Computed tomographic angiography versus conventional angiography for the diagnosis of blunt cerebrovascular injury in trauma patients.

BACKGROUND: Blunt cerebrovascular injuries (BCVI) in trauma patients are rare but potentially devastating injuries, particularly if the diagnosis is delayed. Conventional angiography (CA) has been the screening and diagnostic modality of choice for identifying BCVI. With the advent of high-resolution computed tomography (CT), CT angiography has become a common modality for the screening of BCVI. A liberalized screening approach has suggested that cerebrovascular injuries are missed in many patients; however, no standard BCVI screening protocol exists. Early diagnosis of the BCVI can prevent long-term sequelae. METHODS: In this prospective study, all patients received a CT angiogram (16-slice or 64-slice) at the time of injury assessment and followed 24 hours to 48 hours later with CA of the cerebrovasculature. RESULTS: A total of 158 patients were enrolled in the study. CA identified 32 injuries to the cerebrovasculature in 27 patients; CT detected only 13 true injuries (40.6%) in 12 patients. Of the 32 injuries, 11 were carotid artery injuries and 21 were of the vertebral artery. Seventy-four patients were screened with the 16-slice CT scanner with an overall sensitivity of 29%, and 84 patients were screened with the 64-slice CT scanner with an overall sensitivity of 54%. The combined specificity and sensitivity of 16- and 64-slice CT in detecting BCVI were 0.97 (95% confidence interval: 0.92-0.99) and 0.41 (95% confidence interval: 0.22-0.61), respectively. CONCLUSION: Neither 16- nor 64-slice CT angiography is as accurate as CA as a screening tool for BCVI.