Abstracts

Rationale: Outside epilepsy monitoring units, diagnosing epileptic seizures (ES) is challenging. Mimics such as psychogenic nonepileptic attacks (PNEA) are common and physiologic nonepileptic events (PhysNEE) dictate different treatment.1 Differentiating ES from nonepileptic "episodes" is a learned skill.2 Home videos and hand-held video-camcorders in the evaluation of seizures are promising adjunctive techniques to history and examination (H & P). 3,4 We report the preliminary findings of a multi-center prospective study to determine the feasibility and usefulness of outpatient smartphone videos in epilepsy (OSmartViE). Results: 23 patients [12 Females, age 44; range 19-80] received VEM with SV review by 7 epileptologists and 4 residents. VEM demonstrated 7/23 (30%) with epilepsy, 13/23 (57%) had PNES and 3/23 (13%) had PhysNEE (tremor, syncope) with 29%, 67% and 100% reflecting convulsive (v nonconvulsive) episodes. Resident responses were correct in 26% of ES while epileptologists were correct in 62% of case without difference in PNES (87%, 88%). The quality of the SV was adequate for interpretation in more than 3/4th of SV (75% v 81%) and did not differ between reviewers. Inter-subject differences were present largely based upon technical as opposed to video quality. SV were mostly limited by lack of whole body view and the duration of the ictal recording. Epileptologists had a level of confidence greater than residents but was not significant (7.26/10 v 6.28/10). Conclusions: Secured uploading, exchange, and analysis of SV information are feasible. Most SV brought by patients were convulsive episodes but 70% were not ES. SV yielded a level of confidence similar to routine H & P. Epileptologists only identified ES 62% of the time though were more likely to identify non-epilepsy with greater confidence than trainees despite both having similar accuracy. Funding: This study was supported in part by Mayo Clinic.