Identifying patients at risk for intracranial and extracranial blunt carotid injuries

Abstract

Blunt carotid injuries are rare, often occult, and potentially devastating. Angiographic screening programs have detected this injury in up to 1% of blunt trauma patients. Implementing a liberal angiographic screening program at our hospital is impractical and we want to identify a high-risk group to target for screening. We hypothesize that intracranial and extracranial carotid injuries have different risks, presentations, and outcomes. Patients with intracranial and extracranial carotid injuries were identified from the British Columbia trauma registry. Presentation and outcome were reviewed. To facilitate statistical modeling the analysis was done by matching cases to 5 randomly selected controls. Risk factors for injury were evaluated by univariate and multiple logistic regression. A total of 35 carotid injuries were identified. Thirteen intracranial injuries were identified in 10 patients. Twenty-two extracranial injuries were identified in 18 patients. Sixty-seven percent of patients with intracranial injuries and 31% of those with extracranial injuries died (P = 0.11). Eleven percent of intracranial injuries and 56% of extracranial injuries were occult (P = 0.04). Glasgow outcome scores were 2.04 intracranial and 3.12 extracranial (P = 0.18). For intracranial injuries the multiple variable predictive model had two predictors: Glasgow Coma Score </=8 and facial fractures. For extracranial the predictors were GCS < or =8 and thoracic injury (Abbreviated Injury Score > or =3). Intracranial injuries were frequently detected on initial investigations and have very poor outcomes. Extracranial injuries were more frequently occult and stand to benefit from early detection by screening programs. As independent risk factors for these two injuries differ, limited screening resources should focus on risk factors for occult extracranial injury: namely, low GCS and significant thoracic injury.