Why Are Children With Epileptic Encephalopathies Encephalopathic?

Abstract

The epileptic encephalopathies are devastating conditions characterized by frequent seizures, severely abnormal electroencephalograms (EEGs), and cognitive slowing or regression. The cognitive impairment in the epileptic encephalopathies may be more concerning to the patient and parents than the epilepsy itself. There is increasing recognition that the cognitive comorbidity can be both chronic, primarily due to the underlying etiology of the epilepsy, and dynamic or evolving because of recurrent seizures, interictal spikes, and antiepileptic drugs. Much of scholars' understanding of the neurophysiological underpinnings of cognitive dysfunction in the epileptic encephalopathies comes from rodent studies. Frequent seizures and interictal EEG discharges in rats lead to considerable spatial and social-cognitive deficits. Paralleling these cognitive deficits are dyscoordination of dynamic neural activity within and between the neural networks that subserve normal cognitive processes.

Keywords: active avoidance; coherence; interictal spikes; oscillations; phase; theta rhythm; voltage correlations; water maze.

Fig. 1

Place cell recording. A) Rat with head stage for electrodes with attached light-emitting diode. B) Recording chamber with cue card marked with a “X”. Rat is tethered to cable connecting head stage to the amplifier. C) Example of a place cell firing rate map with corresponding color code for firing rate at right. The yellow signifies areas of the chamber where the rat visited but the cell did not fire action potentials. D) Example of a phase map where the phase of the theta in which the action potential fires is shown.

Fig. 2

Conceptual illustration of the relationship between action potential firing and local oscillations. There are strong neural connections between the hippocampus (Hipp) and PFC. As shown in the bottom two traces, when coherence of the theta rhythm in the PFC and Hipp is high and there is no difference in phase action potentials in both structures are closely aligned. When coherence is low with large differences in phase the action potentials are not aligned and action potentials coming from hippocampus arrive in the PFC at an inopportune time.

Fig. 3

Synaptic integration of spatial information indicated by CA1 theta phase preference and implications for cognitive outcome post FSE. A) CA1 place cell with cell body indicated in the pyramidal cell layer (purple), CA3 inputs indicated at stratum radiatum (red) and entorhinal cortex (blue) inputs indicated at stratum lacunosum molecular (SLM). During foraging on the stable arena, this particular CA1 place cell (field shown on far right) tends to fire action potentials toward the peak of local theta oscillations (middle), suggesting that it is innervated by inputs from the entorhinal cortex (solid blue line at left). B) During active avoidance on the rotating arena, the same CA1 place cell (remapped place field shown on far right) now tends to fire action potentials toward the trough of local theta oscillations (middle), suggesting that it is now innervated by inputs from the CA3 (solid red line at left). C) Illustration of baseline networks states during foraging in relation to cognitive outcome defined by performance in the active avoidance task. The top portion illustrates the mean phase preference of individual CA1 place cells (black lines) of a subset of FSE animals that are unable to learn the active avoidance task (FSE-NL). These cells tend to exhibit preferred phases of firing at different phases of theta that may represent changes in the microcircuitry that underpin phase preference and the routing of spatial information between brain regions. The bottom portion illustrates phase preference of CA1 place cells from FSE animals that are able to learn the active avoidance task (FSE-L). Collectively, these cells tend to prefer to fire toward the peak of theta oscillations (conceptual illustration of corresponding rayleigh vectors are shown at right). This tendency may improve the coincidence of CA1 cell activity with inputs from the entorhinal cortex in a manner that may improve the underlying cellular mechanisms of learning. Ultimately, this alteration of temporal organization could offset the effects of FSE and increase the probability of learning the active avoidance task.