Background: Computed tomography (CT) and magnetic resonance (MR) imaging can help diagnose benign adrenal adenomas, but prior studies rely on nonoperative follow-up as proof of a lesion's benign nature. We examined adrenalectomy tissues to determine if imaging characteristics correlate with histopathologic findings.
Methods: We retrieved data for 196 consecutive adrenalectomies in 192 patients from 2000 to 2008. Imaging results were considered to signify benign adrenal adenoma if one or more of the following was present: Hounsfield units <10 on unenhanced CT, contrast-enhanced CT quantifying absolute contrast washout of >60% or relative contrast washout of >40%, or MR with chemical-shift imaging demonstrating loss of signal intensity on out-of-phase images.
Results: The sensitivity and specificity of preoperative imaging in predicting benign adrenal adenoma were 57 and 94%, respectively. Histopathology confirmed that all 66 adrenal masses with imaging characteristics suggesting benign adenoma were indeed benign lesions and included 61 benign adrenal adenomas and 5 benign nonadenomatous lesions (3 myelolipomas, 1 composite myelolipoma/adenoma, and 1 ganglioliponeuroma). The specificity of imaging in predicting benignity was 100%. Malignant adrenal lesions were diagnosed in 17/130 (13%) masses: 8 metastases, 7 adrenal cortical carcinomas, 1 epithelioid angiosarcoma, and 1 ganglioneuroblastoma. The sensitivity of imaging in identifying malignancy was 100%. No malignancies were diagnosed during postoperative follow-up (mean 6 months, range 0.2-67 months).
Conclusion: CT or MR characteristics predicted the presence of benign lesions with 100% specificity. Every adrenal malignancy had CT or MR results that were inconsistent with benign adenoma (100% sensitivity). To exclude malignancy, adrenal masses with non-benign imaging characteristics should be resected.