能谱 CT 水基物质分离技术检测骶髂关节炎骨髓水肿初探

Primary exploration of energy spectral CT water -based material decomposition technology for the detection of bone marrow edema in sacroiliitis

目的：初步探讨能谱 CT 水基物质分离技术检测骶髂（sacroiliac，SI）关节炎骨髓水肿（bone marrow edema，BME）的价值。方法对9例脊柱关节病 SI 关节炎 BME 患者（研究组）和8例健康志愿者（对照组）SI 关节进行 MRI 和能谱 CT 检查。将能谱 CT 扫描原始图像重建为单能谱图像，运用物质分离与分析软件将单能谱图分解为水基图和钙基图。参照 MRI，分别在水基图和钙基图上测量研究组 BME 区和对照组 SI 关节旁骨髓区的水基-钙基浓度，并比较其差别。利用 ROC 曲线分析水基浓度诊断 SI 关节炎 BME 的最佳浓度、敏感度、特异度、阳性似然率和阴性似然率。结果研究组 SI 关节旁 BME 区水基浓度[（1067．43±10.84）g／L]高于对照组 SI 关节旁骨髓区水基浓度[（1039．43±8．01）g／L]（t ＝－3．13，P ＝0.003）；研究组 SI 关节旁 BME 区的钙基浓度[（68．98±20．53）g／L]与对照组的 SI 关节旁骨髓区钙基浓度[（78．03±26．39）g／L]差异无统计学意义（t ＝1．88，P ＝0．066）。ROC 结果显示以 SI 关节旁水基浓度预测 SI关节旁 BME 的曲线下面积为0．75，具有中等诊断效能。当水基浓度为1052．00 g／L 时，其诊断效能最佳，敏感度为84．00％，特异度为62．50％。结论能谱 CT 水基检测可能对 SI 关节炎 BME 的检出有一定参考作用，具有潜在临床价值。

Objective To explore the value of water -based material decomposition images for detection of bone marrow edema （BME）in sacroiliitis with energy spectral CT.Methods The sacroiliac （SI）joints of nine spondyloarthropathy patients with sacroiliitis （the research group）and eight healthy volunteers （the control group） were underwent MRI and energy spectral CT.The mixed energy image of energy spectral CT was reconstructed to be mono energy image.Then,the mono image was divided into water -based image and calcium image by material divid-ed and analyzed software.The SI para -articular marrow region water -calcium relative concentration of research group was compared with that of control group.The BME diagnosis efficiency and optimal water -based concentration of energy spectral CT was analyzed by receiver operating characteristic （ROC）curve.Then,the sensitivity,specificity, positive likelihood ratio and negative likelihood ration were calculated.Results The water -based concentration of BME in research group （1 067.43 ±10.84）g/L was higher than that of control group （1 039.43 ±8.01）g/L（t =-3.14,P =0.003）.Meanwhile,the calcium -based concentration of research group （68.98 ±20.53）g/L was not＆nbsp;different from control group （78.03 ±26.39）g/L（t =1.88,P =0.066）.ROC curve showed that the diagnosis efficiency was medium as the area of under curve was 0.75.When the optimal concentration of water -based was 1 052.00g/L,the diagnosis efficiency was the best.The sensitivity and specificity was 84.00%,62.50% respectively. Conclusion There are reference value and potential clinical value with energy spectral CT water -based concentra-tion detection for diagnosis BME in patients with sacroiliitis.