Region matters in Spinal Trauma: Diagnostic Performance of Conventional Radiography in a Retrospective Cohort with Nested Accuracy Analysis.

INTRODUCTION

Accurate diagnosis of traumatic spinal fractures is crucial for appropriate treatment. This study explores region-specific performance characteristics of posttraumatic spinal radiography and assessed diagnostic accuracy in a cross-sectional-imaging subgroup.

METHODS

This retrospective study analyzed 474 posttraumatic two-view radiograms of the cervical (n = 220), thoracic (n = 96), and lumbar spine (n = 158) in 380 patients. Cohort characteristics, trauma mechanisms, fracture suspicion rates, and radiation dose metrics were analyzed. In a nested subgroup of 76 patients who underwent additional CT or MRI, diagnostic accuracy of radiography was assessed regarding fracture presence, severity, multi-segmental presence, and posterior edge involvement. Dose-area products were compared across regions.

RESULTS

Fractures were most frequently suspected in lumbar spine (22.2%) versus thoracic (13.5%) and cervical spine radiograms (0.5%) (p < 0.001). Dose-area product was lowest in the cervical spine (2.61 dGy cm2) compared with thoracic (22.0 dGy cm2) and lumbar spine imaging (34.8 dGy cm2) (p ≤ 0.006). In the diagnostic accuracy subgroup (n = 76), overall specificity was high (≥87.5%). Sensitivity was highest for fracture detection (86.4%) and lowest for posterior edge involvement (27.3%). Accuracy was highest in the cervical spine (93.3-99.9%), acceptable in the lumbar spine for fracture detection (86.7%) and multi-segmental injury (95.6%), and lowest in the thoracic spine for fracture severity (43.8%) and posterior edge involvement (56.3%).

CONCLUSION

Diagnostic performance of conventional radiography in spinal trauma is strongly region-dependent. Particularly, thoracic spine radiograms are rather inaccurate regarding overall fracture severity and posterior edge involvement.

IMPLICATION FOR PRACTICE: Cervical and lumbar spine radiography can be confidently used in the initial evaluation of mild spinal trauma due to their high diagnostic accuracy. In contrast, limited reliability of thoracic spine radiographs warrants a lower threshold for complementary CT or MRI to ensure comprehensive fracture assessment.