Abstracts

Rationale: It is well known that seizure freedom decreases over time after TLE surgery and even patients with seizure remission may relapse after antiepileptic drugs withdrawal. This indicates incomplete resection of epileptogenic tissue despite intensive workup. Therefore, additional information may be necessary to better define epileptogenic zone (EZ). In recent years, there have been increasing evidence ISA may be important in identifying EZ. However, literature on ISA has mainly been about ictal ISA (IcISA), and information on IISA is currently very minimal. This pilot study is to see whether IISA studied by visual analysis has a role in assessment of EZ in children with TLE in clinical practice. Methods: We retrospectively studied 11 consecutive patients with medically intractable TLE caused by focal cortical dysplasia (FCD) who underwent extraoperative invasive EEG monitoring (4 stereo-EEG and 7 subdural with/without depth EEG) at the Children's Hospital Colorado between January 2009 and February 2016. Ten patients underwent surgical resection and achieved either seizure freedom or significant seizure improvement (Engel class 1 or 2) for at least 3 months. One patient was found to have EZ on the opposite site of the previous temporal lobectomy and the surgery was not performed due to the risk of severe memory loss. To analyze the IISA, the interictal EEG was selected at least 1 hour before the seizure onset and 1 hour after the seizure offset. The files were viewed on the maximum window of up to 20-30 minutes. The IISA, IcISA and traditional ictal EEG (IcEEG) were analyzed using both conventional (1-70 Hz) and ISA (0.01-0.1 Hz) band-pass filter. To achieve pure ISA, high-pass filter was left open or at 0.01 Hz and the low-pass filter was set at 0.1 Hz. Results: Six pure mesial TLE (mTLE) and 5 neocortical TLE (nTLE) with or without seizure spreading to mesial temporal were identified. IISA and IcISA were noted in all patients, generally less widespread than IcEEG and usually seen at the center of EZ. IISA and IcISA were always concordant. IISA was usually easier to recognize than IcEEG especially the low-voltage fast activity (LFA) EEG pattern, which tend to be more diffuse. IISA is easier to interpret during wakefulness than during sleep. IISA in mTLE is more prominent and consistent than in nTLE. All IISA was identified in EZ, which was resected in 5 out of 6 patients with mTLE and 4 out of 5 patients with nTLE. In one patient with pure mTLE. The IISA and EZ were found in a mesial temporal lobe opposite to the previous temporal lobectomy, therefore, the surgery was not performed due to the risk of severe memory impairment. In one patient with nTLE, the IISA and EZ were noted in a temporal operculum and an anterior hippocampus. This patient underwent left anterior temporal lobectomy only without resection of the temporal operculum due to a possible risk of language impairment. The patient remained seizure free after the surgery. Conclusions: IISA may be one of the biomarkers of EZ in pediatric patients with TLE caused by FCD. In addition to IcEEG and IcISA, visual analysis of IISA is very useful to better define EZ, which may lead to better seizure outcome. Analysis of IISA does not require special EEG equipment and is simple to perform, therefore, it should be considered in intracranial EEG interpretation of pediatric patients with TLE. Funding: None