Keywords:
Neuroradiology spine, MR physics, Musculoskeletal spine, MR-Diffusion/Perfusion, MR, MR-Functional imaging, Imaging sequences
Authors:
F. Cartes-Zumelzu, S. Ingorokva, H. Kostron, G. Feuchtner, C. Kremser, C. Thomé; Innsbruck/AT
DOI:
10.1594/ecr2013/C-0877
Methods and Materials
Diffusion tensor imaging was performed in 24 patients with intraspinal lesions.
Lesions showed different amount of spinal cord compression or infiltration.
All patients underwent MR imaging on a 3.0T magnet (Verio,
Siemens,
Erlangen,
Germany) with a dedicated cervical spine coil.
A standardized MR imaging protocol for the cervical spine was used including sagittal T2 (TR: 4680 TE: 105) and T1-weighted (TR: 510,
TE: 10 ) and axial T2 weighted planes as well as gadolinium enhanced T1 weighted sagittal and axial acquisitions.
For diffusion tensor imaging an experimental multi-slice spin echo EPI sequence (work in progress 511) was used in sagittal and transversal orientation with b-factors of 0 mm2/sec and 900 mm2/sec. Data were acquired for 30 diffusion encoding directions.
To enable a short TE a modified Stejskal-Tanner diffusion encoding scheme was used [10].
Other parameters: 20 slices in sagittal orientation and 35 slices in transversal orientation (3mm slice thickness); FOV = 220mm; acquisition matrix: 128x128; TR/TE=5800ms/74ms; parallel imaging mode GRAPPA with acceleration factor 2.
Additional 3D Tractography were calculated for every patient.
Tractographies were performed of all lesions.Spinal cord tracts were evaluated for colour,
size and course.
Patients were classified into 3 groups according to the fiber course with respect to the lesion (Fig.5).
The tumor size and location was evaluated by inspection of the conventional MR-images.
Furthermore the lesions were rated as resectable or non-resectable (Fig.6)