Indications for obtaining surveillance thoracic and lumbar spine radiographs

J Trauma. 1994 Oct;37(4):673-6. doi: 10.1097/00005373-199410000-00024.


The purpose of this study was to identify risk factors for thoracic/lumbar spine fractures in patients with blunt injuries and subsequently establish indications for obtaining surveillance thoracolumbar radiographs. Retrospective review of all patients with blunt injuries (n = 1485) admitted in 1992 to a level I trauma center with a discharge diagnosis of thoracolumbar spine fracture established entrance criteria for a 4-month prospective study. Relative risk of fracture (RR) was calculated. Retrospective. Seventy-six percent (176 of 233) had radiographs; 21% had fractures; one diagnosed late. Prospective. One hundred percent (167 of 167) had radiographs; 9% (15 of 167) had fractures; none diagnosed late or missed. Forty percent (26 of 65) of patients with fractures had no pain or tenderness; 35% (9) required surgical spinal fixation. Our data define these indications for obtaining thoracolumbar radiographs in patients with blunt injuries: back pain (RR1), fall > or = 10 feet, ejection from motorcycle/motor vehicle crash > or = 50 mph, GCS score < or = 8, (all RR2), and neurologic deficit (RR10). The sensitivity of our surveillance radiography protocol has increased to 100%. The absence of back pain does not exclude significant thoracolumbar trauma.

MeSH terms

  • Adult
  • Female
  • Humans
  • Lumbar Vertebrae / diagnostic imaging*
  • Male
  • Odds Ratio
  • Prospective Studies
  • Radiography
  • Retrospective Studies
  • Sensitivity and Specificity
  • Thoracic Vertebrae / diagnostic imaging*
  • Wounds, Nonpenetrating / diagnostic imaging*