基于不同区域ADC值和1H-MRS代谢物比值鉴别胶质母细胞瘤与原发性中枢神经系统淋巴瘤

Differentiation of glioblastoma from primary central nervous system lymphoma based on ADC values and 1H-MRS metabolite ratio in different regions

目的:探讨基于不同区域表观扩散系数(ADC)值和磁共振波谱(MRS)代谢物比值鉴别诊断胶质母细胞瘤(GBM)与原发性中枢神经系统淋巴瘤(PCNSL)的价值。方法:回顾性分析经手术病理证实的27例GBM和15例PCNSL患者的影像资料,分别测量肿瘤实质区和瘤周水肿区的ADC值和MRS代谢物比值。根据数据是否符合正态分布,采用两样本独立t检验或Mann-Whitney U检验比较GBM组与PCNSL组肿瘤不同区域ADC值及MRS代谢物比值的差异。对组间差异存在统计学意义(P<0.05)的参数采用受试者工作特征(ROC)曲线分析其对GBM与PCNSL的鉴别诊断效能。结果:肿瘤实质区ADC值(ADC^(*))、瘤周水肿区ADC值(ADC^(#))、肿瘤实质区Cho/NAA(Cho/NAA^(*))、肿瘤实质区NAA/Cr(NAA/Cr^(*))、瘤周水肿区NAA/Cr(NAA/Cr^(#))在GBM组与PCNSL组之间的差异均具有统计学意义(P值均<0.05)。ROC曲线分析结果表明基于肿瘤实质区Cho/NAA比值(Cho/NAA^(*))是鉴别GBM与PCNSL的最佳参数,ROC曲线下面积、敏感度、特异度和截断值分别为0.820、92.6%、60.0%和>2.28。结论:肿瘤实质区、瘤周水肿区ADC值和MRS代谢物比值能够有效鉴别GBM与PCNSL,有助于临床医生制定治疗方案。

Objective:To investigate the value of differentiating glioblastoma(GBM)from primary central nervous system lymphoma(PCNSL)based on apparent diffusion coefficient(ADC)values and metabolite ratios of hydrogen proton magnetic resonance spectroscopy(1H-MRS)in different regions.Methods:The MRI imaging data of 27 patients with GBM and 15 patients with PCNSL confirmed by postoperative histopathology were retrospectively analyzed.The ADC values and MRS metabolite ratios were measured in the tumor parenchyma and peritumoral edema regions,respectively.According to whether the data conform to normal distribution,a two-sample independent t-test or Mann-Whitney U test were used to compare the differences in ADC values and MRS metabolite ratios in different tumor regions between the GBM and PCNSL groups.For parameters with statistically significant differences between groups(P<0.05),the receiver operating characteristic(ROC)curve was used to analyze the differential diagnosis ability of GBM and PCNSL.Results:The differences in ADC value in the tumor parenchyma region(ADC^(*)),ADC value in the peritumoral edema region(ADC^(#)),Cho/NAA in the tumor parenchyma region(Cho/NAA^(*)),NAA/Cr in the tumor parenchyma region(NAA/Cr^(*))and NAA/Cr in the peritumoral edema region(NAA/Cr^(#))between GBM group and PCNSL group were statistically significant(all P<0.05).The results of ROC analysis showed that the Cho/NAA ratio in the tumor parenchyma region(Cho/NAA^(*))was the optimal classifier for differentiating between GBM and PCNSL.The area under curve(AUC),sensitivity,specificity and cut-off values were 0.820,92.6%,60.0%,and>2.28,respectively.Conclusion:The ADC values and MRS metabolite ratios in the tumor parenchyma and peritumoral edema region can effectively distinguish GBM from PCNSL,which helps clinicians formulate an appropriate treatment plan for patients.