Objective: The management of large nonsecreting adrenal tumors (at least 4 cm) is still a matter of debate as it is unclear whether imaging, especially 18F-fluorodeoxyglucose (FDG), can be used to characterize their potential malignancy. Moreover, the risk of new hypersecretion in nonoperated tumors is uncertain. Our aim was to better characterize these large adrenal incidentalomas. Methods: Patients followed in our center for a nonsecreting large (at least 4 cm) adrenal incidentaloma, with an initial computed tomography (CT) and 18F-FDG positron emission tomography (PET) CT, were retrospectively included. Patients who were not operated after initial diagnosis had to be followed with clinical, biological, and imaging evaluations for at least 3 years or until delayed surgery. Results: Eighty-one patients were included in the study: 44 patients (54.3%) had initial surgery while 37 were followed, including 21 (25.9%) who were operated after a mean of 19 months. Among the 65 operated patients, 13 (20%) had a malignant lesion (3 with metastasis, and 10 with adrenocortical carcinoma). Unenhanced CT <10 showed 85.6% sensitivity and 78.8% specificity; all had a 18F-FDG uptake ratio >1.5. Among the 24 patients who were followed for at least 3 years, 5 (20.8%) finally presented hypercortisolism (4 subclinical). Conclusion: As expected, large adrenal tumors are at a higher risk of malignancy. The combination of unenhanced CT <10 and 18F-FDG PET ratio <1.5 prove to be reassuring and might lead to a close follow-up rather than immediate surgery. Hormonal follow-up should be focused on the risk of hypercortisolism. Abbreviations: CI = confidence interval; CT = Computed Tomography; ENSAT = European Network for the Study of Adrenal Tumors; ESE = European Society of Endocrinology; FDG = fluorodeoxyglucose; HU = Hounsfield units; PET = positron emission tomography; ROI = regions of interest; SUV = standard uptake value.