Vahe Michael Zohrabian MD, Presenter: Nothing to Disclose

Benjamin Morris Zussman BS, Abstract Co-Author: Nothing to Disclose

Jeffrey John Mathews MD, Abstract Co-Author: Nothing to Disclose

Andrea J. Frangos MPH, Abstract Co-Author: Nothing to Disclose

Richard Joseph Thomas Gorniak MD, Abstract Co-Author: Consultant, Bracco Group

John Lackey MS, Abstract Co-Author: Nothing to Disclose

M. Alex Dresner PhD, Abstract Co-Author: Employee, Koninklijke Philips Electronics NV

Ralph J. Marino MD, Abstract Co-Author: Research Consultant, Asubio Pharmaceuticals, Inc Research Consultant, Geron Corporation Research Consultant, Neuralstem, Inc Research Consultant, Novartis AG Research grant, Geron Corporation Research grant, Pfizer Inc

Song Lai PhD, Abstract Co-Author: Nothing to Disclose

Adam E. Flanders MD, Abstract Co-Author: Research Consultant, Geron Corporation

MR imaging of SCI reveals changes in proportional to neurologic deficit. Experimental evidence suggests that DTI may provide objective evidence of myelin integrity and could be exploited as an analog of neurologic deficit. The purpose of this study is to determine if DTI metrics could be used to discriminate grades of post-traumatic myelopathy in SCI.

31 cervical SCI patients (ASIA A-D) & 14 controls (ASIA E) were evaluated on a 1.5T MR (Philips Achieva) within 24 hours of injury. Standard anatomic axial/sagittal SE T1, T2 and GE imaging was performed. DTI of the entire cervical spinal cord was performed using 4 mm axial single-shot SE EPI technique (TR/TE: 2500/75 msec; 6 directions; B-value: 600; FOV 20 cm; scan time: 3:40). Two observers independently traced whole cord ROIs around each axial DTI image. DTI images were mapped to anatomic locations in the cervical spine relative to areas of intramedullary edema/hemorrhage (injury). Fractional anisotropy (FA), apparent diffusion coefficient (ADC), primary eigenvector (L1) and radial diffusivity (L23) values were derived from voxels contained within the injured segments. Non-parametric testing was used to determine if significant differences exist for DTI values between ASIA classes. Correlation and multiple linear regression were used to identify relationship between DTI metrics and upper extremity motor index scores (UE MIS) corrected for neurologic level of injury (NLI).

90 spinal cords (2 readers x 45 patients) were analyzed. Significant differences (p<.05) were identified in ADC and L1 values for all injuries compared to controls, however no differences were identified for FA and L23.  All four DTI metrics were significantly different between motor complete (ASIA A/B) and motor incomplete (ASIA C/D) subgroups. Correlation analysis showed a relationship between FA, ADC, L1 & L23 and ASIA class. Pairwise comparisons between ASIA subclasses showed sig. differences (p<.008) for ADC, L1 & L23 values between ASIA A/C, A/D, B/D, but not for A/B, B/C or C/D. Regression analysis of MIS found that NLI was a predictor of MIS but no independent DTI values predicted MIS.

The capacity of this DTI characterization to subclassify groups of SCI patients by neurologic deficit may be limited when used in isolation.

Absolute DTI values were not a suitable substitute for an accurate neurologic examination in the setting of SCI.

Zohrabian, V, Zussman, B, Mathews, J, Frangos, A, Gorniak, R, Lackey, J, Dresner, M, Marino, R, Lai, S, Flanders, A, Can Diffusion Tensor Imaging (DTI) Replace the Neurologic Examination in Spinal Cord Injury (SCI)?.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11009213.html