合成MRI联合弥散加权成像评估胶质瘤级别及肿瘤细胞增殖活性

目的 观察合成MRI联合弥散加权成像（DWI）鉴别高、低级别胶质瘤及评估肿瘤细胞增殖活性的价值。方法 回顾性分析69例经病理学证实的胶质瘤患者资料,包括低级别胶质瘤（LGG）27例、高级别胶质瘤（HGG）42例,术前均接受颅脑MRI、合成MRI和DWI。检测肿瘤实质T1值、T2值、质子密度（PD）和表观弥散系数（ADC）,比较LGG与HGG间各参数差异。采用受试者工作特征（ROC）曲线评价各参数单独及联合鉴别LGG与HGG的效能,分析各参数与Ki-67标记指数（Ki-67 LI）的相关性。结果 LGG T1值及PD低于HGG （P均<0.001）,而ADC高于HGG （P<0.001）,二者T2值差异无统计学意义（P=0.328）。单一参数中,以ADC鉴别LGG与HGG的曲线下面积（AUC）为0.901,高于T1值（AUC=0.862）和PD （AUC=0.848）（P均>0.05）;T1值＋PD＋ADC的AUC （0.988）高于任一单一参数（P均<0.05）,其敏感度和特异度分别为92.86%和100%。肿瘤ADC与其Ki-67 LI呈中等负相关（r=-0.617,P<0.001）,T1值和PD与Ki-67 LI呈中等正相关（r=0.441、0.437,P均<0.001）,T2值与Ki-67 LI无明显相关性（r=-0.044,P=0.719）。结论 合成MRI联合DWI可无创评估胶质瘤级别及细胞增殖活性。

Synthetic MRI combined with diffusion weighted imaging for evaluation of glioma grade and tumor cell proliferation

Objective To explore the value of synthetic MRI combined with diffusion weighted imaging (DWI) for distinguishing high grade, low grade glioma and evaluating proliferative activity of tumor cells. Methods Data of 69 cases of histopathologically confirmed glioma, including 27 cases of low-grade glioma (LGG) and 42 cases of high-grade glioma (HGG) were retrospectively analyzed. All patients underwent head MRI, synthetic MRI and DWI before surgery. T1 value, T2 value, proton density (PD) and apparent diffusion coefficient (ADC) of tumor parenchyma were measured, and the differences of the above parameters were compared between LGG and HGG. The receiver operating characteristic (ROC) curve was used to evaluate the efficacy of single and combined parameters for distinguishing LGG and HGG. The correlations of parameters and Ki-67 label index (Ki-67 LI) were analyzed. Results T1 value and PD in LGG group were lower than those in HGG group (both P<0.001), while the ADC in LGG group was higher than that in HGG group (P<0.001). There was no significant difference in T2 value between LGG and HGG (P=0.328). Among the above parameters, the area under the curve (AUC) of single ADC for distinguishing LGG from HGG was 0.901, higher than that of single T1 value (AUC=0.862) and PD (AUC=0.848, both P>0.05). The AUC of T1 value+PD+ADC was 0.988, higher than that of any single parameter (all P<0.05), with sensitivity of 92.86% and specificity of 100%. There was a medium negative correlation between ADC and Ki-67 LI (r=-0.617, P<0.001), a medium positive correlation between T1 value and PD with Ki-67 LI (r=0.441, 0.437, both P<0.001), but no significant correlation between T2 value and Ki-67 LI of tumor (r=-0.044, P=0.719). Conclusion Synthetic MRI combined with DWI could noninvasively distinguish high grade and low grade glioma as well as evaluating proliferative activity of tumor cells.