Background:An important applications of the Dual energy computed tomography DECT in the field of musculoskeletal radiology is the detection of bone marrow edema (BME), using a post-processing software to remove calcium in trabecular bone by using a “virtual non-calcium (VNCa)” subtraction process DECT have been successfully employed in the evaluation of the extent of BME in patients with sacroiliitis (1)Objectives:The aims of this study were i)to evaluate the discriminating capacity of DECT versus MRI in the detection of BME of the sacroiliac joints in patients with axial-SpA and to define the optimal cutoff; ii) to define of inter-observer agreement between radiologistsMethods:All patients underwent a pelvic DECT examination, within 30 days of the MRI imaging, (Somatom Force; Siemens Healthineers, Enlangen, Germany). Each exam was evaluated by two operators: an experienced radiologist and a radiologist in training. The dedicated software also allows the precise calculation of the attenuation values in the region of interest (ROI). On the reformatted color-coded dual-energy virtual non-calcium images bone marrow signal is depicted in green and corresponding to high signal intensity on T2-weighted fat suppression MR images (Fig. 1). With the consent of the two operators, three ROIs were manually positioned for each side of the sacroiliac joints in the subchondral region of the proximal, middle and distal thirds of each joint head, respectively. The interobserver agreement analysis was carried out in the semi-quantitative evaluation of the scores assigned in CT. The accuracy of DECT for the detection of BME compared to MRI was analyzed using the Receiver Operating Characteristics (ROC) curve methodFigure 1.Patient with SpA A. Short tau inversion recovery MR image demonstrating extensive BME in both iliac and sacral subchondral bone, indicating active sacroiliitis. B. CT Semicoronal multiplanar reformatted grayscale image shows signs of structural change. C. Semicoronal multiplanar reformatted color-coded dual-energyVNCaimage reveals bone marrow signal involving both sacroliliac joints, corresponds toBMEon the MR imageResults:56 axial-SpA patients have been evaluated, 30 males and 26 females, a mean age of 48.6 ± 12.3 years, a mean disease duration of 5.5 ± 2.9 years, a mean C-reactive protein level of 3.0 ± 2.5 mg/dl. The inter-rater agreement of readers showed a high statistical significance greater than 0.80, in particular the weighted kappa is 0.815, with a standard error of 0.04 and a 95% variability coefficient between 0.73 and 0.89. Sensitivity, specificity, and positive likelihood ratio in the identification of BME at DECT were 95.8%, 83.3% and 6.67, respectively. The differences in mean CT number (HU) among the four levels of edema category were significant (p<0.0001). The AUC was 0.905 in the differentiation of the presence of BME from no edema (Fig. 2). A cutoff value of –14.8 HU yielded overall sensitivity of 82.86% and specificity of 90.48%, with an LR+ of 8.70, in the detection of BMEFigure 2.Graph shows ROC curves from CT numbers (in Hounsfield units) derived from DECT images in the detection of sacroiliitis with and without BME. AUC was 0.905Conclusion:We confirm the potential of DECT for the detection of BME of the sacroiliac joints in patients affected by SpA. This new method appears to be very useful, not only in the diagnostic phase, but also for the monitoring of patients.References:[1]Carotti M, Salaffi F, Beci G, Giovagnoni A. The application of dual-energy computed tomography in the diagnosis of musculoskeletal disorders: a review of current concepts and applications. Radiol Med. 2019;124(11):1175-1183.Disclosure of Interests:None declared
Fast kilovoltage (KV)-switching dual-energy computed tomography hydroxyapatite (HAP)-water decomposition technique for identifying bone marrow edema in vertebral compression fractures
