SUN-355 Computed Tomography Appearance of Adrenal Glands in Non-Adrenal Causes of Cushing's Syndrome

Abstract

BACKGROUND: Incidental adrenal masses are noted on computed tomography (CT) imaging in 1-5% of the general population, with higher rates in older people. While most are benign, they may be hormonally functional and/or malignant. Characteristics such as size and radiodensity in Hounsfield units (HU) help to make these distinctions, as larger size and increased radiodensity (>20 HU) are associated with malignancy and/or pheochromocytoma. While adrenal causes of Cushing’s syndrome present with masses, limited data exists regarding adrenal morphology by CT in Cushing’s disease (CD) or Ectopic ACTH Syndrome (EAS). OBJECTIVE: To identify the incidence and characteristics of adrenal lesions in patients with non-adrenal causes of Cushing’s syndrome. METHODS: We retrospectively studied 80 patients with CD or EAS admitted under our protocols from 2008 to 2018. Inclusion required at least one intact adrenal gland and an abdominal or adrenal CT scan obtained before surgical cure. The radiology reports were reviewed for the presence of adrenal masses, simple adrenal hyperplasia and/or nodular adrenal hyperplasia. If available, the size and HU of masses were noted. Data are presented as median (range) or %. RESULTS: The age was 48.5 (17-81) years. 78 patients had histologically-proven CD (n=53, 44 women) or EAS (n=25, 13 women); 2 had occult presumed EAS (1 woman). The prevalence of discrete nodules was similar in EAS (15%) and CD (25%, P=0.3). • 13 CD patients had nodules (size 2.0 (0.9-5.0) cm). Of these, 5 were < 10 HU, 3 were > 10 but < 20 HU, and 1 was > 20 HU (not available, N/A=4). • 4 EAS patients had nodules ( size 1.0 (0.9- 1.3) cm). Of these, 2 were > 20 HU and 1 was < 10 HU (N/A=1). Biochemical testing for pheochromocytoma was negative when HU was >20. • Nearly all patients with discrete nodules had contralateral and/or adjacent adrenal hyperplasia. • The rate of nodularity was 15% in patients aged <50 and 28% in those aged ≥50 (P = 0.13). • All patients with a discrete nodule > 1.3 cm had CD. The presence of any nodule, regardless of size, had a positive predictive value for CD of 76% (95% CI: 54-90%). Of 63 patients without discrete nodules, 28 had simple hyperplasia (CD = 15, 28% vs EAS = 13, 48%, P=0.07). Another 12 patients, 8 with CD, had indiscrete nodular hyperplasia. One patient had normal sized glands. Data were N/A for 27 patients. CONCLUSION: The prevalence of adrenal nodules is greater in patients with CD and EAS than in the general population. Discrete nodules > 1.3 cm confirmed a diagnosis of CD, while simple hyperplasia was more likely in EAS. A larger cohort should be studied to determine if these trends are useful in the differential diagnosis of ACTH-dependent Cushing’s syndrome. The increased rate of nodularity in older patients may reflect pre-existing nodules, rather than an effect of Cushing’s syndrome per se. Finally, the reason(s) for differential formation of nodules in CD vs EAS requires further evaluation.