Adrenal incidentalomas: single centre review of hormonal and imaging results in 125 individuals.

A Theodoraki, B Khoo, S Perera, A Schwappach, A Hamda, M Vanderpump, PM Bouloux

Royal Free Hospital, Hampstead. London NW3 2PF

Background: Adrenal masses discovered incidentally during imaging studies (adrenal incidentalomas- AIs) are common and prompt investigations to exclude secretory lesions and malignancy. Uncertainty exists over the best management strategy of AI.

Objective: To monitor the current practice against the local protocol and existing guidelines; to identify the biochemical and imaging outcomes in a single centre cohort.

Methods: Retrospective review of medical records, biochemistry and imaging of all 125 patients referred to our centre between 2005 and 2009 with an AI.

Results: Out of the patients identified (56% women, mean age 62.2 years) 32% were referred by physicians, 30% by surgeons and 38% by General Practitioners. Most AIs were imaged at least once with CT scanning (91.2%), 82% were unilateral, average diameter on imaging was 20.4mm. Hounsfield Units were measured in 77% of those who had a CT scan done. Dedicated adrenal imaging was done in 72% cases, while in 24% non-dedicated adrenal scans were felt to characterise the lesion adequately.  87% of patients had at least one repeat scan done, with a 17 months mean interval between the first and the last scan. 5.5% of AIs increased ≥5mm in size on follow up, 7.8% decreased ≥5mm, while 86.7% remained unchanged. The increasing adenoma size prompted adrenalectomy in one patient only.

Following our local protocol and including only patients who completed all the investigations (82%), 66% patients were diagnosed with non secretory benign adenomas, 2% primary hyperaldosteronism (one unilateral adenoma, one bilateral adrenal hyperplasia), 2% phaeochromocytoma, 2% adrenal metastasis, 5% other diagnosis. 24 patients (23%) failed to suppress on a low dose or overnight dexamethasone suppression test using a cut-off of 50nmol/L; of those 4 were diagnosed with adrenal Cushing’s, while 5 had false positive results. None of the remaining 15 patients had clinical features of Cushing’s syndrome and all but one had normal 24hr urinary free cortisol excretion.  Finally, 7 patients were found to have at least two abnormal tests of the HPA axis and were diagnosed with mild subclinical cortisol secretion. The use of a higher cut-off for post dexamethasone cortisol (83nmol/L) would exclude 4 of these patients, while the recommended by the NIH consensus cortisol cut-off (138nmol/L) would exclude all these patients but one.

Conclusion: The increase in size in AIs is uncommon. 11% of the AIs in our cohort were functional. Altering the post dexamethasone cut-off in accordance to published guidelines radically changes the performance of the suppression test.