ADC全瘤直方图鉴别颅内孤立性纤维瘤/血管外皮瘤与血管瘤型脑膜瘤

目的:通过与局部感兴趣区法(ROI)进行比较,探讨基于全瘤体的表观扩散系数(ADC)直方图分析方法在鉴别颅内孤立性纤维瘤/血管周细胞瘤(SFT/HPC)与血管瘤型脑膜瘤(AM)中的价值。方法:回顾性搜集经手术病理确诊的12例SFT/HPC及16例AM患者的临床和MRI资料。DWI数据经单指数模型进行后处理可获得ADC图,分别采用局部及全瘤ROI直方图分析方法测量肿瘤的ADC值以及ADC直方图参数,包括ADCmean、ADCmedian、ADCmax、ADCmin及第5、10、25、75、90、95百分位数(ADCP5、ADCP10、ADCP25、ADCP75、ADCP90及ADCP95)。采用两独立样本t检验比较两组间各参数的差异,并采用ROC曲线评估其诊断效能。结果:采用局部ROI法测量的ADC值在SFT/HPC与AM间差异无统计学意义(P=0.097)。全瘤直方图分析中,SFT/HPC组中ADCmin(P=0.001)及ADCP5(P=0.045)低于AM组(P<0.05),两组间性别、年龄、发病部位及其它直方图参数的差异均无统计学意义(P>0.05)。ROC曲线分析显示,ADCmin鉴别SFT/HPC与AM的诊断效能最高,曲线下面积为0.86,以569.00×10^-6mm^2/s为截断值时,敏感度为81.3%,特异度为83.3%。结论:基于局部ROI法测量的ADC值不能有效鉴别SFT/HPC与AM,而ADC图全瘤直方图分析法可用于鉴别两类肿瘤,其中以ADCmin的诊断价值较高。

Whole-tumor histogram analysis of ADC in differentiating intracranial solitary fibrous tumor/hemangiopericytoma from angiomatous meningioma

Objective:The purpose was to assess the role of whole-tumor histogram analysis based on apparent diffusion coefficient(ADC)maps in differentiating intracranial solitary fibrous tumor/hemangiopericytoma(SFT/HPC)from angiomatous meningioma(AM).Methods:Pathologically confirmed intracranial SFT/HPC(n=12)and AM(n=16)were retrospectively collected.Diffusion-weighted(DW)images were postprocessed with the mono-exponential model.First,ADCLR value in solid tumor based on local regions of interest(ROI)were measured.Next,whole-tumor ROI were drawn on all slices of the ADC maps to obtain histogram parameters,including the mean ADC(ADCmean),median ADC(ADCmedian),maximum ADC(ADCmax),minimum ADC(ADCmin),as well as the 5th,10th,25th,75th,90th and 95th percentile ADC(ADCP5,ADCP10,ADCP25,ADCP75,ADCP90 and ADCP95).The differences of ADC histogram parameters between SFT/HPC and AM were compared using unpaired t-test.Receiver operating characteristic(ROC)curves analysis was used to determine the diagnostic value of each significant parameter.Results:No significant difference was found on mean ADC value based partial ROI between the two entities(P=0.097).The ADCmin(P=0.001)and ADCP5(P=0.045)were significantly lower in SFT/HPCs than those in AMs,while no significant difference was found on gender,age,tumor location,ADCmean,ADCmedian,ADCmax,ADCP10,ADCP25,ADCP75,ADCP90 and ADCP95 between the two entities(all P>0.05).ROC curves analysis revealed that ADCmin demonstrated the highest area under curves(AUC=0.86)in differentiating SFT/HPC from AM,with optimal cutoff value of 569.00×10^-6mm^2/s,the corresponding sensitivity and specificity were 81.3%and 83.3%,respectively.Conclusion:ADC value of the tumor measured based on local ROI can’t diffe-rentiate between SFT/HPC and AM,while histogram analysis of ADC maps based on whole tumor may be a useful tool for differential diagnosis.The ADCmin value is the most potential parameter in differentiating these two entities.