Abstracts

Rationale: Electrocortical stimulation (ECS) is the gold standard for functional brain mapping during an awake craniotomy. The critical issue is to set aside enough time to identify eloquent cortices by ECS. High gamma activity (HGA) ranging between 80 and 120 Hz on electrocorticogram (ECoG) is assumed to reflect localized corticalprocessing. We developed novel techniques of real-time data processing to visualize HGA at bed-side and in the operation room.Methods: We applied the real-time technique to 30 patients with intractable epilepsy with subdural grid implantation and 20 patients with brain tumors who underwent awake craniotomy. Semantic-ECoG was recorded with word, figure, face recognition and memory tasks with g-Hi amp (g-tech, Graz). The ECoG related to epilptogenic and semantic activities were processed by time-frequency analysis and the all the functional profiles were projected on grid electrodes of individual brain surface.In addition, we performed functional MRI (fMRI) by the similar tasks with semantic-ECoG. After we confirmed the ECoG system worked, we applied the ECoG recording and real-time HGA analysis for awake craniotomy.Results: HGA accumulation clearly indicated seizure onset in real-time in each case. The superior temporal and inferior frontal regions were alternatively activated until 800msec with the word recognition at the bed-side measurements. Because of different electrode configuration of each patient, we made ECoG-fMRI normalization using SPM8. Time-frequency analysis of ECoG showed three major spots with increased HGA in the left frontal, posterior left temporal and bilateral temporal base. The basal temporal-occipital cortex was activated within 250 msec after visual object presentations. Face stimulation evoked significantly higher amplitudes of ECoG than other stimuli . We found a positive correlation between HGA and fMRI activity, suggesting that the HGA might be physiological correlates of fMRI. The frontal lobe showed longer-lasting HGA than the temporal lobe. ECoG-fMRI normalization contributes to neurophysiological basic for fMRI during semantic processing. During awake craniotomy, we analyzed the frequency components of brain oscillation and performed real time HGA mapping to identify eloquent areas(Fig. 1). Real-time HGA mapping during awake craniotomy rapidly indicated the eloquent areas of motor and language functions and significantly shortened the operation time, ompared to electrocortical mapping. However, the HGA dynamics was different between the tasks (Fig. 2). Sensitivity and specificity of HGA mapping were 90.2% and 82.4%, respectivelty.Conclusions: The novel technique should improve reliability of bed-side and intraoperative monitoring and enable the development of rational and objective surgical strategies. Real-time HGA analysis provide clincal merits not only for patients, but also medical doctors for treatments.