What this Trial Is About

Mild traumatic brain injury (mTBI) is one of the most common reasons behind emergency department (ED) visits. A small portion of mTBI patients will develop an intracranial lesion that might require neurosurgical intervention. Several guidelines have been developed to help direct these patients for head Computerized Tomography (CT) scanning, but they lack specificity, mainly focus on ruling out lesions, and do not estimate the risk of lesion development. The aim of this retrospective observational study is to create a risk stratification score that predicts the likelihood of intracranial lesion development, lesion progression, and need for neurosurgical management in patients with mTBI presenting to the ED. Eligible patients are adults (≥ 15 years) with mTBI (defined as admission Glasgow Coma Scale (GCS) 13-15) who presented to the ED within 24 hours of injury to any ED in Stockholm, Sweden between 2010-2020. Reasons for ED visit and Internal Classification of Disease (ICD) codes will be used to screen for patients. Machine-learning models will be applied. The primary outcome will be a traumatic lesion on head CT, defined as a cerebral contusion, subdural haematoma, epidural haematoma, subarachnoid haemorrhage, intraventricular haemorrhage, diffuse axonal injury, skull fracture, traumatic infarction or sinus thrombosis. The secondary outcomes will be any clinically significant lesion, defined as an intracranial finding that led to neurosurgical intervention, discontinuation or reversal of anticoagulant or antiplatelet medication, hospital admission \> 48 hours due to the TBI, or death.

Score Predicting Lesion Development on CT Following Mild TBI