SAT-365 Prevalence and Clinical Features of Adrenal Incidentalomas: A Prospective Study in an Unselected Population

Abstract

Context. The widespread use of cross-sectional imaging in medical practice has increased detection of adrenal tumors, which are most frequently unexpected prior to their discovery (adrenal incidentalomas). Most adrenal incidentalomas (AI) are benign and frequently cause low-grade cortisol excess that may portend adverse clinical consequences. However, this association is subject to confounding by indication since more diagnostic tests are done in diseased patients. Objective. To assess prospectively the prevalence of AI in an unselected adult population performing abdominal CT in our catchment area, and to evaluate the demographic, laboratory and clinical features of AI patients. Design and setting. Monocentric, prospective study. Patients. A consecutive series of 601 unselected patients undergoing abdominal CT as part of their management. Patients with any known adrenal disorder or malignancy were excluded. Results. Of the 601 patients, aged 63.5 ± 14.4 years, 55.4% had hypertension, 15.1% diabetes, 16.8% dyslipidemia and 16% previous CV events. These characteristics are overall comparable to the background general population. We found serendipitously 44 adrenal tumors (7.3%), 43 with radiological features of benign cortical adenoma (density <10 HU) and 1 with indeterminate features, which was a pheochromocytoma. Mass size ranged between 1-5 cm (median 2 cm) and tumors were bilateral in 29.5%. Patients with AI compared to patients without were more frequently suffering from diabetes (31.8% vs 14.2%; p=0.004), showing higher BMI (27.6 ± 6.2 vs. 25.6 ± 4.8; p=0.009) and larger waist (101.2 ± 13.9 cm vs. 95.3 ± 13.9 cm; p=0.007). No differences were found in frequency of hypertension, dyslipidemia or history of CV events. Forty patients underwent a 1-mg overnight dexamethasone test (DST) and 20 (50%) did not suppress cortisol below 1.8 μg/dL (4 had post-DST cortisol >5.0 μg/dL). Post-DST cortisol levels were not different between diabetic and non-diabetic AI patients (3.5 ± 4.0 µg/dL vs. 2.4 ± 2.5 µg/dL). In multivariate analysis, diabetes was significantly associated with AI (p=0.003). Conclusions. We confirm in a prospective study the high prevalence of previously unsuspected adrenal tumors, most of which are benign adenomas. Moreover, we found a significant association between adrenal incidentalomas and diabetes. Since our patients were recruited at the radiology department, this finding is not confounded by selection bias.