GnRH agonist‐induced ovarian hyperstimulation: A diagnostic clue to pituitary adenoma

A 36-year-old woman presented with chronic pelvic pain and dysmenorrhea. Imaging revealed posterior wall adenomyosis, right ovarian endometrioma (6 × 5 × 5.5 cm). She underwent robotic adenomyomectomy and endometriotic cystectomy. A single dose of gonadotropin-releasing hormone (GnRH) agonist (leuprolide acetate 3.75 mg) was administered postoperatively in the luteal phase for endometriosis suppression. One month later, she presented with acute abdominal pain and distension. Imaging showed bilaterally enlarged, multiloculated ovaries (right: 19 × 15 cm; left: 14 × 13 cm) without ascites or solid areas (Figure 1a). Tumor markers were within normal limits. She reported breast heaviness and headaches. Hormonal evaluation revealed grossly elevated estradiol (5121 pg/mL), follicle-stimulating hormone (FSH) (41.36 mIU/mL), and prolactin (117 ng/mL), with suppressed luteinizing hormone (LH) (0.53 mIU/mL). Ovarian hyperstimulation syndrome (OHSS) was diagnosed, but its etiology remained unclear. Given the elevated FSH, absence of any drug for ovulation stimulation and recent GnRH agonist exposure, an FSH-secreting pituitary adenoma was suspected. Magnetic resonance imaging (MRI) of the brain showed a 2.1 × 1.9 × 1.6 cm sellar-suprasellar mass (Figure 1b,c), suggestive of a pituitary macroadenoma. Cabergoline (0.25 mg/day) was started, but patient reported no symptom relief. She underwent endonasal endoscopic resection of the tumor. Intraoperatively, the tumor appeared necrotic and was easily aspirated. Postoperatively, serum FSH and prolactin levels normalized by day 1 (1.25 and 2.92 mIU/mL, respectively). Histopathology of the tumor tissue revealed tiny, fragmented areas of necrotic tissue. Within 72 h of surgery, the patient was symptomatically better. By day 7, her ovaries had significantly reduced in size (right: 9 × 6.1 cm; left: 4.4 × 3.9 cm), and complete regression was seen within three weeks. One-month postoperative MRI showed no residual tumor. The patient remains asymptomatic at 6-month follow-up. Informed consent from the patient was taken for this publication. Gonadotroph adenomas comprise approximately 40% of all pituitary tumors and are typically clinically silent.1 Less than 1% of Pituitary adenomas secrete biologically active FSH, resulting in symptoms like infertility, menstrual disturbances, spontaneous OHSS, testicular enlargement in men, and iso-sexual puberty in children.2 This case illustrates that, although OHSS is most often an iatrogenic complication of assisted reproduction, a single postoperative GnRH agonist dose may unmask a previously silent FSH-secreting pituitary adenoma. Patients with functioning gonadotroph adenomas are known to exhibit abnormal responses to hormonal stimulation.3 In this case, instead of suppression (as the GnRH agonist was administered in the luteal phase), there was a paradoxical rise in FSH following GnRH agonist administration. The elevated FSH led to ovarian hyperstimulation and subsequent supraphysiological estrogen levels. Hyperprolactinemia was likely caused by “stalk effect,” where tumor compression of the pituitary stalk impairs dopamine delivery from the hypothalamus, leading to elevated prolactin levels.4 GnRH agonist was given in the luteal phase, when a typical flare effect is unlikely. The patient stayed well for about three weeks and only developed abdominal pain and distension around four weeks after surgery. This timeline suggests a delayed presentation due to activation of a previously silent FSH-secreting pituitary adenoma rather than an early flare response. GnRH agonists have been previously reported to cause tumor enlargement via edema, hemorrhage, or vascular changes.5 In this case, both GnRH agonist and cabergoline may have contributed to pituitary apoplexy, as suggested by the necrotic intraoperative findings and absence of viable tumor tissue on histopathology. In this patient, OHSS presented with markedly enlarged ovaries and symptoms related to mass effect, but without ascites or pleural effusion. Interestingly, the FSH:LH ratio was elevated, in contrast to the typical reversed ratio seen in OHSS associated with polycystic ovary syndrome (PCOS). Nonetheless, the significantly enlarged ovaries posed a similar risk of adnexal torsion. Medical therapy is largely ineffective for functioning gonadotroph adenomas.6 Dopamine agonists may reduce prolactin but have a limited impact on tumor size or gonadotropin secretion. GnRH antagonists have no established role. Surgical resection remains the mainstay of treatment.7 Endoscopic transsphenoidal surgery provides excellent tumor visualization, minimal morbidity, and rapid resolution of hormonal and mass effects. Authors concur that centrifugation of the received sample could have enhanced the likelihood of detecting adenomatous cells in the fixed histopathologic sections. In this context, the present case also underscores the importance of utilizing adjunct techniques for histopathologic confirmation in future cases, offering valuable insight for readers of this report. However, dramatic post-surgical normalization of FSH and regression of OHSS supported the diagnosis. This case emphasizes the need for clinicians to suspect pituitary adenomas in women presenting with spontaneous OHSS or after GnRH agonist administration. Early multidisciplinary collaboration between gynecologists, endocrinologists, and neurosurgeons is crucial to avoid delayed diagnosis and potential complications such as ovarian torsion. Gonadotroph adenomas carry a high risk of recurrence,8 making long-term surveillance essential. Recurrences may be managed with repeat surgery, radiotherapy, or adjunctive medical therapies.7 LC, SR, and Pooja were gynecologists responsible for clinical management of the patient. JC and PM were neurosurgeons who operated on the patient. KS was the endocrinologist involved in management. LC and JC were responsible for preparation of manuscript. Pooja & PM provided the relevant images. All authors contributed to and approved the final version of the manuscript. No funding received. Authors report no conflict of interests. Data sharing is not applicable to this article as no new data were created or analyzed in this study.