This study aimed to determine if a single-source rapid kilovolt (peak)-switching dual-energy (RSDE) multidetector computed tomography (CT) can differentiate high lipid content (HLC) from low lipid content (LLC) incidental adrenal lesions.
A retrospective intrapatient study of 40 consecutive adults with known hepatic or pancreatic pathology who underwent multiphasic abdominal RSDE for nonadrenal-related clinical indications and had adrenal lesions was done. Arterial phase was acquired with RSDE, conventional unenhanced (CU) images with standard MDCT. RSDE measurements included lesion attenuation in Hounsfield units on simulated monochromatic 140-keV images and density (in milligrams per milliliter) on material decomposition images, using water-iodine and fat-iodine basis pairs. Each variable was correlated with CU Hounsfield units (Pearson coefficient). RSDE lesion values were compared with analysis of variance and Tukey HSD test. Receiver operating characteristic (ROC) analysis was performed to identify RSDE thresholds comparable to 10 HU on unenhanced MDCT.
Twenty-nine HLC and 18 LLC lesions were evaluated in 40 subjects (21 men; mean age, 66.5 years). RSDE variables correlated with CU Hounsfield units, r = 0.90–0.92, P < 0.001. Myelolipomas, HLC, and LLC lesions were different by analysis of variance, P < 0.001 for all dual-energy variables. Excluding myelolipomas from ROC curve analysis, ROC areas for Hounsfield unit 140-keV images, fat(-iodine), and water(-iodine) were 0.929 (0.039), 0.917 (0.046), and 0.912 (0.048), respectively (P < 0.001); using a specificity of 94.4%, 64% of adenomatous lesions had 140 keV values of less than 9.5 HU, 59% had fat(-iodine) values of less than 987 mg/mL, and 50% had water(-iodine) values of less than 994 mg/mL.
There is a strong correlation between RSDE measures and accepted MDCT attenuation values for HLC and LLC adrenal lesions. In some patients undergoing postcontrast RSDE who are found to have incidental adrenal nodules, further unenhanced CT or adrenal-protocol CT or magnetic resonance imaging may not be necessary.
From the* Department of Radiology, University of Alabama at Birmingham, Birmingham, AL; †Department of Ophthalmology, Cleveland Clinic, Cleveland, OH; and ‡Department of Biostatistics, University of Alabama at Birmingham, Birmingham, AL.
Received for publication July 16, 2013; accepted October 3, 2013.
Reprints: Desiree E. Morgan, MD, Department of Radiology, University of Alabama at Birmingham, JTN452, 619 S 19th St, Birmingham, AL 35249-6830 (e-mail: firstname.lastname@example.org).
The authors declare no conflict of interest.