AUTHORS: Rahmanzadeh R, Lu PJ, Weigel M, Galbusera R, La Rosa F, Bach Cuadra M, Schiavi S, Radue EW, Kuhle J, Kappos L, Granziera C

Multiple Sclerosis Journal, 25(7): 683-684, June 2019

ABSTRACT

Introduction: 

Multiple sclerosis (MS) is characterized by areas of focal and diffuse pathology in the central nervous system.
T1 relaxometry (qT1) maps have been shown to be sensitive to diffuse white matter (WM) abnormalities in MS patients. The increase in T1 may be compatible with a number of pathological alterations e.g. myelin and axonal damage.

Objectives: 

This study aimed to investigate the extent of axonal damage in areas of abnormal qT1 in MS patients by using a voxel-wise analysis.

Methods:

 Thirty-two MS patients (22 relapsing-remitting MS and 10 progressive MS) and 20 healthy controls underwent magnetic resonance imaging (MRI) at 3T; qT1 maps and multi-shell diffusion data were acquired. Neurite orientation dispersion and density imaging (NODDI) parameteric maps (neurite density index- NDI and orientation dispersion index- ODI) were computed on diffusion data. Lesion masks were obtained using an automatic method followed by manual correction. Normal appearing WM (NAWM) was obtained by subtracting the lesion mask from a whole brain WM segmentation produced by Freesurfer. NAWM maps were co-registered to a reference brain. In NAWM we (1) excluded voxels that were not present in at least 50 percent of subjects and (2) filled missing data with the group mean value of voxels present in group subjects. Then we computed a voxel-wise comparison of qT1, NDI and ODI maps between patients and controls by using the randomise tool of FSL with Threshold-Free Cluster Enhancement (TFCE) clustering. P values less than 0.01 were considered significant.

Results: 

T1 relaxation times were significantly increased in 15.98% of voxels compared to controls throughout the NAWM. Remarkably, NDI was lower in 50.14% of NAWM voxels and NDI abnormalities were diffuse to the whole brain. 35.72% of voxels with increased T1 relaxation time had reduced NDI. No abnormalities were observed in ODI maps of patients compared to controls.

Conclusion:

 In accordance with previous studies, we showed that T1 relaxation times were increased throughout the NAWM. Extending existing literature, we have provided new evidence that part of those abnormalities is probably due to decreased axonal density in non-lesion areas. Interestingly, NDI maps revealed that the loss of axonal integrity is substantially more extended than the one measured by qT1, suggesting that the combination of qT1 mapping and NODDI should be considered to gather a more comprehensive quantification of NAWM abnormalities in MS.

BibTex

Module: