Optimal thoracic and lumbar spine imaging for trauma: are thoracic and lumbar spine reformats always indicated?

BACKGROUND: Computed tomography (CT) of the thoracic and lumbar (T/L) spine with reformats has become the imaging modality of choice for the identification of T/L spine fractures. The objective of this study was to directly compare chest/abdomen/pelvis CT (CAP CT) with CT with T/L reformats (T/L CT) for the identification of T/L spine fractures.

METHODS: Patients who had both a CAP CT scan (5-mm imaging spacing) and T/L CT reconstruction (2.5-mm image spacing with sagittal and coronal reformats) were selected. A "fracture" group (N = 35) and a "no fracture" group (N = 57) were identified. The type and level of fracture were recorded.

RESULTS: The CAP CT correctly identified all 35 patients with a thoracolumbar fracture (100% sensitivity; 95% confidence interval: 88-100%). A total of 80 separate fracture sites were present in the 35 patients. The CAP CT accurately identified 78 of those fractures (97.5% sensitivity; 95% confidence interval: 90.4-99.6%). The two fractures not identified on the CAP CT were both the transverse process fractures in patients with multiple fractures at different levels.

CONCLUSION: Patients who have a CAP CT do not require reformats for clearance of the T/L spine.