Abstracts

Rationale: An increasing number of studies report about the potential utility of electric source imaging (ESI) to delineate the irritative zone in patients with focal epilepsy, and promote its inclusion in the presurgical planning. However, the number of patients included in these studies has been relatively low so far, and the validation by surgical resection and outcome was often not provided. Here we prospectively studied the yield of ESI in 150 operated patients by comparing the location of the spike source maximum with the resected zone. Methods: we evaluated the influence of two important factors on sensitivity and specificity of ESI: the number of electrodes, and the use of the individual MRI (iMRI) vs. a template MRI (tMRI) for the head model. Low resolution EEG (LR; < 32 electrodes) was available in all 150 patients. In 55 patients, high resolution EEG was recorded (HR; 128-256 electrodes) as well. The yield of ESI was determined in the combinations of the two factors. Results: HR-iMRI led to a sensitivity of 85% and a specificity of 87%, followed by HR-tMRI with 76% sensitivity and 56% specificity. Low resolution EEG was clearly less precise, worst when analyzed with a template MRI (LR-tMRI: sensitivity: 57%, specificity: 68%). These numbers did not change significantly when only the 55 patients were considered where all combinations were available. Conclusions: Our study showed a very high sensitivity and specificity of ESI if optimal technical equipment is used, i.e. => 128 channel EEG and the individual MRI. Our results recommend the use of ESI in the presurgical evaluation of patients with focal epilepsy. Supported by the Swiss National Science Foundation (SPUM consortium 33CM30-124089 and 320030-122073)