河北医科大学学报 ›› 2025, Vol. 46 ›› Issue (6): 656-661.doi: 10.3969/j.issn.1007-3205.2025.06.006

• • 上一篇    下一篇

术前MRI特征联合碱性磷酸酶与前白蛋白比值在肝细胞癌微血管侵犯预测中的价值探讨

  

  1. 1.四川省遂宁市中医院放射科,四川 遂宁 062900;2.四川省遂宁市中心医院放射科,
    四川 遂宁 062900;3.四川省遂宁市第三人民医院放射科,四川 遂宁 062900

  • 出版日期:2025-06-25 发布日期:2025-07-04
  • 作者简介:魏晋(1990-),男,四川兴文人,四川省遂宁市中医院主治医师,医学学士,从事腹部影像诊断研究。

  • 基金资助:
    四川省科技计划项目(2021YFG0129)

The value of preoperative MRI characteristics combined with the ratio of alkaline phosphatase (ALP) to prealbumin (PA) in predicting microvascular invasion in hepatocellular carcinoma

  1. 1.Department of Radiology, Suining Hospital of Traditional Chinese Medicine, Sichuan Province, Suining 
    062900, China; 2.Department of Radiology, Suining Central Hospital, Sichuan Province, 
    Suining 062900, China; 3.Department of Radiology, the Third People′s Hospital of 
    Suining City, Sichuan Province, Suining 062900, China

  • Online:2025-06-25 Published:2025-07-04

摘要: 目的探讨术前MRI特征联合碱性磷酸酶(alkaline phosphatase,ALP)与前白蛋白(pro-albumin,PA)比值在肝细胞癌(hepatocellular carcinoma,HCC)微血管侵犯预测中的价值。
方法回顾性分析2022年1月—2023年10月于四川省遂宁市中医院接受手术治疗的HCC患者152例的病历资料。以术后病理为金标准,将受试者分为微血管侵犯组、非侵犯组。分析HCC微血管侵犯的影响因素,评估术前磁共振(magnetic resonance imaging,MRI)特征、ALP/PA比值预测HCC微血管侵犯效能。
结果152例HCC患者术后经病理确诊微血管侵犯54例,占比35.53%。侵犯组和非侵犯组的CNLC分期Ⅱ期占比分别为63.27%、36.73%,肿瘤包膜不完整占比分别为77.78%、59.18%,T2WI肿瘤信号不均匀占比分别为74.07%、40.82%,肝胆期瘤周低信号占比分别为74.07%、48.98%,ALP/PA分别为(0.72±0.16)、(0.54±0.13),差异有统计学意义(P<0.05)。CNLC分期(OR=6.001,95%CI:2.753~13.082)、肿瘤包膜不完整(OR=4.080,95%CI:1.913~8.702)、T2WI肿瘤信号不均匀(OR=4.660,95%CI:2.246~9.667)、肝胆期瘤周低信号(OR=4.375,95%CI:2.069~9.252)、ALP/PA(OR=3.904,95%CI:1.830~8.328)是HCC微血管侵犯的危险因素(P<0.05)。肿瘤包膜不完整、T2WI肿瘤信号不均匀、肝胆期瘤周低信号、ALP/PA四者联合预测HCC微血管侵犯的曲线下面积为0.842(95%CI:0.761~0.922)。
结论术前MRI特征与ALP/PA比值联合预测HCC微血管侵犯的效能良好。


关键词: 癌, 肝细胞, 微血管侵犯, 磁共振特征

Abstract: Objective To explore the value of preoperative MRI characteristics combined with the ratio of alkaline phosphatase (ALP) to prealbumin (PA) in predicting microvascular invasion in hepatocellular carcinoma. 
Methods A retrospective analysis was performed on medical records of 152 patients with hepatocellular carcinoma (HCC) who underwent surgical treatment at Suining Hospital of Traditional Chinese Medicine from January 2022 to October 2023. Using postoperative pathology as the gold standard, the subjects were divided into the microvascular invasion group and the non-invasion group. Factors influencing microvascular invasion in HCC were analyzed, and the predictive efficacy of preoperative magnetic resonance imaging (MRI) characteristics and the ratio of ALP to PA for microvascular invasion of HCC was evaluated. 
Results Among the 152 HCC patients, 54 patients were confirmed to have microvascular invasion postoperatively, accounting for 35.53%. The proportion of CNLC stage Ⅱ in the invasion and non-invasion groups was 63.27% and 36.73%, respectively, the proportion of incomplete tumor capsule was 77.78% and 59.18%, respectively, and the proportion of heterogeneous tumor signal on T2WI was 74.07% and 40.82%, respectively; the proportion of peritumoral hypointensity in the hepatobiliary phase was 74.07% and 48.98%, respectively, and the ALP/PA scores were (0.72±0.16) and (0.54±0.13), and the difference was significant (P<0.05). CNLC staging (OR=6.001, 95%CI: 2.753-13.082), incomplete tumor capsule (OR=4.080, 95%CI: 1.913-8.702), heterogeneous tumor signal on T2WI (OR=4.660, 95%CI: 2.246-9.667), peritumoral hypointensity in the hepatobiliary phase (OR=4.375, 95%CI: 2.069-9.252), and ALP/PA ratio (OR=3.904, 95%CI: 1.830-8.328) were identified as risk factors for microvascular invasion in HCC (P<0.05). The combined prediction of incomplete tumor capsule, heterogeneous tumor signal on T2WI, peritumoral hypointensity in the hepatobiliary phase, and ALP/PA ratio yielded an area under the curve of 0.842 (95%CI: 0.761-0.922) for predicting microvascular invasion in HCC. 
Conclusion Preoperative MRI characteristics combined with the ALP/PA ratio show good efficacy in predicting microvascular invasion in HCC. 


Key words: carcinoma, hepatocellular, microvascular invasion, magnetic resonance characteristics