Predicting seizures: are we there yet?

Commentary Efforts to predict impending seizures before they actually occur have been the focus of extensive lines of research over decades. Successful seizure prediction would offer the possibility of immediate treatment to preempt the seizure, as well as to provide an opportunity for the patient to take precautionary measures and reduce risk. There is even evidence that simply reducing the unpredictability of seizures would significantly reduce the burden of epilepsy (1). Despite the tremendous potential benefits and years of impressive research, there are still no highly reliable, clinically useful methods available to predict and preempt impending seizures in patients with epilepsy (2). This fact underscores the complexity of epilepto-genic networks, the neurophysiology of seizure generation, and preictal/peri-ictal physiology. Upon this backdrop, Cook et al. studied a " long-term implanted seizure advisory system. " The system—utilizing intracra-nial electrodes connected to an implanted telemetry unit—was designed to predict seizure likelihood and inform the subject of this likelihood using colored advisory lights. While the primary outcome was the safety of the device, the authors developed reliable advisory algorithms for seizure occurrence, representing a significant advance in neurophysiologic seizure prediction. The term " seizure advisory " adds to a list of terms including seizure anticipation, seizure prediction or seizure detection, describing attempts to reliably identify seizures in advance of behavioral manifestations. The time period from the identification of the state of high seizure risk and actual seizure onset is considered the pre-ictal period; identification of the " preictal " BACKGROUND: Seizure prediction would be clinically useful in patients with epilepsy and could improve safety, increase independence, and allow acute treatment. We did a multicentre clinical feasibility study to assess the safety and efficacy of a long-term implanted seizure advisory system designed to predict seizure likelihood and quantify seizures in adults with drug-resistant focal seizures. METHODS: We enrolled patients at three centres in Melbourne, Australia, be-Eligible patients had between two and 12 disabling partial-onset seizures per month, a lateralised epileptogenic zone, and no history of psychogenic seizures. After devices were surgically implanted , patients entered a data collection phase, during which an algorithm for identification of periods of high, moderate, and low seizure likelihood was established. If the algorithm met performance criteria (ie, sensitivity of high-likelihood warnings greater than 65% and performance better than expected through chance prediction of randomly occurring events), patients then entered an advisory phase and received information about seizure likelihood. The …