New ways of looking at an old disease: the reimagination of epilepsy

A standardised definition of SE has long evaded medical terminology with previous definitions relying on ambiguous phrases including a seizure as a “fixed and enduring condition”, or one persisting for a “sufficient length of time.” The authors of this paper have sought to provide a new definition of SE, with a stated goal of providing a structured terminology, but one which could evolve as advances in neurobiology and pathophysiology became apparent. The suggested terms consist of a time point (t ₁) after which time a given seizure is considered to be “an abnormally prolonged seizure”, and a second time point (t ₂) after which time neuronal injury is thought to occur. This second time point provides a degree of impetus about how aggressive one should be in treating an abnormally prolonged seizure. The values of these time points vary depending on underlying seizure type.

This new classification of SE involves four axes: the semiology of the seizure, aetiology, EEG correlate, and the age of the patient. The goals of reclassification are to improve communication between healthcare professionals, improve treatment of patients, permit epidemiological studies, and guide basic research.

The reclassification of SE provides a more reasoned argument. Certainly, improved communication between healthcare professionals is needed in the emergency setting. However, this also highlights a limitation since it is directed at an audience most accepting of the proposed changes. It could be argued that the optimum target for reclassification should be Emergency Physicians who are more likely to encounter SE at presentation. Nevertheless, this more accurate guide to the description, semiology and aetiology of the seizure will undoubtedly aid management. Less helpful in the acute management (but more so with ongoing management) is an EEG correlate, an investigation that may not always be provided in a timely fashion.

Trinka E et al. (2015) Epilepsia 56(10):1515–1523.

The identification of genetic underpinnings of generalised epilepsy has allowed for the development of animal models of these epilepsies. This paper described the situation of a GABA_Aγ2(R43Q) mouse which replicates the human phenotype of absence seizures with spike-and-wave discharges, and reduced threshold to thermal seizures. A pedigree with a pure febrile seizure phenotype was chosen to determine the effect of this mutation on cortical development. A labelled, viral vector was inserted into the thalamus to aid in histological interpretation. A putative area of cortex was studied against a control area in the somatosensory cortex. The authors found an increased density of GABAergic neurons, but also that the ratio between excitatory and inhibitory GABAergic neurons was reduced (suggesting that inhibitory neuron density alone did not predict function). Additionally, changes in the diversity of inhibitory neurons were seen at different cortical layers, especially in layers associated with high levels of expression of the γ2 subunit during embryonic development. This paper suggested that microscopic changes are seen in specific areas of the cortex associated with a specific model of a genetic generalised epilepsy.

Comment. This paper is an important first step in the understanding of the effects of mutations on cortical structures, and provides a springboard for future efforts and replication. However, there are a number of limitations to the paper itself: Firstly the SCN1A mutation is the more well-characterised mutation but fails historically as a mouse model. Secondly, the GABA_Aγ2 mouse was used as its own control for neuron density and type. Thirdly, the area of cortex studied is not well developed in humans. It would have been a useful adjunct to this paper, if mouse MRI data were also included.

However, this paper does provide an interesting insight suggesting that it is not solely inhibitory neuron density that is responsible for changes in cortical excitability, but perhaps the ratio with excitatory neurons. This may provide an explanation of the differences seen in transcranial magnetic stimulation excitability in different epilepsies.

Wimmer VC et al. (2015) Neurology 84:1308–1316.

This paper reports a well-designed, multicentre, randomised, placebo-controlled trial that demonstrates clear efficacy for perampanel in the adjunctive management of patients with refractory generalised epilepsy. The population group and range of current medications was similar to that in clinical practice. There was a 76.5 % reduction in median seizure frequency in the treatment group, with a greater than 50 % response rate seen in 64.2 % of the treatment group. Complete seizure freedom was achieved in 23.5 % of the treatment group. Adverse events (AE) were comparable between the two groups, and serious AE were only seen in a minority of patients, and were manageable with medication withdrawal. One death occurred in each group: SUDEP in the placebo arm, and an accidental drowning in the treatment arm. No biochemical abnormalities were noted in either group. This paper provided justification for inclusion of perampanel into the pharmacopoeia for refractory generalised epilepsy.

The drug was well tolerated and no unexpected AE were identified. One SUDEP death amongst the placebo group continues to raise concerns about the design of trials in generalised epilepsy that have placebo arms, and how to protect these at-risk patients. Earlier trials of perampanel had allowed greater doses (up to 12 mg) that were not seen in this paper (maximum 8 mg). The reason for this was not stated.

The trial is now undergoing an open-label extension and the results of this are eagerly awaited.

French JA et al. (2015) Neurology 85:950–957.