Abstract
Background: There is still debate about the relationship between white matter(WM) lesions and microstructural WM damage measures, such as fractional anisotropy(FA) and mean diffusivity(MD), and grey matter(GM) atrophy in multiple sclerosis(MS), showing inconclusive results in the literature. Multi-shell diffusion weighted imaging(DWI) also allows for neurite orientation dispersion and density imaging(NODDI), thereby providing more information on microstructural damage in the WM.
Aim: To investigate the predictive properties of NODDI in the connected tracts on thalamus atrophy
Methods: At two time points (interval 1.0±0.1 years), 31 subjects with RRMS underwent extensive MRI examination, including T1, FLAIR, and multi shell (interleaved 7 volumes b=0s/mm2, 29 volumes b=500s/mm2, 59 volumes b=2000s/mm2) DWI at 3T (GE Discovery MR750) in a single centre. Lesions were segmented with an optimized version of nicMSlesions and filled with LEAP, and FreeSurfer was used for longitudinal segmentation of left and right thalamus. Multi-shell bedpost and probabilistic tractography of the tracts connecting to the left and right thalami was performed on the first time-point, and registered via DTI-TK templates to the second time point. NODDI was used to obtain neurite density index(NDI) and orientation dispersion index(ODI) in the connecting tracts. FA and MD were calculated from the b=2000 shell. Forward linear regression analysis for annual percentage change in left and right thalamus volume was performed with age, gender, and baseline FA, MD, ODI and NDI of the connected WM tracts as independent factors.
Results: At baseline, subjects with RRMS (disease duration 3.6±1.3 years, 23% male) were 37±8 years old and had a median EDSS of 3.0 (IQR 2.5-3.5). Mean annual percentage change was -1.2±2.4% and -0.8±2.5% in left and right thalamus, respectively. Forward linear regression showed that lower baseline NDI in the connected tracts could predict more negative annual percentage change (i.e. atrophy) in left thalamus (adj. R2=0.129, p=0.027 with β=0.398), but not right thalamus. FA or MD did not have a predictive value.
Conclusion: Lower baseline NDI in the tracts connected to the left thalamus can predict relative thalamic atrophy over time, whereas FA and MD could not, showing the importance of NODDI measures and multi-shell diffusion imaging in addition to basic DTI measures FA and MD. Results should be replicated over a longer follow-up period and for more GM structures.
Disclosure
Funding: Dutch MS Research Foundation, grant number 14-876.
Aim: To investigate the predictive properties of NODDI in the connected tracts on thalamus atrophy
Methods: At two time points (interval 1.0±0.1 years), 31 subjects with RRMS underwent extensive MRI examination, including T1, FLAIR, and multi shell (interleaved 7 volumes b=0s/mm2, 29 volumes b=500s/mm2, 59 volumes b=2000s/mm2) DWI at 3T (GE Discovery MR750) in a single centre. Lesions were segmented with an optimized version of nicMSlesions and filled with LEAP, and FreeSurfer was used for longitudinal segmentation of left and right thalamus. Multi-shell bedpost and probabilistic tractography of the tracts connecting to the left and right thalami was performed on the first time-point, and registered via DTI-TK templates to the second time point. NODDI was used to obtain neurite density index(NDI) and orientation dispersion index(ODI) in the connecting tracts. FA and MD were calculated from the b=2000 shell. Forward linear regression analysis for annual percentage change in left and right thalamus volume was performed with age, gender, and baseline FA, MD, ODI and NDI of the connected WM tracts as independent factors.
Results: At baseline, subjects with RRMS (disease duration 3.6±1.3 years, 23% male) were 37±8 years old and had a median EDSS of 3.0 (IQR 2.5-3.5). Mean annual percentage change was -1.2±2.4% and -0.8±2.5% in left and right thalamus, respectively. Forward linear regression showed that lower baseline NDI in the connected tracts could predict more negative annual percentage change (i.e. atrophy) in left thalamus (adj. R2=0.129, p=0.027 with β=0.398), but not right thalamus. FA or MD did not have a predictive value.
Conclusion: Lower baseline NDI in the tracts connected to the left thalamus can predict relative thalamic atrophy over time, whereas FA and MD could not, showing the importance of NODDI measures and multi-shell diffusion imaging in addition to basic DTI measures FA and MD. Results should be replicated over a longer follow-up period and for more GM structures.
Disclosure
Funding: Dutch MS Research Foundation, grant number 14-876.
| Original language | English |
|---|---|
| Journal | MULTIPLE SCLEROSIS JOURNAL |
| Publication status | Published - Oct 2021 |