The Incidence of Blunt Cerebrovascular Injury Identified on CT Cervical Spine with Contrast

Document Type

Conference Proceeding - Restricted Access

Publication Date

5-8-2026

Abstract

No longer considered rare, blunt cerebrovascular injury (BCVI) is believed to occur in 1-3% of blunt trauma patients. However, no optimal screening regimen that balances both cost and detection of BCVI has been determined. Our level 1 trauma center has adopted the practice of universal whole-body CT imaging in blunt trauma patients, which includes a contrast-enhanced CT of the cervical spine to potentially screen for BCVI. Currently, patients with clinical concern for BCVI, as determined by the expanded Denver Criteria, undergo subsequent CTA of the head and neck to definitively assess for vascular injury. However, visualization of the arterial vasculature on routine CT cervical spine with contrast imaging may negate this need for additional CTA. This study aimed to measure the sensitivity of our CT cervical spine with contrast for detection of BCVI and compare it to the Denver Criteria.

This retrospective study was determined by the Corewell Health West IRB to be exempt. The Butterworth trauma registry was queried to identify all adult patients admitted with blunt trauma from December of 2017 to August of 2025 who were diagnosed with BCVI after undergoing a screening CT of the cervical spine with contrast and subsequent confirmatory CTA. The initial patient list was transferred to a statistician for upload into a secure REDCap. The study team accessed patient charts via EPIC to supplement data from the registry. Primary outcomes were to identify the incidence of BCVI at Butterworth Hospital, as well as to calculate the sensitivity and negative predictive value of the contrast-enhanced CT cervical spine for this condition. Secondary aims were to compare the sensitivity of the screening CT to that of the expanded Denver Criteria and first rib fractures. Numeric outcomes were expressed as mean ± standard deviation and analyzed using t-tests or Wilcoxon Rank-Sum tests.

5,037 blunt trauma patients met inclusion criteria. 148 were diagnosed with BCVI for an incidence of 2.94% (95% CI 2.5%-3.44%). Screening CT identified 109 of 148 cases of BCVI, giving it a sensitivity of 73.6% (66.0%-80.1%). 114 of these patients also met one or more Denver Criteria, including 80 (54.1%) with cervical spine injuries, 45 (30.4%) with upper rib fractures, 31 (20.9%) with basilar skull fractures, 17 (11.5%) with GCS < 6, 8 (5.4%) with mandibular fractures, 6 (4.1%) with occipital condyle fractures, 3 (2.0%) with LeFort II/III fractures, and 2 (1.4%) with thoracic vascular injuries. No patients sustained blunt cardiac rupture or near hanging injuries. Notably, while the Denver Criteria had a higher sensitivity of 77.0% (69.5%-83.3%), these clinical criteria would have missed 34 cases that were identified by our current trauma imaging protocol. 5 of the 148 (3.4%, 1.5%-7.7%) BCVI patients developed strokes; 3 were missed on the screening CT, and one did not meet any of the Denver Criteria.

With an incidence of 2.94% in blunt trauma patients presenting to Butterworth Hospital, the current CT imaging protocol carries a 73.6% sensitivity in detecting these injuries. Additionally, the Denver Criteria was only 77.0% sensitive, thus solely relying on screening via the reformatted cervical spine images or clinical criteria alone is not sufficient. Algorithms should take into consideration both imaging and clinical factors to determine the need for angiography to minimize the risk of missed injury.

Comments

2026 Research Day Corewell Health West, Grand Rapids, MI, May 8, 2026. Abstract 2092

This document is currently not available here.

Share

COinS