Antiepileptic drugs (AEDs) are the first-line and most commonly used treatment for seizures. Treatment with 1 AED, or monotherapy, leads to complete resolution of seizures in approximately 50% of patients.1 Of the other half of patients, some will achieve seizure control with multiple AEDs. Unfortunately, as many as 30% of all patients with epilepsy continue to have seizures despite multiple medications at high doses. These patients are generally referred to surgical centers for possible resective surgery; however, not all patients with medically intractable epilepsy are candidates for resection. There are also several limitations to using AEDs including allergies, behavioral side effects, cognitive limitations, and poor adherence.
When patients have epilepsy that does not respond to AEDs, termed medically refractory epilepsy, have contraindications to AEDs, and are not surgical candidates, implantable devices continue to be an emerging option. The 3 types of implantable devices approved for treatment of people with epilepsy that have proven efficacy in clinical trials are vagus nerve stimulation (VNS), deep brain stimulation (DBS), and responsive neurostimulation (RNS). All of these therapies are considered neuromodulation, which, as defined by the International Neuromodulation Society, is the alteration of nerve activity through targeted delivery of a stimulus, such as electrical stimulation or chemical agents, to specific neurologic sites.
Vagus Nerve Stimulation
Approved by the Food and Drug Administration (FDA) since 1997, VNS has been used with good outcomes for people with medically refractory epilepsy. In VNS, leads are implanted around the left vagus nerve and then connected to a generator that is implanted under the skin in the patient’s chest, just below the left clavicle. The generator produces electrical impulses that modulate the vagus nerve. Although there is no consensus on the mechanism of action for VNS, studies in animal models suggest VNS may modulate noradrenaline release from the locus coeruleus or cause changes in brain blood flow and metabolism in multiple areas including the thalamus.2 After surgical implantation, the device is programmed in the outpatient setting by the patient’s epileptologist. The device is easily interrogated and reprogrammed wirelessly in the office. Frequency, output current, pulse width, signal-on time, signal-off time, and magnet parameters can be set. Magnet settings are an added advantage. Patients or caregivers are given a magnet that they can use to activate the VNS as needed if aura or seizure occurs. Newer versions allow for programming so that a rapid increase in heart rate that is often seen at the start of a seizure triggers the device. A long-term follow-up study has shown responder rates of 37% at year 1 and 43% at years 2 and 3.1 This trend of continued improvement in seizure control is common to all 3 devices. Improved seizure control can be seen 18 months to 2 years following implantation of VNS.2
Intraoperative surgical risk is relatively low for VNS implantation. Local site infection and implant-site pain are also small risks of the device. The most common side effect is overstimulation of the VNS, which can result in local spreading of current causing reversible vocal hoarseness, coughing, and throat pain. These symptoms are reversed by lowering the stimulation and limiting how much vagus stimulation can be delivered. Battery life is approximately 10 years, depending on VNS settings, before the generator device needs to be surgically accessed and replaced. Obstructive sleep apnea has been reported as a complication of VNS.3
Deep Brain Stimulation
Approved by the FDA in 2018, DBS provides electrical stimulation to the anterior nucleus of thalamus (ANT) for treatment of medically refractory epilepsy. In the SANTEa trial, among the first to show evidence for DBS in 2005, patients ages18 to 65 who had partial seizures with or without secondary generalization, failed trials of at least 3 different AEDs, and at least 6 seizures per month were evaluated in this prospective randomized double-blind parallel group design. Similar to the VNS device, the response rate in the SANTE trial was 40.5% with continued improvement noted at the end of 2 years.2 The mechanism of action of DBS is unknown and there is controversy regarding whether the effect of DBS on the thalamus is inhibitory or excitatory.1 Surgical implantation of DBS is more complicated than that of VNS. After burr holes are created in the skull, and using advanced imaging and both stereotactic and functional neurosurgery, DBS electrodes are implanted bilaterally in the ANT. The electrode leads are run subcutaneously to the left chest wall where a generator is implanted. Similar to VNS, the DBS is programmed in the outpatient setting wirelessly with a programming wand by the patient’s epileptologist.
As with any surgery, there is risk of intraoperative death, infection, or hemorrhage. The most common complications are paresthesia, lead repositioning, and superficial site infections. Hemorrhages have been noted but are most commonly asymptomatic.4 However, because DBS is becoming more popular and the surgery is being more commonly performed, the risk is low. Other potential complications are status epilepticus, sudden death, and neuropsychiatric complications like Kluver-Bucy syndrome.5
Approved by the FDA in 2013 as an adjunctive treatment for people with medically refractory epilepsy, RNS is an intracranially implanted device with 2 leads that can both record intracranial electric activity and provide dynamic electric stimulation to prevent seizure progression. Using what is referred to as closed-loop stimulation, the device detects seizure activity in the brain by monitoring electrocorticographic activity and provides electric stimulation directly at the seizure focus to prevent seizure propagation. For proper lead placement, good localization of the epileptogenic zone is essential. As with VNS and DBS, RNS trials have shown a similar pattern of efficacy. In the pivotal clinical trial, there was a 37.9% reduction in seizures in the active group with efficacy improving to 56% by the end of 2 years, and further studies have demonstrated continued improvement with time.1,2
Complications include implant site infection, implant site pain, seizure, worsening memory problems, and headache.1 Intracranial hemorrhage and death remain a possibility; however, the majority of side effects seen are self-limited and mild.
Neuromodulation devices increasingly are an option for patients with medically refractory epilepsy who are not candidates for resective surgery. General neurologists should consider referral to epilepsy centers for patients with intractable epilepsy because patients who refuse or are not candidates for surgery may consider or be candidates for devices. Both patients and physicians must have realistic expectations of efficacy because rates of seizure freedom with these devices are relatively low. Device use does not eliminate the need for AEDs, although it may allow a reduction in AED use and thereby minimize side effects of AEDs. Devices in general are not additive to the side-effect profile commonly seen with AEDs. All 3 approved neuromodulation devices have demonstrated similar efficacy in the initial treatment period with improvement in seizure control that gets better over time. The full extent of this improvement may not yet be known, particularly for DBS and RNS. This progressive improvement may be caused by downstream network changes that evolve with time as a result of the neuromodulation. Currently there are no head-to-head trials to guide selection of a specific device over the others. Determining which device is best for each patient is best done after comprehensive case evaluation and discussion with the patient.
1. Cox JH, Seri S, Cavanna AE. Clinical utility of implantable neurostimulation devices as adjunctive treatment of uncontrolled seizures. Neuropsychiatr Dis Treat. 2014;10:2191-2200.
2. Ben-Menachem, E. Neurostimulation-past, present and beyond. Epilep Curr. 2010;12(5):188-191.
3. Marzec M, Edwards J, Sagher O, Fromes G, Malow BA. Effects of vagus nerve stimulation on sleep-related breathing in epilepsy patients. Epilepsia. 2003;44(7):930-935.
4. Fisher RS. Therapeutic devices for epilepsy. Ann Neurol. 2012;71(2):157-168.
5. Demetriades P, Rickards H, Cavanna AE. Impulse control disorders following deep brain stimulation of the subthalamic nucleus in Parkinson’s disease: clinical aspects. Parkinsons Dis. 2011;2011:658415.