Accepted Article
Article type
: 14 Clinical Question
Title: Clinical Case Seminar: Post-menopausal ovarian androgen excess: challenges in diagnostic work-up and management of ovarian thecosis
Authors: Yaasir Mamoojee 1, Murali Ganguri 1, Norman Taylor2, Richard Quinton 1,3
1. Department of Endocrinology, Newcastle-upon-Tyne Hospitals, UK
2. Steroid Laboratory, Kings College Hospital, London, UK
3. Institute of Genetic Medicine, University of Newcastle-upon-Tyne, UK
Corresponding author:
Dr Richard Quinton
Department of Endocrinology
Royal Victoria Infirmary,
Newcastle-upon-Tyne, UK
NE1 4LP
Email:
Telephone:
+44 191 282 4635
Fax:
+44 191 282 0129
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/cen.13492
Clinical Case Seminar
Clinical features and presentation:
A 72 year old female was referred to the Department of Endocrinology with a 20-year history of hirsutes, mostly affecting her face and torso, the initial onset of which dated shortly after menopause age 52 years. There was no history of subfertility, menstrual irregularity, or hyperandrogenism during reproductive life to suggest prior polycystic ovary syndrome (PCOS), with three live births, two miscarriages and one still-birth.
Hirsutes progressively increased over a 20-year period, accompanied by a gradual change in her facial features, with the onset of androgenic alopecia and voice-deepening developing over the previous 2-3 years. Past medical history included hypertension, stroke (age 58 years) and type 2 diabetes (age 61 years). Examination revealed early hyperandrogenic alopecia, with partly masculinised facial features and early clitoromegaly, albeit no major virilisation or cushingoid features. Weight was 78.7kg, height 1.6m (body mass index -BMI : 28) and blood pressure 125/66 mmHg. Medication comprised sulphonylurea (gliclazide), angiotensin receptor blocker (candesartan), calcium channel blocker (amlodipine) and cholesterol absorption blocker (ezetimibe). She was intolerant of all forms of Metformin, even at low dose modified-release preparation, and low dose statin (Atorvastatin 10 mg OD).
Investigation:
Biochemical hyperandrogenism was confirmed by serum testosterone (T) concentration 4.2 nmol/L (reference range <3.2 nmol/L) and calculated free T 94 pmol/L (reference range<25 pmol/L), with normal levels of 17-hydroxyprogesterone (17-OHP), adrenocorticotropic hormone (ACTH) and dehydroepiandrosterone sulphate (DHEAS). Urine steroid metabolomic profile did not indicate an abnormal adrenal source of androgens. Other laboratory parameters are listed in Table 1. Pelvic sonography revealed a slightly enlarged right ovary (29mm) with normal echotexture and an irregular, cystic and 6mm slightly thickened endometrium (normal post-menopause <5mm), with multiple uterine fibroids.
Management:
Although the presence of virilising features is conventionally taken to indicate risk of tumorous hyperandrogenism, these features had taken nearly 20 years to develop after an unusually long delay in seeking medical advice. Thus, a benign ovarian cause of androgen
Accepted Article
excess - postmenopausal ovarian thecosis (OH) was strongly suspected, particularly given the associated metabolic phenotype. Dexamethasone testing was not performed due to non-suppressed levels of ACTH and normal DHEAS (which in conjunction with absence of cushingoid features effectively rules out an adrenal source of androgen excess - see section below on ‘ACTH-suppression test with low-dose Dexamethasone), but gonadotropin- releasing hormone (GnRH) suppression test (3.75mg intramuscular Leuprorelin acetate®) demonstrated complete suppression of serum LH & FSH and Testosterone levels at 8 weeks, confirming ovarian source of T (Table 2). A diagnosis of OH thus remained probable, although a benign virilising ovarian tumour (VOT) could not be definitively excluded.
After informed discussion, the patient initially elected for medical management with an oral anti-androgen (spironolactone), which maintained serum T within the normal post- menopausal range, but this was not tolerated long-term due to persistent nausea. She was thus referred for bilateral salpingo-oophorectomies. Histology confirmed bilateral OH, with frequent small clusters of thecal cell proliferation within the stroma. Post-operatively her hormonal profile remained normal (Table 2) and hirsutes has progressively improved.
Diagnostic approach to severe post-menopausal hyperandrogenism
The initial evaluation
Post-menopausal hyperandrogenism is rare, with the key initial observation being whether or not virilising features are present (androgenic alopecia, severe hirsutism, deepening of the voice, squaring of the jawline/muscular hypertrophy, clitoromegaly) and, if present, how rapidly these have developed1,2. Serum testosterone (T) concentrations are almost invariably >4 nmol/L (115 ng/dL), but levels >5 nmol/L raise particular concerns about tumorous hyperandrogenism3, especially if accompanied by erythrocytosis4.
If tumorous hyperandrogenism is suspected, based upon the initial presentation and findings, it is crucial to rapidly identify/exclude an obvious adrenal or ovarian neoplasm with high-resolution cross-sectional imaging - initially bypassing much of the investigational algorithm outlined in Figure 1. Otherwise, the next key differential lies between adrenal and ovarian sources of androgens, with adrenal hyperandrogenism being the more readily characterised, both biochemically and radiologically. The relative frequency of various etiologies in post-menopausal hyperandrogenism has not been well researched. This is not surprising given the rarity of the condition and the influence of referral pathways.
Accepted Article
Nevertheless a tertiary referral centre in France found the following etiologies in 22 consecutive patients referred with post-menopausal hyperandrogenism over a 14-year period: ovarian tumours (8), adrenal tumours (2) and non-tumourous cause (12 - of which spontaneous resolution (4), normal ovarian stroma after oophorectomies (3), ovarian cyst with normal stroma (2), Cushing’s disease (1), pseudo-Cushing’s syndrome (1), ovarian hyperthecosis (1))2.
What are the defining characteristics of adrenal hyperandrogenism?
Biochemistry:
The majority of androgen-secreting adrenocortical tumours (ACT) are characterised by concurrently raised levels of dehydroepiandrosterone (DHEA), or its sulphated form DHEAS. In contrast, an isolated rise in serum T with normal or suppressed DHEA/S is commonly seen in primary adrenal Cushing’s syndrome, due to glucocorticoid-induced ovarian hyperandrogenism (stimulation of insulin-mediated ovarian T secretion by excess glucocorticoid) and suppression of adrenocorticotropic hormone (ACTH) secretion8. Autonomous adrenal co-secretion of cortisol and T indicates high risk of adrenocortical carcinoma (ACC)5,6,7.
In adrenal Cushing’s baseline serum ACTH levels are expected to be suppressed (mostly undetectable) and dynamic testing with low-dose dexamethasone test (LDDST) reveals unsuppressed serum cortisol (normal <50 nmol/L (1.8 ug/dL) post-dexamethasone administration)5. Concurrent measurement of serum T during LDDST is also a useful marker in diagnosing extremely late presentations of non-classical congenital adrenal hyperplasia (NCCAH). In congenital adrenal hyperplasia (CAH) serum T production is primarily driven by ACTH, hence suppression of ACTH with dexamethasone (LDDST) leads to a simultaneous decline serum T. Conversely non-suppression of serum T with LDDST excludes ACTH- responsive hyperandrogenism10, as might be observed in NCCAH. Although the most common form of CAH (due to a deficiency in 21-hydroxylase enzyme responsible for catalytic conversion of 17-OHP to 11-deoxycortisol in the cortisol synthesis pathway) is classically diagnosed through elevated serum 17-hydroxyprogesterone (>31 nmol/L) with ACTH stimulation11, urine steroid metabolomics can identify all forms of CAH and, potentially, also contribute to cancer risk stratification in adrenal hyperandrogenism (Figure 1)12.
Accepted Article
Imaging:
Cross-sectional imaging with computed tomography (CT) or magnetic resonance imaging (MRI) can reliably identify and risk-stratify adrenal tumours13,14. Nevertheless, given that the prevalence of non-functioning adrenal lesions increases with age15, clinicians should be alert to the possibility of non-functioning adrenal “coincidentaloma” in post-menopausal females with isolated serum T elevation and negative LDDST result16,17. On the other hand, any adrenal lesion with radiological features of concern should be considered for surgery anyway, even if not suspected to be the principal significant source of excess T secretion.
What are the causes and characteristics of post-menopausal ovarian hyperandrogenism?
The possibility of exogenous androgen use should always be explored. This might be intentional (e.g. testosterone prescribed ‘off-label’ for libido-enhancement as part of post- menopausal HRT), or inadvertent, through use of “body-sculpting” health supplements, marketed as herbal- or protein-based, but in fact containing androgen50, or even from transdermal absorption of their male partner’s T gel51.
However, the two most important causes of ovarian androgen excess in post-menopausal women are ovarian hyperthecosis (OH) and virilising ovarian tumours (VOTs). A minority of women with polycystic ovarian syndrome (PCOS), whether or not diagnosed as such pre- menopausally, continue to experience persistent hyperandrogenism after the climacteric age18. Eliciting a history of post-pubertal androgen excess symptoms (hirsutism, acne), menstrual irregularity, subfertility, weight gain and/or metabolic syndrome may point to undiagnosed pre-menopausal PCOS19, although diagnostic pelvic ultrasonographic criteria may be inapplicable after the menopause. However, many women with OH were unaffected with hyperandrogenism during reproductive life.
Whereas PCOS is a largely clinical diagnosis, albeit characterised by excess T production from luteinised ovarian follicle theca cells20, the diagnosis of OH is fundamentally histological, based on evidence of luteinised theca cells within the ovarian stroma and absence of neoplasm21. The pathophysiology of OH is also less well characterised, but is best explained as a manifestation of the “two cell hypothesis”, wherein persistent T secretion by ovarian thecal cells is unmasked post-menopausally through the loss of granulosa cell-mediated aromatisation of T to estradiol. This appears to be underpinned by synergy between elevated serum levels of insulin and LH, similar to that observed in PCOS22.
Accepted Article
As in PCOS, hyperinsulinaemia reduces hepatic secretion of sex hormone binding globulin (SHBG), thereby increasing bioavailability of androgens. Indeed, an often unappreciated characteristic of women with OH is that they typically exhibit features of the metabolic syndrome and are thus potentially at high risk of cardio-metabolic sequelae23,24,25. To our knowledge co-existence of OH and PCOS based on histological evidence has not been previously documented. Equally there are no reports of PCOS progressing to OH.
VOTs are predominantly benign sex-cord stromal tumours and include Leydig cell, Sertoli cell and steroid cell neoplasms, and ovarian thecomas26. They vary significantly in size and display an array of radiological characteristics on cross-sectional imaging, with the smallest being under the detection thresholds of current technologies.
Can OH and VOTs be reliably identified and differentiated, based on clinical history, biochemical assessment or imaging studies?
Once an adrenal source of T excess has been ruled-out, differentiating between OH and VOTs may be challenging. The initial plausible notion that patients with VOTs present with rapid virilisation has not been substantiated so far2,27,28. In a case control series, there was no significant difference in duration from onset of symptoms to diagnosis, ranging from 4 to 126 months, when comparing patients with OH and VOTs28. The only discriminatory virilising signs between the 2 groups were deepening of the voice and muscular hypertrophy, ascertainment of which in clinical practice relies heavily on patients’ subjective assessment and, possibly, comparison of serial photographs.
Previous studies have noted that patients with VOTs have higher serum T and lower gonadotropin (follicle stimulating hormone, FSH, and luteinising hormone, LH) levels as compared to patients with OH, but with a great deal of overlap in serum T and gonadotropin concentrations between the 2 groups. Indeed, even a high serum T cut-off value >10.9 nmol/L (315.5ng/dL) confers a sensitivity and specificity of only 77% and 91%, respectively in identifying patient with VOTs28. The cut-off value for serum FSH level has been variably studied, with one study reporting a cut-off value of ≤35IU/L (sensitivity 90%, specificity 92%) and another reporting a value of <22.3 IUL (sensitivity 77%, specificity 86%) in discriminating between ovarian tumour and non-tumour sources of post-menopausal androgen excess2,28.
Accepted Article
Most VOTs are unilateral, but bilateral tumours have also been reported29,41, and small VOTs may evade detection with current imaging modalities30. Sonographic measurements of ovarian volumes and symmetry can indicate ovarian pathology in the post-menopausal state, with volumes ≥4.0 cm3 considered pathological and ovarian asymmetry ≥ 2 between sides suggestive of an ovarian tumour31,32. However, sonographically normal ovaries in patients with biochemical evidence of ovarian androgen excess may still harbour VOTs or OH33; even MRI may still miss VOTs28. To confound matters further, VOTs are frequently associated histologically with bilateral OH28. Therefore, presence of nodules or cysts on ovarian imaging does not confirm VOTs, absence of such lesions does not rule out VOTs, unilateral ovarian enlargement may indicate either OH or a VOT, and bilateral ovarian enlargement may indicate OH, with or without VOTs.
Therefore in clinical practice there remains no reliable clinical or diagnostic parameter to confidently differentiate between OH and VOTs. Although OH seems to be strongly associated with insulin-resistance and metabolic syndrome, the latter is sufficiently prevalent to also occur serendipitously with VOTs. Initial reports of PET-CT imaging appear promising, but it is currently not cost-effective 28,34.
How useful is dynamic endocrine testing in differentiating the different causes of post- menopausal androgen excess?
GnRH-analog suppression test:
Based on the observation that adrenal androgen secretion is not LH-responsive, the gonadotropin-suppression test, using long-acting GnRH-analogue injection, can reliably discriminate between adrenal and ovarian source of T excess. Suppression of serum T level by >50% indicates an ovarian source and, particularly in the context of co-existing adrenal incidentaloma, can usefully direct further diagnostic work-up and clinical management17. Failure to suppress serum T level to >50% is highly suggestive of an adrenal source and the subsequent work up should be realigned accordingly. However, the test is unhelpful in differentiating between the two key causes of post-menopausal ovarian androgen excess, OH and VOTs, both of which exhibit gonadotrotopin-dependent T secretion, indicating that VOTs are only semi-autonomous and retain gonadotropin receptor24,35,36,37.
Accepted Article
ACTH-suppression test with low-dose Dexamethasone:
Kaltsas et al (2003)10 found lack of serum T suppression (normalisation or >40% reduction) during 2-day LDDST to be associated with 100% sensitivity and 88% specificity in distinguishing patients with ovarian and adrenal androgen-secreting tumors from those with non-tumorous hyperandrogenism. However, Vaikkakara et al (2008)24 found frequent false positives using LDDST in an insulin-resistant group of women with OH, presumably because glucocorticoid-induced stimulation of insulin-mediated ovarian T secretion counterbalanced any corresponding reduction in ACTH-mediated T secretion. Hence we would not recommend LDDST as a first line dynamic test in assessing post-menopausal hyperandrogenism. Rather, the usefulness of this test can be considered once the results the GnRH-analog suppression test and cross-sectional imaging are available (see flow chart 1).
Can the effect of ameliorating hyperinsulinism differentiate between OH and VOTs?
The triad of post-menopausal ovarian androgen excess, elevated gonadatropins and metabolic syndrome (including hyperinsulinism) has been previously well documented23,24,25. Although hyperinsulinism has been shown to independently stimulate ovarian growth and ovarian androgen production in a subset of pre-menopausal women with severe insulin resistance syndromes35, a co-stimulatory role of gonadotropins in driving ovarian androgen excess in the presence of hyperinsulinaemia has also been described39. Manipulating insulin resistance with rigorous lifestyle changes, weight loss and insulin- sensitisers (e.g. metformin) can restore serum T level to normal post-menopausal range in patients presenting with metabolic syndrome and non-tumourous ovarian hyperandrogenism24,25. The effect of weight loss at reducing hyperinsulinism has been well described40.
Thus in post-menopausal patients with ovarian hyperandrogenism (evidenced by normal adrenal workup and/or suppression of serum T with GnRH-analogue) and features of the metabolic syndrome (overweight, hypertension, dyslipidemia, type 2 diabetes or pre- diabetic glycaemic marker) a rigorous diagnostic trial of weight loss combined with metformin therapy (at maximal tolerated dosage) may be employed. Normalisation of serum T concentration (or reduction by >40%) is likely to indicate OH rather than VOTs.
Accepted Article
Further studies are, however, needed to evaluate the specificity of this proposed test, including characterisation of VOT tissue for expression (or absence) of insulin receptors.
Is there role for steroid metabolomic studies in post-menopausal androgen excess?
Mass-spectrometry-based steroid profiling, typically of “spot” or 24-hour urine samples, has been successfully employed to discriminate between benign and malignant adrenal tumours12. It can also reliably identify congenital defects in adrenal steroidogenesis and use or abuse of exogenous androgens. Steroid metabolomic profiling measures concentrations of steroidogenic hormones, their precursors and metabolites. Even apparently non- secretory adrenocortical cancers exhibit a pattern of predominantly immature, early-stage steroidogenesis, with high sensitivity and specificity. In contrast, this tool has not been validated in androgen excess of ovarian origin. Nevertheless, future predicted reductions in cost and processing time will make steroid metabolomics an attractive and cost-effective diagnostic test as opposed to inpatient dynamic testing.
Can selective ovarian venous sampling aid diagnosis?
Ovarian venous sampling has been trialled in specialised institutions, but success rates at identifying VOTs have been suboptimal, with one study reporting a detection rate of only 66%41,42. Technical difficulties in catheterising the appropriate veins and the frequency of confounders, such as coexisting bilateral OH with unilateral VOT, unilateral OH and bilateral VOTs, makes the formulation of diagnostic algorithms and interpretation of results challenging. Moreover, differentiation of ovarian from adrenal hyperandrogenism can be done far more simply and less-invasively as previously described. Hence we do not recommend this investigative strategy in routine clinical practice.
Are there non-surgical options for post-menopausal ovarian hyperandrogenism?
Whether for OH or VOTs, bilateral laparoscopic salpingo-oophorectomy remains the treatment of choice, being a well-tolerated procedure with low morbidity and mortality43 that simultaneously offers both definitive treatment and “gold standard” confirmatory
Accepted Article
histological diagnosis. It does not, however, address the long-term morbidity and mortality in OH that is associated with hyperinsulinaemia and metabolic syndrome; lifestyle-related and pharmacologic risk-reduction measures need to be separately undertaken post- operatively.
Nevertheless for patients who are not yet ready to consider surgery, or are not fit for it, there are a number of medical management options, either as interim “bridging” measures pending eventual surgery, or as a long-term alternative to it.
As discussed, both OH and VOTs appear to be gonadotropin-dependent; hence successful suppression of gonadotropins with long-acting GnRH analogues is an attractive long-term medical management strategy that is successful in both VOTs and OH24,36 and also in transgender women in the UK44. However, these remain expensive drugs, with an additional cost associated with quarterly nurse-administration, and long-term follow-up studies are lacking. How this strategy impacts (positively, negatively, or neutral) on tumour progression in patients with known VOTs, or whether VOTs develop in patients with presumed OH, is uncertain.
Anti-androgen therapy, as either cyproterone acetate or spironolactone, may also be used to lower serum T level and improve on the clinical manifestations of hyperandrogenism. Cyproterone acetate is a progesterone derivative and is metabolised in the liver. Its use is associated with hepatic dysfunction and depression and there have also been reports of its association with meningiomas, possibly likely to a progesterone-related growth effect45,46,47.
Spironolactone is a mineralocorticoid receptor antagonist that acts as an androgen receptor partial antagonist as well as an estrogen receptor agonist. It is a potassium sparing diuretic and can cause renal dysfunction and hyperkalaemia as well as liver dysfunction.
Anti-androgen agents are all competitive antagonists and as such, favourable outcomes are heavily dependent on the degree of blockage of the androgen receptors against the stimulatory effect of severe hyperandrogenaemia. The degree to T normalisation with these agents is thus likely to be inversely proportional to the pre-therapy serum concentrations and hence variable clinical results may be expected.
Accepted Article
Other potential agents that have not yet been trialled for this indication include dopamine antagonists (through induction of hyperprolactinaemia) and neurokinin 3 receptor antagonists 48,49 .
Conclusions
Laparoscopic bilateral oophorectomy is the preferred procedure for postmenopausal hyperandrogenism overall, because it offers high likelihood of both cure and confirmatory diagnostics. However, there is a role for medical therapy as an alternative, or interim measure. OH is strongly associated with metabolic syndrome and Type 2 diabetes or pre- diabetes, so if these features are present, we favour addressing insulin resistance through lifestyle management in combination with metformin therapy, both for diagnostic and therapeutic purposes, even when surgical management is likely to be the final outcome. Improving insulin sensitivity has the potential to not only normalise serum T (thereby potentially distinguishing between OH and VOTs), but also addresses the cardio-metabolic parameters in this high risk group of patients - hopefully impacting on their long-term cardiovascular outcomes.
References
1. Martin, K.A., Chang, R.J., Ehrmann, D.A., et al. (2008) Evaluation and treatment of hirsutism in premenopausal women: an endocrine society clinical practice guideline. Journal of Clinical Endocrinology and Metabolism 93, 1105-1120.
2. Sarfati, J., Bachelot, A., Coussieu, C. et al. (2011) Impact of clinical, hormonal, radiological, and immunohistochemical studies on the diagnosis of postmenopausal hyperandrogenism. European Journal of Endocrinology 165 :779-788.
3. O’Driscoll, J.B., Mamtora, H., Higginson, J. et al. (1994) A prospective study of the prevalence of clear-cut endocrine disorders and polycystic ovaries in 350 patients presenting with hirsutism or androgenic alopecia. Clin Endocrinol (Oxf) 41:231.
4. Kozan, P., Chalasani, S., Handelsman, D.J. et al. (2014) A Leydig Cell Tumor of the Ovary Resulting in Extreme Hyperandrogenism, Erythrocytosis, and Recurrent Pulmonary Embolism. J Clin Endocrinol Metab 99 (1): 12-17.
5. Nieman, L.K. (2010) Approach to the patient with an adrenal incidentaloma. Journal of Clinical Endocrinology and Metabolism 95, 4106-4113.
Accepted Article
Accepted Article
6. Cordera, F., Grant, C., van Heerden, J. et al. (2003) Androgen-secreting adrenal tumors. Surgery 134, 874- 880.
7. Moreno, S., Montoya, G., Armstrong, J. et al. (2004) Profile and outcome of pure androgen-secreting adrenal tumors in women: experience of 21 cases. Surgery 136(6),1192.
8. Kamenicky, P., Houdoin, L., Ferlicot, S. et al.(2007) Benign cortisol-secreting adrenocortical adenomas produce small amounts of androgens. Clin Endocrinol (Oxf)66(6),778.
9. Nieman, L.K., Biller, B.M., Findling , J.W. et al. (2008) The diagnosis of Cushing’s syndrome: an Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab 93:1526.
10. Kaltsas, G.A., Isidori, A.M., Kola, B.P. et al. (2003) The value of the low-dose dexamethasone suppression test in the differential diagnosis of hyperandrogenism in women. J Clin Endocrinol Metab. 88, 2634-43.
11. Speiser, P.W., Azziz, R., Baskin, L.S., et al. (2010) Congenital Adrenal Hyperplasia Due to Steroid 21- Hydroxylase Deficiency: An Endocrine Society Clinical Practice Guideline .J Clin Endocrinol Metab 95 (9): 4133-4160.
12. Arlt, W., Biehl, M., Taylor, A.E. et al. (2011) Urine steroid metabolomics as a biomarker tool for detecting malignancy in adrenal tumors. J Clin Endocrinol Metab. 96(12):3775-84.
13. Hussain, S., Belledregun, A., Seltzer, S.E. et al. (1985) Differentiation of malignant from benign adrenal masses: predictive indices on computed tomography. Am J Roentgenol 144:61-65.
14. Hennings, J., Hellman, P., Ahlstrom, H. et al. (2009) Computed tomography, magnetic resonance imaging and 11C-metomidate positron emission tomography for evaluation of adrenal incidentalomas. Eur J Radiol. 69:314-323.
15. Young, W.F. (2007) Jr. Clinical practice. The incidentally discovered adrenal mass. N Engl J Med 356:601.
16. Alpañés, M., González-Casbas, J.M., Sánchez, J., et al. (2012) Management of postmenopausal virilization. J Clin Endocrinol Metab 97(8):2584.
17. Marcondes, J.A.M., Curi, D.D.G., Matsuzaki, C.N., et al. (2008) Ovarian hyperthecosis in the context of an adrenal incidentaloma in a postmenopausal woman. Arq Bras Endocrinol Metab 52(7):1184-8.
18. Winters, S.J., Talbott, E., Guzick, D.S. et al. (2000) Serum testosterone levels decrease in middle age in women with the polycystic ovary syndrome. Fertility and Sterility 73, 724-729.
19. Rotterdam ESHRE/ASRM-Sponsored PCOS consensus workshop group. (2004) Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS). Human Reproduction 19, 41-47.
20. Strauss, J.F. (2003). Some new thoughts on the pathophysiology and genetics of polycystic ovary syndrome. Ann. N. Y. Acad. Sci. 997: 42-8.
Accepted Article
21. Scully R. Tumors of the Ovary, Maldeveloped Gonads, Fallopian Tube and Broad Ligament. In Atlast of Tumor Pathology. Edn 3, pp 413-415, Ed H WH. Washington DC: Armed Forces Institute of Pathology 1996.
22. Nagamani M., Osuampke C., Kelver M.E. (1999) Increased bioactive luteinizing hormone levels and bio/immuno ratio in women with hyperthecosis of the ovaries: possible role of hyperinsulinemia. J Clin Endocrinol Metab 84(5):1685-1689.
23. Barth, .J.H, Jenkins, .M, Belchetz , P.E. (1997) Ovarian hyperthecosis, diabetes and hirsuties in post- menopausal women. Clin Endocrinol (Oxf) 46:129-130.
24. Vaikkakara, S., Al-Ozairi, E., Lim, E. et al. (2008) The investigation and management of severe hyperandrogenism pre- and postmenopause: non-tumor disease is strongly associated with metabolic syndrome and typically responds to insulin-sensitization with metformin. Gynecol Endocrinol 24(2):87-92.
25. Al-Ozairi, E., Michael, E., Quinton, R. (2008) Insulin resistance causing severe postmenopausal hyperandrogenism. Int J Gynaecol Obstet. 100(3):280-1.
26. Morgan, L.S. (1990) Hormonally active gynaecologic tumors. Seminars in Surgical Oncology 6, 83-90.
27. Markopoulos, M.C., Kassi, E., Alexandraki, K.I. et al. (2014) Management of endocrine disease: hyperandrogenism after menopause. European Journal of Endocrinology 172, R79-R91.
28. Yance, V.R.V., Marcondes, J.A.M., Rocha, M.P. et al. (2017) Discriminating between virilizing ovary tumors and ovary hyperthecosis in postmenopausal women: clinical data, hormonal profiles and image studies. Eur J Endocrinol. 177(1):93-102.
29. Cserepes, E., Szucs, N., Patkos, P. et al. (2002) Ovarian steroid cell tumor and a contralateral ovarian thecoma in a postmenopausal woman with severe hyperandrogenism. Gynecol Endocrinol.16: 213-216.
30. Horta, M. & Cunha, T. M. (2015) Sex cord-stromal tumors of the ovary: a comprehensive review and update for radiologists. Diagn Interv Radiol. 21(4): 277-286.
31. Merz, E., Miric-Tesanic, D., Bahlmann, F. et al. (1996) Sonographic size of uterus and ovaries in pre- and postmenopausal women. Ultrasound Obstetrics & Gynecology. 7: 38-42.
32. Cohen, H.L., Tice, H.M. & Mandel, F.S. (1990) Ovarian volumes measured by US: bigger than we think. Radiology. 177 :189-192.
33. Brown, D.L., Henrichsen, T.L., Clayton, A.C. et al. (2009) Ovarian stromal hyperthecosis: sonographic features and histologic associations. J Ultrasound Med. 28(5):587-93.
34. Mattsson, C., Stanhope, C.R., Sam, S. et al. (2006) Testosterone-Secreting Ovarian Tumor Localized with (Fluorine-18)-2-Deoxyglucose Positron Emission Tomography. J Clin Endocrinol Metab. 91 (3): 738-739.
35. Marcondes, J.A., Nery, M., Mendonca, B.B. et al.(1997) A virilizing Leydig cell tumor of the ovary associated with stromal hyperplasia under gonadotropin control. Journal of endocrinological investigation 20: 685-689.
Accepted Article
36. Pascale, M.M., Pugeat, M., Roberts, M. et al. (1994) Androgen suppressive effect of GnRH agonist in ovarian hyperthecosis and virilizing tumours. Clinical Endocrinology 41: 571-576.
37. Gomes, L.G., Yance, V.R.V., Rocha, M.P. et al. Agnrh Suppression Test Is Not a Useful Tool in the Differential Diagnosis of Hyperandrogenemia of Ovarian Origin. Poster presented at: ID: SUN-141: Presented at the Endocrine Society’s 98th Annual Meeting and Expo; 2016 April 1-4, Boston, USA.
38. Musso, C., Shawker, T., Cochran, E. et al. (2005) Clinical evidence that hyperinsulinaemia independent of gonadotropins stimulates ovarian growth. Clin Endocrinol (Oxf) 63(1):73-8.
39. Brown, R.J., Joseph, J., Cochran, E. et al. (2017) Type B Insulin Resistance Masquerading as Ovarian Hyperthecosis. J Clin Endocrinol Metab.102 (6): 1789-1791.
40. Al-Mrabeh, A., Hollingsworth, K.G., Steven, S. et al. (2016) Morphology of the pancreas in type 2 diabetes: effect of weight loss with or without normalisation of insulin secretory capacity, Diabetologia 59:1753- 1759.
41. Levens, E.D., Whitcomb, B.W., Csokmay, J.M. et al. (2008) Selective venous sampling for androgen- producing ovarian pathology. Clin Endocrinol (Oxf) 70(4): 606-614.
42. Kaltsas, G.A., Mukherjee, J.J., Kola, B. et al. (2003) Is ovarian and adrenal venous catheterization and sampling helpful in the investigation of hyperandrogenic women? Clin Endocrinol (Oxf) 59(1):34-43.
43. Asante, A., Whiteman, M.K., Kulkarni, A. et al. (2010) Elective oophorectomy in the United States: trends and in-hospital complications, 1998-2006. Obstet Gynecol. 2010 Nov. 116(5):1088-95.
44. Mamoojee, Y., Seal, L.J., Quinton, R. (2017) Transgender hormone therapy: understanding international variation in practice. Lancet Diabetes Endocrinol. 5(4):243-246.
45. Frey, H, & Aakvaag, A. (1981) The treatment of essential hirsutism in women with cyproterone acetate and ethinyl estradiol. Clinical and endocrine effects in 10 cases. Acta Obstet Gynecol Scand 60: 295-300.
46. Willemse, P.H., Dikkeschei, L.D., Mulder, N.H. et al. (1988) Clinical and endocrine effects of cyproterone acetate in postmenopausal patients with advanced breast cancer. Eur J Cancer Clin Oncol. 24: 417-421.
47. Gil, M., Oliva, B., Timoner, J. et al. (2011) Risk of meningioma among users of high doses of cyproterone acetate as compared with the general population: Evidence from a population-based cohort study. Br J Clin Pharmacol. 72(6): 965-968.
48. George, J.T., Kakkar, R., Marshall, J. et al. (2016) Neurokinin B Receptor Antagonism in Women With Polycystic Ovary Syndrome: A Randomized, Placebo-Controlled Trial. J Clin Endocrinol Metab 101: 4313- 4321.
49. Fraser, G.L., Ramael, S., Hoveyda, H.R. et al. (2016) The NK3 Receptor Antagonist ESN364 Suppresses Sex Hormones in Men and Women. J Clin Endocrinol Metab; 101: 417-426.
ccented Article
50. Abbate, V., Kicman, A.T., Evans-Brown, M., Mcveigh, J., Cowan, D.A., Wilson, C., Coles, S.J., Walker, C.J. (2014) Anabolic steroids detected in bodybuilding dietary supplements - a significant risk to public health. Drug Testing & Analysis. DOI 10.1002/dta.1728.
51. de Ronde W. (2009) Hyperandrogenism after transfer of topical testosterone gel: case report and review of published and unpublished studies. Human Reproduction; 24: 425-428.
| Serum parameter | Result | Reference range |
|---|---|---|
| Hb g/L | 133 | 115-165 |
| Hct L/L | 0.40 | 0.36-0.46 |
| Na mmol/L | 145 | 133-146 |
| K mmol/L | 5.0 | 3.5-5.3 |
| Urea mmol/L | 5.7 | 2.5-7.8 |
| Cr pmol/L | 95 | 65-120 |
| Albumin g/L | 40 | 35-50 |
| Bilirubin umol/L | 6 | <21 |
| ALP U/L | 94 | 30-150 |
| ALT U/L | 18 | <140 |
| SHBG nmol/L | 57 | 14-110 |
| Cholesterol mmol/L | 4.6 | - |
| TGL mmol/L | 1.8 | <1.7 |
Accepted. Article
| HDL mmol/L | 1.1 | 1.2-1.8 |
| TSH mU/L | 0.6 | 0.3-4.7 |
| LH IU/L | 33.8 | - |
| FSH IU/L | 61 | - |
| ACTH ng/L | 22 | <48 |
| DHEAS umol/L | 0.9 | 0.3-6.7 |
| E2 pmol/L | <60 | - |
| 17-OHP nmol/L | 2.3 | - |
Accepted Article
| Serum parameter | At Presentation | Post GnRH analogue | Pre-surgical | Post-surgical |
|---|---|---|---|---|
| LH IU/L | 33.8 | 5.9 | 36.6 | 27.7 |
| FSH IU/L | 61.1 | 23.8 | 52.8 | 45.5 |
| T nmol/L | 4.2 | <1 | 3 | <1 |
Accepted Article
Post-menopausal female with serum T >4nmol/L
. Normal serum T by mass spectrometry . Calculated free T normal
Rapid onset virilisation, Cushingoid
. evidence of exogenous androgen
Check FSH, LH, DHEA/S
No further work up.
I LH+FSH
<>/1 LH+FSH
1 DHEA/S
<>/J DHEA/S
Cross-sectional abdominal and pelvic imaging (USS/CT/MRI)
T not suppressed
GnRH-analogue suppression test
If adrenal lesion: . ACTH, LDDST & T
T suppressed >50% or into normal range
. Urine steroid profile
Adrenalectomy
. Adrenal cancer or benign androgen excess
Ovarian source of androgen excess
. Adrenal Cushing’s
Figure 1: Diagnostic approach in post-menopausal females with androgen excess. T: testosterone, FSH: follicle stimulating hormone, LH: luteinising hormone, DHEA/S: dehydroepiandosterone or dehydroepiandosterone sulphate, CT: computed tomography, MRI: magnetic resonance imaging, USS: ultrasound sonography, LDDST: low dose dexamethasone suppression test, GnRH: gonadotropin-releasing hormone.