Vagus nerve stimulation (VNS) is an established treatment of medically refractory partial-onset seizures. Recent data from an open-label multicenter pilot study also suggest a potential clinical usefulness in the acute and maintenance treatment of drug-resistant depressive disorder. Despite the fact that surgery is needed to implant the stimulating device, the option of long-term use largely devoid of severe side effects would give this treatment modality a privileged place in the management of drug-resistant depression. Besides the potential clinical usefulness, VNS can be used as a research tool in epilepsy patients implanted for clinical reasons, allowing neurophysiologic investigations of the parasympathetic system and its interactions with other parts of the central nervous system

Vagus nerve stimulation (VNS) in humans refers generally to stimulation of the left vagus nerve at the cervical level using the NeuroCybernetic Prosthesis (NCP, Cyberonics, Inc., Houston, TX, U.S.A.) system. This equipment consists basically of three parts: 1) the implantable, multiprogrammable bipolar NCP pulse generator, which is similar to a cardiac pacemaker in size and shape; 2) two helical electrodes, which are wrapped around the vagus nerve and are linked to the pulse generator by a bipolar lead; and 3) a programming wand linked to a computer running programming software, which allows noninvasive programming, functional assessment (device diagnostics), and data retrieval. The pulse generator is implanted in a subcutaneous pocket in the left chest wall, whereas the electrodes are attached to the vagus nerve, made accessible by an incision in the neck. Through a subcutaneous tunnel, the electrodes are linked to the pulse generator. The system delivers electrical impulses at frequencies between 1 and 30 Hz, at 0.25 to 3.5 mA, with a pulse width varying from 130 to 1000 microseconds at variable on-off times.

Vagus nerve stimulation has recently gained considerable interest as a potential treatment of relative drug-resistant major depressive episodes

The first clinical application of VNS in humans, however, concerns the treatment of medically refractory seizures. VNS has been used in more than 40000 patients for the treatment of epilepsy since 1988, when the first pilot studies began. The US Food and Drug Administration approved the NCP system in 1997 for management of medically refractory partial-onset seizures. The first implant of an NCP system for treatment resistant depression was performed in 1998 at the Brain Stimulation Laboratory of the Medical University of South Carolina. In 2001, VNS was approved for a CE mark (indicating compliance with safety and environmental regulations) in the member countries of the European Union for treatment of adults with treatment-resistant or treatment-intolerant chronic or recurrent depression, including unipolar and bipolar depression .

Considerable experience with the technique of VNS has been gained from its application as a treatment of drug-refractory epilepsy, including information about its mode of action, its clinical efficacy, its side effect profile, and technical and safety-related aspects.
The precise mechanism by which VNS suppresses seizures is not known. However, the electrical stimuli applied at the vagus nerve must influence the state of excitability of the brain. This effect is mediated by the vagus nerve, a nerve containing approximately 80% afferent nerve fibers. The cell bodies of the afferent cells are in the nodose ganglion and project primarily to the nucleus tractus solitarius (NTS). Neurons of this nucleus project to numerous areas in the forebrain and the brainstem, including indirect projections into the locus ceruleus and diffuse connections to the cortex. Important structures thought to mediate antiepileptic effects receiving projections from the NTS include the amygdala and the thalamus. The cell bodies of the efferent fibers are located in the nucleus ambiguous and the dorsal motor nucleus of the vagus nerve. They provide innervations of the heart, aorta, lungs, gastrointestinal tract, and voluntary striated muscles of the larynx and pharynx. The heart rate is mostly influenced by the right vagal nerve. The build-up of the different parts of the vagal system—afferent and efferent parts with their respective brainstem nuclei (NTS and nucleus ambiguous and the dorsal nucleus of the vagus nerve) and the asymmetric distribution of function with the right vagal nerve preferentially involved in the chronotropic regulation of the heart—has been explained in an evolution-based concept, the polyvagal theory.

Vagus nerve stimulation clearly has effects on brain function. In 21 patients assessed at baseline and 3, 6, and 12 months after implantation of a VNS system, serial electroencephalographic (EEG) studies showed a progressive decrease in the number of spikes on EEG with time. In five patients who showed active spikes and wave activity on baseline EEGs, a progressive increase in the duration of spike-free intervals and a progressive decrease in the duration and frequency of spikes and wave activity was found over time. Acute effects on brain activity were assessed in a blood oxygenation level dependent (BOLD) functional magnetic resonance imaging (BOLD fMRI) study conducted in six adults with recurrent drug-resistant major depression. VNS applied 7 seconds at 20 Hz resulted in a bilateral activation in the frontal cortex, the hypothalamus, and the left globus pallidus. However, at 5 Hz stimulation, the level of activation did not reach statistical significance compared with baseline recordings without stimulation. Brain regions activated by hearing a tone were also greater when VNS was applied at 20 Hz than at 5 Hz. These findings suggest a frequency-dependent activity of VNS.

Treatment efficacy was evaluated prospectively in patients with medically refractory partial-onset seizures who had completed a 3-month, double blind, low-stimulation, controlled multicenter study (E05) initiated by the manufacturer of the device. Efficacy was evaluated by comparing the change in total seizure frequency at 3-month and 12-month VNS treatment with a 3-month preimplantation baseline. The median seizure reductions at 3 and 12 months were 34% and 45%, respectively. Twenty percent of patients had a 75% or greater reduction in seizure frequency at 12 months. Similar results confirming the efficacy of VNS in the treatment of medically refractory epilepsy were found in several other studies. Side effects were generally mild and transient.

Cardiac arrhythmias attributable to VNS in patients undergoing chronic stimulation have not been described. The most common adverse side events at 1 year postimplantation were voice alteration (21%) and dyspnea and neck pain (each 7%).  Isolated complications of VNS such as chronic diarrhea, Horner syndrome, posture-dependent stimulation of the phrenic nerve, worsening of pre-existing obstructive sleep apnea, and exacerbation of preimplantation dysphoric disorders and psychotic episodes with VNS-associated seizure reduction have been reported.

RATIONALE OF USING VAGUS NERVE STIMULATION TO TREAT DEPRESSION
Biologic treatments in depression have a long tradition, best illustrated by the use of electroconvulsive therapy (ECT), which was developed from the 1930s on. Its application improved constantly, and today it is considered one of the most efficient treatments in psychiatry, with an acceptable side effect profile compared with the severity of the symptoms of medically treatment-resistant major depression.
Depression is indeed one of the most burdensome diseases with respect to its global impact on morbidity, mortality, and loss of quality of life. According to a study supported by the World Health Organization, unipolar major depression will rank at the second place of the leading causes of disability-adjusted life years worldwide in 2020, just behind ischemic heart disease. Studies of acute major depression generally show an approximately 50 to 60% response rate to a medical antidepressant treatment after 6 weeks. However, a large proportion of patients with depression do not improve substantially even after two treatment trials using antidepressant drugs of different classes for at least 6 weeks at a standard dose agreed on as being generally effective. ECT is considered to have acute efficacy even in drug-resistant depression, ranging higher than 80%. However, a major problem in the treatment of depression is relapse prevention. In a study completed by 62 patients, at least 50% of patients relapsed within the first year after ECT, and approximately 90% of these relapses occurred in the first 6 months. Relapse prevention after successful ECT might necessitate further maintenance ECT sessions. A well-tolerated, efficacious long-term treatment in medically resistant depression is not yet available. VNS could offer a valuable treatment option because it has a good side effect profile, as assessed in the treatment of epilepsy.

The hypothesis that VNS could be effective in treating depressive disorder was initially based on the following observations:
1. Improved mood and cognition were observed in epilepsy patients treated with VNS.

2. Drugs used to treat epilepsy, including carbamazepine, gabapentin, lamotrigine, and valproate, are also effectively used to treat mood disorders. VNS also is an effective treatment of epilepsy.

3. Positron emission tomography studies show that VNS affects the metabolism and thus the function of limbic structures in a way compatible with antidepressant function.

4. Neurochemical studies in animals and humans reveal that VNS alters concentrations of monoamines within the central nervous system.

5. The vagus nerve is anatomically linked to brain structures related to mood disorders.

Vagus nerve stimulation is an effective, safe, and well-tolerated treatment in patients with long-standing, refractory partial-onset seizures, and may also be beneficial in other types of seizure. However, data indicate that the effect of VNS may be delayed as long as a year and that patients continue to improve during that time.  Problems arising during the implantation procedure are rare and manageable. VNS might not be recommended to patients with cardiac conduction disorder and with sleep apnea.

Vagus nerve stimulation as a treatment of medically resistant depression is a novel development.

At present, preliminary results available concerning the efficacy of VNS as a treatment of medically resistant depression suggest acute and longer-term effectiveness. Besides efficacy, technical aspects need to be investigated, such as how to tailor individual treatment concerning stimulation frequency, strength, and duration. In a recent report, the threshold of eliciting a response in the vagus nerve by VNS was found to be age-dependent, with higher thresholds in young patients. The establishment of predictive factors for treatment response to VNS is important, because this is an invasive treatment. Results from the open-label American multicenter study (D01) involving 60 patients yielded the following variants as significant in univariate analysis: receiving ECT ever in lifetime, response to most recent ECT, number of unsuccessful antidepressant trials in the current major depressive episode, and number of unsuccessful antidepressant trial categories in the current major depressive episode. Of the 13 patients who had not responded to more than seven adequate antidepressant trials, none met response or remitter criteria. It appears that VNS as currently delivered is most effective in patients with moderate but not extreme levels of resistance to conventional antidepressant treatments.

Compared with other treatment modalities, the cost profile of VNS could be beneficial, if effective maintenance therapy could be achieved with either a reduction in the use of concomitant medications or a reduction in other costs, such as hospitalization for depression relapses. However, in the treatment of drug-refractory epilepsy, a seizure-free state is rarely achieved, and VNS is mostly combined with antiepileptic drug maintenance therapy.