In the assessment of thoracolumbar burst fractures, computed tomography(CT) is superior to other imaging modalities, especially when a more definitive assessment of the posterior elements of the spine or the nueral canal is desired. A few authors have reported the relationship between traumatic spinal canal stenosis and neurologic deficits. Authors analysed 64 patients with thoracolumbar burst fractures about the relationship between traumatic spinal canal stenosis and neurologic deficits using plane x-ray findings and CT scans. The results were as follows; 1. There was no difference of kyphotic angle and anterior vertebral height loss between neurologic deficit group on lateral plane x-ray findings. 2. The incidence of injury of the superior endplate was 100%(64/64); of the inferior endplate 53%(36/64); of the posterior element 45.3%(29/64). 3. Twenty(69.0%) of 29 patients with disruption of posterior element had neurologic deficits, while fourteen(40.0%) of 35 patients without disruption had neurologic deficits. 4. Spinal canal ratio of 35.3% or more at the epiconus level, 56.0% or more at the conus medullaris level, and 65.3% or more at the cauda equine level were significant factors for neurologic deficits in thoracolumbar burst fractures. Spinal canal ratio of 22.0% or less at the epiconus level, 34.5% or less at the conus medullaris level, and 43.1% or less at the cauda equine level were not accompanied with neurologic dieficit in thoracolumbar burst fractures. 5. Fifteen(68%) of 22 patients with neurologic deficit in epiconus and conus medullarsis level had the horsehoe or crescent shape of patient spinal canal, eight (66.7%) of 12 patients with neurologic deficit in cauda equine level had the horsehoe or crescent shape of spinal canal.