Plasma Steroid Profiles in Subclinical Compared With Overt Adrenal Cushing Syndrome

Jimmy Masjkur; Matthias Gruber; Mirko Peitzsch; Denise Kaden; Guido Di Dalmazi; Martin Bidlingmaier; Stephanie Zopp; Katharina Langton; Julia Fazel; Felix Beuschlein; Stefan Richard Bornstein; Martin Reincke; Graeme Eisenhofer


J Clin Endocrinol Metab. 2019;104(10):4331-4340. 

In This Article

Abstract and Introduction


Context: Diagnosis of subclinical adrenal hypercortisolism is based on several tests of the hypothalamic-pituitary-adrenal axis to establish mild alterations of cortisol secretion and dysregulated cortisol physiology.

Objective: We assessed whether plasma steroid profiles might assist diagnosis of subclinical Cushing syndrome (SC).

Design: Retrospective cross-sectional study.

Setting: Two tertiary medical centers.

Patients: Of 208 patients tested for hypercortisolism, disease was excluded in 152 and confirmed in 21 with overt adrenal Cushing syndrome (AC) compared to 35 with SC. Another 277 age- and sex-matched hypertensive and normotensive volunteers were included for reference.

Main Outcome Measures: A panel of 15 plasma steroids was measured by mass spectrometry, with classification by discriminant analysis.

Results: Patients with SC had lower plasma concentrations of dehydroepiandrosterone and dehydroepiandrosterone-sulfate than subjects without SC (P < 0.05). The largest increases (P < 0.001) in plasma steroids among patients with SC were observed for 11-deoxycortisol and 11-deoxycorticosterone. Nevertheless, concentrations of 11-deoxycorticosterone, 11-deoxycortisol, and pregnenolone in patients with AC were higher (P < 0.05) than in those with SC. Patients with SC or AC could be distinguished from subjects without disease using this combination of steroids as precisely as with use of measurements of serum cortisol after administration of dexamethasone. The steroid combination provided superior diagnostic performance compared with each of the other routine biochemical tests.

Conclusion: Distinct plasma steroid profiles in patients with SC may provide a simple and reliable screening method for establishing the diagnosis.


Up to 50% of patients with adrenocortical adenomas detected incidentally (adrenal incidentalomas) may have hypercortisolism.[1] In most of these cases, the classical clinical features of Cushing syndrome are absent. Such conditions have been described as subclinical Cushing syndrome (SC; and the term used hereafter in this article),[1] subclinical autonomous glucocorticoid hypersecretion, or subclinical hypercortisolism.[2]

Dysregulated cortisol physiology with mild elevations of cortisol secretion is most often recognized during evaluation of adrenal incidentalomas to exclude hypercortisolism.[2,3] Insulin resistance, hypertension, obesity, dyslipidemia, impaired glucose tolerance, and diabetes mellitus are frequently associated with both SC and overt Cushing syndrome, thereby contributing to cardiovascular complications and high mortality risk.[4] Vertebral fractures may also be features of otherwise asymptomatic cortisol excess in osteoporotic patients with SC.[5]

The aforementioned considerations and findings of a 45% prevalence of SC among patients with unilateral adrenal masses >2.5 cm and CT-imaging attenuation <1 Hounsfield unit indicate the potential importance of recognizing SC.[2] Nevertheless, a consensus on the criteria used to diagnose SC has yet to be reached. Clinical, radiological, and hormonal characteristics of this pathological condition all require consideration. Absence of clinical signs and symptoms related to cortisol hypersecretion in the presence of hypercortisolism provides the generally accepted criteria for SC but is weakened by reliance on recognition of clinical clues.

Findings of a serum cortisol level >1.8 μg/dL after the dexamethasone suppression test (DST) combined with ACTH concentrations <10 pg/mL provide one of several criteria for establishing dysregulated cortisol secretion in SC.[6] Elevated 24-hour urinary outputs of free cortisol (UFC) have been reported in some studies but are not reliable alone for diagnosis of either overt Cushing syndrome or SC.[2,7] Measurement of late-night salivary cortisol is one method of demonstrating loss of diurnal rhythm of cortisol secretion in patients with SC; however, with cutoff points of the late-night and midnight salivary free cortisol (SFC) ranging from 0.13 to 0.55 μg/dL,[8] use of this test alone is also unreliable for diagnosis of either AC or SC. Overall, diagnosis of SC remains difficult and requires a positive DST and at least another hypothalamic-pituitary-adrenal axis aberration.[9]

Use of other adrenal steroids apart from cortisol for diagnosis of SC has not been widely investigated. Altered plasma concentrations of several steroids in patients with adrenocortical adenomas and SC, as measured by liquid chromatography–tandem mass spectrometry (LC-MS/MS), have suggested that steroid profiling may be a useful tool for diagnosis of SC.[10] This has been further supported by an LC-MS/MS steroid profiling study establishing a panel of steroids that can be used to identify patients with Cushing syndrome and discriminate those with and without ACTH-dependent subtypes.[11] Other studies have indicated that urinary steroid profiling may be useful for diagnosis of overt clinical Cushing syndrome.[12,13]

Panels of steroids have been used for more than two decades for diagnosing disorders of adrenal steroidogenesis, particularly with gas chromatography–mass spectrometry methods applied to congenital adrenal hyperplasia and related disorders of sexual development.[14,15] Use of steroid profiles in adrenal cortical disorders has more recently gained traction with methods using LC-MS/MS that also take advantage of advances in computational mathematics. Through such advances, there is potential for a paradigm shift from unidimensional to multidimensional diagnostic approaches using basic multivariate discriminant and principal components analyses to more sophisticated machine-learning methods.[11,16,17]

On the basis of these promising leads and advances, we used LC-MS/MS to analyze the plasma steroid profiles of patients with adrenocortical adenomas associated with SC compared with profiles of patients in whom Cushing syndrome was excluded (EX) and patients with overt adrenal Cushing syndrome (AC). The analysis took advantage of well-characterized reference intervals in a series of 525 hypertensive and normotensive volunteers; all relevant data for volunteers are available by open access.[18] The aim was to establish from this panel a selection of steroids that could serve as a single-test alternative to routine tests for discriminating patients with SC from AC and patients without disease.