Brain Advance Access originally published online on December 9, 2005
Brain 2006 129(2):375-385; doi:10.1093/brain/awh709
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Interictal diffusion MRI in partial epilepsies explored with intracerebral electrodes
1 Epileptology Unit, 2 INSERM U739, Faculté de Médecine Pitié-Salpêtriére, 3 CNRS UPR640 LENA, Departments of 4 Neuroradiology and 5 Neurosurgery, Hôpital de la Pitié-Salpêtrière, Paris, France and 6 Center for Magnetic Resonance Research, University of Minnesota School of Medicine, Minneapolis, MN, USA
Correspondence to: Lionel Thivard, MD, INSERM U739, Faculté de Médecine Pitié-Salpêtrière, 105, Boulevard de l'Hôpital, 75013, Paris, France E-mail: lionel.thivard{at}psl.ap-hop-paris.fr
Patients with refractory partial seizures may benefit from epilepsy surgery. However, invasive investigations are often needed to define the precise location and limits of the epileptogenic zone (EZ). In this study, we asked whether diffusion tensor imaging (DTI) might provide a non-invasive alternative to locate the EZ or at least provide insights to help place intracerebral electrodes for stereo-electroencephalography (SEEG). Whole brain DTI and voxel-based analysis (SPM99) was used to assess diffusion properties objectively in 16 epilepsy patients investigated with SEEG. Epilepsy was symptomatic in two patients and cryptogenic in the 14 remaining patients. The suspected onset of seizures was temporal in 10 patients, frontal in 2 and occipital in 4. Individual maps of apparent diffusion coefficient (ADC) and fractional anisotropy (FA) were calculated and compared to a database of 40 healthy volunteers. Thirteen of 16 patients exhibited diffusion abnormalities. ADC abnormalities were better correlated with SEEG data than FA abnormalities which were usually located at a distance or in the white matter. A significant increase in ADC (P < 0.01) was found in 11 patients and was located in the regions explored with depth electrodes in 7 of them. Surgery outcome was available in 3 of these 7 patients (2 were seizure free and 1 not). DTI specificity was better in extratemporal lobe epilepsy (83%) than in temporal lobe epilepsy (20%). When abnormalities concurred with the SEEG data, the concordance was optimal between the localization of the diffusion abnormalities and the irritative zone defined by SEEG. These encouraging, preliminary results, suggest that DTI examinations may provide accurate spatial data on the location and extent of the epileptogenic network in extratemporal lobe epilepsies.
Key Words: partial epilepsies; diffusion tensor imaging; stereo-electroencephalography; irritative zone
Abbreviations: ADC = apparent diffusion coefficient; DTI = diffusion tensor imaging; EZ = epileptogenic zone; FA = fractional anisotropy; IZ = irritative zone; OZ = onset zone; SZ = spreading zone; SEEG = stereo-electroencephalography; SPECT = single photon emission computed tomography
Received April 4, 2005. Revised October 25, 2005. Accepted October 27, 2005.