目的：探讨3.0T MRI 扩散张量成像（DTI）定量参数在胶质瘤分级中的应用价值。资料与方法回顾性分析经病理证实的51例脑胶质瘤的 DTI 参数图，将51例患者分为低级别胶质瘤（I~II 级，18例）和高级别胶质瘤（III~IV 级，33例），比较低级别与高级别胶质瘤肿瘤实质区、瘤周水肿区和健侧脑组织区各向异性分数（FA）、平均扩散系数（MD）、轴向扩散系数（AD）、横向扩散系数（RD）、λ1、λ2及λ3的差异。结果低级别胶质瘤组肿瘤实质区 rMDt、rADt、rRDt、λ1t、λ2t 及λ3t高于高级别胶质瘤组（t=－3.235~－2.458，P<0.05），rFAt 低于高级别胶质瘤组（t=1.554，P>0.05）；低级别胶质瘤组瘤周水肿区 rFAe 高于高级别胶质瘤组，rMDe、rADe、rRDe、λ1e、λ2e 及λ3e 低于高级别胶质瘤组，其中仅λ1e 差异有统计学意义（t=2.052，P<0.05）。ROC 曲线分析显示，rMDt、rADt、rRDt、λ1t、λ2t、λ3t 及λ1e鉴别胶质瘤分级的曲线下面积分别为0.746、0.710、0.762、0.735、0.722、0.705、0.374，其中 rMDt、rADt、rRDt、λ1t、λ2t、λ3t 的 ROC 曲线下面积在鉴别低级别及高级别胶质瘤中差异均有统计学意义（Z=3.287~4.605，P<0.001）。结论在应用 DTI 定量参数对胶质瘤进行分级诊断中，肿瘤实质区 rMD、rAD、rRD、λ1、λ2、λ3对低级别及高级别胶质瘤的分级诊断具有一定的价值。%Purpose To study the application of MR diffusion tensor imaging (DTI) quantitative parameters in grading of cerebral glioma on a 3.0T scanner. Materials and Methods DTI mapping of 51 cases of cerebral glioma confirmed by pathology were retrospective analyzed. All the cases were divided into two groups: low-grade gliomas (grade I-II, 18 cases) and high-grade gliomas (grade III-IV, 33 cases). Value of fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), radial diffusivity (RD),λ1, λ2, and λ3 of the tumor, peritumoral edema and contralateral tissue area were recorded and compared. Results rMDt, rADt, rRDt, λ1t, λ2t and λ3t of tumor in the low-grade glioma group were higher than those in the high-grade glioma group, and the difference was statistically significant (t=-3.235- -2.458, P<0.05). rFAt was lower in the low-grade glioma group, and the difference was not statistically significant (t=1.554, P>0.05). rFAe of peritumoral edema in the low-grade glioma group was higher than those in the high-grade group, while rMDe, rADe, rRDe, λ1e, λ2e and λ3e were lower in the low-grade group. All differences were not statistically significant except λ1e (t=2.052, P<0.05). ROC analysis showed the area under the curve (Az) of rMDt, rADt, rRDt, λ1t, λ2t, λ3t and λ1e were 0.746, 0.710, 0.762, 0.735, 0.722, 0.705 and 0.374, respectively. Az value of rMDt, rADt, rRDt,λ1t, λ2t, λ3t were statistically different between the low- and high-grade gliomas (Z=3.287-4.605, P<0.001). Conclusion Among DTI quantitative parameters on glioma grading, rMD, rAD, rRD, λ1, λ2, and λ3 of tumor area are helpful in grading gliomas.
展开▼
