Quantitative MRI for Monitoring Metabolic Dysfunction-Associated Steatotic Liver Disease: A Test-Retest Repeatability Study

Abstract

Background: Quantitative magnetic resonance imaging metrics iron-corrected T1 (cT1) and liver fat from proton density fat-fraction (PDFF) are both commonly used as noninvasive biomarkers for metabolic dysfunction-associated steatohepatitis (MASH); however, their repeatability in this population has rarely been characterized.

Purpose: To quantify the variability of cT1 and liver fat fraction from PDFF in patients with biopsy-confirmed metabolic dysfunction-associated steatotic liver disease (MASLD) and MASH.

Study type: Prospective, single center.

Population: Twenty-one participants (female = 11, mean age 53 ± 24 years) with biopsy-confirmed MASLD, including 6 with MASH and fibrosis ≥2.

Field strength/sequence: 3 T; T1 and T2* mapping for the generation of cT1 (shMOLLI: CardioMaps and 2D MDE, T1map-FIESTA and LMS MOST: StarMap, 2D Multi-Echo FSPGR) and magnitude-only PDFF sequence for liver fat quantification (LMS IDEAL: StarMap, 2D Multi-Echo FSPGR).

Assessment: T1 mapping and PDFF scans were performed twice on the same day for all participants (N = 21), with an additional scan 2-4 weeks later for MASH patients with fibrosis ≥2 (N = 6). Whole liver segmentation masks were generated semi-automatically and average pixel counts within these masks were used for the calculation of cT1 and liver fat fraction.

Statistical tests: Bland-Altman analysis for repeatability coefficient (RC) and 95% limits of agreement (LOA) and intraclass correlation coefficient (ICC).

Results: Same-day RC was 32.1 msec (95% LOA: -36.6 to 24.2 msec) for cT1 and 0.6% (95% LOA: -0.5% to 0.7%) for liver fat fraction; the ICCs were 0.98 (0.96-0.99) and 1.0, respectively. Short-term RC was 65.2 msec (95% LOA: -63.8 to 76.5 msec) for cT1 and 2.6% (95% LOA: -2.8% to 3.1%) for liver fat fraction.

Data conclusion: In participants with MASLD and MASH, cT1 and liver fat fraction measurements show excellent test-retest repeatability, supporting their use in monitoring MASLD and MASH.

Level of evidence: 2 TECHNICAL EFFICACY: Stage 2.

Keywords: MASH; MASLD; PDFF; cT1; multiparametric‐MRI.

Figure 1

Consort diagram of subject omissions due to missing data for all phases of the study.

Figure 2

Example cT1 and PDFF maps of a patient with MASLD and no fibrosis and an example patient with confirmed MASH and fibrosis of 2.

Figure 3

Bland‐altman plots showing cT1 and PDFF difference between baseline and follow‐up on the same day (N=21), top row; Baseline and follow‐up on a different day (N=6), bottom row. The dashed lines represent the upper and lower 95% limits of agreement and the mean difference.