Mediastinal-width/chest-width ratio in blunt chest trauma: a reappraisal.

The chest radiographs of 54 patients with blunt chest trauma and suspected aortic rupture were reviewed retrospectively. A mediastinal-width/chest-width ratio was calculated at the level of the aortic arch. There were 44 patients without evidence of aortic rupture on aortography or surgical exploration and 10 patients with surgically confirmed aortic rupture. There was no identifiable ratio of significance in distinguishing those with aortic rupture and those without rupture. The mediastinal-width/chest-width ratio is of insufficient sensitivity and specificity in confirming or excluding aortic rupture to be clinically useful.