Changes of dimension of EEG/ECoG nonlinear dynamics predict epileptogenesis and therapy outcomes

The lack of early biomarkers of epileptogenesis precludes a sound prediction of epilepsy development after acute brain injuries and of the natural course of the disease thus impairing the development of antiepileptogenic treatments. We investigated whether the dimensional changes of nonlinear dynamics in EEG/ECoG signals, that were recorded in the early aftermath of different epileptogenic injuries, provide a measure to be exploited as a sensitive prognostic and predictive biomarker for epilepsy. Using three different models of epilepsy in two rodent species, we report a common and significant decrease of nonlinear dynamics dimension in EEG/ECoG tracings during early epileptogenesis. In particular, the magnitude of this dimensional decrease predicts the severity of ensuing epilepsy, and this measure is modulated by disease-modifying or antiepileptogenic treatments. The broad application of EEG/ECoG monitoring in epilepsy underlines the translational value of these findings for enriching the population of patients at risk for developing epilepsy in clinical investigations.