文献速递腹部技术与安全MR

钆塞酸增强MRI移行期肝实质信号:对廓清及LI-RADS诊断肝细胞癌效能的影响

Gadoxetate-enhanced MRI Transitional Phase Parenchymal Signal: Effect on Washout and LI-RADS Diagnostic Performance for Hepatocellular Carcinoma.
PubMed 文献 · Radiology · 2026-01-01

本回顾性研究纳入438例患者共558个肝脏病灶,比较钆塞酸增强MRI中三种廓清评估方式对LI-RADS诊断肝细胞癌效能的影响。将非周边廓清扩展至移行期、仅在移行期肝实质信号为低信号或等信号时判定廓清,较仅门静脉期判定廓清提高了敏感度(77.8% vs 70.7%,P < .001)且特异度无显著降低(96.2% vs 96.9%,P = .95)。移行期肝实质高信号与低体重指数、低胆红素及高心脏指数相关。

要点

  • 仅在移行期肝实质信号为低信号或等信号时将非周边廓清扩展至移行期,可提高LR-5诊断HCC的敏感度且特异度无显著降低。
  • 移行期肝实质呈高信号者占25.6%(112/438),等信号占51.6%(226/438),低信号占22.8%(100/438)。
  • 移行期肝实质高信号与低体重指数、低胆红素、高心脏指数相关,阅片时需注意其对廓清判定的影响。

核心结论

  • 在钆塞酸增强MRI中,将非周边廓清扩展至移行期并考虑移行期肝实质信号强度,可提高LI-RADS对肝细胞癌的诊断效能。

研究设计与样本

  • 回顾性研究,纳入2019年1月至2020年12月期间接受钆塞酸增强MRI并后续行肝切除或肝移植的患者438例(平均年龄60岁±9.1[SD],345例男性),共558个肝脏病灶,以手术病理为参考标准。

关键影像表现/征象

  • 移行期肝实质信号以门静脉为参照分为低信号、等信号和高信号,分别占22.8%(100/438)、51.6%(226/438)和25.6%(112/438)。

诊断标准与关键数值

  • 廓清判定定义为仅门静脉期(符合LI-RADS 2018版,PVP-only)、扩展至移行期 regardless of 移行期肝实质信号(TP-all)、仅当移行期肝实质为低信号或等信号时扩展至移行期(TP-iso/hypo)三种方式。
  • PVP-only的LR-5敏感度为70.7%(302/427),特异度为96.9%(127/131)。
  • TP-iso/hypo的LR-5敏感度为77.8%(332/427),较PVP-only提高(P < .001);特异度为96.2%(126/131),较PVP-only无显著降低(P = .95)。

临床应用与阅片要点

  • 阅片时在门静脉期评估非周边廓清的同时,可参照移行期肝实质信号:仅在移行期肝实质为低信号或等信号时将廓清扩展至移行期。
  • 移行期肝实质呈高信号时,将廓清扩展至移行期可能影响特异度,需结合移行期肝实质信号强度判断。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

Background At gadoxetate-enhanced MRI, extending nonperipheral washout to the transitional phase (TP) may reduce specificity for hepatocellular carcinoma (HCC) diagnosis. Purpose To compare the diagnostic performance of the Liver Imaging Reporting and Data System (LI-RADS) according to different washout assessments incorporating TP liver parenchymal signal intensity at gadoxetate-enhanced MRI and to identify factors associated with hyperintense TP liver parenchyma. Materials and Methods This retrospective study included patients who underwent gadoxetate-enhanced MRI and subsequent hepatic resection or liver transplant between January 2019 and December 2020. Signal intensity of TP liver parenchyma was assessed relative to the portal vein and categorized as hypo-, iso-, or hyperintense. Washout was defined as (a) confined to the portal venous phase (PVP) only, consistent with LI-RADS version 2018 (PVP-only); (b) extending to the TP regardless of TP liver parenchymal signal intensity (TP-all); and (c) extending to the TP only when TP liver parenchymal signal was hypointense or isointense (TP-iso/hypo). Sensitivity and specificity of LR-5 (ie, definitely HCC) were compared across these definitions using generalized estimating equations. Univariable and multivariable logistic regression analyses were performed to identify predictors of hyperintense TP parenchyma. Results A total of 438 patients (mean age, 60 years ± 9.1 [SD]; 345 men) with 558 hepatic observations were included; 25.6% of patients (112 of 438) showed hyperintense liver parenchyma in the TP, whereas 51.6% (226 of 438) showed isointense liver parenchyma and 22.8% (100 of 438) showed hypointense signal. Compared with PVP-only, TP-iso/hypo showed higher sensitivity (77.8% [332 of 427 patients] vs 70.7% [302 of 427]; P < .001) without evidence of a reduction in specificity (96.2% [126 of 131 patients] vs 96.9% [127 of 131]; P = .95). Low body mass index (adjusted odds ratio [OR], 0.89; P = .003), low bilirubin (adjusted OR, 0.62; P = .03), and high cardiac index (adjusted OR, 2.89; P = .03) were associated with hyperintense TP liver parenchyma. Conclusion Extending nonperipheral washout to the TP at gadoxetate-enhanced MRI while considering TP liver parenchymal signal intensity improved the diagnostic performance of LI-RADS for HCC. © RSNA, 2026 Supplemental material is available for this article. See also the editorial by Kim and Yoo in this issue.

来源:Radiology,查看原文。版权归原作者所有。

引用

  1. Cho SH, Jung YJ, Hong SB, et al. Gadoxetate-enhanced MRI Transitional Phase Parenchymal Signal: Effect on Washout and LI-RADS Diagnostic Performance for Hepatocellular Carcinoma. Radiology. 2026. PMID: 41914861. doi:10.1148/radiol.252892

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。