Abstracts

Rationale: Despite remarkable advances in the presurgical evaluation of drug-resistant focal epilepsy, intracranial EEG remains essential in several cases. The choice between subdural or depth recording depends on the experience and preference of each epilepsy surgery team. As for now, there exists no guidelines for which technique should be used in a given case. Methods: At our institution, we use both techniques, subdural grid recording and stereo-electroencephalography (S-EEG) in children. For each patient, all noninvasive data are extensively discussed and the type of recording is individually planned. We reviewed the charts of the 252 invasive explorations in children aged from 3 months to 15 years, performed between 1995 and 2009. We tried to define clear-cut rules concerning the choice of each method. Results: Subdural electrodes had been placed in 92 children through open craniotomy, with a few additional depth electrodes, positioned with the guidance of intraoperative echography or MRI-based neuronavigation, in order to better explore cortical regions in the depth of a sulcus and the insular cortex when indicated. Complications were meningitis in three, bone infections in two, and subdural hematoma in four children. In the same period, 160 children underwent robot-assisted, stereotactic placement of multiple depth electrodes (6 to 15 electrodes per procedure) for S-EEG, with the occurence of epidural hematoma in two children. There was no postoperative neurological deficit. In children younger than 2.5 years of age, electrodes were exclusively placed through craniotomy, since a minimum skull thickness is required for stereotactically implanted multiple depth electrodes. Only two of the 92 children (2%) who had been explored with subdural electrodes were excluded from resective surgery, whether this was the case in 20% of the children following S-EEG. In older children, S-EEG was the predominant recording technique. However, we prefered subdural grids when exploration of eloquent cortex required functional mapping. In children with very extensive malformations, there are limitations in craniotomy size for complete subdural covering, whereas with S-EEG, it is difficult to insert a sufficient number of electrodes to properly delineate the lesion. Conclusions: Subdural recording is the unique method to be used in infants and provides a better means for delineating the border of a known lesion over the convexity or when functional cortical mapping is required. S-EEG is preferable when the location of the epileptogenic zone is only hypothetical or when the study of a long-distance propagation is mandatory. The two invasive approaches should not be considered as opposites but can rather be complementary ways for an optimal presurgical assessment in complex focal epilepsies. The future challenges are to perform S-EEG in infant surgical candidates and to better combine both techniques.