3D-ASL与DCE-MRI在鉴别诊断脑高级别胶质瘤与脑转移瘤中的应用价值

The application value of 3D arterial spin labeling and dynamic contrast-enhanced MRI in differential diagnosis of high-grade gliomas and brain metastatic tumors

目的:探讨3D-ASL及DCE-MRI序列获得的定量参数在鉴别诊断高级别胶质瘤(HGG)与脑转移瘤(MT)中的应用价值,提高术前诊断准确率。材料与方法:选取2019年6月—2020年11月在青岛大学附属青岛市市立医院经手术病理确诊为HGG、MT的患者共61例(HGG患者22例,MT患者39例),其中男35例,女26例,年龄15~85岁,平均(53.33±17.63)岁。对受试者行颅脑DCE-MRI及3D-ASL扫描,测量肿瘤实质区及瘤周水肿区(<1 cm)各渗透性参数值(K^(trans)、V_(e)、CBF),采用独立样本t检验比较HGG与MT在实质区与瘤周水肿区间的差异。使用受试者工作特征(ROC)曲线评估各参数及联合参数鉴别诊断两种肿瘤的效能,计算最佳阈值及其对应的敏感度、特异性,P<0.05表示差异有统计学意义。结果:HGG在肿瘤实质区、瘤周水肿区的平均K^(trans)、V_(e)、CBF及r CBF均比MT高;在肿瘤实质区,K^(trans)、V_(e)及rCBF在两种肿瘤间无统计学差异(P>0.05),CBF在两种肿瘤间有统计学差异(P<0.001);在瘤周水肿区,K^(trans)、V_(e)、CBF及rCBF在两种肿瘤间有统计学差异(P<0.001),瘤周水肿区K^(trans)、V_(e)及rCBF鉴别两种肿瘤的最佳值分别为0.053、0.483、1.092,所对应的曲线下面积(AUC)分别为0.798、0.735、0.727,3个定量参数联合诊断时,AUC为0.897,所对应的敏感度为0.718,特异性为0.955。结论:3D-ASL与DCE-MRI所得的定量参数有助于鉴别HGG与MT,肿瘤瘤周水肿区比肿瘤实质区更有诊断价值,两种技术联合应用可提高鉴别诊断准确率。

Objective:To explore the application value of quantitative parameters obtained by 3 D-ASL and DCE-MRI sequences in the differential diagnosis of high-grade glioma(HGG)and brain metastatic tumor(MT),so as to improve the accuracy of preoperative diagnosis.Materials and methods:A total of 61 patients with HGG and MT diagnosed by surgery and pathology in Qingdao Municipal Hospital Affiliated to Qingdao University from June 2019 to November 2020 were selected(22 patients with HGG and 39 patients with MT),35 males and 26 females,aged 15~85 years old,with an average age of(53.33±17.63)years old,were proved by pathological examination.DCE-MRI and 3 D-ASL scanning were performed in subjects,the parameters value(K^(trans),Ve,rCBF)of tumor parenchyma and peritumoral edema(<1 cm)were measured.The differences of HGG and MT between tumor parenchyma and peritumoral edema were compared by independent sample t-test.Receiver operating characteristic(ROC)curve was used to evaluate the efficacy of various parameters and combined parameters in the differential diagnosis of two kinds of tumors,and the optimal threshold and its corresponding sensitivity and specificity were calculated.P<0.05 indicated that the difference was statistically significant.Results:Average K^(trans),V_(e),CBF and rCBF of HGG were higher than those of MT in tumor parenchyma and peritumoral edema.In the tumor parenchyma,there were no significant differences in K^(trans),V_(e) and rCBF between the two groups(P>0.05),but there was significant difference in CBF between the two groups(P<0.001).K^(trans),V_(e),CBF and rCBF values were significantly different between two kinds of tumors(P<0.001)in the peritumoral edema.The optimal values of K^(trans),V_(e) and rCBF identified two types of tumors were 0.053,0.483,and 1.092,respectively,and the corresponding the area under curve(AUC)were 0.798,0.735,and 0.727,respectively.When the three quantitative parameters were combined for diagnosis,the AUC was 0.897,the corresponding sensitivity was 0.718,and the specificity was 0.955.Conclusion:The quantitative parameters obtained by 3 D-ASL and DCE-MRI can differentiate HGG from MT,and the peritumoral edema is more valuable in diagnosis than that of the tumor parenchyma region.The combination of the two technologies can improve the accuracy of differential diagnosis.