Comparison of gadoxetate disodium-enhanced MRI sequences for measuring hepatic observation size and its implication of LI-RADS classification.

Purpose: We aimed to determine the optimal image sequence for measurement of hepatic observations on gadoxetate disodium-enhanced MRI in comparison with pathologic measurement, and to evaluate its clinical impact on the Liver Imaging Reporting and Data System (LI-RADS) v2018 classification.

Methods: Two hundred and fifty-three patients (279 hepatic observations) who underwent gadoxetate disodium-enhanced MRI and subsequent hepatectomy were retrospectively included. Two radiologists independently evaluated the visualization score (five-point scale) and size of each observation on six MRI sequences (T1-weighted, T2-weighted, arterial-phase, portal venous-phase, transitional-phase [TP], and hepatobiliary-phase [HBP] images) and assigned a LI-RADS category. Correlations between MRI and pathologic measurements were evaluated using Pearson correlation coefficients. A repeated measures analysis of variance with Bonferroni post hoc comparison tests was used to compare the visualization scores and absolute differences between MRI sequences and pathologic measurements. The LI-RADS classification according the size measurement of each MRI sequence was compared using Cochran's Q test with a post hoc McNemar's test.

Results: Of the MRI sequences, HBP had the highest visualization score (4.1 ± 0.6) and correlation coefficient (r = 0.965). The absolute difference between MRI and pathologic measurement was lowest on TP (2.3 mm ± 2.2), followed by HBP (2.4 mm ± 2.1). In the LI-RADS classifications, HBP did not have any non-visible observations. Regarding LR-3, LR-4, and LR-5, there was no significantly different LI-RADS classification among the six MRI sequences (p ≥ 0.122).

Conclusion: Hepatobiliary-phase images are clinically useful for measuring hepatic observations on gadoxetate disodium-enhanced MRI, especially regarding visibility and correlation with pathologic findings.