Invasive electrocorticographic monitoring of epilepsy patients is performed typically when scalp recordings and imaging data are inconclusive about the location of probable ictal onset. These recordings are performed with a variety of commercially available electrode systems, which include subdural grids, strips, and depth electrodes. Most subdural grid and strip arrays are manufactured with 1 cm distance between the recording elements, with each electrode contact having a diameter of a few millimeters. Schevon et al (Schevon CA, Weiss SA, McKhann G, Goodman RR, Yuste R, Emerson R, Trevelyan AJ. Evidence of an inhibitory restraint of seizure activity in humans. Nature Comm 2012;3:1060.), have now examined extracellular action potential recordings using a small microelectrode array at much smaller spatial and temporal scales than typical clinical recording elements. The authors have demonstrated clear differences in the behavior of neurons within tissue regions affected by the ictal spread vs a penumbra region which resists recruitment in the seizure. These characteristics of an evolving seizure cannot be demonstrated by traditional recording methods, and may imply the role of an active feedforward cortical inhibitory network influencing seizure evolution.
In this report, Schevon et al describe the use of both an animal model of seizure propagation (the low magnesium model investigated in mouse brain slice preparations) and microelectrode recordings in humans to investigate the propagation of epileptiform activity. The authors performed patch clamp recordings in neocortical Layer 5 pyramidal cells (often performed simultaneously with 2 neurons) as well as calcium sensitive dye imaging of the ongoing low magnesium induced network activity. Four human subjects were reported in this work as well. These consisted of patients with refractory epilepsy undergoing invasive monitoring for seizure localization and cortical mapping. The microelectrode arrays consisted of 96 microelectrodes arranged in a 4 mm × 4 mm square lattice array. The implant site for these arrays was kept close to the hypothesized seizure onset region and away from eloquent cortex. The signals from the microelectrodes were bandpass filtered (0.3 Hz–7.5 kHz) so that extracellular action potentials could be reliably detected and sorted.
In the mouse slice preparations the authors were able to study the dynamics of several ictal events by performing whole cell patch clamp recordings in cells separated by ∼600 µm, and discern transitions from predominantly inhibitory PSP barrages to excitatory PSP barrages as ictal propagation proceeded into a given recorded region. Ca2+ imaging was also used in the mouse slices, where failure of propagation into a region of tissue neighboring the ictal onset could also be detected. Similarly, from the human recordings, ictal activity was demonstrated as well as spike firing features associated with the penetration or lack of penetration of spreading depolarization waves through tissue (Figure 1). The propagation speeds of the ictal depolarization waves recorded were on the order of 0.1 to 0.2 mm/sec, slower than disinhibited slice preparations and possibly demonstrating the effects of feed forward inhibition. Additionally, the Fano factor measure was used to measure the variations in spike firing as an ictal event penetrated a region of tissue. Large Fano factor values were demonstrated at ictal onset, but small values as the ictal depolarization spread through and incorporated the measured tissue area. The authors go on to make the hypothesis that given these findings of differences in the regions of ictal penetration and ictal penumbra, the tissue resection needed for seizure control may actually be smaller than plans provided by traditional electrocorticography.
In summary, Schevon et al have examined the spread of ictal activity using whole cell patch clamp recordings in mouse brain slices and in microelectrode array recordings in human epilepsy subjects. They have firmed up evidence for the presence of inhibitory restraint mechanisms present in brain tissue which act to limit the spread of ictal propagation as reported in other animal preparations. As the authors point out, this could well have implications regarding resective surgery for epilepsy. If epilepsy teams can consistently identify boundary regions of tissue demonstrating full ictal involvement vs an ictal penumbra (with an active inhibitory brake) then more precise resections can possibly be performed, and regions of cortex spared.