CT能谱成像在肝细胞肝癌与局灶性结节增生鉴别诊断中的应用研究

Differentiation of Hepatocellular Carcinoma from Focal Nodular Hyperplasia with CT Spectral Imaging

目的探讨CT能谱成像在肝细胞肝癌(HCC)与局灶性结节增生(FNH)中的各能谱参数方面的差异,评估CT能谱成像在HCC与FNH鉴别诊断中临床应用价值。方法搜集行能谱CT扫描且经手术病理证实的31例HCC及7例FNH患者,分别测量其病灶于动脉期和门静脉期从40～140 keV的每隔10 keV的单能量平均CT值,标准差(即病灶的不均质性SD)及碘浓度(IC),并计算病灶能谱曲线的斜率(S),动脉期标准化碘浓度(NIC)及病灶碘摄取比值(IUR)。对上述各定量参数结果行非参数统计学分析,用中位数±四分位数间距表示,组间比较行Mann-Whitney秩和检验分析,并应用受试者工作特征(ROC)曲线分析各定量参数鉴别诊断HCC与FNH的能力。P<0.05为差异有统计学意义。结果动脉期及门静脉期FNH的40～140 keV能量段各平均单能量CT值均显著高于HCC(P<0.05),且两者间IUR_(动脉期),IUR_(门静脉期),S_(动脉期),S_(门静脉期)及NIC_(动脉期)的差异具有统计学意义(P<0.05)。鉴别HCC与FNH的最佳能谱定量参数为门静脉期100 keV的平均CT值,阈值为77.38 HU,ROC曲线下面积(AUC)为0.949,鉴别两者的敏感度与特异度分别为100%和90.3%,其次为IUR_(动脉期),阈值为3.23,AUC为0.903,鉴别两者的敏感度与特异度分别为100%和71%。结论 CT能谱成像多参数分析有助于HCC与FNH的鉴别诊断,能提高检出效能和诊断准确性。

Objective To investigate the value of CT spectral imaging in differentiating hepatocellular carcinoma from focal nodular hyperplasia. Methods Thirty-one subjects with HCC and seven subjects with FNH who were confirmed by pathology underwent spectral CT scan.The mean CT values at 40-140 keV in AP and PP were measured to generate spectral curves.The slopes of spectral curves during AP(Sap),PP(Spp),the iodine concentration,water concentration derived from iodine-based material-decomposition images in AP(ICap),PP(ICpp,) were measured.Iodine concentration(IC) in AP was normalized to that in aorta(NICap).The iodine uptake ratio in AP(IURap),PP(IURpp) were calculated.Quantitative values were recorded as median±interquartile range with non-parametric analysis.Mann-Whitney test was used for comparisons of the parameters between HCC and FNH group.Receiver operating characteristic(ROC) curves were used to determine the diagnostic capability between HCC and FNH group.P<0.05 was considered to be of statistical significance. Results The mean CT values of FNH were significantly higher than HCC in both AP and PP.There were significant differences in IURap,IURpp,NICap,Sap,Spp between HCC and FNH.The CT values in PP and IURap had high sensitivity(100%,100%) and specificity(90.3%,71%) with the area under the ROC curves(0.949,0.903) and cutoff values(77.38 HU,3.23) respectively in differentiating HCC from FNH. Conclusion The mean CT values at 100 keV in PP and quantitative analysis of CT spectral data(especially IURap) could be helpful in differentiation of HCC from FNH.