Yoshizawa Eriko, Yamada Akira, Okajima Yukinori, Notake Tsuyoshi, Shimizu Akira, Soejima Yuji, Fujinaga Yasunari
Department of Radiology, Shinshu University School of Medicine, Matsumoto, Japan.
Medical Data Science Course, Shinshu University School of Medicine, Matsumoto, Japan.
Eur Radiol. 2025 Apr 18. doi: 10.1007/s00330-025-11582-5.
This study aimed to develop a clinically applicable harmonization method for the hepatocellular uptake index (HUI), a quantitative liver function index, using gadoxetate disodium-enhanced (EOB)-MRI, to ensure consistency across diverse MR systems.
This retrospective study, approved by our institutional review board, included consecutive patients who underwent three-dimensional gradient-echo T1-weighted EOB-MRI, HUI measurements, indocyanine green disappearance rate (ICG-PDR), and albumin-bilirubin linear predictor (ALBI-LP) between April 2011 and June 2024. Six different MR systems were used for HUI measurements. A harmonization method using ALBI-LP was developed and validated for estimating liver reserves corresponding to ICG-PDR through statistical analysis of residuals.
A total of 498 patients (mean age, 68.0 years ± 11.6; 320 men) were evaluated. A statistically significant linear correlation was observed between HUI, ICG-PDR, and ALBI-LP in each MR system, leading to the determination of conversion factors for HUI harmonization. The harmonizing equation, harmonized HUI (h-HUI) = HUI･(Slope2'/-1.425)･0.955, was derived, with Slope2' representing the regression slope between HUI and ALBI-LP for each MR system. The standard deviation of the estimation error for ICG-PDR was significantly smaller using h-HUI by ALBI-LP (0.051, [0.048-0.054]) compared to non-harmonized HUI (0.060, [0.056-0.063]) or ALBI-LP (0.060, [0.057-0.064]), and equivalent to h-HUI by ICG-PDR (0.051, [0.045, 0.055]).
The HUI harmonized by the ALBI-LP is a clinically applicable method for ensuring the comparability of MR devices in quantitative liver reserve prediction using gadoxetate disodium-enhanced MR imaging.
Question The accurate prediction of quantitative liver function by hepatocyte-specific contrast-enhanced MRI necessitates the harmonization of MR systems. However, no established method has yet been identified. Findings In quantitative hepatic function assessment, albumin-bilirubin linear predictor can be employed to achieve harmonization between MR systems equivalent to the indocyanine green clearance test. Clinical relevance Quantitative liver function, as measured by the indocyanine green clearance test, can be accurately estimated using the hepatocellular uptake index, harmonized with the albumin-bilirubin linear predictor, across diverse MR systems.
本研究旨在开发一种临床适用的肝细胞摄取指数（HUI）的标准化方法，HUI是一种定量肝功能指标，使用钆塞酸二钠增强（EOB）-MRI，以确保不同MR系统之间的一致性。
这项回顾性研究经我们机构审查委员会批准，纳入了2011年4月至2024年6月期间连续接受三维梯度回波T1加权EOB-MRI、HUI测量、吲哚菁绿消失率（ICG-PDR）和白蛋白-胆红素线性预测指标（ALBI-LP）检查的患者。使用六种不同的MR系统进行HUI测量。通过对残差的统计分析，开发并验证了一种使用ALBI-LP的标准化方法，用于估计与ICG-PDR相对应的肝脏储备。
共评估了498例患者（平均年龄68.0岁±11.6；320例男性）。在每个MR系统中，HUI、ICG-PDR和ALBI-LP之间观察到具有统计学意义的线性相关性，从而确定了HUI标准化的转换因子。得出标准化方程，标准化HUI（h-HUI）=HUI·(Slope2'/-1.425)·0.955，其中Slope2'代表每个MR系统中HUI与ALBI-LP之间的回归斜率。与未标准化的HUI（0.060，[0.056-0.063]）或ALBI-LP（0.060，[0.057-0.064]）相比，使用基于ALBI-LP的h-HUI时，ICG-PDR估计误差的标准差显著更小（0.051，[0.048-0.054]），且与基于ICG-PDR的h-HUI（0.051，[0.045，0.055]）相当。
通过ALBI-LP标准化的HUI是一种临床适用的方法，可确保在使用钆塞酸二钠增强MR成像进行定量肝脏储备预测时，MR设备之间具有可比性。
问题 肝细胞特异性对比增强MRI对定量肝功能的准确预测需要MR系统的标准化。然而，尚未确定既定方法。发现 在定量肝功能评估中，白蛋白-胆红素线性预测指标可用于实现MR系统之间的标准化，等同于吲哚菁绿清除试验。临床意义 使用与白蛋白-胆红素线性预测指标标准化的肝细胞摄取指数，可在不同MR系统中准确估计吲哚菁绿清除试验所测量的定量肝功能。
