【摘要】 目的：探讨表观弥散系数（ADC）的直方图特征鉴别弥漫性胶质瘤高、低级别的价值，并评价其与Ki-67增殖指数的相关性。方法：回顾性分析67例经术后病理证实的弥漫性胶质瘤（其中高级别组48例，低级别组19例）的影像学及病理学资料。比较高、低级别组弥漫性胶质瘤ADC值的多个直方图特征差异，对满足正态分布及方差齐性的特征采用独立样本t检验进行分析，对不满足正态分布或方差不齐的特征采用Mann-Whitney秩和检验，然后绘制ROC曲线以评价其ADC值直方图特征鉴别高、低级别脑胶质瘤的效能。采用Pearson相关分析评价ADC值的直方图特征与Ki-67增殖指数的相关性。结果：高级别胶质瘤的ADC最大值、标准差及不均匀度分别为2.395 （2.359, 3.227）×10^-3mm²/s、0.410 （0.306, 0.534）×10^-3mm²/s、0.305±0.086，大于低级别胶质瘤相对应的直方图特征[分别为2.107 （1.927,2.546）×10^-3mm²/s、0.204（0.171,0.383）×10^-3mm²/s、0.203±0.114],2组间差异具有统计学意义（P值分别为0.025、0.001、0.000）。高级别胶质瘤的ADC值第10百分位数、第25百分位数分别为0.825 （0.749,0.991）×10^-3mm²/s、（1.068±0.246）×10^-3mm²/s，小于低级别胶质瘤相对应的直方图特征[分别为1.191 （0.863,1.313）×10^-3mm²/s、（1.225±0.297）×10^-3mm²/s],2组间差异具有统计学意义（P值分别为0.002、0.029）。ROC曲线示ADC值的不均匀度鉴别高、低级别胶质瘤的ROC曲线下面积最大（0.815），同时对鉴别诊断的灵敏度最高（95.8%），最大值、第10百分位数的特异度最高（均为68.4%）。Pearson相关性分析示ADC值的第10百分位数与Ki-67增殖指数呈负相关（r=-0.415,P=0.002），不均匀度与Ki-67增殖指数呈正相关（r=0.343,P=0.008）。结论：表观弥散系数的直方图特征有助于预测弥漫性胶质瘤的组织学分级、Ki-67增殖指数，对胶质瘤临床治疗方案的选取具有一定的价值。更多还原

【Abstract】 Purpose To explore the clinical value of histogram parameters of apparent diffusion coefficient(ADC in distinguishing high-and low-grade diffuse gliomas, and evaluate its correlation with Ki-67 proliferation index.Methods: A retrospective analysis of the imaging and pathological data of 67 cases of diffuse gliomas confirmed by postoperative pathology（48 cases in high-grade group and 19 cases in low-grade group） were analyzed. The independent samples t test（normal distribution and homogeneity of variance） or Mann-Whitney rank sum test（abnormal distribution or heterogeneity of variance） were used to compare the differences of histogram parameters of ADC values between high-and low-grade gliomas. Receiver operating characteristic（ROC） curve was used to evaluate the efficiency of histogram parameters of ADC value in distinguishing high-and low-grade glioma of the brain. Pearson correlation was used to assess associations between histogram parameters of ADC value and Ki-67 proliferation index. Results: The ADC maximum value, standard deviation and unevenness of high-grade glioma were 2.395（2.359, 3.227）×10^-3mm²/s, 0.410（0.306,0.534）×10^-3mm²/s, 0.305±0.086, which were greater than those of low-grade gliomas [2.107（1.927, 2.546） ×10^-3mm²/s, 0.204（0.171, 0.383） ×10^-3mm²/s, 0.203±0.114], the difference between the two groups was statistically significant（P values were 0.025, 0.001, 0.000, respectively）. The 10 th percentile and 25 th percentile of ADC values for high-grade gliomas were 0.825（0.749, 0.991） ×10^-3mm²/s,（1.068±0.246） ×10^-3mm²/s, which were smaller than those of low-grade gliomas [1.191（0.863, 1.313） ×10^-3mm²/s,（1.225±0.297） ×10^-3mm²/s], the difference between the two groups was statistically significant（P values are 0.002, 0.029,respectively）. In all histogram parameters of ADC value, inhomogeneity had the largest area of ROC curve（0.815）,and the highest sensitivity（95.8%） in differential high-and low-grade brain gliomas. The maximum and the 10 th percentile had the largest specificity（68.4%）. Pearson correlation analysis showed that the 10 th percentile of ADC value was negatively correlated with Ki-67 proliferation index（r=-0.415, P=0.002）, and inhomogeneity was positively correlated with Ki-67 proliferation index（r=0.343, P=0.008）. Conclusion: The histogram parameters of ADC value are helpful to predict the histological grade of glioma and Ki-67 proliferation index, and have certain value in the selection of clinical treatment plan for diffuse gliomas.更多还原