Introduction

The inability to treat patients with refractory epilepsy provides a continuous impetus to investigate novel forms of treatment. Neurostimulation is an emerging treatment for neurological diseases. Electrical or magnetic currents are administered directly to or in the neighborhood of nervous tissue to manipulate a pathological substrate and to achieve a symptomatic, or even curative, therapeutic effect. Depending on the part of the nervous system that is being affected and the way stimulation is being administered, different types of neurostimulation are distinguished.

Neurostimulation is not a new technique. The earliest recorded human effort at neurostimulation may have been that of the Mesopotamian healer Largus, who applied electrical torpedo fish to the human body and evoked an immediate and residual numbness in an extremity. Following the development of the battery by Volta, Faraday and Franklin experimented with electricity, giving rise to devices that could transcutaneously affect nerves and representing the precursors of today’s TENS systems. In the early 20th century an “electreat” device was patented by Charles Kent to treat pain. In the 1950s efforts were made in a number of medical device manufacturing companies in collaboration with universities to miniaturize these devices and make them implantable. Around this time electrodes to insert into the brain and record or stimulate were also designed. Initially the major clinical indications for neurostimulation were psychiatric (e.g., treatment of schizophrenia) and refractory pain. More recent progress in biotechnology in combination with successes of neurostimulation in movement disorders has led to renewed interest to investigate neurostimulation as a therapeutic option in various neurological and psychiatric disorders.

Electrical stimulation of the tenth cranial nerve or vagus nerve stimulation (VNS) is an extracranial but invasive type of stimulation that was developed in the 1980s and is currently routinely available in epilepsy centers around the world. Through an implanted device and electrode, electrical pulses are administered to the afferent fibers of the left vagus nerve in the neck (Figure 21-1). It is indicated in patients with refractory epilepsy who are unsuitable candidates for epilepsy surgery or who have had insufficient benefit from such a treatment.¹ As stimulation is applied to that part of the vagus nerve that passes through the neck, direct intracerebral manipulation is unnecessary. Other cranial nerves are being targeted to treat refractory seizures. Preliminary but promising results are available for noninvasive trigeminal nerve stimulation (TNS).²

Figure 21–1 X-ray of Neurocybernetic Prosthesis System for vagus nerve stimulation.

Transcranial magnetic stimulation (TMS) and direct current stimulation (tDCS) represent different types of extracranial and noninvasive neurostimulation techniques.³ In TMS, a coil that transmits magnetic fields is held over the scalp and allows a noninvasive evaluation of separate excitatory and inhibitory functions of the cerebral cortex. In addition, repetitive TMS (rTMS) can modulate the excitability of cortical networks.⁴ This therapeutic form of TMS is currently being investigated as a treatment option for refractory epilepsy with varying results.⁵ tDCS uses sponge electrodes attached to the patient’s head to deliver electrical currents over longer periods of time (minutes) to achieve changes in cortical excitability that persist even after stimulation has ceased, hence with therapeutic potential in diseases characterized by a disturbed cortical excitability.³

Intracerebral neurostimulation requires accessing the intracranial nervous system as stimulation electrodes are inserted into intracerebral targets for “deep brain stimulation” (DBS) or placed over the cortical convexity for “cortical stimulation” (CS). These modalities of neurostimulation are not entirely new for neurological indications. Some have been extensively used (e.g., for movement disorders and pain).^6,⁷ Moreover, several new indications such as obsessive compulsive behavior and cluster headaches are being investigated with promising results.^8,⁹ In the past, DBS and CS of different brain structures such as the cerebellum, the locus coeruleus, and the thalamus have already been performed. This was done mostly in patients with spasticity or psychiatric disorders who also had epilepsy, but the technique was not fully explored or developed into an efficacious treatment option.⁹^–¹³ The vast progress in biotechnology along with the experience in other neurological diseases in the past 10 years has led to a renewed interest in intracerebral stimulation for epilepsy. Several epilepsy centers around the world have recently reinitiated trials with DBS in different intracerebral structures such as the thalamus, the subthalamic nucleus, the caudate nucleus, and medial temporal lobe structures.¹⁴^–¹⁹ Also, CS is being investigated in a multicenter trial and incorporated in a so-called closed-loop system (the responsive neurostimulator system, RNS).²⁰

VNS on the one hand and TNS, TMS, tDCS, DBS, and CS on the other hand are currently at different levels of availability and clinical applicability. VNS is widely available with over 50,000 patients currently being treated. For therapeutic TNS, proof of concept has been shown. Therapeutic TMS protocols for epilepsy have been developed in centers with a large experience in diagnostic TMS, but at this time, TMS is not a routinely available treatment in epilepsy centers, nor is tDCS. DBS is under investigation in experimental trials in several specialized centers with large experience in refractory epilepsy and functional neurosurgery. Apart from a group of patients that carry implanted devices from the earlier era of neurostimulation for epilepsy, the more recent studies report results in no more than 150 patients worldwide. CS is considered a therapeutic neurostimulation option for specific types of epilepsy (e.g., neocortical epilepsy and results from the multicenter study in the United States have to be awaited). This chapter will focus on VNS and DBS.

Vagus Nerve Stimulation

Electrical stimulation of the tenth cranial nerve or VNS was developed in the 1980s. In the past decade it has become a valuable option in the therapeutic armamentarium for patients with refractory epilepsy, and it is currently routinely available in epilepsy centers worldwide. Through an implanted device and electrode, electrical pulses are administered to the afferent fibers of the left vagus nerve in the neck. It is indicated in patients with refractory epilepsy who are unsuitable candidates for epilepsy surgery or who have had insufficient benefit from such a treatment.¹

Anatomical Basis and Mechanism of Action

The vagus nerve is a mixed cranial nerve that consists of ∼ 80% afferent fibers originating from the heart, aorta, lungs, and gastrointestinal tract and of ∼ 20% efferent fibers that provide parasympathetic innervation of these structures and also innervate the voluntary striated muscles of the larynx and the pharynx.²¹^–²³ Somata of the efferent fibers are located in the dorsal motor nucleus and nucleus ambiguus, respectively. Afferent fibers that are targeted for therapeutic VNS have their origin in the nodose ganglion and primarily project to the nucleus of the solitary tract. At the cervical level, the vagus nerve mainly consists of small-diameter unmyelinated C-fibers (65 to 80%) and of a smaller portion of intermediate-diameter myelinated B-fibers and large-diameter myelinated A-fibers. The nucleus of the solitary tract connects to other brain stem nuclei and has widespread projections to numerous areas in the forebrain, including important areas for epilepsy such as the amygdala and the thalamus. The diffuse pathways of the vagus nerve mediate important visceral reflexes such as coughing, vomiting, swallowing, control of blood pressure, and heart rate. Heart rate is mostly influenced by the right vagus nerve that has dense projections primarily to the atria of the heart.²⁴

Since the first human implant of the VNS therapy device in 1989, over 50,000 patients have been treated with VNS worldwide. As for many antiepileptic treatments, clinical application of VNS preceded the elucidation of its mechanism of action. Following a limited number of animal experiments in dogs and monkeys, investigating safety and efficacy, the first human trial was performed.²⁵ The basic hypothesis on the mechanics of action was based on the knowledge that the tenth cranial nerve afferents have numerous projections within the central nervous system and that in this way, action potentials generated in vagal afferents have the potential to affect the entire organism.²⁶ To date the precise mechanism of action of VNS and how it suppresses seizures remains to be elucidated.

Crucial questions with regard to the mechanism of VNS occur at different levels. Vagus nerve stimulation aims at inducing action potentials within the different types of fibers that constitute the nerve at the cervical level. The question remains, what fibers are responsible and/or necessary for its seizure-suppressing effect? Unidirectional stimulation, activating afferent vagal fibers, is preferred, as epilepsy is considered a disease with cortical origin, and efferent stimulation may cause side effects. The next step is to identify central nervous system structures located on the anatomical pathways from the cervical part of the vagus nerve up to the cortex that play a functional role in the mechanism of action (MOA) of VNS. These could be central gateway or pacemaker function structures such as the thalamus or more specific targets involved in the pathophysiology of epilepsy, such as the limbic system, or a combination of both. Another issue concerns the identification of the potential involvement of specific neurotransmitters. The intracranial effect of VNS may be based on local or regional GABA increases or glutamate and aspartate decreases or may involve other neurotransmitters that have been shown in the past to have a seizure threshold regulating role such as serotonin and norepinephrine.²⁷

Research directed toward the identification of involved fibers, intracranial structures, and neurotransmitter systems has been performed. Animal experiments and research in humans treated with VNS have comprised electrophysiological studies (EEG, EMG, EP), functional anatomic brain imaging studies (PET, SPECT, fMRI, c-fos, densitometry), and neuropsychological and behavioral studies. Also from the extensive clinical experience with VNS, interesting clues concerning the MOA of VNS have arisen. More recently the role of the vagus nerve in the immune system has been investigated.

From the extensive body of research on the mechanism of action, it has become conceivable that effective stimulation in humans is primarily mediated by afferent vagal A- and B-fibers.^28,²⁹ Unilateral stimulation influences both cerebral hemispheres, as shown in several functional imaging studies.^30,³¹ Crucial brainstem and intracranial structures have been identified and include the locus coeruleus, the nucleus of the solitary tract, the thalamus, and limbic structures.³²^–³⁴ Neurotransmitters playing a role may involve the major inhibitory neurotransmitter GABA but also serotoninergic and adrenergic systems.^35,³⁶ An extensive overview on the mechanism of action of VNS can be found in Vonck et al.³⁷

CLINICAL EFFICACY

The first descriptions of the implantable VNS Therapy System for electrical stimulation of the vagus nerve in humans appeared in the literature in the early 1990s.^38,³⁹ At the same time, initial results from two single-blinded pilot clinical trials (phase-1 trials EO1 and EO2) in a small group of patients with refractory complex partial seizures, who were implanted since November 1988 in three epilepsy centers in the United States, were reported.^25,⁴⁰^–⁴² In nine of 14 patients, treated for 3 to 22 months, a reduction in seizure frequency of at least 50% was observed.³⁹ One of the patients was seizure free for more than 7 months. Complex partial seizures, simple partial seizures, as well as secondary generalized seizures were affected.⁴⁰ It was noticed that a reduction in frequency, duration, and intensity of seizures lagged 4 to 8 weeks after the initiation of treatment.²⁵

In 1993, Uthman et al. reported on the long-term results from the EO1 and EO2 study.⁴³ Fourteen patients had now been treated for 14 to 35 months. There was a mean reduction in seizure frequency of 46%. Five patients had a seizure reduction of at least 50%, of whom two experienced long-term seizure freedom. In none of the patients did VNS induce seizure exacerbation.

In the meantime, two prospective multicenter (n = 17) double-blind randomized studies (EO3 and EO5) were started including patients from centers in the United States (n = 12), Canada (n = 1), as well as in Europe (n = 4).⁴⁴^–⁴⁸ In these two studies, patients over the age of 12 with partial seizures were randomized to a HIGH or LOW stimulation paradigm. The parameters in the HIGH stimulation group (output: gradual increase up to 3.5 mA, 30 Hz, 500 μs, 30 s on, 5 min off) were those believed to be efficacious based on animal data and the initial human pilot studies. Because patients can sense stimulation, the LOW stimulation parameters (output: single increase to point of patient perception, no further increase, 1 Hz, 130 μs, 30 s on, 3 hours off) were chosen to provide some sensation to the patient to protect the blinding of the study. LOW stimulation parameters were believed to be less efficacious, and the patients in this group represented an active control group. The results of EO3 in 113 patients were promising with a decrease in seizures of 24% in the HIGH stimulation group versus 6% in the LOW stimulation group after 3 months of treatment.⁴⁵^–⁴⁷ The number of patients was insufficient to achieve Food and Drug Administration (FDA) approval leading to the EO5 study in the United States including 198 patients. Ninety-four patients in the HIGH stimulation group had a 28% decrease in seizure frequency versus 15% in patients in the LOW stimulation group.⁴⁸

The controlled EO3 and EO5 studies had their primary efficacy endpoint after 12 weeks of VNS. Patients who ended the controlled trials were offered enrollment in a long-term (1 to 3 years of follow-up [FU]) prospective efficacy and safety study. Patients belonging to the LOW stimulation groups were crossed-over to HIGH stimulation parameters. In all published reports on these long-term results increased efficacy with longer treatment was found.⁴⁹^–⁵³ In these open extension trials, the mean reduction in seizure frequency increased up to 35% at 1 year and up to 44% at 2 years of FU. After that improved seizure control reached a plateau.⁵²

In the following years, other large prospective clinical trials were conducted in different epilepsy centers worldwide. In Sweden, long-term follow-up FU in the largest patient series (n = 67) in one center not belonging to the sponsored clinical trials at that time reported similar efficacy rates with a mean decrease in seizure frequency of 44% in patients treated up to 5 years.⁵⁴ A joint study of two epilepsy centers in Belgium and the United States included 118 patients with a minimum FU duration of 6 months. They found a mean reduction in monthly seizure frequency of 55%.⁵⁵ Only in a minority of patients (7%) long-term seizure freedom was achieved. In China a mean seizure reduction of 40% was found in 13 patients after 18 months of VNS.⁵⁶ From a clinical point of view, prospective randomized trials investigating long-term efficacy in comparison to other therapeutic options for patients with refractory epilepsy are still lacking. An ongoing multicenter randomized trial called PulSE is currently recruiting patients worldwide and may shed light on the exact position of VNS. On the basis of currently available data the responder rate in patients treated with VNS is not substantially higher compared to recently marketed antiepileptic drugs.

Children

There are no controlled studies of VNS in children, but many epilepsy centers have reported safety and efficacy results in patients less than 18 years of age in a prospective way. All these studies report similar efficacy and safety profiles compared to findings in adults.⁵⁷^–⁶⁰ Rare adverse events, unique to this age group, included drooling and increased hyperactivity.⁶¹ In children with epileptic encephalopathies, efficacy may become evident only after >12 months of treatment.⁶² A recent Korean multicenter study evaluated long-term efficacy in 28 children with intractable epilepsy. In half of the children there was a >50% seizure reduction after a FU of at least 12 months.⁶³ In our own prospective analysis of 118 patients, 13 children with a mean age of 12 years (range: 4 to 17 years) were included with similar efficacy rates and without specific side effects.⁵⁵

Elderly

A study of Sirven et al. included 45 patients who were 50 years of age and older. Thirty-one of 45 patients had a FU of 1 year, with a reported responder rate of 68%, good tolerance, and improvement of quality-of-life scores.⁶⁴

Seizure Type and Syndrome

The clinical studies EO1, EO2, EO3, and EO5 included patients with partial epilepsy. This is a reflection of the fact that patients considered for treatment with VNS were initially evaluated for resective surgery, the preferred treatment for partial epilepsy, but turned out to be unsuitable surgical candidates.

The open-label longitudinal multicenter EO4 study also included patients with generalized epilepsy (n = 24).^65,⁶⁶ In these patients, overall seizure frequency reduction was 46%. Generalized tonic seizures responded significantly better compared with generalized tonic-clonic seizures. Quintana et al.,⁶⁷ Michael and Devinsky⁶⁸ and Kostov et al.⁶⁹ described in a retrospective way that primary generalized seizures and generalized epilepsy syndromes responded equally well to VNS, compared with partial epilepsy syndromes. A prospective study of Holmes et al. in 16 patients with generalized epilepsy syndromes and stable antiepileptic drug (AED) regimens showed an overall mean seizure frequency reduction of 43% after a FU of at least 12 months.⁷⁰ Ben-Menachem et al. included nine patients with generalized seizures in a prospective long-term FU study. Especially important, the patients with absence epilepsy had a significant seizure reduction.⁵⁴

A few studies are available specifically describing the use of VNS in patients diagnosed with Lennox-Gastaut syndrome (LGS). One prospective study in 16 patients with Lennox-Gastaut (FU = 6 months) found that one-quarter of patients had a >50% reduction in seizure frequency, which is comparable to the response rates in the controlled studies that included a few patients with LGS.⁷¹ Other prospective studies reported higher responder rates with a >50% seizure frequency reduction in half of the patients (n = 13, FU = 6 months),⁷² in six of seven patients (FU = 6 months)⁷³ and in seven of nine patients (FU = 1 to 35 months).⁷⁴ A retrospective multicenter study in 46 patients with LGS reported responder rates of 43%.⁷⁵

There have been many reports on various other seizure types and syndromes, such as seizures in patients with hypothalamic hamartomas,⁷⁶ tuberous sclerosis,^77,⁷⁸ progressive myoclonic epilepsy,^79,⁸⁰ Landau-Kleffner syndrome,⁸¹ Asperger syndrome,⁸² epileptic encefalopathies,⁷⁶ and syndromes with developmental disability and mental retardation.⁸³^–⁸⁶ All these studies reported good efficacy with regard to controlling seizures as well as other disease-related symptoms, such as cerebellar dysfunction, behavioral disturbances, and mood disturbances. One study in children with infantile spasms reported less favorable results with long-term benefit in only two to ten patients and with four patients who interrupted VNS due to behavioral problems.⁸⁷ A recent report on the efficacy of VNS in five children with mitochondrial electron transport chain deficiencies described no significant seizure reduction in any of the children.⁸⁸ Also, a study in patients with previous resective epilepsy surgery showed a limited seizure-suppressing effect of VNS,⁸⁹ although another report described improved seizure control in this specific patient group.⁹⁰

SAFETY, SIDE EFFECTS, AND TOLERABILITY

The most prominent and consistent sensation in patients when the vagus nerve is stimulated for the first time, even at low output current levels, is a tingling sensation in the throat and hoarseness of the voice. The tingling sensation may be due to secondary stimulation of the superior laryngeal nerve that branches off from the vagus nerve superior to the location of the implanted electrode but travels along the vagus nerve in the carotid sheath.⁹¹ The superior laryngeal nerve carries sensory fibers to the laryngeal mucosa. Stimulation of the recurrent laryngeal nerve that branches off distally from the location of the electrode and carries motor (Aα) fibers to the laryngeal muscles causes the stimulation-related hoarseness.^92,⁹³

With regard to side effects related to stimulation of vagal efferents, effect on heart rate and gastrointestinal digestion are of major concern. Stimulation of the efferent fibers may induce bradycardia and hypersecretion of gastric acid. The stimulation electrode is always implanted on the left vagus nerve, which is believed to contain fewer sinoatrial fibers than the right. It has been suggested that the electrode be implanted below the superior cardiac branch of the vagus nerve. In the initial pilot trials and controlled randomized trials, extensive internal investigations were performed, including continued monitoring in the long-term extension phases. With regard to potential central nervous system side effects related to stimulation of vagal afferents and their connections in the brainstem and cerebral hemispheres, some studies were performed to evaluate changes in EEG, sleep stages, balance, and cognition. In most studies, systematic AED plasma monitoring was performed. No systematic side effects on heart functioning or other internal organ or cerebral functions were found. There was no effect on AED serum levels.^25,⁴³ Side effects are almost always related to the stimulation on-time and consist of hoarseness and tingling sensation and coughing.

In the long-term extension trials, the most frequent side effects were hoarseness in 19% of patients and coughing in 5% of patients at 2-year follow-up and shortness of breath in 3% of patients at 3 years.⁵² There was a clear trend toward diminishing side effects over the 3-year stimulation period. Ninety-eight percent of the symptoms were rated mild or moderate by the patients and the investigators.⁹⁴ Side effects can usually be resolved by decreasing stimulation parameters. Central nervous system side effects typically seen with AEDs were not reported. After 3 years of treatment, 72% of the patients were still on the treatment.⁵² The most frequent reason for discontinuation was lack of efficacy.

Initial studies on small patient groups treated for 6 months with VNS showed no negative effect on cognitive motor performance and balance.⁹⁵^–⁹⁷ These findings were confirmed in larger patient groups with a FU of 2 years.^98,⁹⁹ Hoppe et al. showed no changes in extensive neuropsychological testing in 36 patients treated for 6 months with VNS.¹⁰⁰