Article
Comparison of the predictive value of computerized tomography and magnetic resonance imaging in "diffuse" head injury
Vergleich des prädiktiven Wertes von Computertomographie und Magnetresonanztomographie bei "diffuser Hirnschädigung"
Search Medline for
Authors
Published: | April 11, 2007 |
---|
Outline
Text
Objective: The diagnosis "diffuse" head injury is commonly based on computerized tomography (CT) findings. Lately magnetic resonance imaging (MRI) has been claimed to be more precise in depicting brain lesions. We therefore investigated comatose patients after head injury to compare the predictive value of CT to MRI.
Methods: 220 comatose patients, i.e. Glasgow Coma score of less than 9, who had suffered a severe head injury, were analysed. After an initial CT the MRI was obained with a 1.5 Tesla scanner within one week of the head injury. The location of lesions as depicted by CT and MRI were evaluated by a neuroradiologist, who was blinded to the clinical findings.The location of lesions were correlated with the 30 day mortality or clinical outcome at the time of discharge from the hospital. Statistical methods included T-square, fisher's exact test and cross tables.
Results: While CT proved to be reliable in depicting supratentorial lesions, MRI showed lesions of the mesencephalon, pons or medulla oblongata, in addition to a more detailed imaging of supratentorial lesions. Statistically CT was not significantly related to outcome. MRI was highly significantly related to the outcome (p<0.05).
Conclusions: MRI demonstrates the inadequacy of CT in detecting traumatic lesions of the brain. While CT is undoubtedly indispensible for indicating lesions in need of neurosurgical interventions, it clearly fails to depict the lesions related to death or later disability. MRI is therefore recommended to predict outcome.