论文部分内容阅读
目的探讨动脉增强分数(AEF)定量彩图对肝脏局灶性结节增生(FNH)与肝细胞癌(HCC)的鉴别诊断价值。方法回顾性搜集行三期MDCT增强扫描后经病理证实的15例FNH患者和18例HCC患者,利用CT Kinetics血流动力学软件计算获得AEF定量彩图。先由两名医师进行双盲法独立计算,分析两名医师计算AEF值的一致性。然后比较两组患者病灶实质部分和周围正常肝组织AEF值的组内差异,以及两组患者病灶实质部分AEF、AEFL/AEFN值的组间差异,并对AEFL/AEFN的鉴别诊断价值进行ROC曲线分析。结果两名医师计算AEF值的一致性良好(ICC=0.95,95%CI=0.92 to 0.97)。FNH组和HCC组病灶实质部分的AEF值[分别为(69.54±7.31)%、(55.63±4.17)%]均高于周围正常肝组织[分别为(44.16±2.63)%、(42.20±3.71)%],差异具有统计学意义(FNH组内t=13.90,P=0.000;HCC组内t=13.29,P=0.000)。FNH组病灶实质部分的AEF值显著高于HCC组(t=6.86,P=0.000);FNH组的AEFL/AEFN值(1.58±0.16)也高于HCC组(1.33±0.13),差异具有统计学意义(t=4.86,P<0.001)。AEFL/AEFN值鉴别FNH和HCC的ROC曲线下面积为0.930,当AEFL/AEFN=1.425时Youden指数最大,对应敏感性和特异性分别为86.7%、88.9%。结论 AEF定量彩图能反映FNH和HCC病灶的血流灌注特点,对两者的鉴别诊断具有重要的参考价值。
Objective To investigate the differential diagnosis of focal hepatic fibrosis (FNH) and hepatocellular carcinoma (HCC) with quantification of arterial augmentation score (AEF). Methods Fifteen patients with FNH confirmed by pathology and 18 patients with HCC were enrolled in the retrospective study. The quantitative AEF images were obtained by CT Kinetics hemodynamics software. Two physicians performed a double-blind, independent calculation and analyzed the consistency of the AEF values calculated by the two physicians. Then compare the intra-group differences of the AEF value of the parenchyma and the surrounding normal liver tissue between the two groups and between the two groups of patients with AEF, AEFL / AEFN value of the lesion, and ROC curves of differential diagnostic value of AEFL / AEFN analysis. Results The two physicians calculated good agreement with AEF values (ICC = 0.95, 95% CI = 0.92 to 0.97). The AEF values in the parenchyma of the FNH group and the HCC group were significantly higher than those in the surrounding normal liver tissue [(69.54 ± 7.31)% and (55.63 ± 4.17)%, respectively] (44.16 ± 2.63 and 42.20 ± 3.71, respectively) %], The difference was statistically significant (t = 13.90 in FNH group, P = 0.000; t = 13.29 in HCC group, P = 0.000). The AEF value in the FNH group was significantly higher than that in the HCC group (t = 6.86, P = 0.000). The AEFL / AEFN in the FNH group (1.58 ± 0.16) was also significantly higher than that in the HCC group (1.33 ± 0.13) Significance (t = 4.86, P <0.001). The area under the ROC curve of FNH and HCC with AEFL / AEFN value was 0.930. The Adenozyme index was the largest when AEFL / AEFN = 1.425, with a corresponding sensitivity and specificity of 86.7% and 88.9%, respectively. Conclusion AEF quantitative colorimetry can reflect the characteristics of perfusion of FNH and HCC lesions, which has important reference value for the differential diagnosis between the two.