Subclinical Hypercortisolism (SCH) Presenting with Cyclic Cortisol Secretion: A Newly Identified Condition That May Confound Classification of Adrenal Incidentalomas

Presentation Number: SUN 379
Date of Presentation: April 2nd, 2017

Rafael Buck Giorgi*1, Marcelo Vieira Correa2, Flavia Amanda Costa Barbosa3 and Claudio E. Kater4
1Federal University of São Paulo, Sorocaba, BRAZIL, 2Federal University os São Paulo, Sao Paulo, BRAZIL, 3Universidade Federal de São Paulo - UNIFESP, São Paulo, Brazil, 4Federal University of São Paulo, São Paulo, SP, Brazil

Abstract

Introduction: Adrenal incidentalomas (AI) occur in 5% of the adult population and are a public health problem. Most AI are “non-functioning” adenomas (NFA), but up to 30% may secrete cortisol autonomously without evident clinical manifestations of Cushing’s syndrome (CS), so-called subclinical hypercortisolism (SCH). Identification of this subset is critical since they are associated with increased cardiovascular mortality. The dexamethasone (Dex) suppression test (DST) is largely used for this purpose, but the dose employed, the response cutoff, validity and reproducibility are not yet resolved. Cyclic cortisol secretion occurs in all forms of classical CS: pituitary adenoma, ectopic ACTH syndrome and adrenal disease. Aim: Identify SCH among AI and investigate whether cyclic cortisol production occurs in SCH using sequential DST. Patients and methods: 232 patients with AI (60M/172F, 24-72y), studied from 2006 to 2016, underwent 422 overnight 1mgDST (from 1 to 8 per patient); DST was validated when 8:00h serum Dex, measured in the same cortisol sample, was >140ng/dL. ACTH, DHEAS and 23:00h salivary  cortisol were also measured in most patients. We define SCH when 8:00h post-validated DST serum cortisol levels are >2.5mg/dL plus an elevated 23:00h salivary cortisol (>250ng/dL) and/or suppressed ACTH (<10pg/mL) or DHEAS (<40mg/dL). Results: 41 patients had 3 or more tests done on follow up (median: 4; range: 3 to 8), with a median interval of 6 months between tests (from 3 to 12 mo). Among these, 18 (44%) had all 3+ tests negative (post-DST cortisol values <2.5mg/dL) being classified as NFA; five patients (12%) had all 3+ tests positive (post-DST cortisol >2.5mg/dL) and were classified as sustained SCH. The remaining 18 (44%) had discordant test responses along time: 10/41 (24.4%) had at least two positive and one negative DST, whereas 8/41 (19.5%) had at least two positive and two negative tests. If we define cyclic SCH by this latter criteria, then at least 19.5% of AI may have a cyclic behavior of cortisol secretion, similar to patients with typical CS. We found no statistical differences between patients with NFA, sustained or cyclic SCH, regarding sex or clinical status, but sustained SCH were older. Conclusion: Extended follow up on patients with an AI studied in a single institution, employing repeated overnight DST with consistent methodology to determine serum cortisol and Dex and a cutoff of 2.5mg/dL, have permitted the identification of a new subset of SCH that have cyclic cortisol secretion, even though the question of test reproducibility has not been consistently verified. These data rise the possibility of misclassification of AI as NFA or SCH based on only one or two DST (or other related tests) prompting improper recommendations. Proof for that are recent reports that patients with “NFA” may behave clinically as SCH, with increased cardiovascular morbidity and mortality.

 

Nothing to Disclose: RBG, MVC, FACB, CEK