| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrine Care |
Institutes of Endocrinology (I.S., T.R.) and Pathology (D.N.), and Department of Neurosurgery (M.H.), Sheba Medical Center, Tel-Hashomer 52621; Fertility Unit (I.B.-H.) and Endocrinology Service (G.H.), Rabin Medical Center, Golda Campus, Petah Tikva 49372; and Department of Molecular Cell Biology (A.A.), The Weizmann Institute of Science, Rehovot 76100, Israel
Abstract
We report a unique case of a 28-yr-old woman with a
gonadotroph adenoma secreting FSH, presented with ovarian
hyperstimulation, without elevation of serum estradiol. She presented
with abdominal pain and large ovaries (both 10 cm in diameter) with
multiple follicular cysts shortly after discontinuing oral
contraceptive pills. She had a supranormal PRL level of 71 µg/liter
(normal, <20), FSH of 8.49.2 IU/liter (normal for follicular phase,
2.410), LH of 0.01 IU/liter (normal, 1.69.3), estradiol of 108
pmol/liter (normal for follicular phase, 80790), and free
-subunit
level of 0.11 µg/liter (normal, <1.8). A nuclear magnetic resonance
study revealed invasive pituitary macroadenoma, 30 mm in diameter.
Dopamine agonist (cabergoline) treatment normalized serum
PRL but had no affect on FSH levels. A transsphenoidal surgery was
performed, and most of the adenoma was resected. One month after
surgery the patient resumed menstruation, and the hormonal profile
included serum FSH of 6.3 IU/liter, LH of 2.1 IU/liter, estradiol of
156 pmol/liter, and PRL of 10 µg/liter. The excised adenoma tissue
exhibited intense immunostaining for FSH and secreted this hormone to
culture medium. Stimulation with TRH (both in vivo
preoperatively and in vitro study of the excised tumor)
had no effect on FSH secretion from the adenoma. Estradiol did not
suppress FSH release from cultured adenoma cells. Patient serum samples
showed significant FSH bioactivity when tested in a human granulosa
cell line.
This case is remarkable because the ovarian hyperstimulation related to the FSH-secreting adenoma was not associated with high levels of serum estradiol, probably due to insufficient LH production by the normal pituitary. Thus, it supports the two-cell, two-gonadotropin theory, that both FSH and LH are necessary for normal ovarian estrogen production.
APPROXIMATELY 40% OF all pituitary
macroadenomas are nonfunctional monoclonal gonadotroph-cell
tumors (1). These tumors may secrete intact
gonadotropins or
-subunits, ß-FSH, and ß-LH. Gonadotroph-cell
tumors usually secrete their products inefficiently and in low
concentrations and, thus, very rarely cause any clinical syndrome other
than sellar mass effects (2, 3). Several cases of men with
FSH-secreting pituitary macroadenomas have been reported
(4, 5, 6). These patients may present with visual
disturbances and hypogonadism due to destruction of normal pituitary
gonadotrophs by the macroadenoma, or because of simultaneous
hypersecretion of PRL by the pituitary adenoma. Postmenopausal women
with FSH-secreting pituitary adenomas do not present with any
characteristic hormonal hypersecretion signs (7). Thus,
gonadotropin overproduction by pituitary adenomas usually does not
result in a clinically recognizable syndrome.
We report a unique case of a 28-yr-old female with a large pituitary macroadenoma secreting intact FSH, who presented with bilateral multicystic enlarged ovaries without elevated serum estradiol levels.
Materials and Methods
Hormone assays
Serum FSH and PRL were quantified with the Immulite two-site
chemiluminescent immunometric assay from Diagnostic Products (Los Angeles, CA; detection limit of FSH, 0.6
IU/liter). Serum LH was quantified preoperatively and postoperatively
with a Delfia fluoroimmunoassay LH kit (detection limit, 0.05
IU/liter), but during the immediate postoperative period with the
Immulite competitive immunoassay (Diagnostic Products;
detection limit, 0.7 IU/liter). 17ß-Estradiol was measured with
Immulite competitive immunoassay and
-subunit with double
antibody RIA kit from Biomerica (Newport Beach, CA). Human FSH in cell
culture medium was measured using FSH immunoradiometric assay from
Diagnostic Products (Coat-A-Count).
Histological studies
The transsphenoidal specimen of the pituitary tumor was fixed in formalin, paraffin-embedded sections were stained with hematoxylin-eosin, and histochemical staining for reticulin was performed. Immunohistochemistry studies included staining for FSH, LH, PRL, GH, ACTH, and TSH. Monoclonal mouse antihuman FSH and LH antibodies were purchased from Zymed Laboratories, Inc. (South San Francisco, CA). Cross-reactivity of these antibodies with human TSH is 0.050.1%, the anti-FSH with human LH is 0.1%, and the anti-LH with FSH is 0.02%.
FSH bioactivity
Human granulosa cell line expressing human FSH receptor and secreting progesterone in response to human FSH in cell culture medium was used for measuring FSH bioactivity in the patient serum. This cell line was prepared using a similar methodology as previously reported (8). Cells were incubated in 24-well tissue cultures plates for 24 h with serum samples of the patient (25100 µl added to 0.5 ml culture medium; in triplicates) or with known concentrations of human recombinant FSH (standard curve). Medium was collected after treatment for later progesterone measurements (RIA, using materials supplied by the NIDDK, NIH).
Pituitary cell culture
A specimen of the pituitary adenoma was minced and enzymatically
dissociated using 0.35% collagenase and 0.1% hyaluronidase (both from
Sigma, St. Louis, MO) for 4560 min. Cell suspensions
were filtered and resuspended in low glucose DMEM supplemented
with 10% fetal bovine serum, 2 mM glutamine, and
antibiotics. For primary cultures,
5 x
104 cells were seeded in 48-well tissue culture
plates (Costar, Cambridge, MA) in 0.5 ml medium and
incubated for 72 h in a humidified atmosphere of 95% air/5%
CO2, at 37 C. The medium was then changed to
serum-free defined low glucose DMEM containing 0.2% BSA, 120
nM transferrin, 100 nM hydrocortisone, 0.6
nM triiodothyronine, 5 U/liter insulin, 3 nM
glucagon, 50 nM PTH, 2 mM glutamine, 15
nM epidermal growth factor, and antibiotics. The cells were
treated for 4 h with either 17ß-estradiol (100 nM),
TRH (10 nM), or GnRH (100 nM) (all from
Sigma), whereas control cell cultures were treated with
vehicle solution. Medium was collected after treatment and stored at
-20 C for later FSH measurements.
Case report
A 28-yr-old woman presented to the Department of
Obstetrics and Gynecology in August 1999 with right lower quadrant
abdominal pain and tenderness. Her past medical history was
unremarkable. She had regular menses before she started taking
contraceptive pills 10 yr ago and while taking the pills, and had never
been pregnant. She had no galactorrhea or headache. Three weeks before
admission she stopped oral contraceptives containing 20 µg
ethinyl estradiol and 150 µg desorgestel (Mercilon;
Organon, Oss, Holland). The abdominal pain was
combined with nausea that gradually increased during the preceding
week. She mentioned one episode of vomiting on the same day. Her vital
signs were normal. Abdominal examination revealed general tenderness
with no sign of peritoneal irritation. On bimanual examination the
uterus was found to be of normal size and not tender. Transvaginal
ultrasound examination (Fig. 1
) revealed
bilateral enlarged ovaries (both 10 cm in diameter) with a sonographic
picture resembling ovarian hyperstimulation, demonstrating multiple
(>20) ovarian cysts ("expanded follicles") on both ovaries. There
was no free fluid in the peritoneal cavity, and a thin (6 mm)
endometrium was demonstrated. Due to the clinical picture (pain,
nausea, vomiting, abdominal tenderness especially on the right side,
and enlarged ovaries), subtorsion of the right adnexa was suspected.
Therefore, a laparoscopy was conducted. During the laparoscopy
bilateral enlarged ovaries (10 cm each) with smooth surface were seen.
There were no signs of torsion, pelvic inflammatory disease,
hydrosalpinges, endometriosis, or pelvic adhesions. A biopsy was taken
from the right ovary, later found to be consistent with a follicular
cyst. The postoperative course was unremarkable. After the procedure
oral contraceptives were restarted and stopped 3 months later when
repeated ultrasound did not show significant ovarian changes.
|
-subunit was 0.11 µg/liter (normal, <1.8).
Magnetic resonance imaging (MRI) of the pituitary was performed
due the hyperprolactinemia and showed invasive pituitary
macroadenoma, 30 mm in diameter, with parasellar extension, and
invasion of the right cavernous sinus and the sphenoid sinus (Fig. 2
). Visual field examination was normal.
A 3-week therapeutic trial of cabergoline (0.5 mg;
Pharmacia & Upjohn, Inc., s.p.A., Italy), twice a
week, without contraceptives normalized PRL, but had no influence on
FSH or LH levels. Stimulation with 400 µg TRH had no effect on serum
FSH and LH levels (Table 1
).
Transsphenoidal pituitary surgery was performed with successful removal
of the intrasellar adenoma. Postoperatively the patient developed
severe hot flashes together with decrease of her serum estradiol levels
(Table 2
). The hot flashes gradually
disappeared in a week, whereas LH and estradiol levels became
detectable (Table 2
). A month after operation the patient resumed
menstrual bleeding, and her hormonal profile included FSH of 6.3
IU/liter, LH of 2.1 IU/liter, estradiol of 156 pmol/liter, PRL of 10
µg/liter, and normal thyroid hormones. Four months after the
operation her midcycle serum FSH measured 3.3 IU/liter, LH measured
0.08 IU/liter, and estradiol was 481 pmol/liter (Table 2
). Ultrasound
evaluation 4 months after the operation disclosed a decrease in ovarian
size to less than 5 cm each and gradual disappearance of the cysts. MRI
done 6 months after the operation revealed residual tumor mass in the
right cavernous sinus.
|
|
|
In vitro studies
Histology.Histological examination of the pituitary tissue
resected showed the characteristic disruption of reticulin fiber
network as seen in pituitary adenomas stained for reticulin (Fig. 3A
). Immunohistochemical studies for
anterior pituitary hormones revealed diffuse and intense cytoplasmic
staining for intact FSH in pituicytes all over the adenoma (Fig. 3B
),
whereas LH, PRL, GH, ACTH, and TSH were not detected.
|
Tissue culture. Tumor tissue obtained from the surgical specimen was kept for 8 h in low glucose DME supplemented with 0.3% BSA and antibiotics, and the concentrations of FSH, LH, and PRL in the medium were analyzed. The FSH level was 60.7 IU/liter, LH was undetectable, and PRL was 1.2 µg/liter. Then, the tumor specimen was cultured for 72 h before a 4-h treatment with 17ß-estradiol, TRH, or GnRH. Estradiol did not suppress in vitro FSH secretion from the adenoma cells, and TRH and GnRH failed to stimulate FSH release to the medium.
Discussion
The patient presented had ovarian hyperstimulation with bilateral enlarged multicystic ovaries associated with an FSH-secreting pituitary macroadenoma. The full-blown ovarian hyperstimulation syndrome is characterized by massive ovarian enlargement with multiple cysts and by an increase in capillary permeability, which may cause ascites or edema. This syndrome is usually associated with extremely high serum estradiol levels after gonadotropin therapy for ovulation induction or for controlled ovarian stimulation for harvesting ova for in vitro fertilization (9). Rare cases of ovarian hyperstimulation were reported associated with severe primary hypothyroidism (10), bilateral granulosa cell tumors (11), and during spontaneous pregnancy in a woman with polycystic ovary disease (12).
Gonadotroph adenomas rarely present with hormonal hypersecretion.
Excessive secretion of FSH is more commonly reported in men with
gonadotroph adenomas, but it does not cause any symptoms. FSH-secreting
tumors are rare in postmenopausal women. The tumor is not associated
with any recognizable hormonal syndrome, and it is difficult to
distinguish between FSH hypersecretion from normal postmenopausal
gonadotrophs or from adenoma cells, unless LH levels are extremely low
(7). Our premenopausal patient had pure FSH-secreting
pituitary adenoma. The resected tumor secreted intact FSH to the
culture medium but not LH and was immunostained for FSH only.
Preoperatively the patient had serum FSH levels in the upper normal
range, LH was very low (also after stimulation with TRH; Table 1
), and
free
-subunits were also very low. Estradiol was within low-normal
follicular phase levels, and pituitary imaging was performed because of
hyperprolactinemia (caused probably by stalk compression). LH was very
low, probably as the result of tumor pressure and destruction of normal
pituitary tissue and gonadotrophs. After transsphenoidal removal of a
large part of the adenoma, this mass effect resolved, and several days
later LH secretion from normal gonadotrophs was increased (Table 2
).
FSH decreased immediately after tumor removal (Table 2
), together with
a rapid decline of estradiol, which was associated with transient hot
flashes. FSH mildly increased later with the expected recovery of
normal pituitary FSH secretion. However, it is clear that some FSH was
still released from the adenoma tissue left in the cavernous sinus
postoperatively. Importantly, the postoperative decrease of ovarian
size confirms the critical role of FSH secreted from the adenoma in
ovarian hyperstimulation development in this patient.
Three premenopausal women (13, 14, 15) and one prepubertal
girl with ovarian hyperstimulation (16) caused by
gonadotroph adenomas have been described before the current case. In
two of these cases FSH was within the normal range (14, 15), and in two others (13, 16) FSH was elevated
(Table 3
). Serum FSH in our patient was
in the normal range, but its bioactivity was enhanced, thus enough to
stimulate ovarian enlargement. FSH-secreting adenomas are known to
produce intact and biologically active FSH, and the bioactivity to
immunoreactivity ratio of secreted FSH may be increased
(17). LH was normal in one woman with ovarian
hyperstimulation (14) and suppressed but detected in all
other reported cases (13, 15, 16). However, unlike the
other patients described who had high estrogen levels and endometrial
hyperplasia (13, 14), our patient did not have elevated
estradiol or thick endometrium (Table 3
). This difference may be
explained by the very low LH levels measured in our reported patient.
FSH alone can induce follicular growth and ovarian enlargement without
stimulating the release of estrogen (18). However, minute
LH is required for steroidogenesis and estrogen production in the
ovary. In women with hypogonadotrophic hypogonadism, treatment with
purified or recombinant FSH to induce ovulation was associated with
inappropriately low serum estradiol, compared with human menopausal
gonadotropin (FSH with LH) treatment that induced very high estrogen
concentrations and endometrial hyperplasia (19, 20). This
phenomenon supports the two-cell, two-gonadotropin hypothesis of
ovarian steroid synthesis (21), confirming that LH
enhances androgen production in theca cells, which are then aromatized
by granulosa cells to estrogens under the control of FSH. In primates,
FSH alone can induce follicular growth and maturation without the
presence of LH (22), whereas LH supports and augments the
regulatory function of FSH during ovarian steroidogenesis. However, as
in our patient, some estradiol may be secreted despite deficient LH
action. This phenomenon is also found in female patients with ovarian
resistance to LH due to mutation in the LH receptor gene
(23), who have estradiol levels that are low or normal for
follicular phase due to impaired LH effect. The patient reported here
is unique because, unlike the other four reported females with
FSH-secreting adenomas and ovarian hyperstimulation
(13, 14, 15, 16) who had at least minute LH secretion enough to
stimulate estradiol hypersecretion, her large tumor abolished LH
secretion and prevented estrogen overproduction.
|
Our patient, similarly to the patient reported by Valimaki et al. (15), developed symptoms of ovarian enlargement shortly after discontinuing the use of contraceptive pills containing 20 µg ethinyl estradiol in combination with progestin. We tested in vitro the feasibility of estradiol to suppress FSH production by the tumor, but were unable to show any effect. Moreover, restarting oral contraceptives was inefficient in suppressing FSH and ovarian enlargement in our patient, confirming the gonadotroph adenoma cell autonomy.
In conclusion, FSH-secreting pituitary adenomas should be considered in
the differential diagnosis of ovarian hyperstimulation, when normal
estradiol concentration is associated with very low LH levels. These
cases may be difficult to diagnose, but the finding of high PRL levels
(Table 3
) may help, indicating the need for pituitary imaging to detect
a gonadotroph adenoma.
Acknowledgments
Received June 7, 2000. Accepted April 25, 2001.
Address all correspondence and requests for reprints to: Ilan Shimon, M.D., Institute of Endocrinology, Chaim Sheba Medical Center, Tel-Hashomer, 52621 Israel. E-mail: i_shimon@netvision.net.il.
Footnotes
Abbreviation: MRI, Magnetic resonance imaging.
References
subunit-secreting pituitary tumors. J Neurosurg 59:585589[CrossRef][Medline]
-subunit-secreting pituitary tumor
treated with bromocrptine. J Clin Endocrinol Metab 61:580584[Abstract]
This article has been cited by other articles:
![]() |
T. Kajitani, S. Liu, T. Maruyama, H. Uchida, R. Sakurai, H. Masuda, T. Nagashima, M. Ono, T. Arase, and Y. Yoshimura Analysis of serum FSH bioactivity in a patient with an FSH-secreting pituitary microadenoma and multicystic ovaries: A case report Hum. Reprod., February 1, 2008; 23(2): 435 - 439. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Ghayuri and J. H. Liu Ovarian Hyperstimulation Syndrome Caused by Pituitary Gonadotroph Adenoma Secreting Follicle-Stimulating Hormone Obstet. Gynecol., February 1, 2007; 109(2): 547 - 549. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Maruyama, H. Masuda, H. Uchida, T. Nagashima, and Y. Yoshimura Follicle Stimulating Hormone-Secreting Pituitary Microadenoma With Fluctuating Levels of Ovarian Hyperstimulation Obstet. Gynecol., May 1, 2005; 105(5): 1215 - 1218. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. P. Mohammad, R. A. Abbud, A. F. Parlow, J. S. Lewin, and J. H. Nilson Targeted Overexpression of Luteinizing Hormone Causes Ovary-Dependent Functional Adenomas Restricted to Cells of the Pit-1 Lineage Endocrinology, October 1, 2003; 144(10): 4626 - 4636. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Murata, H. Ando, T. Nagasaka, I. Takahashi, K. Saito, H. Fukugaki, K. Matsuzawa, and S. Mizutani Successful Pregnancy after Bromocriptine Therapy in an Anovulatory Woman Complicated with Ovarian Hyperstimulation Caused by Follicle-Stimulating Hormone-Producing Plurihormonal Pituitary Microadenoma J. Clin. Endocrinol. Metab., May 1, 2003; 88(5): 1988 - 1993. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |