Purpose: To assess the use of the dual-energy computed tomographic (CT) virtual noncalcium technique in the evaluation of bone marrow edema in vertebral compression fractures. Materials and Methods: This prospective study was approved by the institutional review board; informed consent was obtained from all patients. Sixty-three consecutive patients with 112 thoracic and/or lumbar vertebral compression fractures were studied between January 2011 and April 2012. All patients underwent both dual-energy CT (100 kV and Sn140 kV, where Sn indicates the use of a 0.4-mm tin filter) and magnetic resonance (MR) imaging. Dual-energy CT data were postprocessed by using a three-material decomposition algorithm for generating noncalcium images of the collapsed bodies. Two radiologists evaluated for the presence of abnormal attenuation alterations in the bone marrow by using color-coded maps and measured CT numbers on noncalcium grayscale images. Bone sclerosis and intravertebral air were evaluated with CT scans. MR images served as the reference standard. CT numbers were subjected to receiver operating characteristic curve analysis. Results: MR imaging depicted 46 edematous and 66 nonedematous vertebral compression fractures. Eighty-two bodies were classified as having less than 50% sclerosis and/or air. Significant differences in noncalcium CT numbers between edematous and nonedematous vertebral compression fractures were found for both readers (P <.0001). CT numbers for the diagnosis of bone marrow edema on the basis of MR imaging revealed areas under the receiver operating characteristic curve of 0.799 and 0.841 for readers 1 and 2, respectively (P =.56). Use of a cutoff value of 280 to differentiate edematous vertebral bodies resulted in a sensitivity of 96.3%, specificity of 98.2%, and accuracy of 97.6% in the group of vertebral bodies with less than 50% sclerosis and/or air. Conclusion: Dual-energy CT virtual noncalcium images were able to depict bone marrow in the collapsed vertebral bodies, especially in those with less than 50% sclerosis and/or air.
All Science Journal Classification (ASJC) codes