Other attempts at neuromodulation don’t involve electric or magnetic stimulation but instead involve cooling. Cooling chips have been used. Mechanistically, a current is passed through a chip device and this causes one side of the chip to become cooler and the other side warmer. If the cooler side of the chip is placed on the surface of the brain and the warmer side of the chip on the meningeal side, one can cool the brain and get the heat carried away by the blood vessels of the meninges. Putting a cooling device on an area that is thought to be epileptic can, at least temporarily, shut down seizure activity. This has been shown to be effective in a number of animal models and could, at least in principle, be used to cool human epileptic foci and prevent them from producing seizures.
When to Consider Neuromodulation
Neuromodulation should be considered—and the devices that are now available in the United States are the vagus nerve stimulator and the NeuroPace cortical stimulator—any time that a person is continuing to have seizures after trials of two or three epileptic drugs. Of course the treating physician should also take into account the type of epilepsy that the patient has. So far, the official approval for both vagus nerve stimulation and for the NeuroPace stimulator is for the treatment of focal and secondarily generalized seizures, not for primary generalized seizures, although off-label vagus nerve stimulation has been used for primary generalized epilepsy with some success.
Patient Selection
Neuromodulatory strategies should be reserved for patients whose seizures are medically refractory. How one would define medically refractory varies quite a lot but most neurologists would agree that if a person has not done well with two or three drugs alone, and on a combination of two drugs, then that person qualifies as having medically refractory epilepsy and should be assessed both for neuromodulation and also for epilepsy surgery. Epilepsy surgery could be focal resection or much less frequently corpus callosotomy. In suitable candidates resective epilepsy surgery is preferable to neuromodulation as there is a much higher likelihood of making a patient seizure free.
Neuromodulation Outcomes
It is very hard to make comparisons between outcomes achieved with medication, surgery and neuromodulation, because it is easy to fall into a trap of comparing apples and oranges. If you look at the big picture, one would say that if you take newly diagnosed epilepsy patients, about half of them will have full seizure control on the first drug that you try, provided you’ve picked that drug intelligently. That leaves half the patients who are not doing well on the first drug. With judicious selection of other drugs, alone and in combination, probably another half can be made seizure-free without intolerable side effects. That leaves about 25% of your total population who have medically refractory epilepsy. Of those who are medically refractory, probably half are realistic candidates for epilepsy surgery. Removing areas of epileptogenic brain will probably produce complete seizure control in 50 % of those patients. That leaves you with 15 or 20 % of your original cohort who are medically refractory and have either failed resective epilepsy surgery or are not eligible for it. They would be candidates for one of the stimulator devices. This is probably an underestimate of the population who are candidates for stimulator devices, because a lot of people have side effects on medications even when seizures are controlled, and a lot of people don’t want brain surgery. So perhaps 30% of the population with epilepsy are realistic candidates for neuromodulation.
If you look at the people who are medically refractory and do successive drug trials on them, roughly speaking about a third of them will get 50% better, and maybe 5% will become seizure-free. The results with neuromodulation are not much better. Vagus nerve stimulation will make perhaps 50% of patients 50% better and maybe 5% of patients seizure-free. NeuroPace, the responsive cortical stimulator, will again perhaps make 50% of patients 50% better and maybe 5% or 10% seizure-free. So neuromodulation is somewhat better than trying out any one new drug, and it has the advantage of having no drug-related side effects. However, in terms of effectiveness, neuromodulation and stimulation is not much better than continuing with drug trials. It may be somewhat better and certainly has fewer side effects in terms of sedation, nausea, dizziness, etc.
In contrast to neuromodulation, resective epilepsy surgery probably has a seizure control rate of 50%. So in that sense neuromodulation is still by no means as effective as resective surgery. It is probably somewhat more effective than trying drug after drug, and with a different side effect profile than drugs. Many patients seem to prefer it. It should also be noted that the vagus nerve stimulator has an indication as a treatment for depression and most patients who get one do feel rather more lively.
