RESUMO
Traumatic brain injury (TBI) constitutes injury that occurs to the brain as a result of trauma. It should be appreciated as a heterogeneous, dynamic pathophysiological process that starts from the moment of impact and continues over time with sequelae potentially seen many years after the initial event. Primary traumatic brain lesions that may occur at the moment of impact include contusions, haematomas, parenchymal fractures and diffuse axonal injury. The presence of extra-axial intracranial lesions such as epidural and subdural haematomas and subarachnoid haemorrhage must be anticipated as they may contribute greatly to secondary brain insult by provoking brain herniation syndromes, cranial nerve deficits, oedema and ischaemia and infarction. Imaging is fundamental to the management of patients with TBI. CT remains the imaging modality of choice for initial assessment due to its ease of access, rapid acquisition and for its sensitivity for detection of acute haemorrhagic lesions for surgical intervention. MRI is typically reserved for the detection of lesions that may explain clinical symptoms that remain unresolved despite initial CT. This is especially apparent in the setting of diffuse axonal injury, which is poorly discerned on CT. Use of particular MRI sequences may increase the sensitivity of detecting such lesions: diffusion-weighted imaging defining acute infarction, susceptibility-weighted imaging affording exquisite data on microhaemorrhage. Additional advanced MRI techniques such as diffusion tensor imaging and functional MRI may provide important information regarding coexistent structural and functional brain damage. Gaining robust prognostic information for patients following TBI remains a challenge. Advanced MRI sequences are showing potential for biomarkers of disease, but this largely remains at the research level. Various global collaborative research groups have been established in an effort to combine imaging data with clinical and epidemiological information to provide much needed evidence for improvement in the characterisation and classification of TBI and in the identity of the most effective clinical care for this patient cohort. However, analysis of collaborative imaging data is challenging: the diverse spectrum of image acquisition and postprocessing limits reproducibility, and there is a requirement for a robust quality assurance initiative. Future clinical use of advanced neuroimaging should ensure standardised approaches to image acquisition and analysis, which can be used at the individual level, with the expectation that future neuroimaging advances, personalised to the patient, may improve prognostic accuracy and facilitate the development of new therapies.