P.94 - Stimulated echo DTI of skeletal muscle in Becker muscular dystrophy: a pilot study

Abstract

Diffusion tensor imaging (DTI) is an MRI technique that aims to probe muscle architecture. The technique could be relevant in neuromuscular disorders (NMDs) where differences in fiber size and increased membrane permeability occur. Two approaches are relevant, spin-echo (SE-)DTI and stimulated-echo (STE-)DTI. While STE-DTI has a lower signal to noise ratio (SNR), it allows probing of the long diffusion distances present in skeletal muscle, which SE-DTI cannot. STE-DTI may thus be more sensitive, but has not been applied in NMD, as fat reduces the amount of water signal per voxel, and thus water SNR. We evaluated the feasibility of STE-DTI with fat suppression and signal denoising in Becker muscular dystrophy (BMD) patients. BMD patients (n = 7, age 21–58) and controls (n = 6, age 33–60) were scanned at 3T. Fat suppressed STE-DTI images, (resolution = 2x2x6 mm3) water/fat images (3-point Dixon, resolution = 1x1x10mm3), and noise maps were acquired. DTI data was denoised before computation of fractional anisotropy (FA) and mean diffusivity (MD) maps. FA, MD, fat fraction (FF), and SNR were measured in 6 muscles of the lower leg. In controls, at low SNR < 7, FA values were overestimated. Therefore, data sets with an SNR < 7 (24/42 BMD muscles) were excluded, which corresponded to muscles with significantly higher FF (41 ± 26% vs 10 ± 8 %, p = 0.001). The tibialis anterior (TA) had the highest SNR, allowing the comparison of 7 BMD and 4 controls. FA was higher in BMD (0.30 ± 0.05) than in controls (0.24 ± 0.02, p = 0.016). Sample size was insufficient for group comparison in other muscles. While STE-DTI MRI can be promising, maintaining sufficient SNR is of high importance, especially when imaging muscles with high FF. We were able to use STE and long diffusion time in low fat BMD muscles, likely owing to the use of signal denoising. High FA in the TA could be due to smaller fiber size or fibrosis. Future studies should focus on increasing the SNR, and exploring multiple diffusion times.