L Yuan1, X Wei2, C Xu1, Y Jin1, G Wang1, Y Li2, H Tian1, S Chen1. 1. 1 Department of Neurosurgery, Shanghai Jiao Tong University Affiliated Sixth People's Hospital, Shanghai, China. 2. 2 Department of Diagnostic Radiology, Shanghai Jiao Tong University Affiliated Sixth People's Hospital, Shanghai, China.
Abstract
OBJECTIVE: The aim of this study was to evaluate multisequence 3.0-T MRI in the detection of severe traumatic brain injury (sTBI) and in predicting the outcome. METHODS: 32 patients with sTBI were prospectively enrolled, and multisequence 3.0-T MRI was performed 4-8 weeks post injury. Quantitative data were recorded on each sequence. The ability to display the parenchymal lesions was compared with that of 64-slice spiral CT. The clinical and radiological results were correlated with the Glasgow Outcome Scale Extended scores 6 months after injury. RESULTS: 3.0-T MRI could display more lesions than CT, especially when the lesion was deeply located. The lesion volumes and diffuse axonal injury (DAI) scores were different between good and poor outcome groups on fluid attenuated inversion recovery (p < 0.05). The apparent diffusion coefficient (ADC) values of the splenium of the corpus callosum and brain stem were also different (p < 0.05). Patients with unfavourable outcome showed a significantly higher volume of haemorrhage on susceptibility-weighted imaging than those with favourable outcomes and had haemorrhages generally located more deeply. Logistic regression analysis revealed that the location of haemorrhage and the ADC values of the splenium of the corpus callosum were independent risk factors for poor outcome, with an overall predictive accuracy of 91.4%. CONCLUSION: The joint use of conventional and advanced sequences of 3.0-T MRI can comprehensively detect the pathological changes occurring after sTBI. Haemorrhagic and non-haemorrhagic DAIs in deep structures strongly suggest poor outcome. ADVANCES IN KNOWLEDGE: This article improves the understanding of advanced MRI sequences in the detection of patients with sTBI and prediction of prognosis.
OBJECTIVE: The aim of this study was to evaluate multisequence 3.0-T MRI in the detection of severe traumatic brain injury (sTBI) and in predicting the outcome. METHODS: 32 patients with sTBI were prospectively enrolled, and multisequence 3.0-T MRI was performed 4-8 weeks post injury. Quantitative data were recorded on each sequence. The ability to display the parenchymal lesions was compared with that of 64-slice spiral CT. The clinical and radiological results were correlated with the Glasgow Outcome Scale Extended scores 6 months after injury. RESULTS: 3.0-T MRI could display more lesions than CT, especially when the lesion was deeply located. The lesion volumes and diffuse axonal injury (DAI) scores were different between good and poor outcome groups on fluid attenuated inversion recovery (p < 0.05). The apparent diffusion coefficient (ADC) values of the splenium of the corpus callosum and brain stem were also different (p < 0.05). Patients with unfavourable outcome showed a significantly higher volume of haemorrhage on susceptibility-weighted imaging than those with favourable outcomes and had haemorrhages generally located more deeply. Logistic regression analysis revealed that the location of haemorrhage and the ADC values of the splenium of the corpus callosum were independent risk factors for poor outcome, with an overall predictive accuracy of 91.4%. CONCLUSION: The joint use of conventional and advanced sequences of 3.0-T MRI can comprehensively detect the pathological changes occurring after sTBI. Haemorrhagic and non-haemorrhagic DAIs in deep structures strongly suggest poor outcome. ADVANCES IN KNOWLEDGE: This article improves the understanding of advanced MRI sequences in the detection of patients with sTBI and prediction of prognosis.
Authors: Rainer Scheid; Cristoph Preul; Oliver Gruber; Christopher Wiggins; D Yves von Cramon Journal: AJNR Am J Neuroradiol Date: 2003 Jun-Jul Impact factor: 3.825
Authors: Karen A Tong; Stephen Ashwal; Barbara A Holshouser; Joshua P Nickerson; Christopher J Wall; Lori A Shutter; Renatta J Osterdock; E M Haacke; Daniel Kido Journal: Ann Neurol Date: 2004-07 Impact factor: 10.422