[MRI and surgery of epilepsy]

The evaluation of M.R.I. in the presurgical management of epilepsies requires establishing the significance of the abnormal signals. In 15 cases, correlations among M.R.I., electrophysiological data obtained by chronic intracerebral recordings and histological examination were studied. Abnormalities were detected by M.R.I. in 12 cases (80%). An accurate epileptic-focus topographical orientation was obtained by M.R.I. in 10 cases. But M.R.I. findings did not coincide with functional data obtained by S.E.E.G. recordings. Abnormalities detected by M.R.I. led to rectify the previous hypothesis about the epileptic focus localization in two cases. One false positive and one case of incomplete data would have been misleading if S.E.E.G. had not been performed. M.R.I. abnormalities included two morphological changes and ten high intensity T2 signals. The increased T2 intensity was more often focal (9) than diffuse (1). Among the 8 cases of increased T2 intensity correlated to the epileptic focus, 1 was histologically normal, 3 corresponded to specific lesions (astrocytoma: 1, hamartoma: 2) and 4 corresponded to non specific lesions (gliosis: 2, spongiosis: 2). When M.R.I. was normal at the site of the epileptic focus, either no alteration (2) or non specific lesions (2) were found. Uncus herniation could not be identified with M.R.I. M.R.I. has proven sensitive, detecting small epileptogenic lesions missed by C.T. Moreover in some cases, high T2 signals have been correlated topographically with the epileptic focus and histologically with non specific lesions. These signals are more likely related to a functional phenomenon such as an increase in water content than to the histopathological findings themselves.