Repeated consciousness-impairing seizures weaken the reticular activating system’s connections, possibly leading to the baseline neurocognitive dysfunction these patients experience

How does the localised ictal activity of a temporal lobe seizure generate a global impairment of consciousness? And how do brief temporal lobe seizures beget long-lasting interictal changes in cognition? Dr Englot and his colleagues have pursued these questions over the past decade through several innovative studies, and have convincingly implicated the ascending reticular activating system (ARAS) as an important contributor.1–3 In their most recent JNNP study,4 they further incriminate the ARAS, although now as a source of interictal neurocognitive impairment.

According to the model previously proposed by Englot and Blumenfeld,2 ictal activity in the temporal lobe leads to suppression of the ARAS, which in turn produces sleep-like slow wave activity in the neocortex, correlating with loss of consciousness. In their current study, the authors hypothesised that repeated seizures eventually produce long-lasting changes in ARAS connectivity. These long-lasting connectivity changes might then lead to the neurocognitive dysfunction suffered by patients with temporal lobe epilepsy (TLE).

To assess these changes in connectivity, Englot and colleagues used functional MRI of resting-state brain activity. Leveraging publicly available atlases, they were able to segment ARAS structures in patients with epilepsy and matched controls, and then estimate functional connectivity between each pair of regions. This technique revealed that ARAS connectivity was lower in patients with TLE, and how low correlated with the frequency of consciousness-impairing (that is, ARAS-involving) seizures.

This hypothesis is attractive prima facie. Studies have linked the ARAS to arousal, learning, mood and many other domains affected by TLE.2 If borne out by further study, reduced ARAS connectivity would be a thrifty explanation of the many baseline neurocognitive disturbances patients with TLE experience.

This putative change in ARAS connectivity is in some ways counterintuitive. Why should brief epochs of reduced ARAS activity during seizures lead to long-term depotentiation of ARAS circuits? Many neural circuits in fact exhibit the opposite behaviour, where decreased input leads to eventual hyperexcitability.5 And why, in terms of plasticity, should the decreased ARAS activity during seizures be any different from the far longer decreased ARAS activity seen nightly during normal sleep? What drives this plasticity?

Regardless of mechanism, this study has clear therapeutic implications, suggesting neuromodulation of ARAS targets when resective surgery is not possible. It will also be interesting to see if vagus nerve stimulation or existing forms of neurostimulation modify these same ARAS connections as part of their efficacy.

Overall, this research reaffirms that extratemporal networks, like the ARAS, are intimately involved in TLE. Epilepsy is a network disease, and research should be directed at manipulating these networks as a means to both prevent seizures and improve neurocognitive comorbidities.