SUN-363 Severe Hyperandrogenism in Pre-menopausal Women

Abstract

Background: Severe hyperandrogenism in pre-menopausal women is often due to ovarian and adrenal tumors. Clinical and biochemical picture on initial presentation may inform management. Objective: To describe the presentation, etiology and management of pre-menopausal women with severe hyperandrogenism. Methods: This is a retrospective study of premenopausal patients (XX genotype) age 18 to 55 with total testosterone concentrations > 150 ng/dL (normal 8-60) evaluated between 1996 and 2018. Testosterone was measured by mass spectrometry. Patients on exogenous testosterone, polycystic ovarian syndrome, or unconfirmed diagnosis were excluded. Results: Of 318 patients with testosterone >150 ng/dL, 38 (8.4%) met the inclusion criteria. Patients presented with hyperandrogenism at a median age of 32 (range 18-54) and were diagnosed with ovarian hyperandrogenism (ovarian tumors (n=9), and ovarian hyperthecosis (n=1)), congenital adrenal hyperplasia (n=12), adrenal cortical carcinoma (n=8), adrenal cortical adenoma (n=4), and Cushing disease (n=4). Median testosterone on presentation was 261 ng/dL (range 159-1366), without a significant difference between the median total testosterone in ovarian or adrenal tumors (372 [range 181-1366] vs. 290 [range 159-768], p=NS). Patients complaining of severe hirsutism demonstrated higher median testosterone compared to patients with mild hirsutism (477 ng/dL [range 173-1366] vs. 234 [162-800], p=0.04). Women with ovarian tumors and adrenal tumors presented with a median age of 36.5 years (25-54). The median duration of symptoms was not different between adrenal neoplasms and ovarian neoplasms (1 year [range 0.2-35] vs. 1.5 [range 0.25-8], p=NS). Median adrenal tumor size was 9.7 cm (3.3-19.5) and ovarian tumor size of 4.9 cm (1.1-15.5). Work-up included abdominal CT (n=16), pelvic ultrasound (n=15), and abdominal MRI (n=3). In 2 patients with ovarian tumors, ovarian imaging was inconclusive, and adrenal/ovarian vein sampling was performed. Adrenalectomy was performed in all but 1 patient (with metastatic disease) with adrenal tumors. Of 10 patients with ovarian hyperandrogenism, 1 was treated with bilateral adrenalectomy (ovarian hyperthecosis) and 9 were treated with unilateral oophorectomy. Following the surgery, all patients had improvement in their testosterone levels with a 73-98% decrease. Conclusion: Severe endogenous hyperandrogenism due either adrenal or ovarian tumors was demonstrated in 58% of pre-menopausal women. Age, total testosterone concentrations, and duration of symptoms were similar in patients with ovarian and adrenal tumors. Most patients required both adrenal and ovarian imaging to elucidate the etiology. As ovarian tumor may not be identified on imaging, adrenal/ovarian vein sampling could be considered to avoid bilateral oophorectomy.