摘要
目的 观察光谱CT电子云密度图(EDM)结合CT表现鉴别急慢性椎体骨质疏松性骨折(OVF)的价值.方法 回顾性收集48例同时存在急、慢性OVF患者的胸椎和/或腰椎光谱CT资料,共纳入110个骨折椎体,包括53个急性骨折(急性组)和57个慢性骨折椎体(慢性组).比较组间骨折椎体光谱CT定量参数,包括常规120 kVp混合能量图像(PI,即常规CT图像)和40、70、100 keV虚拟单能量图像(VMI)中的CT值、有效原子序数(Z-eff)及电子云密度(ED),以及常规CT表现的差异;将差异有统计学意义的参数纳入多因素logistic回归分析,筛选鉴别急、慢性OVF的独立危险因素,并构建联合模型.绘制受试者工作特征(ROC)曲线,评估各单一独立危险因素及联合模型鉴别诊断急、慢性OVF的效能.结果 组间光谱CT定量参数,以及终板中断征、骨皮质皱褶、椎体密度增高、椎体内气体影和椎体压缩程度差异均有统计学意义(P均＜0.05).logistic 回归分析结果显示,CTPI(OR=0.855,P=0.005)、ED(OR=16.432,P=0.005)、骨皮质皱褶(OR=0.038,P=0.034)及椎体密度增高(OR=0.025,P=0.013)均为急性OVF的独立危险因素,以之单独鉴别急、慢性OVF的曲线下面积(AUC)分别为0.870、0.889、0.879及0.866,均低于联合模型的0.977(Z=3.47、3.73、2.95、2.71,P均＜0.05).结论 光谱CT EDM结合CT表现能有效鉴别诊断急、慢性OVF.
Abstract
Objective To observe the value of electron density map(EDM)from spectral CT combined with CT features in differentiating acute and chronic osteoporotic vertebral fractures(OVF).Methods Thoracic and/or lumbar spectral CT data of 48 patients with acute complicated chronic OVF were retrospectively analyzed.Totally 110 fractured vertebrae were enrolled,including 53 vertebrae with acute fractures(acute group)and 57 with chronic fractures(chronic group).The quantitative parameters of spectral CT,including CT values of conventional 120 kVp polyenergetic image(PI,i.e.routine CT images)and 40,70,100 keV virtual monoenergetic images(VMI),effective atomic number(Z-eff)and electron density(ED),as well as routine CT finding were compared between groups,and those being significantly different were included in multivariate logistic regression to screen the independent risk factors for acute OVF and construct a combined model.Receiver operating characteristic(ROC)curves were drawn to evaluate the efficacy of each single independent risk factor and the combination for differentiating acute and chronic OVF.Results Significant differences of all spectral CT quantitative parameters,also of routine CT findings including interruption of vertebral endplate,cortical folds,increased vertebral density,gas within vertebral body and vertebral compression degree were found between groups(all P＜0.05).Logistic regression analysis showed that CTPI(OR=0.855,P=0.005),ED(OR=16.432,P=0.005),cortical folds(OR=0.038,P=0.034)and increased vertebral density(OR=0.025,P=0.013)were all independent risk factors for acute OVF.The area under the curve(AUC)of the above single parameters for identifying acute and chronic OVF was 0.870,0.889,0.879 and 0.866,respectively,all lower than that of the combined model(0.977)(Z=3.47,3.73,2.95,2.71,all P＜0.05).Conclusion Spectral CT EDM combined with CT findings could effectively differentiate acute and chronic OVF.
基金项目
东莞市社会发展科技项目(20211800904432)
维普
万方数据