钆塞酸二钠增强磁共振影像学特征联合定量分析在预测肝细胞癌病理分级中的价值

The predictive value of gadolinium-ethoxybenzyl-diethylenetriamine pentaacetic acid enhanced MRI imaging features combined with quantitative parameters for the pathologic grading of hepatocellular carcinoma

目的探讨钆塞酸二钠(Gd-EOB-DTPA)增强磁共振成像(MRI)影像学特征联合定量分析在预测肝细胞癌病理分级中的应用价值。方法回顾性分析2016年6月至2019年6月在苏州大学附属第一医院行Gd-EOB-DTPA增强MRI检查并经穿刺或手术病理证实为肝细胞癌的80例患者的资料,其中男65例、女15例,年龄30~74(59±11)岁。根据术后病理结果,将其分为低分化组(26例,男22例、女4例)及中高分化组(54例,男43例、女11例)。审阅Gd-EOB-DTPA增强MRI图像并记录肝细胞癌定性特征,测量并计算定量指标包括肿瘤大小和动脉期、门静脉期、平衡期、肝胆特异期病灶-肝脏信号比(SIR)及动脉期对比增强比(CER-AP)。对低分化组、中高分化组间临床资料及定性参数进行χ2检验,定量参数进行独立样本t检验。对上述组间差异有统计学意义的定性参数、定量参数、定量联合定性参数、甲胎蛋白(AFP)联合定性定量参数构建二元logistic回归模型,采用受试者工作特征(ROC)曲线分析4种多参数预测模型对肝细胞癌病理分级的诊断效能。结果低分化组与中高分化组间AFP、动脉期瘤周高信号、动脉期环形强化、肝胆特异期瘤周低信号、肿瘤信号的差异有统计学意义(均P<0.05),两组间,肿瘤大小、动脉期病灶-肝脏信号比(SIR-AP)、门脉期病灶-肝脏信号比(SIR-PP)分别为(7.0±3.7)cm比(3.9±2.4)cm、1.11±0.29比1.31±0.32、0.89±0.21比1.03±0.27,差异有统计学意义(均P<0.05)。定性参数、定量参数、定性联合定量参数、AFP联合定性定量参数预测模型诊断肝细胞癌病理分级的ROC曲线下面积(AUC)分别为0.805、0.804、0.855、0.892。定性参数与定性联合定量参数诊断的灵敏度的差异有统计学意义(80.8%比92.3%,P<0.05)。结论Gd-EOB-DTPA增强MRI影像学特征联合定量分析可在术前预测肝细胞癌病理分级,有较大的应用价值。

Objective To investigate the predictive value of gadolinium-ethoxybenzyl-diethylenetriamine pentaacetic acid(Gd-EOB-DTPA)enhanced MRI imaging features combined with quantitative parameters for the pathologic grading of hepatocellular carcinoma(HCC).Methods Eighty patients(65 males,15 females;range 30-74 years,average age(59±11)years old)with HCC who underwent curative resection or biopsy from June 2016 to June 2019 in the First Affiliated Hospital of Soochow University after Gd-EOB-DTPA-enhanced MRI were evaluated retrospectively.According to the pathological results of the postoperative pathology,eighty patients were divided into poorly differentiated group(26 cases,22 males,4 females)and moderately-well differentiated group(54 cases,43 males,11 females).In Gd-EOB-DTPA-enhanced MRI,qualitative parameters were assessed.Quantitative parameters including tumor size,tumor-to-liver signal intensity ratio(SIR)of arterial phase(AP),portal vein phase(PP),equilibrium phase(EP)and hepatobiliary phase(HBP),contrast enhancement ration AP(CER-AP)were measured and calculated.Clinical data and qualitative parameters between poorly differentiated group and moderately-well differentiated group were analyzed by usingχ2 test.Quantitative parameters were analyzed by using independent sample t test.Statistically significant qualitative parameters,quantitative parameters,qualitative combined with quantitative parameters and AFP combined with qualitative and quantitative parameters were included in binary logistic regression model.The receiver operating characteristic(ROC)curves were used to evaluate the diagnostic performance of the four prediction model for the pathologic grading of HCC.Results There was a statistical difference between poorly differentiated group and moderately-well differentiated group in AFP,arterial peritumoral enhancement,arterial rim enhancement,peritumoral hypointensity on HBP and tumor signal(all P<0.05).In two groups,tumor size,SIR-AP and SIR-PP were(7.0±3.7)vs(3.9±2.4)cm,1.11±0.29 vs 1.31±0.32 and 0.89±0.21 vs 1.03±0.27(all P<0.05).AUC of qualitative parameters,quantitative parameters,qualitative combined with quantitative parameters and AFP combined with qualitative and quantitative parameters were 0.805,0.804,0.855,0.892.There was a statistical difference between qualitative parameters and qualitative combined with quantitative parameters in sensitivity(80.8%vs 92.3%,P<0.05).Conclusion Gd-EOB-DTPA-enhanced MRI imaging features combined with quantitative parameters can be used to predict the pathologic grading of HCC preoperatively,which has a great applicative value.