Indications for CT-Angiography of the Vertebral Arteries After Trauma

Joseph P. Drain, BS; Douglas S. Weinberg, MD; James S. Ramey, BS; Timothy A. Moore, MD; Heather A. Vallier, MD


Spine. 2018;43(9):E520-E524. 

In This Article

Abstract and Introduction


Study Design. Retrospective.

Objective. The purpose of this project is to identify factors that predict vertebral artery injury (VAI) in an effort to assess risks and benefits of computed tomography angiography (CT-A) of the neck in the trauma setting. We seek to develop guidelines for practitioners to stratify patients at medium/high risk of VAI from those who are at low risk.

Summary of Background Data. VAI and blunt carotid injury (BCI) together comprise blunt cerebrovascular injury (BCVI). More is known about risk factors for BCI than for VAI, but the neurovascular complications associated with VAI are similarly disastrous. With increasing frequency, trauma providers are using CT-A to screen for BCVI; this test carries risks that include radiation exposure and nephrotoxicity, in addition to higher cost of treatment and longer hospital stay.

Methods. Trauma patients seen over 4 months at an urban, level 1 trauma were analyzed. BCVI screening was conducted in 144/1854 (7.77%) patients. Presence of VAI and several clinical characteristics were recorded. Univariate analysis and binomial logistic regression analysis were conducted at a 95% significance level.

Results. VAI was diagnosed in 0.49% of the study population. Univariate analysis determined six factors associated with positive VAI screening. Regression analysis showed four factors that independently predicted VAI: female sex, decreased Glasgow Coma Scale, cervical spine (c-spine) fracture, and concurrent BCI. A positive c-spine physical examination trended toward predicting VAI without achieving significance.

Conclusion. Several independent predictors of VAI were identified. This study highlights the importance of identifying patients at a higher risk for VAI and indicating CT-A of the neck versus those who are at low risk and can be evaluated without undergoing advanced imaging, as CT-A appears unnecessary for most trauma patients.

Level of Evidence: 3


Blunt cerebrovascular injury (BCVI), composed of blunt carotid injury (BCI) and vertebral artery injury (VAI), is a feared complication of head and spine trauma. The clinical consequences of BCVI can be devastating: up to 80% of patients have neurologic morbidity, and mortality is as high as 40%.[1] BCI specifically carries a 15% mortality rate, and 16% of surviving patients have major neurologic morbidity[2] after excluding patients who died from massive brain injury or were comatose on admission. The morbidity and mortality of VAI similarly includes both stroke and death;[3] incidence of stroke after VAI is as high as 24%.[4] Published incidence of BCVI in hospitalized trauma patients in the United States is approximately 0.1%;[1] however, most are diagnosed after they develop neurologic abnormalities. When asymptomatic patients are screened for BCVI, incidence increases to 1% of all blunt trauma patients (with BCI present in 0.86%[2] and VAI in <1%)[4] and increases to 2.7% among those with an Injury Severity Score of 16 or higher,[2] suggesting that BCI and VAI can remain clinically silent in the trauma population. The importance of screening is underscored by the fact that medical treatment of these injuries has been shown to reduce the rate of neurologic sequelae.[1]

Because of the consequences of these injuries, patient characteristics predictive of BCVI have been investigated. Looking specifically at BCI, four independent predictors have been demonstrated by multivariate regression analysis:[5] Glasgow Coma Scale (GCS) score of 6 or less, petrous bone fracture, diffuse axonal brain injury, and LeFort II or III fracture; patients with just one of these factors had a 41% risk of BCI. But comparatively less is known about VAI: the study that demonstrated BCI risk factors was able to show just one factor, the presence of cervical spine (c-spine) fracture, as an independent predictor of VAI.[1,5]

To bridge the knowledge gap between known risk factors for BCI and VAI, we aim to uncover additional risk factors specific to VAI. Using a registry of patients seen in an urban level 1 trauma center, we will perform univariate and multivariate regression analysis to determine characteristics that independently predict VAI. This can provide better insight into a high-risk patient, which we hope contributes to the refinement of VAI screening protocols as part of overall screening for BCVI.