In the acute phase of TBI, secondary ischaemic injury induced by hypotension and oedema or by haemorrhage maybe superimposed on the primary, blunt or penetrating, parenchymal injury. In these patients the focal injury is studied using T2-weighted and Fluid Attenuated Inversion Recovery (FLAIR) sequences, the ischa-emic injury with diffusion-weighted sequences (DWI), and haemorrhage using T1-, T2*-weighted and FLAIR sequences: in all cases the examination time should be as short as possible.
In the chronic phase, patients exhibit focal lesions with or without haemoglobin degradation products (i.e. diffuse axonal injury, DAI) and phenomena of anatomical and/or functional deafferentation. The MR protocol for clinically stabilized patients thus aims at detecting focal lesions such as signs of stationary tissue distress and DAI using sequences sensitive to gliosis (T2-weighted and FLAIR) and to haemoglobin degradation products (T2*-weighted sequences), and at evidencing direct or indirect signs of deafferentation.
Current guidelines mandate the use of a number of technical features when imaging patients with TBI, including acquisitions in at least two spatial planes, because structures such as corpus callosum and brain-stem are better appreciated in sagittal and coronal planes; slice thickness not greater than 5 mm; spin echo (SE)-turbo SE (TSE), proton density (Pd) and T2-weighted; SE-TSE, T1-weighted; and gradient echo (GE) T2*-weighted and FLAIR sequences.
T2- and PD-weighted sequences afford optimal visualization of oedema, which is found in the majority of patients with TBI. T1-weighted sequences are useful to document recent intra- and extra-axial blood extravasation. T2-weighted FLAIR sequences are very sensitive to thin extracerebral blood collections in brain contusion, DAI and subarachnoid haemorrhage [11, 12]. T2*-weighted sequences, which are particularly sensitive to the effects of magnetic susceptibility induced by haemoglobin degradation products, are superior to T2-weighted sequences in identifying lesions with a haemorrhagic component, such as contusion and DAI [13-15]. A recently proposed ultrafast multi echo-planar sequence (time of acquisition 4 s), t-PEPSI [9,10], is sensitive to static magnetic field inhomogene-ities and is optimized for functional neuroimaging as an alternative to T2*-weighted sequences . In this study, these two sequences did not detect a significantly different total number of DAI lesions; the T2*-weighted sequence proved to be significantly superior to t-PEPSI in depicting temporal lesions, as could be expected from the presence in the EPI sequences of geometric distortion and magnetic susceptibility artefacts due to anatomical structures such as the petrous bone in this region. Ultrafast multi EPI sequences thus dramatically reduce imaging time, and their application should be recommended for the study of clinically unstable or uncooperative patients.
Medium-field magnets like those currently employed in clinical practice (1.5 T) allow the acquisition of a protocol with these features in about 20 min.
Results Obtained with Low- and Medium-Field MR
Despite the high sensitivity of MR to focal traumatic lesions, the correlations between clinical picture and MR parameters are not satisfactory. Gentry et al. documented an inverse correlation between GCS score at the time of trauma and number of DAI lesions detected on MR; a low GCS score was also associated with DAI le sions in corpus callosum and brainstem. A low correlation was obtained between GCS score and cerebral contusion and epidural and subdural haematoma [17-20]. Evaluation of these data must however account for the above-mentioned limitations of the clinical scales and for the use of lesion load as the sole MR parameter in TBI patients.
Implementation of volumetric techniques in longitudinal studies directed at improving the correlation between MR parameters and clinical picture showed that neuron loss progresses in the weeks and months following the trauma resulting in focal brain atrophy, which is associated with a more severe outcome [21,22].
Was this article helpful?