Those interested in applications of magnetic resonance imaging (MRI) of the spinal cord, particularly after spinal cord injury.Trauma to the spinal cord leads to a level dependent immediate functional loss affecting the white matter and grey matter at the lesion site. Recent MRI investigation assessing quantitatively the cross-sectional spinal cord area (SCA) have demonstrated that these changes also occur immediately above the level of lesion; and the magnitude of atrophy correlated with impairment¹. To date it is unclear to which degree changes to the structural integrity of the grey (GM) and white matter (WM) contribute to cervical cord atrophy and whether these remote changes do also occur below the level of lesion (i.e. lumbosacral enlargement). Recent advances in optimizing high-resolution imaging sequences and post-processing methods have made it now possible to robustly visualize and segment gray matter and dorsal column (DC) at the cervical cord and lumbosacral enlargement level². By applying these optimized MRI sequences to the cervical cord and lumbosacral enlargement we aimed to simultaneously characterize the magnitude of trauma-induced retrograde/anterograde fibre degeneration as well as spinal neuron atrophy and to compare the magnitude of change at both levels after chronic SCI.Structural MRI data was acquired on a clinical 3T Siemens Skyra system using a high resolution (0.25 x 0.25 x 2.5 mm³) multi-echo data image combination (MEDIC) sequence that provides high image contrast between GM and WM. Following parameters were applied in five patients and five healthy controls: field of view (FOV) of 162x192 mm², time of repetition (TR) of 44 ms, time of echo (TE) of 19 ms and flip angle α=11. 8 slices of 5 mm thickness centred at the cervical and the lumbosacral enlargement and perpendicular to the cord have been extracted from the MEDIC images. SCA, GM and DC were segmented using a combination of active surface model and fuzzy connector segmentation as implemented in JIM 6.0 (Xinapse systems, www.xinapse.com). Contours were manually edited where necessary. Subsequently, mean SCA, GM and DC areas were calculated and averaged across all slices. We used STATA to assess group differences with a two-sample t-test (p<0.05).All SCI patients had tetraplegia and two of them were classified AIS A (i.e. complete). Results are given in mean values (±SEM) and illustrated in Figure 2. Lower SCA, GM and DC areas were observed in the patient group at both spinal levels. At the cervical level, cord area was lower by 16.9% (controls: 86.3 (±6.3) mm2 vs patients: 71.7 (±4.1) mm², p=0.046), the GM area by 13.1% (16.9 (±0.3) mm² vs 14.7 (±0.3) mm², p<0.001), the DC area by 22.9% (22.2 (±0.3) mm² vs 17.1 (±1.6) mm, p=0.032). At the lumbar level, cord area was smaller by 17.1% (controls: 64.0 (±3.6) mm² vs patients: 53.0 (±4.3) mm², p=0.044), the DC area by 42.5% (14.7 (±1.6) mm² vs 8.4 (±0.9) mm², p=0.011) and the GM area was the same (18.1 (±1.5) mm² vs 17.9 (±2.0) mm², p=n.s.).This pilot study shows for the first time the feasibility of simultaneously segmenting the injured spinal cord at both the cervical and lumbar cord level. Besides a significantly reduced cord and dorsal column area above (i.e. cervical) and below the injury (i.e. lumbar), we provide detailed quantitative insights into spatially localized degenerative retrograde/anterograde as well as neuronal changes within the spinal WM and GM across the spinal axis, respectively. Diffusion weighted imaging acquired at the same level will provide more detail into the tissue integrity underlying these pathological processes in future analyses.The Wellcome Trust Centre for Neuroimaging has an institutional research agreement with and receives support from Siemens Healthcare.