Screening and detection of blunt vertebral artery injury in patients with upper cervical fractures: The role of cervical CT and CT angiography

Objective: To evaluate the clinical utility of nonenhanced CT (NECT)-based screening criteria and CTA in detection of blunt vertebral artery injury (BVAI) in trauma patients with C1 and/or C2 fractures. Methods: We retrospectively reviewed the clinical records of all blunt trauma patients with C1 and/or C2 fractures between 8/2006 and 9/2011. Cervical CTA was prompted by cervical fractures involving/adjacent to a transverse foramen, and/or subluxation on NECT. Two neuroradiologists independently reviewed the CTA studies, and graded the BVAI. Results: 210 patients were included; of these, 124 underwent CTA (21/124 with digital subtraction angiography, DSA), and 2 underwent DSA only. Overall, 30/126 suffered BVAI. Among 21 patients who underwent both CTA and DSA, there was 1 false negative and 1 false positive (both grade 1). There was strong interobserver agreement regarding CTA-based BVAI detection (kappa = 0.93, p < 0.001) and grading (kappa = 0.90, p < 0001). Only 3/30 BVAI patients suffered a posterior circulation stroke; none of the patients who had a negative CTA or were not selected for CTA, based on NECT screening criteria, suffered symptomatic stroke. While C1/C2 comminuted fracture was more common in patients with high grade BVAI (p = 0.039), simultaneous C3–C7 comminuted fracture increased the overall BVAI risk (p = 0.011). Conclusion: CTA reliably detects symptomatic BVAI in patients with upper cervical fractures. Utilization of NECT-based screening criteria such as transverse foraminal involvement or subluxation may be adequate in deciding whether to perform CTA, as no patients who were not selected for CTA suffered a symptomatic stroke. However, CTA may miss lower grade, asymptomatic BVAI