People with epilepsy report that the unpredictable nature of seizures is one of epilepsy’s most disabling features. Despite the perception that seizures are unpredictable, studies have demonstrated that seizures are often associated with predictors in at least 4 categories: trigger factors (measurable precipitants or exposures); premonitory features (symptoms preceding the seizure); self-prediction (the ability of some individuals to accurately predict an impending seizure); and electrographic events on intracranial EEG. Successful seizure forecasting would offer the possibility of immediate treatment to preempt the seizure, and provide an opportunity for a patient to take precautionary measures to reduce risk, including perhaps the risk of SUDEP. This presentation will summarize data on seizure triggers and premonitory features and concentrate on the many years of work by Sheryl Haut and speaker Michael Privitera on seizure self-prediction using prospective patient electronic diaries, demonstrating that many patients can self-predict seizures with surprising accuracy. New data from long term intracranial EEG monitoring is revealing short and long-term EEG patterns that improve seizure forecasting. Commercially available wearable devices can monitor long term physiological parameters. Integration of all these seizure prediction factors is the goal of the Epilepsy Foundation’s Seizure Gauge Initiative. Understanding when seizures are most likely to occur could guide us to more effectively use medications or other treatments, and potentially reduce the incidence of SUDEP.

Learning Objectives