Authors:
But plain films may still be a good starting place. Unlike for cervical spine injury, there are no decision tools to predict thoracolumbar spine fractures or guide imaging decisions. Some diagnostic pathways, such as from the Eastern Association for the Surgery of Trauma, recommend screening for thoracolumbar fractures using computed tomography (CT), while others recommend starting the evaluation with radiographs. Researchers performed a systematic review and meta-analysis to assess the diagnostic accuracy of history, physical exam findings, and imaging tests for thoracolumbar fracture. They identified 48 studies, which were predominantly retrospective and of variable quality. The analysis included 211,463 patients. The meta-analysis found isolated historical and physical exam features were inaccurate. The most useful physical exam finding was palpable spine deformity, which could rule in but not rule out fracture (positive likelihood ratio [+LR], 15.3; negative likelihood ratio [−LR], 0.93). Similarly, plain radiographs could rule in fractures but lacked sensitivity (+LR, 24.97; −LR, 0.43). CT was the only test that adequately ruled in and ruled out thoracolumbar fractures (+LR 81.12; −LR 0.04). Only three studies compared radiographs with CT for identifying clinically significant fractures, with sensitivities of radiographs varying from 33% to 100%. |
|
|
Leave a Reply Cancel reply
You must be logged in to post a comment.
COMMENT
This analysis is not informative regarding the accuracy of history and physical examination because we don’t know how many patients had a quick history and physical and no radiographs and were never included in the study. So, caveat lector. History and physical examination remain the mainstay of diagnosis for thoracolumbar spine fractures, but this study does affirm the unsurprising conclusion that CT is more sensitive than plain films. On the other hand, even that conclusion is limited by the fact that many of the injuries missed by plain films may have been clinically unimportant. Again, caveat lector.