[MRI exploration of partial epilepsy]

Investigative strategies for partial epilepsy have evolved greatly since the advent of morphological MRI. MRI currently provides the most extensive and sensitive information for detecting and localizing brain anomalies. Technical progress has included new sequences (3D acquisition, inversion-recovery, fast spin-echo), new imaging planes (through the greater diameter of the hippocampus for temporal epilepsy, orbitao-meatal for extra-temporal epilepsy), high-resolution imaging, and better signal-noise ratio in addition to new procedurs such as NMR spectroscopy, diffusion/perfusion imaging, or functional MRI. One of the principal aims of MRI is to detect potentially epileptogenic brain anomalies useful for the detection and localization of the epileptic focus. MRI sensitivity is much greater than brain computed tomography, which excepting specific cases, is no longer contributive. Hippocampal sclerosis is a typical example of an epilepsy-related brain anomaly (mesial temporal lobe epilepsy) whose diagnosis has been changed by MRI. This entity was described by Bouchet and Cazauvieilh in 1825 who related anatomic postmortem findings to the clinical expression. Reliable diagnosis of hippocampal sclerosis can now be achieved with MRI, contributing to the diagnosis of mesial temporal lobe epilepsy. MRI can also contribute precious prognostic information, useful for both drug and surgical treatment. It has been demonstrated that the degree of response to drug treatment is not univocal and that certain lesions such as hippocampal sclerosis are generally associated with drug resistance. Sensitivity to drug therapy also varies with the type of cerebral dysgenesia. It has also been demonstrated that the a brain lesion identified at MRI is generally a factor of good prognosis after surgery. Continuing improvements in MRI techniques, widespread availability of technical facilities and new derived techniques contribute greatly to the usefulness of MRI for investigating partial epilepsy.