Abstracts

Rationale: RATIONALE: There is expanding interest, recognition, and knowledge of autoimmune epilepsy (AE). However, little is known about role of long-term EEG and electrographic characteristics of pediatric patients with AE. We aimed to determine the clinical features and the utility of long-term video EEG monitoring on detection of electrographic abnormalities among AE pediatric patients. Methods: METHODS: We conducted a retrospective study of patients with definite or probable AE identified through the Neuro Immunology Program database of Children National Medical Center (CNMC) started in 2014, with case presentations dating back to January 2009. We reviewed the charts for clinical data, immunologic profiles, therapeutic interventions, EEG, MRI, and epilepsy outcomes. EEG findings were identified by clinical neurophysiologists. Results: RESULTS: Fifteen patients with AE were identified (6 males, 9 females; mean age 10.65.8 yrs [range 3-19 yrs]). Nine (60.0%), all with anti-NMDAR antibodies, had surface autoantibodies, 2 (13.3%) had intracellular autoantibodies (1 anti-Hu, 1 anti-Ma2), 4 patients (26.7%) had clinically suggestive AE, but no antibodies. The most common presenting symptoms were fever (60.0%), altered mental status (53.3%) and focal seizures (46.7%). Abnormal CSF findings were found in 11 patients (73.3%), typically mild pleocytosis (median 20, range 3-318 cell/mm3); abnormal MRI findings were identified in 6 patients (40.0%). EEG focal slowing occurred more frequently in autoantibody-positive patients (90.0% vs. 25.0%, p=0.04). There were no other differences in EEG when grouping subjects according to the presence or absence of antibodies or when considering those with surface vs intracellular antibodies. Compared to routine vEEG (53.3%), during active disease, long-term vEEG more frequently detected epileptiform discharges (57.1% vs 0%, p=0.02), diffuse slowing (80% vs 25%, p=0.02), and abnormal sleep patterns (69.2% vs 0%, p = 0.01). Otherwise, follow up long-term video EEG demonstrated extreme delta brushes (14.3%), NCSE (14.3%), and OIRDA (7.1%), while none were found on routine EEG. All patients received immunosuppressive therapy: IVIG (93.3%), steroids (86.7%), plasma exchange (60.0%), and second-line immunosuppressive drugs (66.7%; rituximab, cyclophosphamide, or both). Seven patients (7 of the 10 patients who developed epilepsy) had seizures detected during long-term EEG monitoring when they had clinically suspected AE. Conclusions: CONCLUSIONS: Long-term vEEG is useful for detecting epileptiform abnormalities, sleep disruption, and non-convulsive seizures in AE patients, and is more sensitive than CSF and MRI for detecting abnormalities in AE. The ascertainment of epileptic phenomena and/or specific EEG findings such as extreme delta brush may provide useful information for diagnosis and management of such patients. Long-term vEEG monitoring may help determine epilepsy prognosis and should also be explored as a biomarker of disease recurrence risk in AE patients. Funding: None