Lu Hongyue, Li Zhi, Liang Ziheng, Liu Yuqi
Medical school, Kunming University of Science and Technology, Kunming, Yunnan, China.
Department of Radiology, The First People's Hospital of Yunnan Province, Kunming, Yunnan, China.
J Orthop Surg Res. 2025 Jan 9;20(1):28. doi: 10.1186/s13018-024-05341-3.
In-depth investigation of the diagnostic performance of dual-energy CT (DECT) virtual non-calcium (VNCa) technique for sacroiliac joint bone marrow edema (BME) in patients with ankylosing spondylitis(AS).
A total of 42 patients with AS)who underwent sacroiliac joint MRI and DECT scans on the same day at our Rheumatology and Immunology Department between August 2022 and June 2023 were selected. Using MRI as the reference standard, the presence of BME on the iliac and sacral surfaces was evaluated, resulting in the categorization of patients into BME-positive and BME-negative groups. Subsequently, the DECT scan data was processed using the "Bone Marrow" algorithm to generate VNCa color-coded images of the bone marrow. The diagnostic performance of DECT in detecting BME was assessed through visual qualitative evaluation and objective quantitative analysis.
Visual qualitative assessment analysis showed good agreement between the results of BME analysis on virtual non-calcium images and MRI images by both physicians (Kappa > 0.61). The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of visual qualitative assessment for BME by Physicians A and B were as follows: iliac facet [(92.5%, 88.6%, 88.1%, 92.9%, 90.5%), (90.0%, 90.9%, 90.0%, 90.9%, 90.5%)], sacral facet [(88.4%, 87.8%, 88.4%, 87.8%, 88.1%), (90.7%, 85.3%, 86.7%, 89.7%, 88.1%)].In terms of objective quantitative analysis, the CT values of the edematous areas on the iliac and sacral surfaces were (-41.4 ± 15.9) Hu and (-38.8 ± 19.7) Hu, respectively, while the CT values of normal bone marrow areas were (-79.6 ± 18.2) Hu and (-72.8 ± 14.8) Hu, respectively. The CT values of the edematous areas were higher than those of the non-edematous areas. Based on the receiver operating characteristic curve analysis, the area under the curve for the iliac and sacral surfaces were 0.90 and 0.89, respectively. The optimal CT cutoff values were - 57.4 Hu and - 56.8 Hu, with corresponding sensitivities of 92.5% and 86.4% and specificities of 90.7% and 87.8%.
The DECT VNCa technique has a high diagnostic efficacy in the diagnosis of BME in the sacroiliac joints in ankylosing spondylitis in terms of visual qualitative assessment and objective quantitative analysis.
深入研究双能CT（DECT）虚拟去钙（VNCa）技术对强直性脊柱炎（AS）患者骶髂关节骨髓水肿（BME）的诊断性能。
选取2022年8月至2023年6月期间在我院风湿免疫科于同一天接受骶髂关节MRI和DECT扫描的42例AS患者。以MRI作为参考标准，评估髂骨和骶骨表面BME的存在情况，将患者分为BME阳性组和BME阴性组。随后，使用“骨髓”算法处理DECT扫描数据，生成骨髓的VNCa彩色编码图像。通过视觉定性评估和客观定量分析评估DECT检测BME的诊断性能。
视觉定性评估分析显示，两位医生对虚拟去钙图像和MRI图像上BME分析结果的一致性良好（Kappa＞0.61）。医生A和医生B对BME视觉定性评估的敏感度、特异度、阳性预测值、阴性预测值和准确度如下：髂骨小关节[（92.5%，88.6%，88.1%，92.9%，90.5%），（90.0%，90.9%，90.0%，90.9%，90.5%）]，骶骨小关节[（88.4%，87.8%，88.4%，87.8%，88.1%），（90.7%，85.3%，86.7%，89.7%，88.1%）]。在客观定量分析方面，髂骨和骶骨表面水肿区域的CT值分别为（-41.4±15.9）Hu和（-38.8±19.7）Hu，而正常骨髓区域的CT值分别为（-79.6±18.2）Hu和（-72.8±X14.8）Hu。水肿区域的CT值高于非水肿区域。基于受试者工作特征曲线分析，髂骨和骶骨表面的曲线下面积分别为S0.90和0.89。最佳CT截断值分别为-57.4 Hu和-56.8 Hu，相应的敏感度分别为92.5%和86.4%，特异度分别为90.7%和87.8%。
在视觉定性评估和客观定量分析方面，DECT VNCa技术对强直性脊柱炎骶髂关节BME的诊断具有较高的诊断效能。
