# Quantitative MRI Findings and Their Relationship to Muscle Histopathology and Ambulatory Clinical Function in Duchenne Muscular Dystrophy

## TL;DR

This study shows that MRI can accurately measure muscle fat in Duchenne muscular dystrophy and relates it to disease severity and walking ability.

## Contribution

The study demonstrates a strong correlation between MRI-measured fat fraction and histopathological findings in DMD, along with unique fat infiltration patterns.

## Key findings

- MRI fat fraction (FF) strongly correlates with histopathological fatty infiltration (ρ = 0.98, p < 0.001).
- FF in most lower limb muscles correlates with ambulatory function tests (ρ = 0.27–0.73).
- Thigh muscle FF progression follows non-linear patterns with an inflection point at 7.3 years in DMD.

## Abstract

Iterative decomposition of water and fat with echo asymmetry and least‐squares estimation quantitation (IDEAL‐IQ), a quantitative 6‐point Dixon magnetic resonance imaging (MRI) sequence, has been increasingly used for quantifying muscle fat fraction (FF) in neuromuscular disorders. However, its utility for correlating FF with disease severity and mapping spatiotemporal disease progression in Duchenne muscular dystrophy (DMD) requires further investigation.

In 10 patients with dystrophinopathies (seven DMD and three BMD) we correlated the muscle FF with the histopathological fatty infiltration. IDEAL‐IQ MRI images of 19 individual lower limb muscles were acquired from 133 DMD patients and 41 healthy controls. Ambulatory function tests, including three timed function tests (TFTs) and North Star Ambulatory Assessment (NSAA), were performed. We investigated the spatial distribution of fatty infiltration across five axial MRI slices. Disease progression patterns were analysed using a piecewise linear model and a normal cumulative distribution function model.

A significant positive correlation between muscle FF on MRI and histopathologic fatty infiltration was observed (ρ = 0.98, p < 0.001). A mild to moderate correlation was noted for the FF of 19 individual muscles, excluding the sartorius and gracilis muscles, with TFTs (ρ = 0.27–0.60, p < 0.001) and NSAA (ρ = −0.32–−0.73, p < 0.001). All the eight individual muscles, except the tibialis posterior muscle, showed an inhomogeneous fatty infiltration pattern with higher FF in muscle ends compared to bellies. Thigh muscle FF aligned well with both piecewise linear and sigmoidal models, showing non‐linear increases with disease duration. The piecewise linear model identified an average inflection point for thigh muscle FF at 7.3 (range, 5.2–7.6) years, with average annual FF increase of 1.7% before and 6.7% after the point in DMD. The average μ of DMD patients for thigh muscles was 11.4 years (±2.8), occurring 7.6 years earlier compared to BMD patients.

Our study demonstrates a significant relationship between muscle FF on IDEAL‐IQ MRI and histopathologic muscle fatty infiltration, as well as ambulatory clinical function in DMD. DMD patients exhibit a distinct and inhomogeneous fat infiltration pattern of lower limb muscles.