The role of fluorodeoxyglucose-positron emission tomography/computed tomography in differentiating between benign and malignant adrenal lesions

Abstract

This retrospective study was designed to investigate the clinical role of whole-body positron emission tomography/computed tomography (PET/CT) by using 2-[18F]fluoro-2-deoxy-D-glucose (FDG), for the evaluation of adrenal lesions and to find the best index to distinguish benign from malignant lesions in various cancer patients. A total of 81 patients (55 male and 26 female, age range: 31-81 years, mean: 61.5) who had confirmed primary malignancies (lung cancer in 47 patients, gastrointestinal malignancies in 13 patients, malignant melanoma in one patient, renal cell cancer in three patients, mesothelioma in two patients, breast carcinoma in nine patients, cervical cancer in one patient, ovarian cancer in two patients, pheochromocytoma in one patient, unknown primary in two patients) underwent PET/CT examinations for cancer screening, staging, restaging, and detection of suspected recurrence. Of the 81 patients, 104 adrenal lesions (34 benign and 70 malignant adrenal lesions) were shown by CT. On visual analysis of PET/CT imaging, adrenal uptake was based on a three-scale grading system. For final assessment standards of references for adrenal malignant lesions was based on biopsy (n=2), interval growth, or reduction after chemotherapy. An adrenal lesion, which remained unchanged on clinical and imaging follow-up of at least 7 months (mean follow-up time 19.31 months±6.46, range 7-30 months), was decided as a benign lesion. In adrenal malignant lesions maximum standardized uptake value (SUVmax) (8.82±4.47) was higher than that of adrenal benign lesions (3.02±1.15, P<0.0001). In the differentiation of adrenal benign and malignant lesions, a CT threshold of 10 Hounsfield units corresponded to a sensitivity of 64.7%, specificity of 98.6%, and accuracy of 87.5%. An SUVmax cut-off value of 2.5 corresponded to a sensitivity of 100%, specificity of 38.2%, and accuracy of 80%. An SUVmax cut-off value of 4.2 corresponded to a sensitivity of 88.6%, specificity of 88.2%, and accuracy of 88.5%. The ratio of tumor SUVmax to liver SUVmean was 3.61±1.77 for adrenal malignant lesions whereas it was 1.20±0.38 for adrenal benign lesions (P<0.0001). T/L SUV ratio cut-off value of 1.8 corresponded to a sensitivity of 87%, specificity of 91%, and accuracy of 88.5%. T/L SUV ratio cut-off value of 1.68 corresponded to a sensitivity of 90%, specificity of 91.1%, and accuracy of 90.4%. 2-[18F]fluoro-2-deoxy-D-glucose-PET/CT improves the diagnostic accuracy in the differentiation of benign from malignant adrenal lesions in various cancer patients. Combined information obtained from PET/CT (SUVmax, T/L SUV ratio, visual analysis) and unenhanced CT (size, Hounsfield units measurement) is recommended for better differentiation.