基于MRI预测肿块型肝内胆管细胞癌分化程度的研究价值

Predicting the pathological differentiation grade of mass-forming intrahepatic cholangiocarcinoma based on MRI

目的:探讨磁共振成像(MRI)定性、定量参数在肿块型肝内胆管细胞癌细胞分化程度的预测价值。方法:回顾性分析2016年3月-2022年3月81例经手术病理证实为肿块型肝内胆管细胞癌(MICC)患者临床、病理、MRI资料。将患者分为中分化组(29例)和低分化组(52例),采用t检验或χ^(2)检验比较两组在临床、MRI定性以及定量参数方面的差异,将单因素分析中具有统计学差异的变量纳入Logistic回归进行分析用来建立MRI模型。受试者操作特征(ROC)曲线被用于分析模型的诊断效能。结果:单因素分析显示两组间肝内转移灶、动态强化模式、肝胆期强化模式、扩散加权成像(DWI)、直径的差异有统计学意义(P值分别为0.032、<0.001、0.035、0.017、0.001)。动态强化模式、直径是预测MICC分化程度的独立预测因子,所得MRI模型ROC曲线下面积0.806,敏感度71.2%,特异度79.3%。结论:动态强化模式、直径是预测MICC分化程度的独立危险因素,MRI模型具有良好的诊断效能。

Objective:To investigate the value of MR features and quantitative parameters for predicting the differentiation grade of mass-forming intrahepatic cholangiocarcinoma(MICC).Me-thods:The clinical,pathological and MRI data of 81 patients with surgical pathologically confirmed as MICC were retrospectively analyzed between March 2016 and March 2022.The patients were divided into moderately differentiated(n=29)and poor-differentiated group(n=52).The clinical,MRI features and quantitative parameters were compared by t-test or χ^(2) test.Univariate and multivariate logistic regression analysis were performed to establish MRI-based model.The receiver operating characteristics(ROC)curve was used to analyze the diagnostic performance.Results:Univariate analysis showed intrahepatic metastasis,dynamic enhancement pattern,hepatobiliary phase enhancement pattern,diffusion-weighted imaging(DWI)and diameter differed between two groups(P=0.032,<0.001,0.035,0.017 and 0.001,respectively).The dynamic enhancement patterns and diameter were independent predictors for MICC.The area under ROC curve,sensitivity and specificity of MRI model were 0.806,71.2%and 79.3%,respectively.Conclusion:Dynamic enhancement patterns and diameter are independent risk factors for predicting MICC and MRI-based model shows good diagnostic performance.