Objective: To investigate whether integrated fluorodeoxyglucose-positron emission tomography/computed tomography (FDG-PET/CT) can differentiate benign from adrenal malignant lesions on the basis of maximum standardized uptake value (SUV(max)), tumor/liver (T/L) SUV(max) ratio, and CT attenuation value (Hounsfield Units; HU) of unenhanced CT obtained from FDG-PET/CT data.
Methods: We studied 30 patients with 35 adrenal lesions (16 adrenal benign lesions, size 16 +/- 5 mm, in 15 patients; and 19 adrenal malignant lesions, 24 +/- 12 mm, in 15 patients) who had confirmed primary malignancies (lung cancer in 23 patients, lymphoma in 2, esophageal cancer in 2, hypopharyngeal cancer in 1, prostate cancer in 1, and 1 patient in whom lesions were detected at cancer screening). All patients underwent PET/CT at 1 h post FDG injection. Diagnosis of adrenal malignant lesions was based on interval growth or reduction after chemotherapy. An adrenal mass that remained unchanged for over 1 year was the standard used to diagnose adrenal benign lesions. Values of FDG uptake and CT attenuation were measured by placing volumetric regions of interest over PET/CT images. Adrenal uptake of SUV(max) >/= 2.5 was considered to indicate a malignant lesion; SUV(max) < 2.5 was considered to indicate a benign lesion. In further analysis, 1.8 was employed as the threshold for the T/L SUV(max) ratio. Unenhanced CT obtained from PET/CT data was considered positive for adrenal malignant lesions based on a CT attenuation value >/= 10 HU; lesions with a value < 10 HU were considered adrenal benign lesions. Mann-Whitney's U test was used for statistical analyses.
Results: SUV(max) in adrenal malignant lesions (7.4 +/- 3.5) was higher than that in adrenal benign lesions (2.1 +/- 0.5, p < 0.05). The CT attenuation value of adrenal malignant lesions (27.6 +/- 11.9 HU) was higher than that of adrenal benign lesions (10.1 +/- 12.3 HU, p < 0.05). In differentiating between adrenal benign and malignant lesions, a CT threshold of 10 HU corresponded to a sensitivity of 57%, specificity of 94%, accuracy of 74%, positive predictive value of 92% and negative predictive value of 65%. An SUV(max) cut-off value of 2.5 corresponded to a sensitivity of 89%, specificity of 94%, accuracy of 91%, positive predictive value of 94% and negative predictive value of 88%. The T/L SUV(max) ratio was 1.0 +/- 0.2 for adrenal benign lesions and 4.5 +/- 3.0 for adrenal malignant lesions. And T/L SUV(max) ratio cut-off value of 1.8 corresponded to a sensitivity of 85%, specificity of 100%, accuracy of 91%, positive predictive value of 100% and negative predictive value of 83%.
Conclusions: FDG-PET/CT with additional SUV(max) analysis improves the diagnostic accuracy of adrenal lesions in cancer patients.