Objectives: To develop a preoperative risk scoring system to predict microvascular invasion (MVI) in intrahepatic cholangiocarcinoma (ICCA) by incorporating clinical and gadoxetate-enhanced MRI features.

Materials and methods: We retrospectively enrolled 197 consecutive patients with ICCA who underwent preoperative gadoxetate-enhanced MRI and curative surgical resection between 2009 and 2016. The patients were randomly divided into a development set (n = 139) and a validation set (n = 58). Two radiologists independently reviewed the MRI features. A risk scoring system to predict MVI was developed using multivariable logistic regression analysis, and its diagnostic performance was validated. Recurrence-free survival (RFS) and overall survival (OS) were analyzed based on the MVI risk.

Results: Gadoxetate-enhanced MRI features that were independently associated with MVI included tumor multiplicity (odds ratio (OR) 3.37, 95% confidence interval (CI): 1.37-8.28, p = 0.008), arterial-phase peritumoral enhancement (OR 4.47, 95% CI: 1.87-10.70, p = 0.001), and hepatobiliary-phase peritumoral hypointensity (OR 2.96, 95% CI: 1.29-6.77, p = 0.010). Using these features, the area under the receiver operating characteristic curve of the scoring system was 0.80 (95% CI: 0.72-0.87) in the development cohort and 0.84 (95% CI: 0.74-0.94) in the validation cohort. ICCA patients at high risk for MVI exhibited significantly shorter RFS and OS compared to those at low risk for MVI in both cohorts (p ≤ 0.031).

Conclusion: The risk scoring system based on gadoxetate-enhanced MRI features effectively predicted the risk of MVI in patients with ICCA, facilitating the preoperative identification of patients who are likely to have poorer survival outcomes following curative resection.

Key points: Question Microvascular invasion (MVI) is a well-established adverse prognostic factor in intrahepatic cholangiocarcinoma (ICCA). However, the preoperative prediction of MVI using gadoxetate-enhanced MRI remains unexplored. Findings A risk scoring system based on gadoxetate-enhanced MRI was developed, effectively predicting MVI in ICCA and linking high MVI risk to significantly poorer survival outcomes following resection. Clinical releveance This scoring system enables the accurate assessment of MVI risk in ICCA and provides valuable prognostic insights for patients undergoing curative surgery. Its use can help clinicians develop personalized treatment strategies to improve patient outcomes.

Keywords: Prognosis; Cholangiocarcinoma; Contrast media; Liver neoplasms; Magnetic resonance imaging.

© 2025. The Author(s), under exclusive licence to European Society of Radiology.

目标：通过结合临床和钆塞酸增强MRI特征，开发一种预测肝内胆管癌（ICCA）微血管侵犯（MVI）的术前风险评分系统。

材料与方法：回顾性纳入2009年至2016年间接受术前钆塞酸增强MRI和根治性手术切除的连续197例肝内胆管癌患者。将患者随机分为开发组（n=139）和验证组（n=58）。两名放射科医生独立审查了MRI特征。使用多变量逻辑回归分析建立了预测MVI的风险评分系统，并对其诊断性能进行了验证。根据MVI风险分析无复发生存期（RFS）和总生存期（OS）。

结果：与MVI独立相关的钆塞酸增强MRI特征包括肿瘤多重性（优势比(OR) 3.37，95%置信区间(CI): 1.37-8.28, p = 0.008）、动脉期瘤周强化（OR 4.47，95% CI: 1.87-10.70, p = 0.001）和肝胆期瘤周低信号强度（OR 2.96，95% CI: 1.29-6.77, p = 0.010）。利用这些特征，在开发队列中评分系统的曲线下面积为0.80（95% CI: 0.72-0.87），在验证队列中为0.84（95% CI: 0.74-0.94）。高MVI风险的ICCA患者在两个队列中的无复发生存期和总生存期均显著短于低MVI风险的患者 (p ≤ 0.031)。

结论：基于钆塞酸增强MRI特征的风险评分系统可有效预测肝内胆管癌患者MVI的风险，有助于术前识别可能在根治性切除术后预后较差的患者。

关键点：问题：微血管侵犯（MVI）是肝内胆管癌（ICCA）公认的不良预后因素。然而，使用钆塞酸增强MRI术前预测MVI的研究尚未开展。发现：开发了一种基于钆塞酸增强MRI的风险评分系统，能够有效预测ICCA中的MVI，并将高MVI风险与根治术后显著较差的生存结果联系起来。临床相关性：此评分系统可准确评估ICCA患者的MVI风险，并为接受根治手术的患者提供有价值的预后信息。其使用有助于临床医生制定个性化的治疗策略，以改善患者结局。

关键词：预后；胆管癌；对比剂；肝脏肿瘤；磁共振成像。

© 2025. The Author(s), under exclusive licence to European Society of Radiology.