34 Cortical and Deep Brain Stimulation

10.1055/b-0034-84145

34 Cortical and Deep Brain Stimulation

Kossoff, Eric H., Jallo, George I.

Throughout Part II of this textbook, the risks and benefits of surgical procedures to either remove an epileptic focus or palliate intractable multifocal epilepsy have been discussed. As the reader is well aware, there are situations in which a child has epilepsy that is generalized in onset, focal but localized to a region of the brain that would cause significant functional morbidity, or the child is at too high of a surgical risk because of medical co morbidities. In addition to this, as is discussed in Chapter 36, surgical failures do occur, and further nonsurgical (but nonpharmacological) therapies are requested by parents.

What other options do parents and their children have? At this time, the two major nonpharmacological therapies that are reliably proven to reduce seizure frequency are the ketogenic diet and vagus nerve stimulation. Both of these therapies have had an incredible growth in both clinical and research interest over the past decade, even considering the near doubling in the number of anticonvulsant drugs available.

One of the most exciting areas of research in epilepsy is the use of electricity to abort seizures, not just by continuous stimulation of the vagus nerve but also through direct brain stimulation to the cortex as well as deeper brain structures. Although innovative and likely more advantageous than vagus nerve stimulation, primarily because of closer proximity to the epileptic foci or circuits, the time involvement and preparation of both the epileptologist and neurosurgeon are exponentially greater to use these new technologies. This chapter discusses the current methods under investigation to provide electrical stimulation to the brain. Most of the research to date involves adults, but limited data do exist for children, and future directions of these therapies in the pediatric age range will be presented at the conclusion of this chapter.

Brain Stimulation Regions

Subcortical Structures

Perhaps the most frequently studied regions of the brain for direct neurostimulation have been the subcortical structures, most notably the thalamus, caudate, and cerebellum. Advances in safety and accuracy of the stereotactic placement of deep brain electrodes, especially for movement disorders such as Parkinson disease, has led to a strong interest in this technique to reach brain structures that animal research has identified as involved in the spread of seizures. These structures, although not typically the exact epileptic focus, have widespread excitatory and inhibitory connection and are safe to stimulate without causing a loss of function.

Cerebellum and Caudate

Both controlled and uncontrolled trials of stimulation to the cerebellum and caudate nucleus have shown mixed results.1–4 The caudate nucleus is a theoretically attractive target for stimulation because of its connectivity with the thalamus and neocortex. Only one study from 1997 showed benefits, but it has not been confirmed to date with a controlled trial.4 The cerebellum also has excitatory connections to the thalamus, which can be theoretically inhibited by neurostimulation. One relatively large study of 32 adults from 1992 reported 23 patients (85%) with seizure reduction as a result of cerebellar stimulation.1 A double-blind study from 2005 of five patients demonstrated a mean 67% seizure reduction in patients with active therapy versus 7% reduction in those with therapy not turned on (p = .02).3 Interestingly, seizure improvement continued to occur in patients over the 2-year study period.

Thalamus

Studies of the caudate and cerebellum have been less frequent in the past decade, as alternative subcortical structures have been identified as perhaps more promising. The thalamus specifically appears to be the most likely subcortical structure to result in seizure reduction when stimulated because of its widespread inhibitory connections with the cortex and evidence for possible indirect stimulation by vagus nerve stimulation over the past two decades of use. One of the first thalamic nuclei studied was the centromedian nucleus, because of its projections to the entire cerebral cortex from ascending brainstem regions. The efficacy of this approach is unclear, however, with both positive (for Lennox-Gastaut syndrome)5 and negative (for complex partial seizures) re-sults.6 The subthalamic nucleus has also been stimulated in small studies, with only modest results.7,8 One case report included a child with a focal centroparietal dysplasia who received benefit from stimulation of this nucleus.9

In recent years, there has been significant interest in stimulation of the anterior nucleus of the thalamus. This small structure is distant from basal vascular structures yet has connections to the limbic system via the circuit of Papez, which connects the hippocampus, fornix, and mammillary bodies to the cingulum and parahippocampal cortex in a functional loop.10 The first pilot study of five patients who underwent bilateral anterior thalamic stimulation demonstrated a mean seizure reduction of 54%; however, patients who were implanted but not stimulated had similar responses interestingly.11 Other small studies have mostly replicated these results; one of the more recent studies of four adults demonstrated a slightly higher mean reduction of 76% in monthly seizure frequency.12

On the basis of these early results, this thalamic nucleus has been recently investigated in a multicenter study for intractable partial epilepsy sponsored and just completed by Medtronics, Inc., titled SANTE (Stimulation of the Anterior Nucleus of the Thalamus for Epilepsy). The surgical procedure to implant these electrodes is described well in a recent review by Halpern, et al13 and the device in an adult subject is depicted in Fig. 34.1 . For inclusion, patients were required to be between the ages of 18 and 65 years, have failed 12 to 18 months trial of at least two anticonvulsant drugs, and have at least six partial and disabling seizures per month. Study results should be forthcoming soon.