LI-RADSv2018在肝细胞癌微血管侵犯中的预测研究:基于钆贝葡胺增强MRI

Prediction of Microvascular Invasion in Hepatocellular Carcinoma by LI-RADSv2018:Based on Gd-BOPTA Enhanced MRI

目的:探讨基于钆贝葡胺(Gd-BOPTA)增强MRI的肝脏成像报告和数据系统(LI-RADS)对肝细胞癌(HCC)患者微血管侵犯(MVI)的预测价值。方法:该回顾性研究纳入了2016年1月至2018年12月期间在我院接受根治性切除术和Gd-BOPTA增强MRI的196例HCC患者。根据LI-RADSv2018成像特征进行MRI特征评估,采用单因素和多因素逻辑回归分析确定MVI的独立危险因素,并构建列线图模型、绘制受试者工作特征(ROC)曲线、计算曲线下面积(AUC)、采用校准曲线评估模型。结果:经单因素和多因素逻辑回归分析确定了与MVI显著相关的独立危险因素:甲胎蛋白(AFP)>20 ng/mL(OR=4.122, 95%CI1.785~9.516)、病灶含脂(OR=0.079, 95%CI 0.022~0.285)、晕状强化(OR=7.072, 95%CI 2.634~18.992)、非光滑边缘(OR=3.877, 95%CI 1.727~8.700)、瘤内动脉(OR=4.278, 95%CI 1.516~12.072)、肝胆期瘤周低信号(OR=3.758, 95%CI 1.461~9.664)。基于以上独立因子构建列线图模型,绘制ROC显示曲线下面积(AUC)为0.850 (95%CI 0.791~0.909),该模型显示出较好的诊断效能,校准曲线结果表明该列线图预测的MVI概率与实际的MVI概率一致。结论:基于Gd-BOPTA增强MRI可在术前有效地预测肝细胞癌的微血管侵犯,为临床决策和个性化治疗计划的制订提供重要信息。

Purpose:To assess the predictive value of microvascular invasion(MVI)in patients with hepatocellular carcinoma(HCC)using gadobenate dimeglumine(Gd-BOPTA)enhanced MRI based on the liver imaging reporting and data system(LI-RADS).Methods:The data from 196 HCC patients who underwent resection and Gd-BOPTA enhanced MRI between January 2016 and December 2018 in our hospital were retrospectively analyzed.MRI features were evaluated according to LI-RADSv2018 criteria.Logistic regression analyses were performed to determine the predictive significance of these features for MVI.A nomogram predictive model was developed,and its performance was assessed using receiver operating characteristic(ROC)curve analysis.Results:Univariate and multivariate logistic regression analyses identified independent risk factors significantly associated with MVI:AFP>20 ng/mL(OR=4.122,95%CI 1.785-9.516),fat presence in mass(OR=0.079,95%CI 0.022-0.285),coronal enhancement(OR=7.072,95%CI 2.634-18.992),non-smooth margin(OR=3.877,95%CI 1.727-8.700),internal arteries(OR=4.278,95%CI 1.516-12.072),and peritumoral hypointensity on hepatobiliary phase(OR=3.758,95%CI 1.461-9.664).Using these factors,a nomogram predictive model was constructed.The ROC curvedemonstrated an area under the curve of 0.850(95%CI 0.791-0.909),indicating strong diagnostic efficacy.Calibration curve results confirmed that the predicted MVI probability from the nomogram aligned well with actual MVI probability.Conclusion:Gd-BOPTA enhanced MRI is a valuable tool for predicting microvascular invasion in hepatocellular carcinoma patients prior to surgery.These findings support its role in guiding clinical decision and personalized treatment strategies.