DTI全肿瘤直方图鉴别肿块型肝内胆管细胞癌与单发少血供肝转移瘤的价值

赵  莹1; 刘爱连1; 陈丽华1; 郭  妍2; 李  烨1; 宋清伟1; 牛  淼1; 杨伟萍1

doi:10.12117/jccmi.2020.04.008

中国临床医学影像杂志 ›› 2020, Vol. 31 ›› Issue (4) : 261-266. DOI: 10.12117/jccmi.2020.04.008

赵莹1，刘爱连1，陈丽华1，郭妍2，李烨1，宋清伟1，牛淼1，杨伟萍1

作者信息 +

Value of whole-tumor histogram analysis of diffusion tensor imaging in differentiating intrahepatic mass-forming cholangiocarcinoma and solitary hypovascular hepatic metastases

ZHAO Ying1, LIU Ai-lian1, CHEN Li-hua1, GUO Yan2, LI Ye1, SONG Qing-wei1, NIU Miao1, YANG Wei-ping1

Author information +

文章历史 +

摘要

目的：探讨DTI定量参数ADC和FA信号强度的全肿瘤直方图鉴别肿块型肝内胆管细胞癌（Intrahepatic mass-forming cholangiocarcinoma，IMCC）与单发少血供肝转移瘤（Solitary hypovascular hepatic metastases，SHHM）的价值。方法：回顾性收集我院接受肝脏MRI检查的患者资料，经手术病理证实的IMCC 24例，经病理或影像随访证实的SHHM 29例。在GE AW 4.6工作站，使用Functool软件重建DTI图像，生成ADC及FA图，测量两组病灶的ADC和FA值。将ADC及FA图导入Omni-Kinetics软件（GE Healthcare），在包含肿瘤实质的层面逐一勾画感兴趣区（Regions of interest，ROIs），可自动生成覆盖全肿瘤的ADC和FA信号强度的各直方图参数值，包括最小值、最大值、平均值、第10、25、50、75、90百分位数、标准差、平均偏差、相对偏差、偏度、峰度。比较两组的ADC及FA值、ADC及FA信号强度直方图各参数值的差异。应用ROC曲线分析诊断效能。结果：IMCC组与SHHM组的ADC及FA值差异无统计学意义（P＞0.05）。IMCC组ADC信号强度的相对偏差大于SHHM组（13.76（5.44，26.87）×102和2.41（1.12，5.10）×102，P＜0.001）；IMCC组FA信号强度的平均偏差及相对偏差均大于SHHM组（78.84（70.22，115.37）和67.99（60.92，89.39），22.09（9.29，59.62）×102和3.16（1.19，11.32）×102，P＜0.05）；IMCC组FA信号强度的平均值和第10百分位数小于SHHM组（P＜0.05）。其余参数差异无统计学意义（P＞0.05）。ADC和FA信号强度的相对偏差鉴别IMCC和SHHM的曲线下面积较大，分别为0.828和0.848，灵敏度和特异度分别为79.25%（82.8%）和87.5%（69.0%）。结论：DTI的ADC及FA值难以鉴别肿块型肝内胆管细胞癌与单发少血供肝转移瘤，而ADC及FA信号强度的全肿瘤直方图分析方法可为两者的鉴别提供多种定量参数。

Abstract

Objective: To investigate the value of whole-tumor histogram analysis of apparent diffusion coefficient(ADC) and fractional anisotropy(FA) signal intensity derived from diffusion tensor imaging(DTI) in differentiating intrahepatic mass-forming cholangiocarcinoma(IMCC) and solitary hypovascular hepatic metastases(SHHM). Methods: The data of liver MR scanning in our hospital were retrospectively collected, 24 cases of IMCC were pathologically confirmed, and 29 cases of SHHM were confirmed by pathology or follow-up imaging. ADC and FA maps were derived using Functool software on GE AW4.6 workstation, where ADC and FA values were measured. ADC and FA maps were transferred to Omni-Kinetics software(GE Healthcare), and ROIs covering the entire tumor were drawn on each slice of ADC and FA signal intensity maps. Histogram related parameters based on ADC and FA signal intensity, including min intensity, max intensity, mean value, the 10th, 25th, 50th, 75th and 90th percentiles, standard deviation, mean deviation, relative deviation, skewness, and kurtosis, were generated automatically after 3D ROIs covering the whole tumor were delineated by the readers. Comparison of the above parameters between the two groups was tested. Receiver operating characteristic(ROC) curves were plotted to analyze diagnostic efficiency. Results: The ADC and FA values between IMCC and SHHM groups were not statistically different(P>0.05). There was a significant difference in relative deviation of ADC signal intensity between the IMCC(13.76(5.44, 26.87)×102 and SHHM (2.41(1.12, 5.10)×102 groups(P<0.05). Moreover, a significant difference was observed in mean deviation and relative deviation of FA signal intensity between the IMCC(78.84(70.22, 115.37) and 22.09(9.29, 59.62)×102) group and SHHM (67.99(60.92, 89.39) and 3.16(1.19, 11.32)×102) group(P<0.05). The mean value and 10th percentile of FA signal intensity of IMCC group were statistically lower than those of SHHM group(P<0.05). The remaining parameters were not statistically different between two groups. The area under ROC curve of relative deviation of ADC and FA signal intensity were 0.828 and 0.848, respectively. The sensitivity/specifcity of ADC and FA were 79.2%(82.8%) and 87.5%(69.0%), respectively. Conclusion: Routine ADC and FA values of DTI are not efficient for identifying IMCC and SHHM. However, whole-tumor histogram analysis method based on ADC and FA signal intensity can provide multiple quantitative parameters for differentiating IMCC and SHHM.

导出引用

赵莹1，刘爱连1，陈丽华1，郭妍2，李烨1，宋清伟1，牛淼1，杨伟萍1. DTI全肿瘤直方图鉴别肿块型肝内胆管细胞癌与单发少血供肝转移瘤的价值[J]. 中国临床医学影像杂志. 2020, 31(4): 261-266 https://doi.org/10.12117/jccmi.2020.04.008

ZHAO Ying1, LIU Ai-lian1, CHEN Li-hua1, GUO Yan2, LI Ye1, SONG Qing-wei1, NIU Miao1, YANG Wei-ping1. Value of whole-tumor histogram analysis of diffusion tensor imaging in differentiating intrahepatic mass-forming cholangiocarcinoma and solitary hypovascular hepatic metastases[J]. Journal of China Clinic Medical Imaging. 2020, 31(4): 261-266 https://doi.org/10.12117/jccmi.2020.04.008

中图分类号： Ｒ735.7 Ｒ445.2

参考文献

[1]Khan SA, Thomas HC, Davidson BR. et al. Cholangiocarcinoma[J]. Lancet, 2005, 366(9493): 1303-1314. [2]Maithel SK, Gamblin TC, Kamek I, et al. Multidisciplinary approaches to intrahepatic cholangiocarcinoma[J]. Cancer, 2013, 119(22): 3929-3942. [3] 中华医学会外科学分会胃肠外科学组，中华医学会外科学分会结直肠外科学组，中国抗癌协会大肠癌专业委员会，等. 中国结直肠癌肝转移诊断和综合治疗指南（2018版）[J]. 中华消化外科杂志，2018，17（6）： 527-539. [4]Qian H, Li S, Ji M, et al. MRI characteristics for the differential diagnosis of benign and malignant small solitary hypovascular hepatic nodules[J]. Eur J Gastroenterol Hepatol, 2016, 28(7): 749-756. [5]陈丽华，刘爱连，宋清伟，等. 磁共振扩散张量成像鉴别诊断肝内胆管细胞癌与肝细胞癌[J]. 中国医学影像技术，2017，33（7）：993-997. [6]Hu XX, Yang ZX, Liang HY, et al. Whole-tumor MRI histogram analyses of hepatocellular carcinoma: Correlations with Ki-67 labeling index[J]. J Magn Reson Imaging, 2017, 46(2): 383-392. [7]Joo I, Lee JM, Yoon JH. Imaging Diagnosis of Intrahepatic and Perihilar Cholangiocarcinoma: Recent Advances and Challenges[J]. Radiology, 2018, 288(1): 7-13. [8]Ringe KI, Husarik DB, Sirlin CB, et al. Gadoxetate disodium-enhanced MRI of the liver: part 1, protocol optimization and lesion appearance in the noncirrhotic liver[J]. AJR, 2010, 195(1): 13-28. [9]Lazaridis KN, Gores GJ. Cholangiocarcinoma[J]. Gastroenterology, 2005, 128(6): 1655-1667. [10]Chung YE, Kim MJ, Park YN, et al. Varying appearances of cholangiocarcinoma: radiologic-pathologic correlation[J]. Radio-Graphics, 2009, 29(3): 683-700. [11]Vanderveen KA, Hussain HK. Magnetic Resonance Imaging of cholangiocarcinoma[J]. Cancer Imaging, 2004, 4(2): 104-115. [12]Lincke T, Zech CJ. Liver metastases: Detection and staging[J]. Eur J Radiol, 2017, 97: 76-82. [13]Faletti R, Battisti G, Discalzi A, et al. Can DW-MRI, with its ADC values, be a reliable predictor of biopsy outcome in patients with suspected prostate cancer?[J]. Abdom Radiol(NY), 2016, 41(5): 926-933. [14]陈丽华，刘爱连，马春梅，等. DTI评估大鼠肝纤维化及早期肝硬化的实验研究[J]. 磁共振成像，2014，5（5）：367-371. [15]Tosun M, Inan N, Sarisoya HT, et al. Diagnostic performance of conventional diffusion weighted imaging and diffusion tensor imaging for the liver fibrosis and inflammation[J]. Eur J Radiol, 2013, 82(2): 203-207. [16]王叶，吴南洲，王佳，等. 扩散加权成像在鉴别肝脏转移瘤和肝内胆管细胞癌中的价值[J]. 中国医学影像学杂志，2013，21（3）：214-217. [17]Moriya T, Saito K, Tajima Y, et al. 3D analysis of apparent diffusion coefficient histograms in hepatocellular carcinoma: correlation with histological grade[J]. Cancer Imaging, 2017, 17(1): 1. [18]Chandarana H, Rosenkrantz AB, Mussi TC, et al. Histogram analysis of whole-lesion enhancement in differentiating clear cell from papillary subtype of renal cell cancer[J]. Radiology, 2012, 265(3): 790-798. [19]Woo S, Cho JY, Kim SY, et al. Histogram analysis of apparent diffusion coeffcient map of diffusion-weighted MRI in endometrial cancer: a preliminary correlation study with histological grade[J]. Acta Radiol, 2014, 55(10): 1270-1277. [20]Cho SH, Kim GC, Jang YJ, et al. Locally advanced rectal cancer: post-chemoradiotherapy ADC histogram analysis for predicting a complete response[J]. Acta Radiol, 2015, 56(9): 1042-1050. [21]宋敏，李子平，孙灿辉，等. 结直肠癌MSCT表现与病理分型关系的探讨[J]. 影像诊断与介入放射学，2008，17（4）：162-166.