Abstracts

Newer magnetic resonance imaging (MRI) sequences can detect cerebral abnormalities not identified on visual inspection of conventional MRI, but the neuropathological nature of these changes remains unknown. The aim of the study was to determine if particular quantitative MR parameters are associated with particular structural tissue abnormalities., Nine patients with temporal lobe epilepsy were imaged at 1.5T using standard presurgical volumetric and quantifiable research sequences: magnetisation transfer, double inversion recovery and FFT2. The resected temporal lobe was cut into tissue blocks that were registered with the pre-operative volumetric MRI data according to our previously described method. Stereologically obtained neuronal densities and field fraction of GFAP immunohistochemistry were obtained in grey and white matter and field fraction of synaptophysin and MAP2 immunohistochemistry were obtained in white matter. All analyses were performed in the middle temporal gyrus and compared with quantitative MRI data of the equivalent region., Direct comparisons between quantitative parameters from MRI and pathology could be performed in all cases. Pearson cross correlations did not reveal any significant correlations between any of the neuropathological and MR measures., It is possible that the MRI sequences used were not sensitive to the neuropathological parameters chosen in the study. There might also be limitations to the coregistration of MRI and resected tissue as well as the quantification protocols for both MRI and neuropathology. Further studies using higher field strength MRI and other neuropathological measures are needed to clarify the correlations between MRI and pathology., (Supported by Wellcome Trust, UK, grant number 066185.)