This article is more than 5 years old. Information may no longer be current.
Diagnostic PET/CT should be used with caution in high-risk adrenal cases
Click Here to Manage Email Alerts
ORLANDO, Fla. — In patients at high-risk for adrenal malignancy, 18F-fluorodeoxyglucose-PET/CT imaging should be used with clinical judgment because of suboptimal sensitivity and specificity in this population, according to study findings presented here.
In a retrospective study, Danae Delivanis, MD, a second-year endocrine fellow at Mayo Clinic, and colleagues examined the performance of 18F-fluorodeoxyglucose(18FDG)-PET/CT imaging in a population at high risk for adrenal malignancy. The researchers evaluated data from 352 patients (62.5% men; median age, 68 years; age range, 18-91 years) referred for adrenal biopsies due to suspected or confirmed extra-adrenal malignancy between 1994 and 2014. Malignant adrenal lesions were present in 223 cases and benign lesions in 129. Adrenal histopathology yielded the reference standard in all patients.
Danae Delivanis
“We do have a relatively large number [of studies on PET/CT imaging for adrenal lesions] as well as an optimal reference standard,” Delivanis told Endocrine Today. “But other studies lack clinical follow-up, or they don’t mention how they defined ‘benign’ or ‘malignant,’ compared to other published studies that are characterized by small sample sizes and lack of an optimal reference standard (eg, having used only clinical follow-up or repeat imaging. Here there are only patients who had adrenal biopsy performed or adrenalectomy. So we had a good reference standard.”
Unenhanced CT revealed Hounsfield unit (HU) measurements greater than 10 HU for all malignant lesions, with 100% sensitivity, 33% specificity, 73% positive predictive value and 100% negative predictive value. In 91 patients who underwent 18FDGPET/CT, imaging revealed 44 metastases, three lymphomas, one adrenocortical carcinoma and 43 adrenal adenomas. FDG uptake was higher in malignant (SUV max: median, 10.1; range, 1.9-29.4) than in benign lesions (SUV max: median, 3.7; range, 1.4-24.5; P < .001). Adrenal/liver ratio in malignant lesions was higher than in benign lesions (median, 3; range, 0.5-13.4 vs. median, 1.15; range 0.6-6.6; P < .001). Diagnosis of adrenal malignancy was most accurate with an adrenal/liver ratio cutoff of 1.8 and had 83.3% sensitivity, 83.7% specificity, 85.1% positive predictive value and 81.8% negative predictive value.
“From this study we can conclude that noncontract CT HU of 10 or below is a good initial imaging approach as it excludes a malignant lesion,” Delivanis told Endocrine Today. “For lesions bigger than 10 HU, PET/CT could be considered in a population at high risk for adrenal malignancy. However, both sensitivity and specificity of PET/CT is not perfect; therefore careful clinical judgement is warranted.” – by Jill Rollet
Reference:
Delivanis D, et al. Abstract #144. Presented at: The American Association for Clinical Endocrinologists Annual Scientific & Clinical Congress; May 25-29, 2016; Orlando, Fla.
Disclosure: Delivanis reports no relevant financial disclosures.