Contributions
Abstract: P809
Type: Poster Sessions
Abstract Category: Pathology and pathogenesis of MS - MRI and PET
Introduction: development of MR techniques sensitive to spinal cord (SC) pathology is mandatory to better understand the mechanisms underlying disability in Multiple Sclerosis (MS). Inhomogeneous Magnetization Transfer (ihMT) is a new technique particularly specific to myelin and appears sensitive to MS-related brain pathology.
Aims: assessment of the sensitivity of ihMT to MS-related pathology of SC. To do so, we used a multi-parametric MR protocol with high-resolution anatomical imaging, Diffusion Tensor imaging (DTI) and MT/ihMT combined with dedicated SC templates and atlases, allowing for automatic metrics quantification such as radial diffusivity (RD), MT ratio (MTR) and ihMT ratio (ihMTR).
Methods: 20 MS patients (14 remittent MS and 6 secondary progressive MS) and 20 age-matched healthy controls (HC) underwent the MR protocol at 3T. Clinical assessment included Expanded Disability Status Scale, Medical Research Council Scale, Time Walked Test, Nine-Hole Peg Test. Post-processing, based on the SC Toolbox, included registration to the MNI-Poly-AMU and AMU40 templates for SC, white and gray matter (WM/GM) automatic segmentation, cross sectional area (CSA) measurements from C1 to C6 cervical levels and metrics quantification within specific WM tracts and anterior GM at C2 and C5. Normal appearing tissue (NAT) and manually-segmented lesions were analyzed separately. Accuracy of WM/GM AMU40 segmentation in MS was also evaluated, in regards to manual segmentation.
Results: AMU40 showed satisfactory results for WM/GM segmentation (0.89 and 0.75 Dice score). Decrease of WM and GM CSA (13.4% p< 0.0001 and 4.3% p< 0.05 in average, respectively) was observed in MS compared to HC, suggesting WM and GM atrophies. All DTI/ihMT metrics were altered in patients compared to HC with more important ihMTR variations in anterior GM and WM (-23.4% and -24.9% respectively p< 0.0001) compared to MTR (-8.5% and -7.7% p< 0.01) and RD (+8.9% and +20.4% p< 0,01). Lesions were also associated with more significant ihMTR changes than NAT (-33% vs. -21% in WM p< 0.001). Finally, stronger correlations with all clinical scores were evidenced with ihMTR compared to all other metrics. In conclusion, this study highlights the higher ihMT sensitivity toward SC microstructural changes in both NAT and lesions as compared to conventional MT and DTI. Combination of ihMT with automatic GM/WM SC segmentation is promising to further study the topography and progression of SC pathology in MS.
Disclosure: The authors have nothing to disclose.
Abstract: P809
Type: Poster Sessions
Abstract Category: Pathology and pathogenesis of MS - MRI and PET
Introduction: development of MR techniques sensitive to spinal cord (SC) pathology is mandatory to better understand the mechanisms underlying disability in Multiple Sclerosis (MS). Inhomogeneous Magnetization Transfer (ihMT) is a new technique particularly specific to myelin and appears sensitive to MS-related brain pathology.
Aims: assessment of the sensitivity of ihMT to MS-related pathology of SC. To do so, we used a multi-parametric MR protocol with high-resolution anatomical imaging, Diffusion Tensor imaging (DTI) and MT/ihMT combined with dedicated SC templates and atlases, allowing for automatic metrics quantification such as radial diffusivity (RD), MT ratio (MTR) and ihMT ratio (ihMTR).
Methods: 20 MS patients (14 remittent MS and 6 secondary progressive MS) and 20 age-matched healthy controls (HC) underwent the MR protocol at 3T. Clinical assessment included Expanded Disability Status Scale, Medical Research Council Scale, Time Walked Test, Nine-Hole Peg Test. Post-processing, based on the SC Toolbox, included registration to the MNI-Poly-AMU and AMU40 templates for SC, white and gray matter (WM/GM) automatic segmentation, cross sectional area (CSA) measurements from C1 to C6 cervical levels and metrics quantification within specific WM tracts and anterior GM at C2 and C5. Normal appearing tissue (NAT) and manually-segmented lesions were analyzed separately. Accuracy of WM/GM AMU40 segmentation in MS was also evaluated, in regards to manual segmentation.
Results: AMU40 showed satisfactory results for WM/GM segmentation (0.89 and 0.75 Dice score). Decrease of WM and GM CSA (13.4% p< 0.0001 and 4.3% p< 0.05 in average, respectively) was observed in MS compared to HC, suggesting WM and GM atrophies. All DTI/ihMT metrics were altered in patients compared to HC with more important ihMTR variations in anterior GM and WM (-23.4% and -24.9% respectively p< 0.0001) compared to MTR (-8.5% and -7.7% p< 0.01) and RD (+8.9% and +20.4% p< 0,01). Lesions were also associated with more significant ihMTR changes than NAT (-33% vs. -21% in WM p< 0.001). Finally, stronger correlations with all clinical scores were evidenced with ihMTR compared to all other metrics. In conclusion, this study highlights the higher ihMT sensitivity toward SC microstructural changes in both NAT and lesions as compared to conventional MT and DTI. Combination of ihMT with automatic GM/WM SC segmentation is promising to further study the topography and progression of SC pathology in MS.
Disclosure: The authors have nothing to disclose.