Administration of Metformin to a Diabetic Woman with Extreme Hyperandrogenemia of Nontumoral Origin: Management of Infertility and Prevention of Inadvertent Masculinization of a Female Fetus
Nicholas J. Sarlis,
Stacie J. Weil and
Lawrence M. Nelson
Developmental Endocrinology Branch (S.J.W., L.M.N.), National
Institute of Child Health and Human Development; and Molecular and
Cellular Endocrinology Branch (N.J.S.), National Institute of Diabetes,
Digestive, and Kidney Diseases, National Institutes of Health,
Bethesda, Maryland 20892
Address all correspondence and requests for reprints to: Dr. Nicholas J. Sarlis, Molecular and Cellular Endocrinology Branch/National Institute of Diabetes, Digestive, and Kidney Diseases, National Institutes of Health, Building 10, Room 8D12C, 10 Center Drive-MSC 1758, Bethesda, Maryland 20892-1758. E-mail:
njsarlis{at}box-n.nih.gov
 |
Introduction
|
---|
The polycystic ovary syndrome (PCOS) is
characterized by chronic hyperandrogenemic anovulation and is
considered to be a common cause of infertility because it affects
515% of premenopausal women (1, 2, 3). Metformin has been shown, in
several studies, to significantly lower androgen levels and initiate
menstrual cyclicity in women with PCOS (4, 5, 6, 7, 8). The mechanism of action
of metformin in PCOS is unknown, but it is believed to involve the
combined effects of decreased insulin action on the ovary (lowering
ovarian androgen production rates) and on the liver [resulting in an
increase in sex hormone binding globulin (SHBG) levels] (4, 9, 10, 11).
Though metformin is effective in women with PCOS with mild to moderate
elevations of androgen serum levels (2, 4, 5, 6, 7, 8), its effects in the
considerably smaller subset of PCOS cases associated with extreme
hyperandrogenemia [testosterone (T) > 200 ng/dL (6.9 nmol/L)] and
maternal virilization have not been studied. Such extreme levels of
hyperandrogenemia during pregnancy are usually caused by an ovarian
tumor or luteoma of pregnancy and can be treated effectively by surgery
(12). However, this is not the case when these high levels are caused
by PCOS. In fact, although the possibility of fetal masculinization
exists in the presence of severe maternal hyperandrogenism, there are
presently no effective preventive measures when this is found in women
with PCOS (13).
 |
Clinical Presentation
|
---|
In August 1995, a 29-yr-old woman, gravida 2 para 0,
presented for evaluation of virilization and infertility. After an
unremarkable adolescence and history of two elective early abortions at
age 19 and 20, the patient, at age 22, had noted the onset of rapidly
worsening hirsutism, cephalic male-pattern baldness, deepening of her
voice, and an increase in muscle mass. By age 25, she was shaving her
facial hair twice daily and had experienced a 30-kg increase in weight
over the proceeding 3 yr. The patient had presented with oligomenorrhea
at age 23, and had been amenorrheic since age 24. Thereafter,
menstruation occurred only after progesterone-induced withdrawal. The
patient also had a decade-long history of mild hypertension. On
presentation, she was an obese black woman with an android body
habitus. Her height was 168.8 cm; weight, 88.5 kg; body mass
index, 31.1 kg/m2; and blood pressure, 148/93 mm Hg. The
patient was hirsute and had clitoromegaly and acanthosis nigricans. She
had markedly elevated serum androgen levels (Fig. 1
) and a normal fasting blood glucose.
Baseline serum TSH and PRL, an ACTH stimulation test for serum
cortisol, and 24-h excretion rates for urinary free cortisol and
17-hydroxysteroids were within normal limits. Computed tomography and
ultrasound evaluation of the abdomen and pelvis showed no evidence of
tumor. The patient was treated with oral contraceptives and was
reevaluated in February 1996. By that time, she had demonstrated a
significant reduction in serum androgen levels (Fig. 1
). The diagnosis
of PCOS was established.

View larger version (24K):
[in this window]
[in a new window]
|
Figure 1. Free and total T levels over time. The range
of normal values (NIH laboratory) for free T in both nonpregnant and
pregnant women are 319 pg/mL (10.566.0 pmol/L). The excursions in
total T levels exactly paralleled these seen in free T levels, with a
maximum pre-pregnancy value of 432 ng/dL (15.0 nmol/L) in August 1995
and a dramatic response to metformin therapy. The range of normal
values for total T for nonpregnant women is 2080 ng/dL (0.702.78
nmol/L). MET, metformin; OCP, oral contraceptive pill.
|
|
Subsequently, oral contraceptives were discontinued because the patient
desired a pregnancy. She was placed on a weight reduction regimen,
including a short period of treatment with phentermine/fenfluramine
combination, but was unable to lose weight. Three months after the
discontinuation of oral contraceptives, she remained amenorrheic. In
October 1996, the patient again had significantly elevated androgen
levels (Fig. 1
) and was now meeting diagnostic criteria for diabetes
mellitus, by an oral glucose tolerance test (despite normal fasting
blood glucose levels). She began metformin (500 mg p.o. b.i.d.)
and continued this medication for 5 months, with resultant marked
decreases of serum androgens. Within 2 months after starting metformin,
the patient was experiencing normal menstrual cycles.
In February 1997, the patient discontinued the metformin therapy on her
own, and she became pregnant within a month. In June 1997, at 10.5
weeks of gestation, a relapse of marked hyperandrogenemia was
documented (Fig. 1
). Because of the possibility of masculinization of a
female fetus by this degree of excess androgens, and after extensive
counseling with regard to the possible risks and benefits of metformin
administration during pregnancy, the patient was again placed on
metformin, 500 mg p.o. b.i.d. at 14 weeks of gestation (August 1997).
Both total and free T levels decreased dramatically (Fig. 1
). At 29.5
weeks of gestation, the patient developed preeclampsia. Because of a
rapid deterioration of her condition, primary cesarean section was
performed at 30 weeks of gestation. A nonmasculinized 1,335-g female
infant was delivered. The patient discontinued the metformin therapy in
the immediate postpartum period for 3 months (while she was nursing her
infant). During that time, she maintained normal blood glucose levels.
After the discontinuation of breast-feeding, the patient was again
placed on metformin, 500 mg p.o. b.i.d.; and she continues to take this
medication to date, with excellent metabolic response (normalization of
total and free T levels and glycemia, and resumption of menses).
 |
Discussion
|
---|
Metformin has been shown, in several studies, to significantly
lower androgen levels in women with PCOS and to initiate menstrual
cyclicity (4, 5, 6, 7, 8). These findings were not reproduced in one recent
study of morbidly obese women with PCOS (14). In this study, neither
plasma androgen or insulin levels decreased significantly after
metformin therapy, probably because of the overwhelming metabolic
effect(s) of extreme obesity in the study population (average body mass
index, 40.0 kg/m2). Another study was also unable to show a
metformin-induced decrease in indices of insulin resistance in patients
with PCOS (15). However, resumption of menses shortly after initiation
of metformin therapy has been reported in a significant portion of
patients with PCOS (4, 5, 6, 7, 8). It is believed that metformin leads to
decreases in serum free androgen levels, thus permitting resumption of
menstrual cyclicity (2, 4).
Our case raises two important issues relevant to the presence of
extremely high androgen levels (associated with maternal virilization)
during pregnancy: 1) the indeterminate, yet existent, risk of
masculinization of a female fetus; and 2) the risk of metformin therapy
during pregnancy. With regard to issue 1, the possibility of fetal
masculinization exists in maternal congenital adrenal hyperplasia,
maternal ovarian tumors, exogenous androgen administration to the
mother, luteoma of pregnancy, and (rarely) in placental aromatase
deficiency and in mothers with PCOS (12, 13, 16, 17, 18, 19). The exact
incidence of this side effect of maternal androgens varies greatly
among case reports and published small series (12, 20, 21). Excess
androgen exposure during fetal development is relevant not only to
genital differentiation but also to central nervous system functional
mechanisms underlying subsequent expression of sociosexual behavior and
gender identity (22, 23). Whereas fetal congenital adrenal hyperplasia
is the most common cause of ambiguous genitalia in female neonates,
luteoma of pregnancy is the most frequent cause for maternal
virilization during pregnancy, as approximately 5060% of female
fetuses born to virilized mothers exhibit signs of masculinization
(20). Interestingly, virilization of the fetus has not proven to be
tightly correlated with maternal androgen levels. In general, such a
risk is increased if the mother herself shows clinically significant
manifestations of virilization (12). The timing of androgen exposure
during gestation has been shown to affect the neonatal phenotype, as
expected (12, 13).
With regard to issue 2 (i.e. the effects of metformin during
pregnancy), metformin has been rated by the Food and Drug
Administration as a category B medication (24). Mouse embryos exposed
to metformin at doses of 5002,550 mg daily have shown no major
malformations in the offspring (25). A South African group has
published several series describing women with type II diabetes
mellitus on metformin therapy throughout pregnancy; no adverse effects
were reported, with regard to fetal and neonatal development (26).
In conclusion, we report a case where maternal hyperandrogenism during
pregnancy was not attributable to common causes, i.e.
ovarian tumor or luteoma of pregnancy (which can all be treated
effectively, either medically or surgically), but rather was due to
PCOS. This case is of additional interest because of significant
increases in maternal androgen levels at the early stages of pregnancy,
a phenomenon that is quite rare in normal pregnancy (27, 28) but may be
more frequent in PCOS (13); there is a paucity of information with
regard to this phenomenon in women with PCOS. Our case suggests that
metformin may be effective for anovulatory women with PCOS and extreme
hyperandrogenemia. These patients also might be potentially maintained
on metformin throughout pregnancy, aiming at the prevention of
inadvertent fetal masculinization (feto-protective treatment).
Received December 9, 1998.
Revised January 12, 1999.
Accepted February 3, 1999.
 |
References
|
---|
-
Hull MG. 1987 Epidemiology of infertility and
polycystic ovarian disease: endocrinological and demographic studies. Gynecol Endocrinol. 1:235240.[Medline]
-
Nestler JE. 1997 Role of hyperinsulinemia in the
pathogenesis of the polycystic ovary syndrome and its clinical
implication. Semin Reprod Endocrinol. 15:111117.[Medline]
-
Speroff L. 1994 Anovulation and the polycystic
ovary. In: Speroff L, Glass R, Kase N, eds. Clinical gynecology
endocrinology and infertility. Baltimore: William & Wilkins;
457482.
-
Nestler JE, Jakubowicz DJ, Evans WS, et al. 1998 Effects of metformin on spontaneous and clomiphene-induced
ovulation in the polycystic ovary syndrome. N Engl J Med. 338:18761881.[Abstract/Free Full Text]
-
Velazquez E, Mendoza S, Hamer T, Sosa F, Glueck C. 1994 Metformin therapy in polycystic ovary syndrome reduces
hyperinsulinemia, insulin resistance, hyperandrogenism, and systolic
blood pressure, while facilitating normal menses and pregnancy. Metabolism. 43:647654.[Medline]
-
Velazquez E, Acosta A, Mendoza S. 1997 Menstrual
cyclicity after metformin therapy in polycystic ovary syndrome. Obstet
Gynecol. 90:392395.[Abstract/Free Full Text]
-
Diamanti-Kandarakis E, Kouli C, Tsianateli T, Bergiele
A. 1998 Therapeutic effects of metformin on insulin resistance and
hyperandrogenism in polycystic ovary syndrome. Eur J Endocrinol. 138:269274.[Medline]
-
Morin-Papunen LC, Koivunen RM, Ruokonen A,
Martikainen HK. 1998 Metformin therapy improves the menstrual
pattern with minimal endocrine and metabolic effects in women with
polycystic ovary syndrome. Fertil Steril. 69:691696.[CrossRef][Medline]
-
Nestler J, Jakubowicz D. 1996 Decreases in ovarian
cytochrome P450c17 activity and serum free testosterone after reduction
of insulin secretion in polycystic ovary syndrome. N Engl J
Med. 335:617623.[Abstract/Free Full Text]
-
Nestler JE, Jakubowicz DJ. 1997 Lean women with
polycystic ovary syndrome respond to insulin reduction with decreases
in ovarian P450c17-alpha activity and serum androgens. J Clin
Endocrinol Metab. 82:40754079.[Abstract/Free Full Text]
-
Laatikainen TJ, Apter DL, Paavonen JA, Wahlstrom
TR. 1980 Steroids in ovarian and peripheral venous blood in
polycystic ovarian disease. Clin Endocrinol (Oxf). 13:125129.[Medline]
-
McClamrock HD, Adashi EY. 1992 Gestational
hyperandrogenism. Fertil Steril. 57:257270.[Medline]
-
Ben-Chetrit A, Greenblatt EM. 1995 Recurrent
maternal virilization during pregnancy associated with polycystic
ovarian syndrome: a case report and review of the literature. Hum
Reprod. 10:30573062.[Abstract]
-
Ehrmann D, Cavaghan M, Imperial J, Sturis J, Rosenfield
R, Polonsky K. 1996 Effects of metformin on insulin secretion,
insulin action and ovarian steroidogenesis in women with polycystic
ovary syndrome. J Clin Endocrinol Metab. 82:524530.[Abstract/Free Full Text]
-
Acbay O, Gundogdu S. 1996 Can metformin reduce
insulin resistance in polycystic ovary syndrome? Fertil Steril. 65:946949.[Medline]
-
Shozu M, Akasofu K, Harada T, Kubota Y. 1991 A new
cause of female pseudohermaphroditism: placental aromatase deficiency. J Clin Endocrinol Metab. 72:560565.[Abstract]
-
Magendantz HG, Jones DED, Schomberg DW. 1972 Virilization during pregnancy associated with polycystic ovary disease. Obstet Gynecol. 40:156162.[Medline]
-
Fayez JA, Bunch TR, Miller GL. 1974 Virilization in
pregnancy associated with polycystic ovary disease. Obstet Gynecol. 44:511521.[Medline]
-
Bilowus M, Abassi V, Gibbons MD. 1986 Female
pseudohermaphroditism in a neonate born to a mother with polycystic
ovarian disease. J Urol. 136:10981100.[Medline]
-
Manganello P, Adams L, Harris R, Ornvold K. 1995 Virilization during pregnancy with spontaneous resolution postpartum: a
case report and review of the English literature. Obstet Gynecol Surv. 50:404410.[CrossRef][Medline]
-
Shortie BE, Warren MP, Tsin D. 1987 Recurrent
androgenicity in pregnancy. A case report and literature review. Obstet
Gynecol. 70:462464.[Abstract]
-
Michael R, Zumpe D, Bonsail R. 1992 The interaction
of testosterone with the brain of the orchidectomized primate fetus. Brain Res. 570:6874.[CrossRef][Medline]
-
Sato T, Koizumi S. 1991 Effects of fetal androgens
on childhood behavior. Acta Paediatr Jpn. 33:639644.[Medline]
-
Anonymous. 1998 Metformin. In: Physicians desk
reference. Montvale, NJ: Medical Economics Co., Inc.; 795800.
-
Denno K, Sadler T. 1994 Effect of biguanide class
of oral hypoglycemic agents on mouse embryogenesis. Teratology. 49:260266.[Medline]
-
Coetzee E, Jackson W. 1980 Pregnancy in established
noninsulin-dependent diabetics. A five-and-a-half year study at Groote
Schuur Hospital. S Afr Med J. 58:795802.[Medline]
-
Mizuno M, Lobotsky J, Lloyd CW, Kobayashi T, Murasawa
Y. 1968 Plasma androstenedione and testosterone during pregnancy
and in the newborn. J Clin Endocrinol Metab. 28:11331142.[Medline]
-
Bammann BL, Coulan CB, Jiang N-S. 1980 Total and
free testosterone during pregnancy. Am J Obstet Gynecol. 137:293298.[Medline]