Diagnostic accuracy of different clinical screening criteria for blunt cerebrovascular injuries compared with liberal state of the art computed tomography angiography in major trauma.

BACKGROUND: Blunt cerebrovascular injuries (BCVI) can significantly impact morbidity and mortality if undetected and, therefore, untreated. Two diagnostic concepts are standard practice in major trauma management: Application of clinical screening criteria (CSC) does or does not recommend consecutive computed tomography angiography (CTA) of head and neck. In contrast, liberal CTA usage integrates into diagnostic protocols for suspected major trauma. First, this study's objective is to assess diagnostic accuracy of different CSC for BCVI in a population of patients diagnosed with BCVI after the use of liberal CTA. Second, anatomical locations and grades of BCVI in CSC false negatives are analyzed. METHODS: The hospital database at University Hospital Münster was retrospectively searched for BCVI diagnosed in patients with suspicion of major trauma 2008 to 2015. All patients underwent a diagnostic protocol including CTA. No BCVI risk stratification or CSC had been applied beforehand. Three sets of CSC were drawn from current BCVI practice management guidelines and retrospectively applied to the study population. Primary outcome was false-negative recommendation for CTA according to CSC. Secondary outcome measures were stroke, mortality, mechanism of injury, multivessel BCVI, location and grade of BCVI. RESULTS: From 4,104 patients with suspicion of major trauma, 91 (2.2%) were diagnosed with 126 BCVI through liberal usage of CTA. Sensitivities of different CSC ranged from 57% to 84%. Applying the set of CSC with the highest sensitivity, false-negative BCVIs were found more often in the petrous segment of the carotid artery (p = 0.01) and more false negatives presenting with pseudoaneurysmatic injury were found in the vertebral artery (p = <0.01). CONCLUSION: This study provides further insight into the common debate of correct assessment of BCVI in trauma patients. Despite following current practice management guidelines, a large number of patients with BCVI would have been missed without liberal CTA usage. Larger-scale observational studies are needed to confirm these results. LEVEL OF EVIDENCE: Diagnostic study, Level III.