Decreased CSF-flow artefacts in T2 imaging of the cervical spine with periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER/BLADE).

INTRODUCTION

The cervical spine is prone to artefacts in T2 MR-imaging due to patient movements and cerebrospinal fluid flow. The periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER/BLADE) acquisition method was developed to reduce motion artefacts. We sought to determine if T2-BLADE is superior to T2-TSE with conventional k-space reading.

METHODS

Twenty-five patients were examined using a 1.5 T MR-scanner. T2-weighted imaging of the cervical spine in sagittal and axial orientation using conventional or BLADE k-space reading was performed. Spinal cord, subarachnoid space, vertebrae and discs were evaluated by two independent observers using a scale from 0 (non-diagnostic) to 3 (excellent). Interobserver correlation was assessed as Cohen's kappa. Results of Mann-Whitney U test with p < 0.05 were regarded as significant. Furthermore, the investigators were asked for subjective evaluation in consensus.

RESULTS

Overall interobserver accuracy of κ = 0.91 was obtained. Comparison of sagittal images showed better values for all investigated structures in T2-BLADE: spinal cord (TSE/BLADE: 1.52/2.04; p < 0.001), subarachnoid space (1.36/2.06; p < 0.001) and vertebrae/discs (1.66/2.86; p < 0.001). Comparison of axial images showed better values in T2-BLADE for spinal cord (1.68/1.86; p = 0.149) and vertebrae/discs (1.0/1.96: p < 0.001) while subarachnoid space was better to be evaluated in conventional T2-TSE (1.94/1.12; p < 0.001). In sagittal orientation, motion- and CSF-flow artefacts were reduced in T2-BLADE. In axial orientation, however, CSF-flow artefacts were pronounced in T2-BLADE.

CONCLUSION

The image quality of the sagittal T2-BLADE sequences was significantly better than the T2-TSE and acquired in less time. In axial orientation, increased CSF-flow artefacts may reduce accuracy of structures in the subarachnoid space.