The reliability of nonreconstructed computerized tomographic scans of the abdomen and pelvis in detecting thoracolumbar spine injuries in blunt trauma patients with altered mental status.

BACKGROUND: Computerized tomography, traditionally utilized to evaluate and detect visceral abdominal and pelvic injuries in multiply injured patients with altered mental status, also has been useful for detecting thoracolumbar spine fractures and dislocations. The purpose of the present study was to test the reliability of nonreconstructed computerized tomography of the abdomen and pelvis as a screening tool for thoracolumbar spine injuries in blunt trauma patients with altered mental status.
METHODS: The study consisted of fifty-nine consecutive patients with altered mental status who were admitted to a Level-II trauma center. Each patient had a nonreconstructed computerized tomographic scan of the abdomen and pelvis (5-mm slices), and of the chest when indicated, as well as anteroposterior and lateral radiographs of the thoracolumbar spine. Reconstructed computerized tomographic scans dedicated to the spine (< or =2-mm slices) were completed. With use of the reconstructions as the gold standard, sensitivity and specificity with 95% confidence intervals were calculated to assess the diagnostic accuracy of using the nonreconstructed computerized tomographic scans and the radiographs.
RESULTS: Reconstructions of the spine detected seventy-two thoracolumbar spine fractures, whereas nonreconstructed computerized tomographic scans of the abdomen and pelvis detected fifty-eight and those of the chest detected sixteen. With use of the reconstructions as the standard, computerized tomography of the chest, abdomen, and pelvis had a sensitivity of 89% (95% confidence interval, 65% to 96%) and a specificity of 85% (95% confidence interval, 65% to 96%) for the detection of all fractures, compared with 37% and 76% for plain radiographs, respectively. Computerized tomography of the chest, abdomen, and pelvis was 100% sensitive and specific for the detection of whether a patient had any fracture at all, whereas radiographs were 54% sensitive and 86% specific. No fractures that were missed on nonreconstructed computerized tomography required surgery or other interventions.
CONCLUSIONS: Nonreconstructed computerized tomography detected fractures of the thoracolumbar spine more accurately than plain radiographs did and is recommended for the diagnosis of thoracolumbar spine fractures in acute trauma patients with altered mental status. Reconstructions do not need to be ordered unless an abnormality that is found on the nonreconstructed computerized tomographic scan needs additional elucidation.