QUANTITATIVE SPINAL CORD MRI IN MULTIPLE SCLEROSIS

Background: The spinal cord (SC) is a compact structure with a functional anatomic organization that makes it an ideal substrate to study structure-function relationships in multiple sclerosis (MS). Despite these benefits, the SC has been challenging to study using conventional MRI measures due to limited clinical-radiological correlations and technical difficulties. Quantitative MRI techniques in the SC, including diffusion tensor imaging (DTI) and magnetization-transfer imaging have demonstrated increased sensitivity to underlying tissue microstructural properties in comparison to conventional MRI techniques, and have the potential to provide new insights into the extent and type of tissue damage mediating clinical disability in MS tissue. Methods: 133 MS patients underwent 3-tesla cervical SC-MRI, clinical assessment, and optical coherence tomography on an annual basis over a median duration of 726 days. Quantitative SC-MRI indices, including SC-cross-sectional area (CSA), fractional anisotropy (FA), mean diffusivity (MD), perpendicular diffusivity (λ⊥), parallel diffusivity (λ||), and magnetization-transfer ratio (MTR) were calculated across segments C3-C4. Cross-sectional relationships between clinical disability, SC-MRI indices, and retinal layers were assessed. Longitudinal change in MRI-indices over the follow-up period, and its relationship to clinical disability were assessed. The impact of SC volume normalization by subject height, SC length, and intracranial volume on the ability to detect group differences and clinical-radiological correlations were assessed. Results: Quantitative SC-MRI indices demonstrate independent associations with system-specific and global clinical dysfunction beyond what can be detected by conventional MRI. Significant correlations exist between SC-MRI and retinal layers, and both exhibit independent relationships with clinical dysfunction. There are measurable longitudinal changes in SC-MRI indices, and subject-specific trajectories of SC-MRI index change are relevant to disability at follow-up. SC normalization by SC length was consistently the best strategy to accentuate group differences and to strengthen clinical-radiological correlations. Interpretation: Quantitative SC-MRI measures provide clinically relevant information in MS patients beyond that which can be gleaned from conventional, lesion-based measurements alone. Longitudinal changes in SC MRI-measures are detectable over 2 years, and subject-specific trajectories of SC-MRI index change are relevant to disability progression, warranting further investigation. Further development of quantitative SC-MRI techniques will expand their practical utility in the clinical setting