The aim of the study was to compare multidetector CT venous enhancement level and washout characteristics of adrenal adenoma and pheochromocytoma, with the goal of defining a venous enhancement level predictive of pheochromocytoma.
Retrospective review of medical records between 2002 and 2012 was performed to identify adrenal masses measuring less than 4 cm. Inclusion criteria for adrenal adenomas was venous phase contrast-enhanced computed tomography (CT), confirmatory adrenal CT (precontrast ± washout), 1 to 2 years stability, and absence of clinical indicators of pheochromocytoma. All pathologically proven pheochromocytomas with venous phase CT imaging were evaluated. Nodule size and attenuation (venous ± precontrast, delayed) were recorded. Student t test analysis compared venous enhancement levels.
One hundred eighty-three subjects with 200 adenomas were compared with 22 subjects with 26 pheochromocytomas. The mean (SD) venous enhancement level for all adenomas (58  Hounsfield units [HU]) and lipid-poor adenomas (76 ) was lower than that of pheochromocytomas (111  HU, P < 0.01). No adenomas enhanced greater than 130 HU, compared with 38% (10/26) of the pheochromocytomas. A threshold of 130 HU to identify pheochromocytoma was 38% sensitive and 100% specific for pheochromocytoma. Of the 17 pheochromocytomas with washout imaging, rapid washout was identified in all (10/10, 100%) that enhanced greater than 130 HU on the venous phase, compared with 43% (3/7) that enhanced less than 130 HU.
An indeterminate adrenal lesion that enhances greater than 130 HU on multidetector CT cannot be assumed to be an adenoma. Hypervascular pheochromocytoma (>130 HU) mimics adenoma washout pattern; absolute venous phase enhancement level must be considered.
From the Department of Radiology, Johns Hopkins Hospital, Baltimore, MD.
Received for publication July 21, 2015; accepted September 1, 2015.
Correspondence to: Pamela T. Johnson, MD, Department of Radiology, Johns Hopkins Hospital, 601 N Caroline St, Room 4223D, Baltimore, MD 21287 (e-mail: firstname.lastname@example.org).
The authors declare no conflict of interest.