- Email: [email protected]
Isolated luteinizing hormone (LH) elevation in a woman with secondary amenorrhea: a clue to the diagnosis of an inhibin B–producing thecoma and insights into the influence of inhibin B on LH
CASE REPORT Isolated luteinizing hormone (LH) elevation in a woman with secondary amenorrhea: a clue to the diagnosis of an inhibin B–producing thecoma and insights into the influence of inhibin B on LH Lois E. Donovan, M.D.,a,b Philippa H. Brain, M.D.,b and M aire A. Duggan, M.D.b,c a Division of Endocrinology and Metabolism, Department of Medicine, University of Calgary, Calgary; b Department of Obstetrics and Gynecology, University of Calgary, Calgary; and c Department of Pathology and Laboratory Medicine, Foothills Medical Centre, Calgary, Alberta, Canada
Objective: To review the diagnostic possibilities that exists when the workup of amenorrhea reveals an isolated LH elevation; and to examine the effect of inhibin B on LH levels in vivo. Design: Case report. Setting: University hospital. Patient(s): A 20-year-old woman presented with secondary amenorrhea. Her FSH measurement was low, and the LH level was elevated. The recognition that this was an unusual pattern led to the diagnosis of a rare but very treatable inhibin B–producing thecoma, despite the fact that results on the initial pelvic ultrasound examination performed 10 months after presentation of amenorrhea were relatively unremarkable. Intervention(s): Surgical removal of an ovarian thecoma. Main Outcome Measure(s): Gonadotropins, E2, inhibin B, menstrual bleeding, and fertility. Result(s): Removal of the ovarian thecoma resulted in a normalization of FSH, LH, and inhibin B levels and a return of spontaneous menses 28 days later. Pregnancy occurred with the third postoperative menstrual cycle, followed by the delivery of a healthy full-term girl. Conclusion(s): Inhibin B–producing sex cord granolosa–stromal cell tumors should be considered in women who present with amenorrhea with isolated LH elevations, even in the setting of a previously normal pelvic ultrasound report. Diagnostic considerations that arise in the workup of amenorrhea when there is an isolated elevation in LH that is accompanied by normal or low FSH levels are reviewed. This rare clinical presentation provides the opportunity to observe the impact of inhibin B on gonadotropins in vivo. (Fertil Steril 2010;94:1097.e9–e12. 2010 by American Society for Reproductive Medicine.) Key Words: Isolated LH elevation, inhibin B, thecoma, amenorrhea
Gonadotropin production and release is regulated by GnRH. Ovarian feedback from estrogen (E) and P, was well as inhibins, partially regulates gonadotropin release. Inhibins are heterodimeric glycoproteins with structural homology to transforming growth factor b family (1). Inhibin consists of an alpha subunit joined to a beta A subunit (inhibin A) or a beta B subunit (inhibit B). Inhibin B is secreted primarily by the granulosa cells of developing follicles. Inhibin A seems to be secreted mainly by the corpus luteum, thus it is considered a marker of dominant follicle and corpus luteum activity (2). It is generally recognized that inhibins suppress FSH release. The influence of inhibin B on LH remains an area of ongoing controversy. Recently a significant negative association was Received January 22, 2010; revised February 3, 2010; accepted February 4, 2010; published online March 21, 2010. L.E.D. has nothing to disclose. P.H.B. has nothing to disclose. M.A.D. has nothing to disclose. Reprint requests: Lois E. Donovan, M.D., Foothills Medical Centre, North Tower Room 716, 1403 29th St. NW, Calgary, Alberta, Canada T2N 2T9 (FAX: 403-944-3199; E-mail: [email protected]).
0015-0282/$36.00 doi:10.1016/j.fertnstert.2010.02.016
observed in the follicular phase between inhibin B and LH (3). This has led Robertson et al. to ‘‘postulate that an inhibin B:LH feedback mechanism may exist in the human female’’ (3). Here, we describe a 20-year-old woman with an inhibin B–producing thecoma. This patient provided the rare opportunity to study the impact of inhibin B on gonadotropins in vivo, in the hypoestrogenized state. This patient case prompted a review of the diagnostic possibilities that arise in the workup of amenorrhea when there is an isolated elevation in LH that is accompanied by normal or low FSH levels.
CASE REPORT A 20-year-old woman with a 1-year history of secondary amenorrhea was referred for endocrinology consultation after her family doctor discovered a 1-mm nonenhancing lesion in the pituitary gland. Menarche occurred at 12 years of age. She had experienced regular menses every 28 days with 6 days of menstrual flow, until moving to Canada from Holland and marrying 2 years earlier. Then she
Fertility and Sterility Vol. 94, No. 3, August 2010 Copyright ª2010 American Society for Reproductive Medicine, Published by Elsevier Inc.
1097.e9
FIGURE 1 Intraoperative gross appearance of the thecoma.
Donovan. Thecoma causing isolated LH elevation. Fertil Steril 2010.
began to develop a less-regular pattern of menstrual bleeding that was described at times as 1 week of menstrual bleeding followed by a week of no menstrual bleeding and then a recurrence of menstrual bleeding 14 days after the start of the previous menstrual bleed. Menstrual bleeding completely stopped 1 year before her presentation to the endocrinologist. She had been sexually active without contraceptive use since her marriage. She had never been pregnant. She admitted to hot flushing that had subsided 3 months before presentation. She had previously been well and reported stability in weight. She did not participate in any regular physical exercise. She denied headaches, galactorrhea, or visual field changes. Family history of autoimmune disorders was absent. She was a well-looking normotensive woman of ideal body weight (body mass index, 22.6 kg/m2). She was not hirsute or virulized, nor did she have any acne. Results on thyroid examination were normal. She was Tanner stage 5 for breast development and pubic hair, with no galactorrhea. Abdominal examination failed to reveal any palpable masses or organomegaly. Laboratory investigations revealed negative serum b-hCG, normal PRL (6 mg/L; normal range, 0–25 mg/L), TSH (2.01 mU/L; normal range, 0.2–5.0 mU/L), and T (1.0 nmol/L; normal range, 0.5–2.6 nmol/L). She had a low serum E2 level of 31 pmol/L (follicular phase reference range, 69–905 pmol/L). Her FSH level was low (1.0 IU/L). Luteinizing hormone was elevated at 41 IU/L, into the postmenopausal range (16–54 IU/L). Gonadotropin measurements were repeated 1 week later to ensure that midcycle peak was not the explanation for her elevated LH. However, the pattern of low FSH (1.6 IU/L) and elevated LH (44.6 IU/L) persisted. Two months earlier the family doctor had obtained a pelvic ultrasound examination and magnetic resonance imaging of her sella. The pelvic ultrasound examination revealed that the uterus was retroflexed but otherwise unremarkable. The right ovary showed a follicle measuring 19 mm. The left ovary was unremarkable. No free fluid or pelvic mass were seen. Magnetic resonance imaging of the sella showed a 1-mm nonenhancing mass in the pituitary gland at the insertion of the pituitary stalk, likely representing a very tiny vein or cyst.
1097.e10
Donovan et al.
Androstenedione was measured to assess ovarian responsivity to LH and was found to be normal at 3.1 nmol/L. Pelvic ultrasound examination was repeated despite the relatively unremarkable pelvic ultrasound report from 3 months earlier. This demonstrated a solid-appearing structure adjacent to the right side of the uterus measuring 3.0 3.7 4.6 cm. This was thought to represent a mass arising from the right ovary or a pedunculated fibroid arising from the uterus. The 19-cm follicle arising from the right ovary described on the previous ultrasound examination was not seen. The endometrial echo was quite thin. Her inhibin B level was found to be elevated at 552 pg/mL (follicular phase, 16–290 pg/mL). Retrospective review of the initial pelvic ultrasound results did not alter the original interpretation. Gynecology consultation 6 weeks later found a very prominent solid mass anterior to the uterus on the right measuring 5 cm, which was indistinguishable from a fibroid. Left adnexa was normal. A repeat ultrasound examination that day revealed that the mass had grown to 5.5 3.8 4.5 cm. Three days later the patient underwent laparotomy and right oophorectomy (Fig. 1). The solid ovarian tumor had a yellow whorled appearance, measured 5.4 4.0 3.2 cm, and weighed 38.3 g. The tumor was a thecoma and composed of a circumscribed proliferation of elongated cells with pale cytoplasm. The cells were immunopositive with inhibin alpha (mouse monoclonal antibody, clone R1; AbD Serotec, Cedarlane Laboratories, Burlington, Ontario) and formed broad interconnecting bands separated by hyalinized stroma (Fig. 2). Atypia was minimal, and the mitotic rate was less than three per 10 high-power fields. She had a return of regular menstrual bleeding 28 days after surgical removal of the right ovary. Five weeks postoperatively her inhibin B level was normal at 102 pg/mL (follicular phase, 16–290 pg/mL), as were the gonadotropin levels: FSH 4 IU/L (follicular phase, 2–10 IU/L), LH 6 IU/L (follicular phase, 1–13 IU/L). She had a second menstrual withdrawal bleed 28 days after the first and then went on to conceive. Her pregnancy was uneventful. She delivered a healthy girl weighing 8 lb 13 oz by spontaneous vaginal delivery at 40 weeks’ gestation.
DISCUSSION Diagnostic Considerations for Patients with Amenorrhea and Isolated LH Elevations Determining the etiology of this women’s amenorrhea was challenging. The initial pelvic ultrasound after 10 months of amenorrhea showed a lack of concerning findings. Her clinical presentation and persistently markedly elevated LH with low FSH and low E2 helped to exclude the more common etiologies of secondary amenorrhea, such as polycystic ovarian syndrome, hypothalamic amenorrhea, or premature ovarian failure. With these more common etiologies of amenorrhea ruled out, her gonadotropin profile prompted a review of other etiologies of amenorrhea associated with low or normal FSH and high LH. The diagnostic possibilities are limited and rare. They include [1] inactivating mutation of the LH receptor gene, [2] FSH beta mutation, [3] LH-secreting pituitary adenoma, [4] hCG or heterophilic antibodies in the serum of the patient interfering with the LH immunoassay, and [5] inhibin-secreting sex cord–stromal tumor. Ovarian resistance to LH caused by an inactivating mutation of the LH receptor gene is a rare cause of amenorrhea associated with an isolated LH elevation. A phenotypical female with karyotype XX and an LH gene receptor mutation was reported by Latronico el al. (4). Her LH measurement was similar to our patient’s at 37 IU/L (follicular phase, 7.1 3.0). However, unlike our patient,
Thecoma causing isolated LH elevation
Vol. 94, No. 3, August 2010
FIGURE 2 (A) Thecoma formed of elongated cells with strands of hyalinized stroma (hematoxylin and eosin; original magnification, 200). (B) Inhibin alpha–positive cells (diaminobenzidine-based immunoperoxidase technique; original magnification, 400).
secrete inefficiently and do not generally cause recognizable clinical syndromes (6). Rare reports of gonadotropin-producing pituitary adenomas have been associated with elevated E and hyperstimulation syndrome, rather than the hypoestrogenization state of our patient (7, 8). Had our workup for this patient not have led so quickly to a diagnosis, suppressability of the LH to GnRH agonist could have helped to exclude the diagnosis of an LH-secreting pituitary adenoma. Human chorionic gonadotropin from pregnancy may cross-react in some LH assays. However, we would not expect a low E2 level in pregnancy. Heterophilic antibodies in the serum of the patient that interfere with the immuonoassay are another consideration when abnormally high gonadotropin values cannot easily be explained by the clinical presentation (9). The use of a heterophilic antibodyblocking agent and/or nonlinear measurement of sample dilutions are laboratory techniques that could help support the suspicion of heterophilic antibodies. This was not done in our patient, but given her clinical and LH response to surgery we can conclude that her LH elevations were not the result of heterophilic antibodies interfering with the immunoassay.
Comparison with Other Reported Cases of InhibinSecreting Sex Cord–Stromal Tumors
Donovan. Thecoma causing isolated LH elevation. Fertil Steril 2010.
whose FSH was low, Latronico et al.’s patient had an FSH level that was at mid-normal range, measuring 8.7 IU/L (normal range, 3.2–10). Their patient had spontaneous gonadarche at age 13 years and Tanner stage 5 pubic hair and breast development. She was evaluated at age 22 years after having only one spontaneous menstrual bleed at age 20 years. Although LH receptor mutation was thought to be a possibility in our patient, her history of established regular menses for 6 years before her menstrual dysfunction, as well as her unremarkable family history, prompted the search for alternative explanations for her secondary amenorrhea. Additionally, her normal androstenedione level suggested normal ovarian LH responsivity. High LH and undetectable FSH have been described in women with FSH beta mutations (5). Women with these mutations have delayed puberty, incomplete breast development, primary amenorrhea, and low E2 levels. Although this diagnosis fit the biochemical picture of our patient, her clinical picture was out of keeping with a diagnosis of FSH beta mutation. An LH-secreting pituitary adenoma was also thought to be an unlikely diagnosis in our patient because gonadotropin adenomas Fertility and Sterility
Ovarian sex cord–stromal tumors comprise a variety of benign and malignant tumors that develop from cells surrounding the oocytes. They include granulosa cell tumors, thecomas, and fibromas. They constitute 5% of primary ovarian neoplasms (10, 11). Most produce steroid hormones, so signs of E or androgen excess (i.e., precocious puberty, abnormal uterine bleeding, endometrial hyperplasia, or carcinoma, or hirsutism and virilization, respectively) are more common clinical presentations in patients with these tumors (12). Amenorrhea and infertility are rare as the only presenting complaint of patients with sex cord–stromal tumors. Ovarian sex cord–stromal tumors presenting as amenorrhea with confirmed inhibin elevations were first described in 1989 by Lappohn et al. (13). Other published case reports of granulosa cell tumors that present with amenorrhea in the absence of hirsutism and virilization generally demonstrate low FSH and normal LH measurements (14). Our patient is similar to a 37-year-old woman described by Meyer et al. (15), who was found to have an inhibin B–producing ovarian fibrothecoma. Meyer’s patient also had an isolated elevation in LH (22.7 mIU/mL; normal range, 1.9–12.5 mIU/mL) and low E2 (31 pg/mL). However, the LH was not as elevated as in our patient, whose LH was 41–44.6 IU/L. It is impossible to make direct comparisons between different LH assays; however, the relative increase above normal range was greater in our patient. We postulate that the degree of LH elevation that results in response to hypoestrogenization is influenced by age, just as TSH response to hypothyroidism is more robust in younger vs. older individuals. Gonadotropin response to GnRH has been shown to be influenced by age in humans (16, 17). Another woman, also 37 years old (18), was found to have a 7-cm granulosa cell tumor. She presented with secondary amenorrhea and gonadotropin levels similar to those of the patient described by Meyer (LH 27.2 IU/L and FSH 3.6 IU/L). However, unlike Meyer’s and our patient, preoperative inhibin levels were not obtained. Inhibin levels measured postoperatively were not elevated, and the gonadotropin levels normalized postoperatively (18). The explanation of why reported cases of inhibin B–secreting sex cord–stromal ovarian tumors vary in their LH level is likely a result of varying tumor secretion of E2. Agha-Hosseini et al. (19) have described a 26-year-old women with a granulosa cell tumor. She presented with 4 years of secondary amenorrhea and in contrast to our
1097.e11
patient had low levels of both LH and FSH. Her E2 measured 100 pg/ mL, and menstrual withdrawal bleeding was induced with medroxyprogesterone acetate. Unlike our patient, her estrogenized state may explain why her LH was low. The authors hypothesized that inhibins suppressed both her LH and FSH. However, our patient case suggests that inhibin B may not completely suppress LH, at least in the hypoestrogenized state. Interestingly, both inhibin A and B measurements in Agha-Hosseini’s patient were found to be elevated, 255 ng/ML (normal range, 4–150 ng/mL) and 1,124 ng/mL (normal range, 15–200 ng/mL), respectively. Her gonadotropin levels and inhibin A and B levels normalized, and she had a spontaneous menses 18 days after surgical removal of a granulosa cell tumor. Inhibin B, not inhibin A, was measured in our patient because preoperatively we suspected a granulosa cell tumor. Inhibin B seems to be the predominant form of inhibin secreted by granulosa cell tumors (20). Mom et al. (20) showed that inhibin B measurement had better sensitivity (89%) and specificity (100%) compared with inhibin A (67% and 100%, respectively) in the diagnosis of granulosa cell tumors. Inhibin B also seemed to be the better marker of granulose cell tumor in Mom’s study because the median inhibin B level elevation in the setting of granulose cell tumor was 60-fold vs. inhibin A, which was just sevenfold. On the basis of the LH elevation that our patient demonstrated preoperatively, we postulate that E status is a more potent regulator of LH than inhibin B. Taken together, these case examples support the hypothesis that in humans inhibin B does not completely suppress LH, at least in the hypoestrogenized state, and that E status is a potent regulator of LH.
It is interesting that, other than our case, the only other woman in the English literature reported to have a confirmed inhibin B–producing sex cord–stromal tumor that presented with reversible marked LH elevations and hypoestrogenization had a fibrothecoma (15) and not a granulosa cell tumor. Approximately 30% of granulosa–stromal cell tumors do not produce E2. It has been postulated that the inability to produce E2 may be related to a relative lack of theca cells to produce androstenedione, the precursor for E2 production. Ovarian thecoma is mostly a begin tumor; however, when associated with hyperestrogenization, endometrial carcinoma should be excluded (11).
CONCLUSIONS This human example of inhibin B excess supports our current understanding of the inhibitory effect of inhibin B on FSH secretion in humans. This patient example cannot refute Robertson et al.’s hypothesis (3) that an inhibin B–LH feedback mechanism exists in the human female; however, it provides a clinical example that, at least in the hypoestrogenized state, LH may remain elevated in the setting of high inhibin B. This case study also underscores the importance of measuring inhibin B and repeating a pelvic ultrasound examination despite previously ‘‘normal findings,’’ if amenorrhea is persistent, when biochemistry demonstrates an isolated LH elevation in the setting of hypoestrogenized amenorrhea. Acknowledgments: The authors thank the patient for allowing them to write up her medical history.
REFERENCES 1. Burger HG, Farnworth PG, Findlay JK, Gurusinghe CJ, Healy DL, Mamers P, et al. Aspects of current and future inhibin research. Reprod Fertil Dev 1995;7:997–1002. 2. Tsigkou A, Luisi S, Reis FM, Petraglia F. Inhibins as diagnostic markers in human reproduction. Adv Clin Chem 2008;45:1–29. 3. Robertson DM, Hale GE, Jolley D, Fraser IS, Hughes CL, Burger HG. Interrelationships between ovarian and pituitary hormones in ovulatory menstrual cycles across reproductive age. J Clin Endocrinol Metab 2009;94:138–44. 4. Latronico AC, Anasti J, Arnhold IJ, Rapaport R, Mendonca BB, Bloise W, et al. Brief report: testicular and ovarian resistance to luteinizing hormone caused by inactivating mutations of the luteinizing hormonereceptor gene. N Engl J Med 1996;334:507–12. 5. Berger K, Souza H, Brito VN. d’Alva CB, Mendonca BB, Latronico AC. Clinical and hormonal features of selective follicle-stimulating hormone (FSH) deficiency due to FSH beta-subunit gene mutations in both sexes. Fertil Steril 2005;83:466–70. 6. Chanson P, Brochier S. Non-functioning pituitary adenomas. J Endocrinol Invest 2005;28:93–9. 7. Castelo-Branco C, del Pino M, Valladares E. Ovarian hyperstimulation, hyperprolactinaemia and LH gonadotroph adenoma. Reprod Biomed Online 2009;19: 153–5.
1097.e12
Donovan et al.
8. Knoepfelmacher M, Danilovic DL, Rosa Nasser RH, Mendonca BB. Effectiveness of treating ovarian hyperstimulation syndrome with cabergoline in two patients with gonadotropin-producing pituitary adenomas. Fertil Steril 2006;86:719.e15–8. 9. Berglund L, Holmberg NG. Heterophilic antibodies against rabbit serum causing falsely elevated gonadotropin levels. Acta Obstet Gynecol Scand 1989;68: 377–8. 10. Cronje HS, Niemand I, Bam RH, Woodruff JD. Review of the granulosa-theca cell tumors from the emil Novak ovarian tumor registry. Am J Obstet Gynecol 1999;180:323–7. 11. Aboud E. A review of granulosa cell tumours and thecomas of the ovary. Arch Gynecol Obstet 1997;259: 161–5. 12. Outwater EK, Wagner BJ, Mannion C, McLarney JK, Kim B. Sex cord-stromal and steroid cell tumors of the ovary. Radiographics 1998;18:1523–46. 13. Lappohn RE, Burger HG, Bouma J, Bangah M, Krans M, de Bruijn HW. Inhibin as a marker for granulosa-cell tumors. N Engl J Med 1989;321: 790–3. 14. Kurihara S, Hirakawa T, Amada S, Ariyoshi K, Nakano H. Inhibin-producing ovarian granulosa cell tumor as a cause of secondary amenorrhea: case report and review of the literature. J Obstet Gynaecol Res 2004;30:439–43.
Thecoma causing isolated LH elevation
15. Meyer AC, Papadimitriou JC, Silverberg SG, Sharara FI. Secondary amenorrhea and infertility caused by an inhibin-B-producing ovarian fibrothecoma. Fertil Steril 2000;73:258–60. 16. Shaw ND, Srouji SS, Histed SN, McCurnin KE, Hall JE. Aging attenuates the pituitary response to gonadotropin-releasing hormone. J Clin Endocrinol Metab 2009;94:3259–64. 17. Lambalk CB, de Boer L, Schoute E, Popp-Snyders C, Schoemaker J. Post-menopausal and chronological age have divergent effects on pituitary and hypothalamic function in episodic gonadotrophin secretion. Clin Endocrinol (Oxf) 1997;46:439–43. 18. Datta M, Winter-Roach BA, Fitzgerald C. Secondary amenorrhoea associated with an isolated elevated serum leuteinising hormone: an unusual presentation of a granulosa cell tumour. J Obstet Gynaecol 2006;26: 830–2. 19. Agha-Hosseini M, Aleyaseen A, Safdarian L, Kashani L. Secondary amenorrhea with low serum luteinizing hormone and follicle-stimulating hormone caused by an inhibin A- and inhibin B-producing granulosa cell tumor. Taiwan J Obstet Gynecol 2009;48:72–5. 20. Mom CH, Engelen MJ, Willemse PH, Gietema JA, ten Hoor KA, de Vries EG, et al. Granulosa cell tumors of the ovary: the clinical value of serum inhibin A and B levels in a large single center cohort. Gynecol Oncol 2007;105:365–72.
Vol. 94, No. 3, August 2010
Objective: To review the diagnostic possibilities that exists when the workup of amenorrhea reveals an isolated LH elevation; and to examine the effect of inhibin B on LH levels in vivo. Design: Case report. Setting: University hospital. Patient(s): A 20-year-old woman presented with secondary amenorrhea. Her FSH measurement was low, and the LH level was elevated. The recognition that this was an unusual pattern led to the diagnosis of a rare but very treatable inhibin B–producing thecoma, despite the fact that results on the initial pelvic ultrasound examination performed 10 months after presentation of amenorrhea were relatively unremarkable. Intervention(s): Surgical removal of an ovarian thecoma. Main Outcome Measure(s): Gonadotropins, E2, inhibin B, menstrual bleeding, and fertility. Result(s): Removal of the ovarian thecoma resulted in a normalization of FSH, LH, and inhibin B levels and a return of spontaneous menses 28 days later. Pregnancy occurred with the third postoperative menstrual cycle, followed by the delivery of a healthy full-term girl. Conclusion(s): Inhibin B–producing sex cord granolosa–stromal cell tumors should be considered in women who present with amenorrhea with isolated LH elevations, even in the setting of a previously normal pelvic ultrasound report. Diagnostic considerations that arise in the workup of amenorrhea when there is an isolated elevation in LH that is accompanied by normal or low FSH levels are reviewed. This rare clinical presentation provides the opportunity to observe the impact of inhibin B on gonadotropins in vivo. (Fertil Steril 2010;94:1097.e9–e12. 2010 by American Society for Reproductive Medicine.) Key Words: Isolated LH elevation, inhibin B, thecoma, amenorrhea
Gonadotropin production and release is regulated by GnRH. Ovarian feedback from estrogen (E) and P, was well as inhibins, partially regulates gonadotropin release. Inhibins are heterodimeric glycoproteins with structural homology to transforming growth factor b family (1). Inhibin consists of an alpha subunit joined to a beta A subunit (inhibin A) or a beta B subunit (inhibit B). Inhibin B is secreted primarily by the granulosa cells of developing follicles. Inhibin A seems to be secreted mainly by the corpus luteum, thus it is considered a marker of dominant follicle and corpus luteum activity (2). It is generally recognized that inhibins suppress FSH release. The influence of inhibin B on LH remains an area of ongoing controversy. Recently a significant negative association was Received January 22, 2010; revised February 3, 2010; accepted February 4, 2010; published online March 21, 2010. L.E.D. has nothing to disclose. P.H.B. has nothing to disclose. M.A.D. has nothing to disclose. Reprint requests: Lois E. Donovan, M.D., Foothills Medical Centre, North Tower Room 716, 1403 29th St. NW, Calgary, Alberta, Canada T2N 2T9 (FAX: 403-944-3199; E-mail: [email protected]).
0015-0282/$36.00 doi:10.1016/j.fertnstert.2010.02.016
observed in the follicular phase between inhibin B and LH (3). This has led Robertson et al. to ‘‘postulate that an inhibin B:LH feedback mechanism may exist in the human female’’ (3). Here, we describe a 20-year-old woman with an inhibin B–producing thecoma. This patient provided the rare opportunity to study the impact of inhibin B on gonadotropins in vivo, in the hypoestrogenized state. This patient case prompted a review of the diagnostic possibilities that arise in the workup of amenorrhea when there is an isolated elevation in LH that is accompanied by normal or low FSH levels.
CASE REPORT A 20-year-old woman with a 1-year history of secondary amenorrhea was referred for endocrinology consultation after her family doctor discovered a 1-mm nonenhancing lesion in the pituitary gland. Menarche occurred at 12 years of age. She had experienced regular menses every 28 days with 6 days of menstrual flow, until moving to Canada from Holland and marrying 2 years earlier. Then she
Fertility and Sterility Vol. 94, No. 3, August 2010 Copyright ª2010 American Society for Reproductive Medicine, Published by Elsevier Inc.
1097.e9
FIGURE 1 Intraoperative gross appearance of the thecoma.
Donovan. Thecoma causing isolated LH elevation. Fertil Steril 2010.
began to develop a less-regular pattern of menstrual bleeding that was described at times as 1 week of menstrual bleeding followed by a week of no menstrual bleeding and then a recurrence of menstrual bleeding 14 days after the start of the previous menstrual bleed. Menstrual bleeding completely stopped 1 year before her presentation to the endocrinologist. She had been sexually active without contraceptive use since her marriage. She had never been pregnant. She admitted to hot flushing that had subsided 3 months before presentation. She had previously been well and reported stability in weight. She did not participate in any regular physical exercise. She denied headaches, galactorrhea, or visual field changes. Family history of autoimmune disorders was absent. She was a well-looking normotensive woman of ideal body weight (body mass index, 22.6 kg/m2). She was not hirsute or virulized, nor did she have any acne. Results on thyroid examination were normal. She was Tanner stage 5 for breast development and pubic hair, with no galactorrhea. Abdominal examination failed to reveal any palpable masses or organomegaly. Laboratory investigations revealed negative serum b-hCG, normal PRL (6 mg/L; normal range, 0–25 mg/L), TSH (2.01 mU/L; normal range, 0.2–5.0 mU/L), and T (1.0 nmol/L; normal range, 0.5–2.6 nmol/L). She had a low serum E2 level of 31 pmol/L (follicular phase reference range, 69–905 pmol/L). Her FSH level was low (1.0 IU/L). Luteinizing hormone was elevated at 41 IU/L, into the postmenopausal range (16–54 IU/L). Gonadotropin measurements were repeated 1 week later to ensure that midcycle peak was not the explanation for her elevated LH. However, the pattern of low FSH (1.6 IU/L) and elevated LH (44.6 IU/L) persisted. Two months earlier the family doctor had obtained a pelvic ultrasound examination and magnetic resonance imaging of her sella. The pelvic ultrasound examination revealed that the uterus was retroflexed but otherwise unremarkable. The right ovary showed a follicle measuring 19 mm. The left ovary was unremarkable. No free fluid or pelvic mass were seen. Magnetic resonance imaging of the sella showed a 1-mm nonenhancing mass in the pituitary gland at the insertion of the pituitary stalk, likely representing a very tiny vein or cyst.
1097.e10
Donovan et al.
Androstenedione was measured to assess ovarian responsivity to LH and was found to be normal at 3.1 nmol/L. Pelvic ultrasound examination was repeated despite the relatively unremarkable pelvic ultrasound report from 3 months earlier. This demonstrated a solid-appearing structure adjacent to the right side of the uterus measuring 3.0 3.7 4.6 cm. This was thought to represent a mass arising from the right ovary or a pedunculated fibroid arising from the uterus. The 19-cm follicle arising from the right ovary described on the previous ultrasound examination was not seen. The endometrial echo was quite thin. Her inhibin B level was found to be elevated at 552 pg/mL (follicular phase, 16–290 pg/mL). Retrospective review of the initial pelvic ultrasound results did not alter the original interpretation. Gynecology consultation 6 weeks later found a very prominent solid mass anterior to the uterus on the right measuring 5 cm, which was indistinguishable from a fibroid. Left adnexa was normal. A repeat ultrasound examination that day revealed that the mass had grown to 5.5 3.8 4.5 cm. Three days later the patient underwent laparotomy and right oophorectomy (Fig. 1). The solid ovarian tumor had a yellow whorled appearance, measured 5.4 4.0 3.2 cm, and weighed 38.3 g. The tumor was a thecoma and composed of a circumscribed proliferation of elongated cells with pale cytoplasm. The cells were immunopositive with inhibin alpha (mouse monoclonal antibody, clone R1; AbD Serotec, Cedarlane Laboratories, Burlington, Ontario) and formed broad interconnecting bands separated by hyalinized stroma (Fig. 2). Atypia was minimal, and the mitotic rate was less than three per 10 high-power fields. She had a return of regular menstrual bleeding 28 days after surgical removal of the right ovary. Five weeks postoperatively her inhibin B level was normal at 102 pg/mL (follicular phase, 16–290 pg/mL), as were the gonadotropin levels: FSH 4 IU/L (follicular phase, 2–10 IU/L), LH 6 IU/L (follicular phase, 1–13 IU/L). She had a second menstrual withdrawal bleed 28 days after the first and then went on to conceive. Her pregnancy was uneventful. She delivered a healthy girl weighing 8 lb 13 oz by spontaneous vaginal delivery at 40 weeks’ gestation.
DISCUSSION Diagnostic Considerations for Patients with Amenorrhea and Isolated LH Elevations Determining the etiology of this women’s amenorrhea was challenging. The initial pelvic ultrasound after 10 months of amenorrhea showed a lack of concerning findings. Her clinical presentation and persistently markedly elevated LH with low FSH and low E2 helped to exclude the more common etiologies of secondary amenorrhea, such as polycystic ovarian syndrome, hypothalamic amenorrhea, or premature ovarian failure. With these more common etiologies of amenorrhea ruled out, her gonadotropin profile prompted a review of other etiologies of amenorrhea associated with low or normal FSH and high LH. The diagnostic possibilities are limited and rare. They include [1] inactivating mutation of the LH receptor gene, [2] FSH beta mutation, [3] LH-secreting pituitary adenoma, [4] hCG or heterophilic antibodies in the serum of the patient interfering with the LH immunoassay, and [5] inhibin-secreting sex cord–stromal tumor. Ovarian resistance to LH caused by an inactivating mutation of the LH receptor gene is a rare cause of amenorrhea associated with an isolated LH elevation. A phenotypical female with karyotype XX and an LH gene receptor mutation was reported by Latronico el al. (4). Her LH measurement was similar to our patient’s at 37 IU/L (follicular phase, 7.1 3.0). However, unlike our patient,
Thecoma causing isolated LH elevation
Vol. 94, No. 3, August 2010
FIGURE 2 (A) Thecoma formed of elongated cells with strands of hyalinized stroma (hematoxylin and eosin; original magnification, 200). (B) Inhibin alpha–positive cells (diaminobenzidine-based immunoperoxidase technique; original magnification, 400).
secrete inefficiently and do not generally cause recognizable clinical syndromes (6). Rare reports of gonadotropin-producing pituitary adenomas have been associated with elevated E and hyperstimulation syndrome, rather than the hypoestrogenization state of our patient (7, 8). Had our workup for this patient not have led so quickly to a diagnosis, suppressability of the LH to GnRH agonist could have helped to exclude the diagnosis of an LH-secreting pituitary adenoma. Human chorionic gonadotropin from pregnancy may cross-react in some LH assays. However, we would not expect a low E2 level in pregnancy. Heterophilic antibodies in the serum of the patient that interfere with the immuonoassay are another consideration when abnormally high gonadotropin values cannot easily be explained by the clinical presentation (9). The use of a heterophilic antibodyblocking agent and/or nonlinear measurement of sample dilutions are laboratory techniques that could help support the suspicion of heterophilic antibodies. This was not done in our patient, but given her clinical and LH response to surgery we can conclude that her LH elevations were not the result of heterophilic antibodies interfering with the immunoassay.
Comparison with Other Reported Cases of InhibinSecreting Sex Cord–Stromal Tumors
Donovan. Thecoma causing isolated LH elevation. Fertil Steril 2010.
whose FSH was low, Latronico et al.’s patient had an FSH level that was at mid-normal range, measuring 8.7 IU/L (normal range, 3.2–10). Their patient had spontaneous gonadarche at age 13 years and Tanner stage 5 pubic hair and breast development. She was evaluated at age 22 years after having only one spontaneous menstrual bleed at age 20 years. Although LH receptor mutation was thought to be a possibility in our patient, her history of established regular menses for 6 years before her menstrual dysfunction, as well as her unremarkable family history, prompted the search for alternative explanations for her secondary amenorrhea. Additionally, her normal androstenedione level suggested normal ovarian LH responsivity. High LH and undetectable FSH have been described in women with FSH beta mutations (5). Women with these mutations have delayed puberty, incomplete breast development, primary amenorrhea, and low E2 levels. Although this diagnosis fit the biochemical picture of our patient, her clinical picture was out of keeping with a diagnosis of FSH beta mutation. An LH-secreting pituitary adenoma was also thought to be an unlikely diagnosis in our patient because gonadotropin adenomas Fertility and Sterility
Ovarian sex cord–stromal tumors comprise a variety of benign and malignant tumors that develop from cells surrounding the oocytes. They include granulosa cell tumors, thecomas, and fibromas. They constitute 5% of primary ovarian neoplasms (10, 11). Most produce steroid hormones, so signs of E or androgen excess (i.e., precocious puberty, abnormal uterine bleeding, endometrial hyperplasia, or carcinoma, or hirsutism and virilization, respectively) are more common clinical presentations in patients with these tumors (12). Amenorrhea and infertility are rare as the only presenting complaint of patients with sex cord–stromal tumors. Ovarian sex cord–stromal tumors presenting as amenorrhea with confirmed inhibin elevations were first described in 1989 by Lappohn et al. (13). Other published case reports of granulosa cell tumors that present with amenorrhea in the absence of hirsutism and virilization generally demonstrate low FSH and normal LH measurements (14). Our patient is similar to a 37-year-old woman described by Meyer et al. (15), who was found to have an inhibin B–producing ovarian fibrothecoma. Meyer’s patient also had an isolated elevation in LH (22.7 mIU/mL; normal range, 1.9–12.5 mIU/mL) and low E2 (31 pg/mL). However, the LH was not as elevated as in our patient, whose LH was 41–44.6 IU/L. It is impossible to make direct comparisons between different LH assays; however, the relative increase above normal range was greater in our patient. We postulate that the degree of LH elevation that results in response to hypoestrogenization is influenced by age, just as TSH response to hypothyroidism is more robust in younger vs. older individuals. Gonadotropin response to GnRH has been shown to be influenced by age in humans (16, 17). Another woman, also 37 years old (18), was found to have a 7-cm granulosa cell tumor. She presented with secondary amenorrhea and gonadotropin levels similar to those of the patient described by Meyer (LH 27.2 IU/L and FSH 3.6 IU/L). However, unlike Meyer’s and our patient, preoperative inhibin levels were not obtained. Inhibin levels measured postoperatively were not elevated, and the gonadotropin levels normalized postoperatively (18). The explanation of why reported cases of inhibin B–secreting sex cord–stromal ovarian tumors vary in their LH level is likely a result of varying tumor secretion of E2. Agha-Hosseini et al. (19) have described a 26-year-old women with a granulosa cell tumor. She presented with 4 years of secondary amenorrhea and in contrast to our
1097.e11
patient had low levels of both LH and FSH. Her E2 measured 100 pg/ mL, and menstrual withdrawal bleeding was induced with medroxyprogesterone acetate. Unlike our patient, her estrogenized state may explain why her LH was low. The authors hypothesized that inhibins suppressed both her LH and FSH. However, our patient case suggests that inhibin B may not completely suppress LH, at least in the hypoestrogenized state. Interestingly, both inhibin A and B measurements in Agha-Hosseini’s patient were found to be elevated, 255 ng/ML (normal range, 4–150 ng/mL) and 1,124 ng/mL (normal range, 15–200 ng/mL), respectively. Her gonadotropin levels and inhibin A and B levels normalized, and she had a spontaneous menses 18 days after surgical removal of a granulosa cell tumor. Inhibin B, not inhibin A, was measured in our patient because preoperatively we suspected a granulosa cell tumor. Inhibin B seems to be the predominant form of inhibin secreted by granulosa cell tumors (20). Mom et al. (20) showed that inhibin B measurement had better sensitivity (89%) and specificity (100%) compared with inhibin A (67% and 100%, respectively) in the diagnosis of granulosa cell tumors. Inhibin B also seemed to be the better marker of granulose cell tumor in Mom’s study because the median inhibin B level elevation in the setting of granulose cell tumor was 60-fold vs. inhibin A, which was just sevenfold. On the basis of the LH elevation that our patient demonstrated preoperatively, we postulate that E status is a more potent regulator of LH than inhibin B. Taken together, these case examples support the hypothesis that in humans inhibin B does not completely suppress LH, at least in the hypoestrogenized state, and that E status is a potent regulator of LH.
It is interesting that, other than our case, the only other woman in the English literature reported to have a confirmed inhibin B–producing sex cord–stromal tumor that presented with reversible marked LH elevations and hypoestrogenization had a fibrothecoma (15) and not a granulosa cell tumor. Approximately 30% of granulosa–stromal cell tumors do not produce E2. It has been postulated that the inability to produce E2 may be related to a relative lack of theca cells to produce androstenedione, the precursor for E2 production. Ovarian thecoma is mostly a begin tumor; however, when associated with hyperestrogenization, endometrial carcinoma should be excluded (11).
CONCLUSIONS This human example of inhibin B excess supports our current understanding of the inhibitory effect of inhibin B on FSH secretion in humans. This patient example cannot refute Robertson et al.’s hypothesis (3) that an inhibin B–LH feedback mechanism exists in the human female; however, it provides a clinical example that, at least in the hypoestrogenized state, LH may remain elevated in the setting of high inhibin B. This case study also underscores the importance of measuring inhibin B and repeating a pelvic ultrasound examination despite previously ‘‘normal findings,’’ if amenorrhea is persistent, when biochemistry demonstrates an isolated LH elevation in the setting of hypoestrogenized amenorrhea. Acknowledgments: The authors thank the patient for allowing them to write up her medical history.
REFERENCES 1. Burger HG, Farnworth PG, Findlay JK, Gurusinghe CJ, Healy DL, Mamers P, et al. Aspects of current and future inhibin research. Reprod Fertil Dev 1995;7:997–1002. 2. Tsigkou A, Luisi S, Reis FM, Petraglia F. Inhibins as diagnostic markers in human reproduction. Adv Clin Chem 2008;45:1–29. 3. Robertson DM, Hale GE, Jolley D, Fraser IS, Hughes CL, Burger HG. Interrelationships between ovarian and pituitary hormones in ovulatory menstrual cycles across reproductive age. J Clin Endocrinol Metab 2009;94:138–44. 4. Latronico AC, Anasti J, Arnhold IJ, Rapaport R, Mendonca BB, Bloise W, et al. Brief report: testicular and ovarian resistance to luteinizing hormone caused by inactivating mutations of the luteinizing hormonereceptor gene. N Engl J Med 1996;334:507–12. 5. Berger K, Souza H, Brito VN. d’Alva CB, Mendonca BB, Latronico AC. Clinical and hormonal features of selective follicle-stimulating hormone (FSH) deficiency due to FSH beta-subunit gene mutations in both sexes. Fertil Steril 2005;83:466–70. 6. Chanson P, Brochier S. Non-functioning pituitary adenomas. J Endocrinol Invest 2005;28:93–9. 7. Castelo-Branco C, del Pino M, Valladares E. Ovarian hyperstimulation, hyperprolactinaemia and LH gonadotroph adenoma. Reprod Biomed Online 2009;19: 153–5.
1097.e12
Donovan et al.
8. Knoepfelmacher M, Danilovic DL, Rosa Nasser RH, Mendonca BB. Effectiveness of treating ovarian hyperstimulation syndrome with cabergoline in two patients with gonadotropin-producing pituitary adenomas. Fertil Steril 2006;86:719.e15–8. 9. Berglund L, Holmberg NG. Heterophilic antibodies against rabbit serum causing falsely elevated gonadotropin levels. Acta Obstet Gynecol Scand 1989;68: 377–8. 10. Cronje HS, Niemand I, Bam RH, Woodruff JD. Review of the granulosa-theca cell tumors from the emil Novak ovarian tumor registry. Am J Obstet Gynecol 1999;180:323–7. 11. Aboud E. A review of granulosa cell tumours and thecomas of the ovary. Arch Gynecol Obstet 1997;259: 161–5. 12. Outwater EK, Wagner BJ, Mannion C, McLarney JK, Kim B. Sex cord-stromal and steroid cell tumors of the ovary. Radiographics 1998;18:1523–46. 13. Lappohn RE, Burger HG, Bouma J, Bangah M, Krans M, de Bruijn HW. Inhibin as a marker for granulosa-cell tumors. N Engl J Med 1989;321: 790–3. 14. Kurihara S, Hirakawa T, Amada S, Ariyoshi K, Nakano H. Inhibin-producing ovarian granulosa cell tumor as a cause of secondary amenorrhea: case report and review of the literature. J Obstet Gynaecol Res 2004;30:439–43.
Thecoma causing isolated LH elevation
15. Meyer AC, Papadimitriou JC, Silverberg SG, Sharara FI. Secondary amenorrhea and infertility caused by an inhibin-B-producing ovarian fibrothecoma. Fertil Steril 2000;73:258–60. 16. Shaw ND, Srouji SS, Histed SN, McCurnin KE, Hall JE. Aging attenuates the pituitary response to gonadotropin-releasing hormone. J Clin Endocrinol Metab 2009;94:3259–64. 17. Lambalk CB, de Boer L, Schoute E, Popp-Snyders C, Schoemaker J. Post-menopausal and chronological age have divergent effects on pituitary and hypothalamic function in episodic gonadotrophin secretion. Clin Endocrinol (Oxf) 1997;46:439–43. 18. Datta M, Winter-Roach BA, Fitzgerald C. Secondary amenorrhoea associated with an isolated elevated serum leuteinising hormone: an unusual presentation of a granulosa cell tumour. J Obstet Gynaecol 2006;26: 830–2. 19. Agha-Hosseini M, Aleyaseen A, Safdarian L, Kashani L. Secondary amenorrhea with low serum luteinizing hormone and follicle-stimulating hormone caused by an inhibin A- and inhibin B-producing granulosa cell tumor. Taiwan J Obstet Gynecol 2009;48:72–5. 20. Mom CH, Engelen MJ, Willemse PH, Gietema JA, ten Hoor KA, de Vries EG, et al. Granulosa cell tumors of the ovary: the clinical value of serum inhibin A and B levels in a large single center cohort. Gynecol Oncol 2007;105:365–72.
Vol. 94, No. 3, August 2010