients (P < 0.0024). The children had normal birthweight and no malformations were
observed. A later follow-up of the children showed a normal intellectual and social development. The sex ratio of the
offspring differed significantly, with 25% boys in the CAH group compared with 56% among controls (P < 0.016).
CAH women had more gynaecological morbidity during menopause. CONCLUSIONS: Pregnancy and delivery
rates are reduced in women with CAH mainly due to psychosocial reasons. The outcome of children did not differ
from controls. The unexpected sex ratio in children born to mothers with CAH warrants further research.
Keywords: congenital adrenal hyperplasia; 21-hydroxylase mutation; pregnancy; sex ratio; gestational diabetes
Introduction
Congenital adrenal hyperplasia (CAH) caused by 21-hydroxylase
deficiency is an autosomal recessive disorder characterized by
impaired or total loss of activity in the enzyme required for the
synthesis of cortisol and in 75% of cases also a failure to synthesise aldosterone. CAH is clinically divided into classic CAH
that includes salt-wasting (SW), simple virilizing (SV) and
non-classic (NC) disease. All three types are highly correlated
with the spectrum of 21-hydroxylase mutations and to the
severity of the disease (Wedell et al., 1994; Pinto et al.,
2003). In classic CAH, lifelong glucocorticoid and often also
mineralocorticiod therapy are mandatory, whereas in the
milder NC form, treatment is given when patients have
symptoms such as hirsutism, oligo-amenorrhoea or infertility.
If corticosteroid supplementation is insufficient, androgen
production from the adrenals will increase and suppress
gonadotrophin secretion from the pituitary that will lead to a
disturbance of the ovarian cycle resulting in anovulation and
infertility.
Until now, there has been a scarcity of reports on the
outcome of pregnancies in women with CAH. Traditionally,
low pregnancy rates have been reported especially in
women with the SW form of classical CAH (Mulaikal et al.,
1987; Jääskeläinen et al., 2000; Lo and Grumbach, 2001;
Stikkelbroeck et al., 2003; Gastaud et al., 2007). The reasons
for this low fertility in women with CAH have been discussed
in several articles (Meyer-Bahlburg, 1999; Jääskeläinen et al.,
2000; Otten el al., 2005). The literature on the outcome of
# The Author 2008. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology.
All rights reserved. For Permissions, please email: [email protected]
1607
Hagenfeldt et al.
pregnancies with regard to the progression of the pregnancies
and the health of the children is even sparser. ‘Normal child’
is usually the only information given in most reports. More
information is needed from long term follow-up studies
where the outcomes of the pregnancies are related to the genotype and not only to the phenotype of women with CAH.
In this follow-up study from Sweden, 62 adult women with
CAH due to 21-hydroxylase deficiency have been investigated
to provide information on the influence of the genotype on
menstrual pattern, fertility preferences and fertility and pregnancy outcome.
Statistical methods
The women with CAH were individually matched with the control
women and thus data were considered as paired. In comparisons
between the patients and their controls, McNemar test or the sign
test was performed for categorical data and the Wilcoxon matched
pair test for continuous data. For the outcome of term pregnancies,
statistical methods for independent samples were used. The
Chi-square test or Fisher’s exact test was used for categorical data
and the Mann–Whitney U-test for continuous data. A P-value of
,0.05 was considered statistically significant.
Results
Materials and Methods
The follow-up study was part of a collaborative project between
Stockholm and Göteborg in Sweden to evaluate adult women with
CAH from an endocrinological, gynaecological, psychosocial and
psychosexual perspective. The performance of the study was as previously described (Falhammar et al., 2007a). Sixty-two women, 18 –
63 years old, were recruited either by their physician or through the
National CAH patient organization. An additional 17 CAH women
were contacted, 4 declined to participate due to personal reasons
and 13 could not take part because of different practical reasons.
The proportion of different forms of the disease in these 17 women
was similar to those followed-up. Sixty-two women born on the
same day as the CAH patients were recruited as controls from the
Swedish Population Registry. They were examined within 1 year
from their age-matched CAH patients. All but two of the CAH
women and the controls were of North-European, Caucasian ethnicity.
Half of all women were younger than 30 years.
The women with CAH were analysed for mutations of the
21-hydroxylase gene with the mildest form of the two alleles as representative of the genotype. Fourteen women had a SW, potentially lifethreatening form of CAH (corresponding to no enzyme activity, i.e.
null/null), 15 had an I2splice-mutation (SW—a few percent of
normal enzyme activity), 27 had mutations corresponding to the SV
form of CAH and 6 had the NC form. Age of onset and initial symptoms varied with the severity of the mutations and has been previously
described (Falhammar et al., 2007a).
All women gave their signed informed consent to participate in the
study and were examined as outpatients between October 2002 and
January 2005. The controls were seen within 1 year after the patients
visit. A medical history was taken and all participants answered a
semi-structured questionnaire (120 items) concerning treatment,
earlier surgery, fertility, sexuality, pregnancies and social life. Only
the women with CAH underwent a gynaecological examination to
evaluate their present situation with regard to virilization and postoperative status as reported by us recently (Nordenskjöld et al.,
2008). With the written permission of the CAH patients, all records
of earlier treatment within the healthcare system were obtained and
reviewed. Among women with a history of pregnancy, CAH and controls, the records from the antenatal clinic as well as records from the
delivery wards were collected. The details of the outcome of the pregnancies were extracted from the records and compared with the information given by the women themselves. The process was facilitated
by the fact that all antenatal clinics and delivery wards in Sweden
have been using the same forms for more than three decades. In
order to inform on the post-natal development of the children born
to CAH mothers, one structured telephone or personal interview was
performed 2– 3 years after the first examination.
The Research Ethics Committee of Karolinska Institutet and of
Göteborg University, Sweden, approved the study.
1608
Menstrual pattern, social situation and fertility
There was no difference in age at menarche or first pregnancy
between the women with CAH and their matched controls
(Table I). However, significant differences were observed
between age at sexual debut, number of women living in a
steady heterosexual relationship and number of women who
had attempted pregnancy—all factors having an influence on
fertility outcome. Irregular menstruation, an indication of anovulatory cycles, was similar in women with CAH and controls;
28% of those women in whom the menstrual cycles could be
assessed (Table I). The number of women experiencing menopause (five women) or secondary amenorrhoea (three women)
was the same in the two groups. The use of hormonal contraceptives was similar among the patients and the controls, 17
and 24 women, respectively. Differences were observed
among the women with CAH when their mutation status was
taken into consideration (Table II). Notably, less than one-third
of the women with the null/null mutation were living in heterosexual relations, whereas 81% of the women with the SV
mutation were married or living in a steady heterosexual
relationship. Bi/homosexuality was more common among
the women with SW than among those with SV. Twenty-three
of the 29 women with SW answered the questions on this issue
and 8 considered themselves bi/homosexual. Twenty-three of
the 27 women with SV answered the same questions and 2 considered themselves as bi/homosexual. However, most of these
women were single without sexual partners. As was presented
Table I. Factors affecting fertility in women with CAH and their
age-matched controls (number and median and range).
CAH n ¼ 62
Controls
n ¼ 62
Age, years (range)
30 (18–63)
31 (19–63)
Menarche age, years (range)
13.1(8–20) 12.8 (10–16)
Sex debut age, years
18.7
16.6
No steady relationship n(%)
24 (39%)
11 (18%)
Pregnancy attempt n
19
41
Never pregnant n(%)
46 (74%)
21 (34%)
Age first pregnancy, years
30.0 (21–39) 27.7 (19–42)
(range)
Irregular menstruation n (%)§ 11/39 (28%)
9/32 (28%)
P-value
NA
NS
,0.0094a
,0.031b
,0.0098b
,0.001b
NS
NS
NA, not applicable; NS, not significant.
§Excluding menopausal women, women having secondary amenorrhoea or
using hormonal contraception.
a
Wilcoxon matched pairs.
b
McNemar.
Fertility, pregnancy and congenital adrenal hyperplasia
Table II. Fertility factors correlated to 21-hydroxylase gene mutation in
women with CAH.
Mutation
Number
Number of
women with
children (%)
No heterosexual
relationship (%)
Age <30 years
Sex debut, age
(number)
Null/null
I2splice
14
0 (0)
10 (71%)
SV
NC
15
2 (13%)
27
9 (33%)
6
3 (50%)
8 (53%)
5 (19%)
2 (33%)
8
7
12
3
17.5 (n ¼ 12) 20.6 (n ¼ 8) 19.0 (n ¼ 25) 16.4 (n ¼ 5)
SV, simple virilizing; NC, non-classic.
by us recently, 49 of the women with CAH had been corrected
surgically with clitorido- and/or vaginoplasty. In the SW
group, all 29 patients were operated, in the SV group 20/27
and none of the NC patients. Whether operated or not, the
women with CAH were often less satisfied with the function
and appearance of their genitals than their age-matched controls (Nordenskjöld et al., 2008).
When comparing pregnancy history, significantly fewer
women with CAH had ever been pregnant, 16 compared with
41 in the control group (P , 0.0001) (Table III). A total of
31 pregnancies were seen among the 62 women with CAH
compared with 76 among the 62 women in the control group
(P , 0.0001). Only 25 children were born in the CAH group
compared with 54 in the control group (P , 0.0056). The frequency of spontaneous abortions did not differ between the
patients and the controls, but there were more induced abortions in the control group (P , 0.004). All induced abortions
were performed on social grounds.
When the number of children born is related to the mutation
status of the women with CAH, the severity of the mutation
clearly influenced the results. There were no children born from
SW women with the null/null mutation, 3 children from 15
SW women carrying the I2splice mutation, 19 children from
27 women with SV and 3 children from 6 women with NC
CAH. One pregnancy of a woman with the null/null mutations
ended in a therapeutic abortion, another woman with the same
mutation had a spontaneous abortion.
Nineteen patients had attempted pregnancy. Three of these
belong to the oldest age group and only got superficial evaluation and treatment .20 years ago with Clomiphene without
success. Two patients had unwanted pregnancies; one resulted
in a spontaneous and the other one in a therapeutic abortion.
Among the remaining 14 patients, four became pregnant spontaneously and did not need any evaluation. Five women had
irregular menstruation and ovulation, an increase in the glucocorticosteroid dose or addition of fludrocortisone induced ovulation and pregnancy. No other evaluation or treatment was
necessary. In five women, the pregnancies were the result of
infertility investigation and relevant treatment. In this group
no male factors were found, they all had normal sperm count.
No further genetic or hormonal investigation was performed
according to the clinical routine procedure. One woman with
SV was treated with gonadotrophins and insemination with
her husband’s sperm. The treatment resulted in two pregnancies, each with one child. Another woman with SV gave
birth to two children, each after IVF treatment. Both these
women had increased serum progesterone levels in the follicular phase and polycystic ovary syndrome (PCOS) like ovaries.
Because of anovulation, PCO like ovaries and high progesterone levels, one woman (NC) underwent a period of infertility
treatments including several cycles of gonadotrophin treatment
and IVF, but without result. She finally got pregnant on her
usual corticosteroid doses with metformin added to decrease
insulin resistance (Falhammar et al., 2008). Another two
CAH women, one SV and one NC, received treatment for moderate endometriosis. None of the control women had undergone
infertility evaluation or treatment.
Outcome of the pregnancies
The length of pregnancy was significantly shorter in the women
with CAH than in their age-matched healthy controls (P ,
0.022) (Table IV). The reason for this is the high frequency
of elective Caesarean section among the patients. Thus, 84%
of the deliveries were elective or acute Caesarean sections in
the women with CAH compared with only 9% in the control
group. The main reason for a Caesarean section in the patient
group was earlier genital surgery but in two non-operated
patients, a cephalic-pelvic disproportion. The pregnancy
length was normal in the four pregnancies that ended in a spontaneous vaginal delivery and in the three pregnancies that
started with normal labour but ended in an acute Caesarean
section. An unexpected finding was the incidence of gestational
diabetes among the women with CAH. Three women experienced this complication during five pregnancies. The first
Table IV. Outcome of term pregnancies in women with CAH and
age-matched controls.
Table III. Outcome of pregnancies in women with CAH and their
age-matched controls.
CAH n ¼ 62 Controls n ¼ 62
CAH n ¼ 62 Controls n ¼ 62 P-value
Ever pregnant, n
16
Total number of pregnancies, n 31
Term pregnancies, n
25
Spontaneous abortion, n
2
Therapeutic abortion, n
4
41
76
54
4
18
,0.0001a
,0.0001b
,0.0056b
NS
,0.004c
Number of children, n
25
54
Length of pregnancy (weeks)
38.3
39.2
Caesarean section, n (%)
21/25 (84%) 5/54 (9%)
Gestational diabetes/Number 5/25 (20%) 0/54 (0)
of pregnancies, (%)
Pre-eclampsia, n (%)
0/25 (0)
2/54 (3.7%)
,0.0056a
,0.022b
,0.001c
,0.0024c
NS
NS, not significant.
Wilcoxon matched pairs.
Mann–Whitney test.
c
Fisher’s exact test.
a
a
b
b
Mc Nemar.
Wilcoxon matched pairs.
c
Sign test.
P-value
1609
Hagenfeldt et al.
Table V. Treatment before and during pregnancy in the CAH women who gave birth.
Treatment
Total number of women,
n ¼ 14
Number of children,
n ¼ 25
Dose change during pregnancy
Prednisolone (5– 17.5 mg)1fludrocortisone (0.1–0.15 mg) N
Cortison-acetate (37.5 mg)1fludrocortisone (0.1 mg) N
Hydrocortisone (30– 40 mg)1fludrocortisone (0.1– 0.2 mg) N
Cortison-acetate (37.5 mg) N
Dexamethazone (1 mg) N
Prednisolone (5– 6.25 mg) N
7
1
1
2
1
2
12
1
2
4
1
5
One decrease, one increase
no
Hydrocortisone dose increased
no
no
no
woman (SV) developed gestational diabetes during each of her
three pregnancies and received insulin treatment during the last
two. Her corticosteroid dose before and during the pregnancies
was moderate. The second woman (SV), while on a high prednisolone dose when pregnancy started, developed insulin
dependent diabetes during the last trimester of her first pregnancy. Finally, the third woman (NC) was treated before the
pregnancy with corticosteroids in a moderate dose with metformin added to decrease insulin resistance. She discontinued
metformin in early pregnancy and developed mild gestational
diabetes during the last part of the pregnancy and was treated
with insulin (Falhammar et al., 2007c). All these three
women had a normal BMI (20 – 24 kg/m2) before pregnancy
and a normal weight increase during the pregnancy. There
were no cases of diabetes in the control group. No cases of preeclampsia were registered among the women with CAH, but in
almost 4% among the control women. The medications of the
CAH mothers are presented in Table V. Nine of the 14 women
received fludrocortisone in addition to their corticosteroid
medication. Only 2 of the 14 women, one woman during two
pregnancies, needed an increase in the corticosteroid dose
during the last trimester of their pregnancies. In one woman,
the prednisolone dose was decreased.
Outcome of the children
There were no differences concerning the outcome of the children born to mothers with different mutations. Thus, the birthweight and the future development did not differ between
children born to mothers with SW, SV or NC. Furthermore,
the weights of the first-born child, the weight of the girls and
the boys were similar among children born to mothers with
CAH or to the control women (Table VI). All children with
the exception of one in the CAH group and one in the
control group had an Apgar score of 10 at 5 min. None of the
children were small for gestational age (SGA). One child
from a mother with CAH was prematurely born in the 34
week, weighed 2270 g, and developed infant respiratory distress syndrome, which warranted a 3 weeks stay in the neonatal
intensive care unit. This boy was diagnosed as having classical
CAH at the age of four and has since developed normally on
corticosteroid treatment. Two children in the control group
were born before 37 full weeks, weighing 2270 and 2790 g,
both of them developed normally post-natally. There were no
malformations among the children of the CAH mothers; none
of the girls had any signs of virilization. One child in the
control group had a small ventricular septum defect.
1610
An unexpected finding was the difference in the sex ratio
among the children with significantly fewer boys than girls,
being born to CAH mothers than to their age-matched controls
(Table VI).
Later follow-up of the children
The results of the structured personal or telephone interview
with the 14 women with CAH who had born children (n ¼ 25)
are presented in Table VII. With few exceptions, all children
were healthy and had remained so since the post-natal
period. One boy of a SV mother was diagnosed at the age of
4 with CAH, I173N/I2splice, and developed well on treatment.
All other children had normal growth according to their growth
channel. Two girls were dyslectic; one of these had results
above average in school and was now a university undergraduate, whereas the other girl performed less than average in
school. All other children had normal intellectual development
and school performance. Three children had allergic problems,
childhood eczema or bronchial asthma.
Other gynaecologic morbidity
Three women with CAH, all SW with the null/null mutation,
had long standing secondary amenorrhoea, probably due to
inadequate compliance with their corticosteroid medication
as they all had high serum levels of 17-OH-progesterone and
testosterone (Falhammar et al., 2007a). Three of the control
women also had secondary amenorrhoea, one due to ongoing
breastfeeding, and one as a result of treatment for Mb
Hodgkin 10 years earlier and one due to PCOS.
Of the seven women with CAH in the perimenopausal and
menopausal age group, three had spontaneous menopause
without any major problems. Four had uterine fibroids and
Table VI. Outcome of term pregnancies concerning children born to women
with CAH and their age-matched controls (number or median and range).
CAH n ¼ 62
Controls n ¼ 62
Number of children, n
25
54
Fetal weight, gram
(range)
First born child
3339
3347
Girls
3303 (2892– 3905) 3455 (2732– 4190)
Boys
3212 (2270– 3825) 3359 (2270– 3975)
Sex ratio, female/
19/6 (25%)
23/31 (56%)
male, n (percentage of
males)
a
Wilcoxon matched pairs.
Chi-square.
b
P-value
,0.0056a
NS
NS
NS
,0.016b
Fertility, pregnancy and congenital adrenal hyperplasia
Table VII. Children born to 14 CAH mothers—later development.
Age group (years)
Number
Sex ratio Female/male
Number with normal development
Other findings
2 –6
7 –9
11– 13
16– 20
31– 39
Total
6
6
5
4
4
25
4/2
5/1
4/1
3/1
3/1
19/6
6
6
5
2
3
22
Two with allergy#
One allergy
One boy CAH*, one girl dyslexia, IQ above average
One woman dyslexia and low IQ
#One girl with Henoch– Schönleińs purpura at 3 years of age.
*Diagnosed at 4 years of age, early puberty, otherwise healthy.
menometrorraghia that resulted in hysterectomy, endometrial
ablation or treatment with high doses of Depot-ProveraR or
use of a Levo-Norgestrel device prior to the menopause. All
of the control women in the same age group reported having
had a spontaneous menopause or perimenopause without any
problems.
Discussion
The low frequency of pregnancies among women with CAH
compared with their age-matched controls is consistent with
other previous reports (Mulaikal et al., 1987; Jääskeläinen
et al., 2000; Krone et al., 2001; Lo and Grumbach, 2001; Stikkelbroeck et al., 2003; Gastaud et al., 2007). There were no
differences in spontaneous abortions between the patients and
the control women, but significantly more therapeutic abortions
among the control women. The number of term pregnancies
among the CAH women was clearly related to the severity of
the mutation. In the group with the most severe mutation,
there was no term pregnancy, and in the women with the SV
and NC mutation, the fertility was almost normal. These findings are also consistent with the literature. Lo and Grumbach
(2001) reviewed the literature published 1956 – 2000 and
reported on 105 pregnancies resulting in the birth of 73 children in women with virilizing adrenal hyperplasia. Of the 20
women with a clinically diagnosed SW disease, only a few
had delivered children. Additionally, Krone et al. (2001)
reported on 18 of 119 CAH women who had given birth to
31 children, of which only one by a woman with SW. That
report also included a mutational analysis of the 18 women
who became pregnant (one SW (null/null), 12 SV and 5
NC). Moreover, Dumic et al. (2005) reported on three
women, one SV and two carrying an I2splice mutation,
giving birth to four children. Hoepffner et al. described nine
pregnancies in six women with classic CAH, also including
the mutation status of the subjects. Two women with null/
null mutations became pregnant and each delivered one
child. One woman with the I2splice mutation gave birth to
two, and three women with a SV mutation gave birth to four
children (Hoepffner et al., 2004). These reports are in accordance with our findings of a lower pregnancy rate among the
CAH women with the most severe mutations in the
21-hydroxylase gene.
Several authors have discussed the reasons for the low fertility among women with CAH (Meyer-Bahlburg, 1999; Jääskeläinen et al., 2000; Otten et al., 2005). A number of factors
have been suggested such as delayed psychosexual development, low sexual activity, adrenal overproduction of androgens
and progestins, PCOS, neuro-endocrine factors and genital
surgery. Some of these factors are obviously interrelated. In
the present study, the age of sexual debut was significantly
later among the women with CAH than in their age-matched
controls. A majority of the CAH women had undergone
genital surgery and many were dissatisfied with their genitals
(Nordenskjöld et al., 2008). Irregular menstruation as an indication of anovulation was of the same frequency in patients and
controls, 28%. This is lower than the 61% reported among
CAH women in a recent study (Bachelot et al., 2006). The
number of women living in a steady heterosexual relationship
was also significantly lower. Thus, among the women with SW
(null/null and I2splice), 71% and 53%, respectively, were
singles when compared with 19% in the SV group. A low interest for getting married and performing the traditional child care
role, especially among CAH women with SW, has been discussed by Meyer-Bahlburg (1999). This is probably one
important factor contributing to the low pregnancy rate. It
should also be pointed out that half of the CAH patients with
severe mutations were ,30 years old. Thus, they may not
yet have made a decision with regard to their reproductive
future. In Sweden, the mother’s average age at the birth of
the first child is 30 years (Medical Birth Registry of
Sweden, 2007). Hence, the most plausible explanation for the
observed low fertility among the women with CAH presented
here is probably that 70% of them had not yet tried to get pregnant. All the women with CAH, who had tried and had experienced difficulties, had all been successfully treated, with the
exception of the three women in the oldest age group. This
concurs with the recent report from Hoepffner et al. (2004)
indicating a good fertility outcome also in SW women when
adequately treated. Furthermore, modern methods of assisted
reproduction technologies are now available also to women
living in a lesbian relationship. Thus, women with CAH who
wish to have children should have better prospects today,
than during earlier decades.
The progress and outcome of the pregnancies with regard to
the women’s health were normal except in five cases of gestational diabetes among three women—20% of all pregnancies—
while none of the control pregnancies were complicated with
diabetes. The frequency of gestational diabetes is related to
the stringency of screening for diabetes during pregnancy. It
should be equivalent in patients and controls in this study
because they all attended public antenatal care clinics in
1611
Hagenfeldt et al.
Sweden that use the same guidelines for care. Gestational diabetes in this study is more frequent than reported from the
Swedish Medical Birth Registry, 1.4– 4% (Medical Birth Registry of Sweden, 2007). The explanation may be found in the
glucocorticoid medication. Earlier reports on our data of nonpregnant women with CAH indicated that many of them
were receiving glucocorticoids in doses higher than recommended (LWPES/ESPE, 2002). This is supported by the
findings of the extensively suppressed androgen levels and
high frequency of osteopenia/osteoporosis among women
with CAH (Falhammar et al., 2007a,b). One of the three
women who developed gestational diabetes was treated with
a high dose of prednisolone (17 mg daily) when conceiving
and a second one was treated with metformin for PCOS and
insulin resistance before pregnancy. The only other case of
gestational diabetes in a woman with CAH was reported by
Zacharin. This woman was severely obese before pregnancy
(BMI 38 kg/m2) and on a high glucocorticosteroid dose
(Zacharin, 1999). All of our three women who developed diabetes had a normal BMI. Gestational diabetes was not seen
among any of the 18 women with CAH giving birth to 31 children in Krone’s report, although the doses of glucocorticoids
are comparable. There was however no information on diabetes
screening procedures in that report (Krone et al., 2001). No
case of pre-eclampsia was observed among our pregnant
CAH women as opposed to findings in earlier studies (Lo
et al., 1999; Dumic et al., 2005).
It has been debated whether corticosteroid doses should be
increased during pregnancy. (Garner, 1998; Amadori, 2006).
In our data, the dose was increased in only 3 out of 25 pregnancies. In comparison, in the data presented by Krone et al.
(2001), the doses were increased in one-third of the pregnancies. The decision to do this or not should always be based
on the individual needs of the women, and hence it can differ
between different patient groups (Lo et al., 1999). All
women, who had undergone genital surgery one or several
times, were offered elective Caesarean section hence the rate
of Caesarean section was higher than in controls. Altogether,
only four CAH women were delivered vaginally, none had
had genital surgery. The rate of Caesarean section is thus
higher than previously reported (Krone et al., 2001; Lo and
Grumbach, 2001).
The outcome of the children born to the 14 women with
CAH was excellent; the mean birthweight of both boys and
girls were equal to the birthweight of the children among the
control women. None of the 19 girls in our study were observed
with any signs of virilization and there were no malformations.
Only one child was born prematurely, and there were no children that were SGA. This is in contrast to the only other report
on this issue describing five SGA children among 31 term pregnancies of CAH women (Krone et al., 2001).
The most unexpected finding was the significant difference
in the sex ratio of the children. In the group of children born
to the control women, there were 56% boys compared with
25% boys born to the women with CAH. It is possible that
this is a chance finding due to the fewer pregnancies in the
patient group. However, when looking into data available in
the literature, the same tendency is observed. In the paper by
1612
Lo and Grumbach (2001) concerning 74 children, the sex
was not reported in 41 of them. However, among 33 children,
there were 22 girls and 11 boys. Furthermore, in three recent
publications, another 26 girls and 17 boys were described
(Krone et al., 2001; Hoepffner et al., 2004; Dumic et al.,
2005). If these figures are combined with our data, the sex
ratio of all children born to CAH mothers so far published in
the literature will be 68 girls to 34 boys. The finding is intriguing and the explanation for this is not clear. One hypothesis
is that the hormonal status of the CAH mothers at the time of
implantation may have an impact on the uterine milieu that
favours female sex in the offspring.
The late follow-up of the children was consistent with the
report by Krone et al. (2001), i.e. a normal postnatal development. One of 25 children in the present study was diagnosed
with CAH of the SV type.
Finally, few of our women had reached the age of perimenopause or menopause. It was however obvious that among them
there was a high rate of gynaecological morbidity in comparison with their age-matched controls. This has not been
observed in other clinical reports, mainly due to the fact that
women with CAH in these age groups have not been
followed-up before. The low number of cases makes it impossible to hypothesise on the reasons for this difference but the
finding needs to be kept in mind in future follow-up studies.
Conclusions
In this large follow-up study of 62 women with CAH, only 14
women had given birth to 25 children compared with 54 children born to 41 age-matched control women. The reduced fertility in the women with CAH is clearly related to the severity
of the mutation. The main reason may be that fewer patients
than controls are living in a heterosexual relationship and
fewer have tried to get pregnant. Another factor could be that
the women in our study with the more severe mutations are
still rather young and may not have reached a decision with
regard to their reproductive future. However, all pregnancies
that occurred were normal except for an increase in gestational
diabetes among the CAH patients, a finding that calls for
screening of this complication during a CAH pregnancy. The
outcome of the children was excellent and did not differ
from that of the control children. An unexpected finding was
the uneven sex ratio between the children of mothers with
CAH with significantly fewer boys than girls, an observation
with some support in the literature. A late follow-up of the children indicated a normal development. The high frequency of
gynaecological morbidity among perimenopausal and menopausal women with CAH need further attention in the future.
Acknowledgements
The authors wish to acknowledge the excellent help from Anette Härström RN and Ingrid Hansson RN for practical assistance and Elisabeth Berg BSc, Department of Learning, Informatics, Management
and Ethics, Karolinska Institutet for statistical support.
Conflict of interest: none declared.
Fertility, pregnancy and congenital adrenal hyperplasia
Funding
The Swedish Research Council; HRH Crown Princess Lovisa
Foundation; Foundation Frimurarna; the Samariten Foundation, Sven Jerring Foundation and Göteborg Medical
Society.
References
Amadori P. Reduced need of glucocorticoid therapy in a woman with
congenital adrenal hyperplasia due to 21-hydroxylase deficiency during
pregnancy. J Endocrinol Invest 2006;29:848– 850.
Bachelot A, Plu-Burreau G, Thibaud E, Laborde K, Pinto G, Samara D,
Nihoul-Fekete C, Kuttenn F, Polak M, Touraine P. Long-term outcome of
patients with congenital adrenal hyperplasia due to 21-hydroxylase
deficiency. Horm Res 2006;67:268–276.
Dumic M, Janjanin N, Ille J, Zunec R, Spehar A, Zlopasa G, Francetic I, New
MI. Pregnancy outcome in women with classical congenital adrenal
hyperplasia due to 21-hydroxylase deficiency. J Pediatr Endocrinol Metab
2005;18:887–895.
Falhammar H, Filipsson H, Holmdahl G, Janson PO, Nordenskjöld A,
Hagenfeldt K, Thorén M. Metabolic profile and body composition in adult
women with congenital adrenal hyperplasia due to 21- hydroxylase
deficiency. J Clin Endocrinol Metab 2007a;92:110–116.
Falhammar H, Filipsson H, Holmdahl G, Janson PO, Nordenskjöld A,
Hagenfeldt K, Thorén M. Fractures and bone density in adult women with
21-hydroxylase deficience. J Clin Endocrinol Metab 2007b;92:4643– 4649.
Falhammar H, Thorén M, Hagenfeldt K. A 31-year old woman with infertility
and polycystic ovaries was diagnosed with non-classical adrenal hyperplasia
due to a novel mutation in the CYP21 gene. J Endocrinol Invest
2008;31:176–180.
Garner P. Congenital adrenal hyperplasia in pregnancy. Semin Perinatol
1998;22:446–456.
Gastaud F, Bouvattier C, Duranteau L, Brauner R, Thibaud E, Kutten F,
Bougneres P. Impaired sexual and reproductive outcomes in women with
classical forms of congenital adrenal hyperplasia. J Clin Endocrinol Metab
2007;92:1391–1396.
Hoepffner W, Schulze E, Bennek J, Keller E, Willgerodt H. Pregnancies in
patients with congenital adrenal hyperplasia with complete or almost
complete impairment of 21-hydroxylase activity. Fertil Steril
2004;81:1314–1321.
Jääskeläinen J, Hippeläinen M, Kiekara O, Voutilainen R. Child rate,
pregnancy outcome and ovarian function in females with classical
21-hydroxylase deficiency. Acta Obstet Gynecol Scand 2000;79:687– 692.
Krone N, Wachter I, Stefanidou M, Rosher AA, Schwarz HP. Mothers with
congenital adrenal hyperplasia and their children: outcome of pregnancy,
birth and childhood. Clin Endocrinol 2001;55:523– 529.
Lo JC, Grumbach MM. Pregnancy outcome in women with congenital
virilizing adrenal hyperplasia. Endocrinol Metab Clin North Am
2001;30:207– 229.
Lo JC, Schwitzgebel VM, Blake Tyrell J, Fitzgerald PA, Kaplan SL, Conte FA,
Grumbach MM. Normal female infants born of mothers with classic
congenital adrenal hyperplasia due to 21-hydoxylase deficiency. J Clin
Endocrinol Metab 1999;84:930–936.
LWPES/ESPE. Consensus statement on 21-hydroxylase deficiency from
the Lawson Wilkins Paediatric Endocrine Society and the European
Society for Paediatric Endocrinology. J Clin Endocrinol Metab
2002;87:4048–4053.
Medical Birth Registry of Sweden. www.socialstyrelsen.se/statistics.
Meyer-Bahlburg HFL. What causes low rate of child-bearing in congenital
adrenal hyperplasia? J Clin Endocrinol Metab 1999;84:1844– 1847.
Mulaikal RM, Migeon CJ, Rock JA. Fertility rates in female patients with
congenital adrenal hyperplasia due to 21-hydroxylase deficiency. N Engl J
Med 1987;316:178– 182.
Nordenskjöld A, Holmdahl G, Frisén L, Falhammar H, Filipsson H, Thorén M,
Janson PO, Hagenfeldt K. Type of mutation and surgical procedure affect
long term quality of life in women with congenital adrenal hyperplasia.
J Clin Endocrinol Metab 2008;93:380–386.
Otten BJ, Stikkelbroeck MML, Claahsen-van der Grinten HL, Hermus ARMM.
Puberty and fertility in congenital adrenal hyperplasia. In: Delemarre-van
der Waal HA (ed). Abnormalities in Puberty. Scientific and Clinical
Advances. Endocr Dev Basel Karger 2005;8:54–66.
Pinto G, Tardy V, Trivin C, Thalassinos C, Lortat-Jacob S, Nihoul-Fekete C,
Morel Y, Brauner R. Follow-up of 68 children with congenital adrenal
hyperplasia due to 21-hydroxylase deficiency: relevance of genotype for
management. J Clin Endocrinol Metab 2003;88:2624–2633.
Stikkelbroeck NM, Hermus AR, Braat DD, Otten BJ. Fertility in women with
congenital adrenal hyperplasia due to 21-hydroxylase deficiency. Obstet
Gynecol Surv 2003;58:275– 284.
Wedell A, Thilén A, Ritzen M, Stengler B, Luthman H. Mutational spectrum of
the steroid 21-hydroxylase gene in Sweden: implications for genetic
diagnosis and association with disease manifestation. J Clin Endocrinol
Metab 1994;78:1145– 1152.
Zacharin M. Fertility and its complications in a patient with salt
losing congenital adrenal hyperplasia. J Pediatr Endocrinol Metab
1999;12:89–94.
Submitted on October 11, 2007; resubmitted on March 9, 2008; accepted on
March 12, 2008
1613