Harmonization of quantitative liver function evaluation using gadoxetate disodium-enhanced magnetic resonance imaging

Area of Science:

• Radiology and Imaging
• Hepatology
• Medical Physics

Background:

• Quantitative liver function assessment is crucial for clinical decision-making.
• Gadoxetate disodium-enhanced MRI offers a non-invasive method to evaluate liver function via the hepatocellular uptake index (HUI).
• Variability in MR systems hinders the comparability of HUI measurements, necessitating harmonization.

Purpose of the Study:

• To develop and validate a clinically applicable harmonization method for the hepatocellular uptake index (HUI) in gadoxetate disodium-enhanced MRI.
• To ensure consistent and comparable quantitative liver function assessment across diverse MR imaging systems.
• To establish a harmonization technique that correlates HUI with established measures of liver reserve, such as indocyanine green disappearance rate (ICG-PDR).

Main Methods:

• A retrospective study included 498 patients undergoing EOB-MRI, HUI, ICG-PDR, and albumin-bilirubin linear predictor (ALBI-LP) measurements on six different MR systems.
• A harmonization method was developed using ALBI-LP to estimate liver reserves corresponding to ICG-PDR.
• Statistical analysis of residuals was employed to validate the harmonization method and derive conversion factors for HUI.

Main Results:

• A statistically significant linear correlation was found between HUI, ICG-PDR, and ALBI-LP across all MR systems.
• A harmonizing equation (h-HUI = HUI･(Slope2'/-1.425)･0.955) was derived, where Slope2' is the regression slope between HUI and ALBI-LP for each system.
• Harmonized HUI (h-HUI) using ALBI-LP demonstrated a significantly smaller estimation error for ICG-PDR compared to non-harmonized HUI, achieving comparability with h-HUI derived from ICG-PDR.

Conclusions:

• The ALBI-LP-based harmonization of HUI provides a clinically applicable method for standardizing quantitative liver reserve prediction using EOB-MRI.
• This harmonization ensures the comparability of MR devices, enabling reliable liver function assessment across different imaging platforms.
• The developed method allows for accurate estimation of quantitative liver function, equivalent to the indocyanine green clearance test, using harmonized HUI across diverse MR systems.