MRI伦勃朗视觉感受图像特征对脑胶质瘤分级的预测价值  

Predictive value of MRI VASARI features in brain glioma grading

作　　者：张惠柯 陈传亮 史大鹏 薛丹丹 ZHANG Huike;CHEN Chuanliang;SHI Dapeng;XUE Dandan(Department of Radiology,Henan Provincial People’s Hospital,People’s Hospital of Zhengzhou University,Zhengzhou 450003,China)

机构地区：[1]郑州大学人民医院河南省人民医院放射科

出　　处：《中华实用诊断与治疗杂志》2019年第9期907-910,共4页Journal of Chinese Practical Diagnosis and Therapy

基　　金：国家重点研发计划课题(2016YFC1303204)

摘　　要：目的探讨常规MRI伦勃朗视觉感受图像(visually accessible rembrandt images,VASARI)特征在脑胶质瘤分级及Ki-67、上皮膜抗原(epithelial membrane antigen,EMA)表达预测中的价值。方法脑胶质瘤患者77例,其中高级别胶质瘤41例(高级别组),低级别胶质瘤36例(低级别组),采用免疫组织化学法检测手术切除肿瘤标本Ki-67、EMA阳性表达率。77例术前均行常规MRI平扫及增强扫描,提取VASARI特征。比较2组VASARI特征及Ki-67、EMA阳性表达率,多因素logistic回归分析VASARI特征对脑胶质瘤分级及Ki-67、EMA阳性表达的预测价值。结果高、低级别脑胶质瘤VASARI特征在强化程度、强化百分率、强化边缘厚度、扩散、软脑膜侵犯、深部脑白质侵犯、卫星灶和肿瘤大小上差异均有统计学意义(P<0.05);高级别组Ki-67、EMA阳性表达率(92.7%、37.9%)明显高于低级别组(5.6%、20.7%)(P<0.05);年龄增加(OR=16.380,95%CI:2.882~93.087,P=0.002)和强化程度增加(OR=4.576,95%CI:1.668~12.552,P=0.003)为患高级别脑胶质瘤的独立危险因素;有深部脑白质侵犯(OR=5.437,95%CI:1.516~19.497,P=0.009)为Ki-67阳性表达的独立危险因素;强化强度等8个图像特征均非EMA阳性表达的独立危险因素(P>0.05)。结论高、低级别组脑胶质瘤患者在常规MRI上可量化提取病变VASARI特征,鉴别不同组织病理分级,并可通过深部脑白质侵犯来预测脑胶质瘤组织中Ki-67的表达情况。Objective To explore the predictive value of conventional MRI visually accessible rembrandt images(VASARI)features in glioma grading and the expressions of cell proliferation antigen Ki-67and epithelial membrane antigen(EMA).Methods Seventy-seven cases of glioma included 36cases of low-grade glioma and 41cases of high-grade glioma,and were detected the positive expression rates of Ki-67and EMA by immunohistochemical method.Before operation,routine MRI scanning and enhanced scanning were performed,and VASARI features were extracted.The VASARI features and the positive rates of Ki-67and EMA were compared between two groups.Multivariate logistic regression analysis was used to analyze the predictive value of the image features to the grading of glioma and the positive expressions of Ki-67and EMA.Results There were significant differences in the 8image features between high-and low-grade gliomas including enhancement quality,proportion enhancing,thickness of enhancing margin,diffusion,pial invasion,and deep white matter invasion,satellites and lesion size(P<0.05).The positive rates of Ki-67and EMA were significantly higher in high-grade glioma(92.7%,37.9%)than those in low-grade glioma(5.56%,20.7%)(P<0.05).Age increase(OR=16.380,95%CI:2.882-93.087,P=0.002)and enhancement quality increase(OR=4.576,95%CI:1.668-12.552,P=0.003)were the independent risk factors for high-grade glioma.Deep white matter invasion(OR=5.437,95%CI:1.516-19.497,P=0.009)was the independent risk factor for Ki-67positive.None of these eight image features was the independent risk factor for EMA positive(P>0.05).Conclusion The VASARI features of high-and low-grade gliomas can be quantitatively extracted by conventional MRI,different grades can be identified pathological by multiple image features,and the expression of glioma Ki-67can be predicted by deep white matter invasion.

关 键 词：脑胶质瘤 KI-67 上皮膜抗原 常规MRI 伦勃朗视觉感受图像特征 胶质瘤分级 

分 类 号：R73[医药卫生—肿瘤]