Spinal cord injury (SCI) is characterized by loss of motor and sensory function below the level of injury and affects about 273,000 persons in the United States¹. Despite the relatively low incidence of pediatric SCI², mortality and cost of life long care are high³. Studies of treatment effectiveness are lacking largely due to small clinical population and lack of valid outcome instruments that yield reliable scores in children. The precision of the neurological clinical evaluation is poor in children under 6 years of age⁴. In response to the limitations of existing methods, we have developed and tested an inner field-of-view (FOV) diffusion tensor imaging (DTI) as a method to evaluate the spinal cord (SC) in typically developing (TD) and children with SCI. Few studies have examined the utility of DTI in children^5,6, but none of these studies have examined the diffusion characteristics along the entire cervical and thoracic SC. The purpose of this study was to (a) investigate the feasibility of obtaining reliable DTI parameters along the entire cervical and thoracic SC in TD healthy children and children with SCI using an inner FOV sequence, (b) examine the reproducibility of DTI parameters, (c) determine whether microstructural changes quantified by DTI are associated with clinical neurological deficits.Subjects: Twenty-two TD children (mean age, 11.03yrs) without evidence of SC pathology and 15 patients (mean age, 11.42yrs) with chronic SCI were recruited. Written informed child assent and parent consent were obtained under the protocol approved by Institutional review board. The International Standards for Neurological Classification of SCI (ISNCSCI) were used to define the clinical level and severity of injury in SCI patients. Imaging: Subjects underwent 2 identical scans (minimum time between scans=2h) using 3.0T Verio MR scanner (Siemens, Erlangen, Germany) with 4-channel neck matrix and 8-channel spine matrix coils. The protocol consisted of conventional T1- and T2-weighted structural scans and axial DTI scans based on inner FOV sequence described previously⁷. Manual shim volume adjustments were also performed prior to data acquisition. DTI images were acquired axially using 2 overlapping slabs, to cover the cervical (C1-upper thoracic region) and thoracic (upper thoracic-L1) SC. The imaging parameters included: 3 averages of 20 diffusion directions, 6 b0 acquisitions, b=800s/mm², voxel size=0.8x0.8x6mm³, axial slices=40, TR=7900ms, TE=110ms, and acquisition time=8:49min and no gating. Data Analysis: A central mask was applied to the raw DTI images to eliminate the anatomy outside the SC. A mean b0 image was calculated, generated from the co-registration of all 6 b0 acquisitions. The diffusion weighted images were corrected for motion using a rigid body correction algorithm⁸. After motion correction, tensor estimation was done on a voxel-by-voxel basis from the axial DTI images using in-house software developed in MATLAB. For robust diffusion tensor estimation, a non-linear fitting algorithm using an implementation of the RESTORE technique was used⁹. Regions of interest were manually drawn on the whole cord on grayscale fractional anisotropy (FA) maps at every axial slice along the cervical and thoracic SC for both scans. DTI parameters were quantified at each intervertebral disk level and mid-vertebral body level of the cervical and thoracic SC in all subjects. Statistical Analysis: Analysis of covariance for repeated measures was performed to compare data from TD and SCI. Test-retest reliability was calculated using the intra-class correlation coefficient according to method of Shrout and Fleiss¹⁰. Association between DTI and ISNCSCI values were evaluated by Spearman partial correlation coefficients. A p value ≤ 0.05 was considered statistically significant.The images obtained with inner FOV sequence showed excellent delineation of both cervical and thoracic SC with minimal distortions (Fig. 1). FA values were significantly lower while radial diffusivity (RD) was significantly higher along the cervical and thoracic SC in patients with SCI compared to TD, however, mean diffusivity (MD) and axial diffusivity (AD) values were not statistically significant (Table 1, Fig. 2). There was a strong reliability for all DTI parameters along the cervical and thoracic SC in all subjects (Table 2). MD, AD and RD showed the greatest number of correlations with ISNCSCI followed by FA indicating that better neurological function is associated with greater unidirectional diffusion (Table 3).This is the first study in pediatric subjects investigating the DTI values and it’s reproducibility along the entire cervical and thoracic SC regions. The observed significant DTI changes between TD and SCI suggest that FA and RD appears to be the most sensitive parameter in assessing the state of SC in chronic phase of SCI. This study demonstrates that DTI has a potential to be used as an imaging biomarker for evaluating the extent of injury, which may be useful to prognosticate as well as monitor patients with SCI.