摘要目的:探讨钆塞酸二钠增强MRI检查在肝癌大块肝切除术中评估肝功能的价值。方法:采用单中心回顾性研究的方法。收集北京大学第三医院普通外科2016年6月至2019年6月收治的212例诊断为肝癌拟行大块肝切除患者的临床资料，所有纳入患者均采用钆塞酸二钠增强MRI进行术前评估，其中男性135例，女性77例，年龄（63.1±10.3）岁（范围：43~78岁），体重指数（23.9±3.6）kg/m 2（范围：17.8~33.1 kg/m 2）。所有患者肝功能Child-Pugh分级为A级，其中合并肝炎病史141例，单纯乙肝患者128例，单纯丙肝患者11例，两者均有2例。由两名医师共同收集并分析钆塞酸二钠增强MRI的影像学资料，计算增强MRI中感兴趣区域的相对强化比率。采用相关系数分析相对强化比率与肝脏切除术后肝功能衰竭发生风险的相关性。采用受试者工作特征（ROC）曲线评估相对强化比率在预测肝切除术后肝功能衰竭风险的诊断效能。 结果:术后共有42例（19.8%）患者符合国际肝脏外科研究小组（ISGLS）肝切除术后肝功能衰竭标准，其中A级31例，B级9例，C级2例。相对强化比率与总体患者术后肝功能衰竭及肝切除术后严重肝功能衰竭风险存在显著相关性（ P<0.05）。ROC曲线评估结果显示，钆塞酸二钠增强MRI相对强化比率与术后肝功能衰竭风险的ROC曲线下面积为0.818，灵敏度72.9%，特异度83.3%，临界值73.5%，95 %CI为0.75~0.89；相对强化比率与ISGLS B~C级肝功能衰竭风险的ROC曲线下面积为0.924，灵敏度97.0%，特异度90.9%，临界值61.5%，95 %CI为0.79~0.90。 结论:对肝癌拟行大块肝切除的患者，术前应用钆塞酸二钠增强MRI可以良好地评估肝脏储备功能，并可较好地预测肝切除术后肝功能衰竭的风险。

abstractsObjective:To investigate the value of Gd-EOB-DTPA-enhanced MRI in evaluating liver function and predicting the risk of post-hepatoectomy liver failure in patients with major resection of liver cancer.Methods:A total number of 212 patients were included from June 2016 to June 2019 at Department of General Surgery, Peking University Third Hospital with a retrospectively collected data.All patients underwent Gd-EOB-DTPA-enhanced MRI for diagnosis and preoperative evaluation of liver function.There were 135 males and 77 females, with age of (63.1±10.3) years old (range: 18 to 86 years old) . Relative enhancement ratio (RER) of the region of interest on Gd-EOB-DTPA-enhanced MRI was acquired by two independent researcher and then conducted the comparison of RER among the patients with or without post-hepatoectomy liver failure (PHLF) .Preoperative evaluation demonstrated that 141 cases infected by hepatitis virus, 128 cases with hepatitis B alone and 11 cases with hepatitis C alone, 2 cases had both of hepatitis B and C, and all patients were grade A judged by Child-Pugh score. The relationship between RER and PHLF was evaluated by Pearson correlation analysis and the diagnostic value of RER in predicting PHLF was test by receiver operating characteristic curve.Results:PHLF occurred in 42 patients according to ISGLS standard. Among them, 31 cases had level A liver failure, 9 cases had level B liver failure and 2 had level C failure. There was a significant correlation between RER and overall level of PHLF and RER was also significantly associated with severe B to C level of PHLF ( P<0.05) .The further receiver operating characteristics curve analysis showed that the diagnostic accuracy of RER on overall PHLF was 0.818 (sensitivity 72.9%, specificity 83.3%, cut-off value 73.5%, 95 %CI: 0.75 to 0.887) and on severe PHLF was 0.924 (sensitivity 97.0%, specificity 90.9%, cut-off value: 61.5%, 95 %CI: 0.79 to 0.90) . Conclusion:For patients who planned to undergo major resection of liver cancer, preoperative Gd-EOB-DTPA-enhanced MRI can help with the assessment of liver function and predicting the risk for post-hepatectomy liver failure.