多b值ADC直方图评价高级别胶质瘤治疗后进展的初步研究

Evaluation of Tumor Progression of High-grade Glioma After Chemoradiotherapy by Histogram of Multiple b Value Diffusion Weighted Imaging

目的比较多b值扩散加权成像(DWI)中表观扩散系数(ADC)直方图第5百分位数(C_5)评价高级别胶质瘤(HGG)治疗后进展的差异,以确定最佳b值。资料与方法回顾性分析2014年9月-2015年10月河北医科大学第二医院经病理检查确诊并符合纳入标准的41例HGG术后放化疗患者。治疗前、后行增强T1WI、FLAIR和多b值DWI(b=0、1000、2000、3000 s/mm^2)。根据随访结果将患者分为进展组与非进展组,分别获得5组对应于FLAIR高信号区域的ADC直方图,累计频率法计算C_5。比较进展与非进展组不同ADC直方图C_5的差别,确定肿瘤进展的C_5阈值及其准确度。结果进展组所有ADC图C_5均低于非进展组,差异有统计学意义(P<0.01)。高b值ADC图C_5诊断HGG治疗后进展的准确度明显高于低b值者,5组[ADC_((1000/0))、ADC_((2000/0))、ADC_((3000/0))、ADC_((3000/1000))、ADC_((3000/2000))]C_5的受试者工作特性曲线下面积分别为0.717、0.832、0.909、0.933、0.937。当ADC_((3000/2000))图C_5取405.6 mm/s^2时,判断肿瘤进展的敏感度、特异度、阳性和阴性预测值分别为90.9%、89.7%、89.9%、91.0%。结论 ADC直方图C_5值可有效鉴别HGG治疗后进展情况,且高b值者具有更高的准确度。

Pupose To explore the role of histogram analysis of apparent diffusion coefficient（ADC） maps obtained from multiple b value diffusion weighted imaging（DWI） in the assessment of progression of high-grade glioma（HGG） treated with chemoradiotherapy so as to determine the optimal b value. Materials and Methods Forty-one consecutive patients with HGG proved histopathologically who had undergone concurrent chemoradiotherapy at Second Hospital of Hebei Medical University from September 2014 to October 2015 were enrolled in the study. All the subjects underwent diffusion weighted MR imaging before and after therapy with b values of 0, 1000, 2000 and 3000 s/mm^2. Based on the clinical and radiographic follow-ups, the patients were divided into progression and nonprogression groups. ADC maps were calculated according to hyperintense FLAIR lesions after completion of chemoradiotherapy. The fifth percentile（C5） in terms of cumulative histograms in different b-value ADC maps in multiple b value DWI was calculated, and the C5 of each ADC map between progression and non-progression groups was compared. Moreover, receiver operating characteristic analysis was used to determine the best cutoff values and diagnostic accuracy for predictors in the differentiation of true progression from non-progression. Results The C5 of all different b value ADC maps were significantly lower in the progression group than those in the non-progression group（P〈0.01）. In terms of the accuracy of assessing the progression after therapy, the C5 in the high b value ADC maps was significantly higher than that in the low b value ADC maps. The area under the receiver operating characteristic curve（AUC） of the C5 was 0.717, 0.832, 0.909, 0.933 and 0.937 respectively in the 5 ADC maps [ADC（（1000/0））, ADC（（2000/0））, ADC（（3000/0））, ADC（（3000/1000）） and ADC（（3000/2000））]. When the cutoff value of C5 was 405.6 mm/s^2 in ADC（（3000/2000）） map, the sensitivity, specificity, positive predictive value and negative predictive value were 90.9%, 89.7%, 89.9% and 91.0%, respectively. Conclusion The C5 in ADC map can effectively differentiate tumor progression of HGG, and that of high b values have higher accuracy.