Description and computational modeling of the whole course of status epilepticus induced by low dose lithium–pilocarpine in rats

The main purposes of this study were (1) to describe the whole course of status epilepticus induced by a low dose lithium–pilocarpine model in rats, including depth-EEG from the hippocampus, ECoG from cortex and gross behaviors, and (2) to investigate the possible changes of the intrinsic neural network in the hippocampus during the status epilepticus by model simulation. The course of the induced status epilepticus was divided into baseline, pre-ictal, episodic, onset, continuous, and convalescence stages. At the pre-ictal stage, the main component of the spectrum of the depth-EEG shifted before seizure activity first appeared in the hippocampus at the episodic stage and propagated to the cortex at the onset stage. Model simulation indicated that the changes of depth-EEG of the whole course could be simulated by changes of the loop gains and the inputs of the neural network. The excitatory and inhibitory loops were first enhanced in the pre-ictal stage. The isolated seizures in the episodic stage were caused by variations of inputs from other pyramidal cells. The gain of slow inhibitory loop gradually decreased, leading to the onset of status epilepticus. Different types of waveforms in the status epilepticus corresponded to the different ratio of excitatory and inhibitory influence, caused by different levels of inhibition. Finally, in the convalescence stage, the gains of the excitatory loop and the inhibitory loop both changed toward the base line, recovering the balance. The changes of the parameters in the model were compatible with the experimental results in the literature.

► The time courses of depth-EEG, ECoG, and behaviors were observed during status epilepticus.
► Status epilepticus induced by a low dose of lithium–pilocarpine could be divided into five stages.
► A computational neural model reproduced depth-EEG in all stages of status epilepticus.
► The changes of the neural networks in the hippocampus can be simulated by the model.