Diagnostic challenges in epilepsy: seizure under-reporting and seizure detection

Abstract

Epileptic seizures vary greatly in clinical phenomenology and can markedly affect the patient's quality of life. As therapeutic interventions focus on reduction or elimination of seizures, the accurate documentation of seizure occurrence is essential. However, patient self-evaluation compared with objective evaluation by video-electroencephalography (EEG) monitoring or long-term ambulatory EEG revealed that patients document fewer than 50% of their seizures, on average, and that documentation accuracy varies significantly over time. For good clinical practice in epilepsy, novel and feasible seizure detection techniques for ambulatory long-term use are needed. Generalised tonic-clonic seizures can already be detected reliably by methods that rely on motion recording (eg, surface electromyography). However, the automatic detection of other seizure types, such as complex partial seizures, will require multimodal approaches that combine the measurement of ictal autonomic alterations (eg, heart rate) and of characteristic movement patterns (eg, accelerometry). Innovative and feasible tools for automatic seizure detection are likely to advance both monitoring of the outcome of a treatment in a patient and clinical research in epilepsy.