Medicare Advantage

Seizures are considered paroxysmal disorders (i.e., characterized by abnormal cerebral neuronal discharge) and occur when there is errant electrical discharge activity in the brain. Seizures cause different physical symptoms depending on the location of the electrical activity in the brain. They may be mild to severe, ranging from causing a slight tingling sensation or momentary confusion to causing complete unconsciousness. Classification and subtypes of seizures are commonly diagnosed by electroencephalography (EEG). Seizures are classified as simple partial, complex partial, and generalized onset, and are distinguished on the basis of the level of consciousness.

Partial seizures affect only a portion, or one hemisphere, of the brain and are subdivided into the following:

Generalized seizures are caused by errant electrical activity in multiple locations, or large areas, that may involve both the left and the right hemispheres of the brain. This type of seizure has a more complex set of symptoms, with impaired awareness leading to complete loss of consciousness. The subtypes of generalized seizures are usually classed as follows:

Epilepsy is diagnosed in individuals who have a predisposition for recurrent, unprovoked seizures. During the past 10 years, new medications and surgical techniques have emerged in the treatment of epilepsy. Despite these advances, 20% to 50%  of individuals with epilepsy have breakthrough seizures or experience adverse effects of antiseizure medications. Current peer-reviewed medical literature supports VNS as a useful palliative procedure for individuals who have severe generalized epilepsy with atonic or tonic-clonic seizures. Peer-reviewed literature also supports the use of VNS as an effective treatment for medically refractory seizures other than partial-onset seizures.

The VNS system is composed of an implantable generator, a bipolar electrical lead (array) consisting of helical electrodes and an anchor tether, along with an external programming device that is used to change stimulation settings. Revisions to the VNS system may include procedures such as replacing the battery-powered generator. Battery longevity depends on stimulation parameters, with newer battery-powered generators lasting up to 10 years or more. The generator is surgically implanted subcutaneously, at the lateral border of the left pectoral muscle, or under the arm. An incision is also made to the left side of the neck, to access the vagus nerve, for the placement of the electrodes. The lead (array) is threaded subcutaneously to connect the electrodes, which are attached to the vagus nerve, to the generator. Once implantation is completed, the generator is programmed to stimulate the vagus nerve at regular intervals.

The basic principles of VNS when used as a treatment for epilepsy are that vagal visceral afferents (nerves that convey impulses from sense organs and other receptors to the brain or spinal cord) have a diffuse central nervous system projection, and the activation of these pathways has a widespread effect upon neuronal excitability. The exact mechanism of VNS on neuronal excitability is not fully known. Adverse effects of VNS therapy include headache, neck pain, cough, and voice alterations.

Some of the benefits of using VNS may include less-severe or shorter seizures, a reduction in seizure frequency, improved recovery periods after seizures, and a lessening of seizure clusters. Individuals undergoing VNS must be aware that seizure control improves over time, and that although VNS may reduce the frequency and magnitude of seizure activity, the need remains for an ongoing, concurrent, antiseizure medicinal regimen.

On July 16, 1997, the US Food and Drug Administration (FDA) approved the NeuroCybernetic Prosthesis (NCP®) System/VNS Therapy® System (LivaNova USA, Inc., formerly Cyberonics, Inc., Houston, TX), a VNS device, for the adjunctive treatment of medically refractory partial-onset seizures in adults and children 12 years of age or older. Studies since the original FDA approval have focused on the use of VNS in younger children. On June 23, 2017, the FDA approved the use of the VNS Therapy® System as an adjunctive therapy for seizures in individuals 4 years of age or older with partial-onset seizures that are refractory to antiepileptic medications.

In 2011, Englot et al. performed a meta-analysis of VNS efficacy in epilepsy, identifying 74 clinical studies with 3321 individuals suffering from intractable epilepsy. These studies included three blinded, randomized controlled trials (RCTs) (Class I evidence); two nonblinded RCTs (Class II evidence); 10 prospective studies (Class III evidence); and numerous retrospective studies. After VNS, seizure frequency was reduced by an average of 45%, with a 36% reduction in seizures at 3 to 12 months after surgery and a 51% reduction after more than 1 year of therapy. At the last follow-up, seizures were reduced by 50% or more in approximately 50% of the individuals, and VNS predicted a 50% or greater reduction in seizures with a main effects odds ratio (OR) of 1.83 (95% confidence interval [CI], 1.80–1.86). Individuals with generalized epilepsy and children benefited significantly from VNS despite their exclusion from initial approval of the device. Furthermore, posttraumatic epilepsy and tuberous sclerosis were positive predictors of a favorable outcome. The authors concluded that "VNS is an effective and relatively safe adjunctive therapy in patients with medically refractory epilepsy not amenable to resection. However, it is important to recognize that complete seizure freedom is rarely achieved using VNS and that a quarter of patients do not receive any benefit from therapy."

Five randomized controlled trials (RCTs) (N=439) have evaluated VNS (Michael et al. 1993; Ben-Menchem et al./VNS Study Group 1994, 1995; Handforth et al. 1998; DeGiorgio et al. 2005; Klinkenberg et al. 2012). Four of the studies compared high-frequency VNS that was thought to be therapeutic versus low-frequency VNS at levels that were thought to be subtherapeutic (Michael et al. 1993; Ben-Menchem et al./VNS Study Group 1994, 1995; Handforth et al. 1998; Klinkenberg et al. 2012). One study compared rapid versus medium versus slow cycle VNS (DeGiorgio et al. 2005). The studies generally included people with drug-resistant partial epilepsy with VNS as an add-on treatment. The blinded treatment phase ranged from 12 to 20 weeks in the five studies. Four studies reported the outcome of response (≥50% reduction in seizure frequency) and the risk ratio ranged from 1.49 to 8.27 in the three studies (Michael et al. 1993; Ben-Menchem et al./VNS Study Group 1994, 1995; Handforth et al. 1998) that favored high-frequency VNS; the risk ratio was statistically significantly different from the null in one study. One study (Klinkenberg et al. 2012) reported a risk ratio that did not favor high-frequency VNS for the response outcome but was not statistically significant.

In 2014, Ryvlin et al. reported in an RCT the long-term quality of life outcomes for 112 individuals with medication-resistant focal seizures, which supported the beneficial effects of VNS for this group.

The evidence for the efficacy of VNS for generalized seizures in adults is primarily from observational data, including registries and small cohort studies. Englot et al. (2016) examined freedom from seizure rates and predictors across 5554 individuals enrolled in the VNS Therapy Patient Outcomes Registry. The registry was established in 1999, after the 1997 FDA approval of VNS, and is maintained by the manufacturer of the device, LivaNova (formerly Cyberonics). Data were prospectively collected by 1285 providers from 978 centers (911 in the United States and Canada and 67 internationally) at preoperative baselines and various intervals during therapy. During active data collection, participation in the registry included approximately 18% of all implanted VNS devices. The database was queried in January 2015, and all seizure outcomes reported with the 0- to 4-, 4- to 12-, 12- to 24-, and 24- to 48-month time ranges after VNS device implantation were extracted and compared with patient preoperative baseline. Available information was tracked at each time point of data submission for the following outcomes: patient demographics, epilepsy etiology and syndrome, historical seizure types and frequencies, quality of life, physician global assessment, current antiepileptic drugs, medication changes, malfunctions, battery changes, and changes in therapy. At each observation point, responders were defined as having a 50% or greater decrease in seizure frequency compared with baseline and nonresponders as less than a 50% decrease. A localized epilepsy syndrome such as partial-onset seizures was recorded in 59% of the registry participants, generalized epilepsy in 27%, and 11% had a syndromic etiology (e.g., Lennox-Gastaut). From 0 to 4 months the responder rate and seizure freedom rate was 50% and 7%, respectively; at 4 to 12 and 12 to 24 months, 55% and 8%; at 24 to 48 months, approximately 60% and 9%. These rates did not differ statistically from participants with predominantly partial seizures.

In 2013, Garcia-Navarrete et al. evaluated outcomes after 18 months of follow-up for a prospective cohort of 43 individuals with medication-resistant epilepsy who underwent VNS implantation. Seizure types were heterogeneous, but 52% had generalized epilepsy. Pharmacotherapy was unchanged during the study. Twenty-seven (63%) individuals were described as “responders”, defined as having a 50% or greater reduction in seizure frequency compared with the year before VNS implantation. The difference in reduction of seizure frequency was not statistically significant between individuals with generalized and focal epilepsy.

The evidence for VNS for pediatric seizures consists of a variety of small noncomparator trials (Hornig et al. 1997; Patwardhan et al. 2000; Cukiert et al. 2013), prospective observational studies (Murphy et al. 1999; You et al. 2007), and retrospective case series (Frost et al. 2001; Healy et al. 2013; Terra et al. 2014; Yu et al. 2014). As in the adult studies, there is heterogeneity of seizure etiologies: mixed, syndromic, and idiopathic; there is also generalized and limited information on concomitant antiepileptic drug requirement. Some studies have defined pediatric individuals as less than 12 years of age and others have defined them as less than 18 years and may have included individuals as young as 2 to 3 years of age. Study subpopulations may have had prior failed resective surgery. Complete freedom from seizures is the exception, and the primary reported endpoint is 50% or more reduction in seizure frequency, determined over varying lengths of follow-up. There is an overlap of authors for multiple studies suggesting utilization of VNS in specialized clinical care environments. Multiple studies have some form of innovator device company sponsorship.

The evidence on the efficacy of VNS for treatment of medically refractory seizures consists of RCTs, meta-analyses, and numerous uncontrolled studies. RCTs and meta-analyses of RCTs have reported a significant reduction in seizure frequency with VNS for individuals with partial-onset seizures. The uncontrolled studies and case series have consistently reported reductions of clinical significance, defined as a 50% or more reduction in seizure frequency in both adults and children over almost two decades of publications. Interpretation of all outcomes and results were limited by the variety of comparators (when used), variability in length of follow-up, limited published data on antiepileptic medication requirements, mixed seizure etiologies, and history of prior failed resective surgery. There is an overlap of authors across multiple studies, suggesting utilization of VNS in specialized clinical care environments. Multiple studies have some form of innovator device company sponsorship.

REFRACTORY DEPRESSION

Several systematic reviews and meta-analyses have assessed the role of VNS in TRD. A 2008 systematic review (Daban et al.) of the literature for VNS of treatment-resistant depression (TRD) identified one randomized trial. VNS was found to be associated with a reduction in depressive symptoms in the open-label studies. However, results from the only double-blind trial were considered inconclusive. Daban et al. (2008) concluded that further clinical trials are needed to confirm efficacy of VNS in TRD.

One randomized study (D-02) that compared VNS therapy with a sham control (implanted but inactivated VNS) showed a nonstatistically significant result for the principal outcome (Rush et al. 2005; FDA Summary of Safety and Effectiveness Data 2005). Fifteen percent of VNS subjects responded versus 10% of control subjects (p=0.31). The Inventory for Depressive Symptomatology Systems Review score was considered a secondary outcome and showed a difference in outcome that was statistically significant in favor of VNS (17.4%) compared with sham treatment (7.5%; P=0.04).

Rush et al. (2005) reported results of a 10-week, blinded RCT comparing adjunctive VNS with sham in 235 outpatients with nonpsychotic major depressive disorder or nonpsychotic, depressed phase, bipolar disorder. The individuals were treatment-resistant defined as those who had not responded adequately to between 2 and 6 research-qualified medication trials for the current episode of depression. The primary outcome was response rates (50% or more reduction from baseline on the Hamilton Rating Scale for Depression). There was no statistically significant difference in response rates at 10 weeks in VNS versus sham (15% vs. 10%; P=0.25).

Aaronson et al. (2013) reported on results from an active-controlled trial in which 331 individuals with a history of chronic or recurrent bipolar disorder or major depressive disorder, with a current diagnosis of a major depressive episode, were randomized to 1 of 3 VNS current doses (high, medium, low). Individuals had a history of failure to respond to at least 4 adequate dose/duration of antidepressant treatment trials from at least two different treatment categories. After 22 weeks, the current dose could be adjusted in any of the groups. At follow-up visits 10, 14, 18, and 22 weeks after enrollment, there were no statistically significant differences between the dose groups for the study’s primary outcome, change in Inventory of Depressive Symptomatology (IDS) score from baseline. However, mean IDS scores improved significantly for each group from baseline to the 22-week follow-up. At 50-week follow-up, there were no significant differences between the treatment dose groups for any of the depression scores used. Most individuals completed the study; however, there was a high rate of reported adverse events, including voice alteration in 72.2%, dyspnea in 32.3%, and pain in 31.7%. Interpretation of the IDS improvement over time is limited by the lack of a no-treatment control group. Approximately 20% of the individuals included had a history of bipolar disorder; as such, the results might not be representative of most individuals with treatment-resistant unipolar depression.

The observational study that compared individuals participating in the RCT with individuals in a separately recruited control group (D-04 vs. D-02, respectively) evaluated VNS therapy out to 1 year and showed a statistically significant difference in the rate of change of depression score (George et al. 2005; FDA Summary of Safety and Effectiveness Data 2005). However, issues such as unmeasured differences among individuals, nonconcurrent controls, differences in sites of care between VNS therapy individuals and controls, and differences in concomitant therapy changes raise concern about this observational study. Analyses performed on subsets of individuals cared for in the same sites, and censoring observations after treatment changes, generally showed diminished differences in apparent treatment effectiveness of VNS and almost no statistically significant differences (De Ferrari et al. 2011). Patient selection for the randomized trial and the observational comparison trial may be of concern. VNS is intended for treatment-refractory depression, but the entry criteria of failure of 2 drugs and a 6-week trial of therapy might not be a strict enough definition of treatment resistance. Treatment-refractory depression should be defined by thorough psychiatric evaluation and comprehensive management. It is important to note that patients with clinically significant suicide risk were excluded from all VNS studies. Given these concerns about the quality of the observational data, these results did not provide strong evidence for the effectiveness of VNS therapy.

In 2020, McAllister-Williams et al. reported on results of a subgroup of 156 participants with treatment-resistant bipolar depression from the above-described FDA-required post-marketing surveillance study (Aaronson et al [2017]). Compared to the overall population in the primary study, cumulative first-time response rates were similar in this bipolar depression subgroup (63% vs. 39%; p not reported). Median time-to-initial response was not significantly different between groups (13.7 vs. 42.1 months; Hazard Ratio [HR]=1.7; 95% CI, 1 to 2.7). Median time-to-relapse from initial response in the first year was also not significantly different between groups (15.2 vs. 7.6 months; HR=0.7; 95% CI, 0.3 to 1.4). Based on MADRS item 10, the mean reduction in suicidality score across the study visits was reportedly significantly greater in the VNS group than in the no VNS group (P<0.001 as per F-test). However, the validity of this finding is unclear as by 60 months, it excluded data from an unacceptably high (n=100, 64%) and imbalanced (59% in VNS group vs. 73% in no VNS group) number of individuals with unavailable suicidality data. It was additionally subject to the same important limitations as described above for the primary study.

Several case series (Marangell et al. 2002; Rush et al. 2000; Sackeim et al. 2001) published before the randomized trials showed rates of improvement with VNS, as measured by a 50% improvement in depression score, of 31% at 10 weeks to greater than 40% at 1 to 2 years, but there were some losses to follow-up. Natural history, placebo effects, and patient and provider expectations make it difficult to infer efficacy from case series data.

Other case series do not substantially strengthen the evidence supporting VNS. A case series by Bajbouj et al. (2010), which followed individuals for 2 years, showed that 53.1% (26/49) met criteria for treatment response and 38.9% (19/49) met criteria for remission. A small 2008 study (Marangell et al.) of 9 patients with rapid-cycling bipolar disorder showed improvements in several depression rating scales over 40 weeks of observation. In a 2014 case series (Tisi et al.) that included 27 patients with TRD, five patients demonstrated complete remission after 1 year, and 6 patients were considered responders.

Adverse events of VNS therapy included voice alteration, headache, neck pain, and cough, which are known from prior experience with VNS therapy for seizures. Regarding specific concerns for depressed individuals (e.g., those with mania, hypomania, suicide, or worsening depression), there does not appear to be a greater risk of these events during VNS therapy (FDA Summary of Safety and Effectiveness Data 2005).

On July 15, 2005, the FDA approved the VNS Therapy System, NeuroCybernetic Prosthesis (NCP®) System (LivaNova USA, Inc., formerly Cyberonics, Inc., Houston, TX) for use as an adjunctive long-term treatment for chronic or recurrent depression in individuals 18 years of age or older who have experienced a major depressive episode and have not adequately responded to four or more antidepressant treatments. Despite FDA approval, current published peer-reviewed literature does not support the use of VNS for the treatment of depression. The clinical trials for the use of VNS for TRD do not demonstrate the effectiveness of VNS therapy on health outcomes in an investigational setting, nor do these trials demonstrate improved health outcomes with the use of VNS when compared to other treatment modalities for TRD. Additional research is recommended. 

OTHER CONDITIONS

VNS therapy has also been investigated for use in other conditions; however, the available published peer-reviewed evidence is limited and not sufficient to permit conclusions of effectiveness.

Systematic Reviews

Case Series

Zannad et al. (2015) reported on results from the NEural Cardiac TherApy foR Heart Failure (NECTAR-HF) RCT, a randomized, sham-controlled trial, with outcomes from VNS in individuals with severe LV dysfunction despite optimal medical therapy. Ninety-six individuals were implanted with a vagal nerve stimulator and randomized in a 2:1 manner to active therapy (VNS ON) or control (VNS OFF) for 6 months. Programming of the generator was performed by a physician unblinded to treatment assignment, while all other investigators and site study staff involved in the endpoint data collection were blinded to randomization. Sixty-three patients were randomized to the intervention, of whom 59 had paired pre-post data available, while 32 were randomized to control, of whom 28 had paired data available. The analysis was a modified intention-to-treat. For the primary endpoint of change in LV end-diastolic diameter from baseline to 6 months, there were no significant differences between groups (p=0.60 between-group difference in LV end-diastolic diameter change). Other secondary efficacy endpoints related to LV remodeling parameters (i.e., LV function and circulating biomarkers of heart failure) did not differ between groups, with the exception of 36-Item Short-Form Health Survey Physical Component Summary score, which showed greater improvement in the VNS ON group than in the control group (from 36.3 to 41.2 in the VNS ON group vs. from 37.7 to 38.4 in the control group; p=0.02). Subject blinding was found to be imperfect, which might have biased the subjective outcome data reporting. 

The evidence on VNS for treatment of chronic heart failure consists of a systematic review including 4 RCTs and 2 uncontrolled studies. A meta-analysis of 4 RCTs found significant improvements in New York Heart Association functional class, quality of life, 6-minute walk test, and N-terminal-pro brain natriuretic peptide levels in individuals treated with VNS compared to sham. The uncontrolled studies consistently reported improvements on a variety of measures, including LV function, 6-minute walk test and quality of life. However, lack of a no-VNS comparator group precludes drawing conclusions based on findings from the uncontrolled studies.

Handforth et al. (2003) studied VNS in nine individuals with essential tremor resulting in no improvement in upper extremity tremors. The authors of the study reported that VNS is not likely to have any clinically meaningful effect in essential tremor treatment.

VNS has been investigated for use in fibromyalgia. Lange et al. (2011) conducted a single-arm Phase I/II trial of 14 individuals with fibromyalgia. At three months, five individuals met criteria for improvement of symptoms, at 11 months, eight individuals met efficacy criteria. However, the results do not provide sufficient evidence for use of VNS for fibromyalgia. 

Vagal nerve stimulation is being studied for treating chronic headaches. Mauskop (2005) evaluated VNS in five individuals with severe, refractory chronic cluster and migraine headaches. One individual had excellent results with regaining the ability to work, and two individuals reported significant improvement of their headaches. Small sample size prevents drawing conclusions on the effects of VNS for the treatment of headache.

On January 14, 2015, the FDA approved the MAESTRO® Rechargeable System (EnteroMedics, St. Paul, MN), consisting of a rechargeable neuroregulator device implanted into the lateral chest wall with flexible leads placed laparoscopically around the vagus nerve, for use in weight reduction in individuals aged 18 years through adulthood who have a Body Mass Index (BMI) of 40 to 45 kg/m2, or a BMI of 35 to 39.9 kg/m2 with one or more obesity-related co-morbid conditions, and have failed at least one supervised weight management program within the past five years. The approval is based on a 12-month randomized, double-blind, sham controlled clinical trial involving 239 participants. Despite FDA approval, several limitations in this study include the use of vagal nerve block compared with a sham control device did not meet either of the pre-specified co-primary efficacy objectives. The co-primary effectiveness endpoints were to determine whether the vagal nerve block was superior in mean percentage excess weight loss to sham by a 10-point margin with at least 55 percent of individuals in the vagal block group achieving a 20 percent loss and 45 percent achieving a 25 percent loss, even though weight loss in the vagal block group was greater than in the sham device group. In addition, the treatment arm participants could possibly feel the stimulation device, while the control arm doesn’t feel anything, so individuals may lose motivation to participate and not adhere to diet program, which would limit their weight loss, causing weight loss in the vagal block group to be greater than in the sham device group. Because the study did not compare outcomes with established surgical obesity treatments (e.g., gastric banding, Roux-en Y gastric bypass), because of the short-term follow-up period without long-term outcomes of vagal nerve stimulation for morbid obesity, at this time there is inadequate data to permit scientific conclusions regarding the efficacy of this procedure.

The commercial availability of the MAESTRO® Rechargeable System is unclear. On the FDA's Weight-Loss and Weight-Management Devices webpage (content noted as current as of 04/27/2020), the Maestro® Rechargeable System is described as "no longer marketed as of September 2018". Additionally, on the ReShape LifesciencesTM website (previously EnteroMedics), the MAESTRO® Rechargeable System, is not listed among their current portfolio of medical devices to treat obesity and metabolic disease. However, updates to the MAESTRO® Rechargeable System were noted in the FDA Premarket Approval database (P130019) subsequent to September 2018, including updates to the circuit assembly and application firmware of the mobile charger (01/25/2019) and approval of modifications to the follow-up schedule for the post-approval study protocol.

Bodenlos et al (2007) suggested that VNS might affect food cravings of individuals with chronic, TRD and thus induce weight loss in obese individuals. Several limitations in the study such as small sample size, lack of randomization, heterogeneity of groups with respect to depression, prevent drawing conclusions about the impact of VNS on eating behavior. The study findings need to be validated in large, well-designed controlled studies to evaluate the impact of VNS on eating behavior and obesity.

The published literature on VNS for obesity consists of 2 randomized, controlled trials (RCTs), both of which were industry-sponsored, multicenter, double-blind, and sham-controlled (Sarr et al, 2012; Ikramuddin et al, 2014). Although both trials included a sham treatment group, protocols differed. In the 2012 Vagal Blocking for Obesity Control (EMPOWER) trial (Sarr et al, 2012), all participants [n=294] had devices implanted and leads placed. However, external controllers were programmed differently such that if the controllers were worn for 10 hours a day, the total charge delivered was 3.9 coulombs (C) to individuals in the treatment group [n=192 to active Maestro® device plus 15 weight management counseling sessions] and a negligible amount (0.0014 C), to the sham group [n=102 to inactive Maestro® device plus 15 weight management counseling sessions]. In the 2014 trial (Ikramuddin et al, 2014) to Evaluate the Safety and Efficacy of vBloc Therapy Delivered by the Maestro® Rechargeable System for the Treatment of Obesity (ReCharge), all participants (n=239) had devices implanted [n=162 to active sham Maestro® device plus 17 weight management counseling sessions], but no leads were placed in the sham group [n=77 to inactive sham Maestro® device plus 17 weight management counseling sessions].

The primary efficacy outcomes were not met in either RCT. The difference in mean percent excess weight loss (EWL) was the sole primary efficacy outcome in the Vagal Blocking for Obesity Control (EMPOWER) study and a coprimary outcome in the Evaluate the Safety and Efficacy of vBloc Therapy Delivered by the MaestroRechargeable System for the Treatment of Obesity (ReCharge) study. This outcome was evaluated in both trials using a superiority margin of 10% (i.e., the efficacy objective would be met only if there was a >10% difference between groups in EWL). FDA documents (Summary of Safety and Effectiveness Data (SSED): Maestro Rechargeable System, 2015) have indicated the unattained 10% margin was considered to indicate a clinically meaningful difference in weight loss between active and sham treatment groups.

For the ReCharge trial, however, in addition to the primary efficacy analysis, the authors conducted a post hoc analysis that evaluated the difference in EWL between groups using a 2-sided t-test with no superiority margin. In this post hoc analysis, the difference between groups (8.5% EWL; 95% confidence interval [CI], 3.1% to 13.9%) was statistically significant. (The difference between groups in percent EWL in the Vagal Blocking for Obesity Control (EMPOWER) study was 1%.)

Both trials met their primary safety outcomes, which related to serious adverse events. However, there were frequent nonserious adverse events. Most were of mild or moderate severity.

Additional information on the ReCharge trial design and findings has been reported in the FDA documents (Summary of Safety and Effectiveness Data (SSED): Maestro Rechargeable System, 2015). The trial was designed to evaluate primary endpoints at 12 months and to follow individuals to 5 years postimplant. Individuals were blinded until 12 months and unblinding began once all individuals had completed the 12-month follow-up. After the 12-month follow-up, sham individuals had the option to cross over into the active treatment group. At 18 months, follow-up data (n=159) were reported for 117 (72%) individuals initially assigned to the active treatment group and 42 (55%) assigned to the sham treatment group. The number of individuals in the sham group who crossed over to active treatment and the timing of unblinding was not reported. At 18 months, the mean percent EWL was 25.3% in the active treatment group and 11.7% in the sham group; the mean between-group difference was 13.5% (95% CI, 5.7% to 21.3%). In this analysis, the treatment group maintained the weight loss they achieved at 12 months, and the control group gained weight. Nearly half of the individuals initially randomized to the sham group were not included in the 18-month analysis, which limits the ability to draw conclusions about these data. In addition, the 18-month analysis could have been biased by unblinding, which occurred after all individuals completed the 12-month follow-up. In the 12-month sham intervention phase of the trial, individuals in both groups experienced decreased hunger, increased cognitive restraint, and decreased food intake. It is likely that unblinding could have had an impact on these factors. The FDA documents also reported longer-term safety data. Analyses of data up to 48 months from the Vagal Blocking for Obesity Control (EMPOWER) trial and 18-month data from the ReCharge trial did not identify any deaths or unanticipated serious adverse events. There were 13 surgical explants through 12 months (5 in the active treatment group, 8 in the sham group) and an additional 16 explantations between 12 and 18 months. Reasons for explant included the patient decision, pain, and need for magnetic resonance imaging (MRI).

Eighteen-month follow-up data from the ReCharge trial were published by Shikora et al. (2015). They reported on a larger proportion of the patient population than that discussed in the FDA documents: in addition to the 159 (67%) of 239 randomly assigned individuals who completed the 18-month follow-up, the 2015 analysis included 30 individuals who missed the 18-month analysis but had a visit at 16 or 17 months. The additional individuals included 11 from the active treatment group and 19 from the sham group, comprising 188 individuals (79% of those originally randomized). At 18 months, the mean percent EWL noted was 23.5% (95% CI, 20.8% to 26.3%) in the active treatment group and 10.2% (95% CI, 6.0% to 14.4%) in the sham group. The mean between-group difference in percentage EWL was 13.4% (95% CI, 8.4% to 18.4%). The authors also evaluated the potential impact of blinding on outcomes and found no statistically significant effect; their findings were similar to the analysis restricted to individuals who remained blinded at 18 months. The percentages of EWL at 18 months in this 2015 analysis of ReCharge trial data were also like those previously reported in the FDA documents, although this sample size was larger, reducing potential bias from missing data. However, because this post hoc analysis incorporated 16- and 17-month data in addition to 18-month data, the authors considered these results preliminary or hypothesis-generating.

Twenty-four-month outcomes from the ReCharge trial were published by Apovian et al. (2017). The investigators noted that the sham arm was no longer a valid comparator at 24 months due to crossovers, dropouts, and patient unblinded at 12 months. There was no prespecified statistical analysis plan for assessments after the 12-month primary outcome assessment, including those in this 2017 article. A total of 103 (43%) individuals of 239 randomized individuals completed the 24-month follow-up. Their mean EWL was 21% (95% CI, 16% to 26%) and mean total weight loss was 8% (95% CI, 6% to 10%). No serious treatment-related adverse events were reported in the 18- to 24-month time period. The analysis lacked a blinded comparison group, and, like the 18-month data, was post hoc.

In 2018, the U.S. Preventive Services Task Force (USPSTF) updated recommendations for screening and management of obesity in adults (Curry et al. 2018). The Task Force recommended screening all adults for obesity and referring those with a body mass index of 30 kg/m2 or higher to intensive, multicomponent behavioral interventions. VNS and other surgical interventions were not addressed in the recommendations or literature review.

The clinical efficacy of VNS on tinnitus was studied by De Ridder et al. (2014) in a case series involving 10 individuals that suggests that VNS may be associated with clinical improvements in individuals with tinnitus. The findings of this study need to be validated in large, well-designed controlled studies to evaluate the impact of VNS on tinnitus.

Currently, VNS is being investigated to augment recovery from a traumatic brain injury. It is proposed that early stimulation of the vagus nerve accelerates the rate and extent of behavioral and cognitive recovery after fluid percussion brain injury in rats. Shi et al. (2013) received FDA approval to conduct a pilot prospective randomized trial to demonstrate objective improvement in clinical outcome by placement of a VNS in individuals who are recovering from severe traumatic brain injury. If this study demonstrates that VNS can safely and positively impact outcome, then a larger randomized prospective crossover trial will be proposed. 

VNS-REHAB studies:

A feasibility study demonstrated that the Vivistim System improved upper limb function with no significant device complications (Dawson et al., 2016). A non-implant control group and Cyberonics (LivaNova) Lead was utilized for this study (Dawson et al., 2016). A MicroTransponder Lead was designed for the next pilot study that looked at outcomes of VNS when provided in the home and clinic to determine VNS paired with rehabilitation was acceptably safe and feasible in participants with upper limb motor deficit after chronic ischemic stroke (Kimberley et al., 2018). The VNS group (n=8) showed greater improvement in function compared to the control group (n=9) for individuals with chronic stroke (anywhere from 4 months and 5 years post-stroke). One year follow-up data from the pilot study showed no VNS treatment-related serious adverse events (Dawson et al., 2020). 

In the VNS-REHAB randomized sham-controlled trial, participants with moderate to severe arm weakness, at least 9 months after ischemic stroke, were assigned to receive rehabilitation paired with active VNS (n=53) or rehabilitation paired with sham stimulation (Control) (n=55) (Dawson et al., 2021). All participants were implanted with a VNS device and received six weeks of in-clinic therapy followed by a home exercise program. The primary outcome was the change in impairment measured by the Fugl-Meyer Assessment Upper Extremity (FMA-UE) score on the first day after completion of in-clinic therapy. The secondary outcomes measures were 1) clinically meaningful response on FMA-UE score at day 90, 2) change in day 90 WMFT-Functional and FMA-UE score. The researchers defined a clinically meaningful response as a six 6 point or greater improvement in FMA-UE score based on previous research (Page et al., 2012).

The primary outcome (change in FMA-UE score from baseline to the first day after in-clinic therapy) was significantly higher in the VNS group than the Control group (VNS: 5.0, SD 4.4, Control: 2.4, SD 3.8; p=0.001; between group difference 2.60, 95% CI 1.03 to 4.2) (Dawson et al., 2021). A clinically meaningful response on the FMA-UE score occurred in more participants in the VNS group compared to the control group at day 90 following completion of in-clinic therapy (47% versus 24%, P=0.01; between group difference 24%, 95% CI 6 to 41%) (Dawson et al., 2021).

The FMA-UE score was also significantly increased in the VNS group compared to the Control group at 90 days (VNS: 5.8, SD 6.0, Control: 2.8, SD 5.2; p=0.008; between group difference 2.96, 95% CI 0.83 to 5.08) (Dawson et al., 2021). Similar was found for the WMFT-Functional score at 90 days after the end of clinic therapy (VNS: 0.46, SD 0.40, Control: 0.16, SD 0.30; between group difference 0.30, 95% CI 0.16 to 0.43) (Dawson et al., 2021). Participants were unblinded after the day-90 follow up and controls then received VNS, meaning that any long-term data will lack a control comparator.

A total of 334 adverse events (163 VNS, 171 Control) were reported in 85 (78%) participants, where the majority were mild (Dawson et al., 2021). A total of 21 (40%) participants in the active VNS group and 24 (55%) controls reported an adverse event rated that were at least possibly, probably, or definitely related to surgery/device implantation (Dawson et al., 2021). The most common AE associated with the surgery was pain (22 %; 6 possible, 3 probable, 13 definite). Surgery-related adverse events occurred at rates lower than previously reported VNS therapy for epilepsy and depression (Liu et al., 2022). 

Post-hoc analyses: 

A post-hoc analysis of the VNS-REHAB RCT showed the response to VNS treatment was similar across subgroups (e.g., age (≥62 years), time from stroke (>2 years), severity (baseline FMA-UE score >34)) (Dawson et al., 2023). Another post-hoc analysis of the VNS-REHAB RCT showed VNS paired with rehabilitation resulted in significant improvements in specifically wrist and hand impairment compared to control (Lin et al., 2024). Another post-hoc analysis of the VNS-REHAB RCT found that dosing parameters (time over which task repetitions were executed and number of VNS stimulations) were not associated with FMA-UE responder status at the end of 6 weeks (Beovich et al., 2024). The most recent post-hoc analysis of the VNS-REHAB RCT found that neuroimaging measures were associated with the extent of gains on the primary endpoint across all individuals (n=80) where less white matter hyperintensities (r = -0.25, p = 0.028) at baseline was associated with a larger Fugl-Meyer score change (Schwarz et al., 2025). Similarly, an exploratory analysis of the pilot and feasibility data showed higher cerebrospinal fluid volume was associated with less improvement in FMA-UE in the control but not VNS group (Dickie et al., 2019). A secondary analysis of the VNS-REHAB trial, reported that paired-VNS improved distal UE impairment in chronic stroke to a greater degree than intensive rehabilitation alone, but proximal improvements were equally responsive to either treatment (Vora et al., 2024). Other post-hoc analyses suggests that the effect of VNS observed in the VNS-REHAB trial are likely to be consistent in a wide range of stroke survivors with moderate to moderately severe upper extremity impairment; however, this should be clarified by further studies (Dawson et al., 2023; Kimberley et al., 2025).

Kimberley et al. (2025) conducted an unblinded, partial crossover, pooled 1-year post hoc analysis of data from the fully blinded, sham-controlled VNS-REHAB trial. The study evaluated the long-term effects of VNS paired with upper extremity rehabilitation in individuals with chronic ischemic stroke occurring 9 months to 10 years prior to enrollment. Participants received 18 sessions of intensive, task-specific rehabilitation (3x/week for 6 weeks) and 3 months of self-directed home exercise, with either active (n=53) or sham (n=55) VNS. After the blinded phase, control participants crossed over to active VNS. All participants continued self-initiated VNS therapy for one year. Outcome measures were assessed using the Fugl-Meyer Assessment UE (FMA-UE), Wolf Motor Function Test (WMFT), and patient reported outcomes including Motor Activity Log (Quality of Movement and Amount of Use), Stroke Impact Scale (Activities of Daily Living and Hand), EQ-5D, and Stroke-Specific Quality of Life. At 1-year follow-up, 74 participants (69%) completed the observation phase and showed significant improvements across all measures. Four serious adverse events (dyspnea, infection, acute infarct) occurred post-blinded phase and were deemed unrelated to the intervention (3 in VNS group, 1 in control). Study limitations included participant attrition, unblinding, incomplete data reporting, lack of diversity metrics, unmonitored home exercise adherence, small sample size, and short follow-up duration.

Meta-analysis/systematic reviews:

Two meta-analyses determined VNS may be an effective and safe adjunct to standard rehabilitation for post-stroke recovery; however, its clinical significance and long-term efficacy and safety remain unclear (Ananda et al., 2023; Zhao et al., 2022). Another meta-analysis suggested that VNS and rehab was no better than rehab on mental health or activities of daily living. Combinations of phase of stroke, specific parameters of VNS and VNS intervention frequency highly influenced outcomes (Gao et al., 2023). 

Another meta-analysis analyzed eight RCTs involving 266 individuals where five studies used direct VNS and three transcutaneous auricular VNS. The results revealed that VNS enhanced upper extremity function via FMA-UE (SMD = 0.73; 95% CI: 0.48 to 0.99; P < 0.00001) and WMFT (SMD = 0.82; 95% CI:0.52 to 1.13; P < 0.00001) in comparison to the control group but showed no significant change on long-term effects of FMA-UE. More high-quality multicentric RCTs are needed to validate this conclusion (Wei et al., 2023). 

However, these meta-analyses on VNS have inappropriately combined studies of fundamentally different interventions, outcome measures, and participant selection, which do not conform to methodological best practices and, hence, cannot be used to deduce the relative efficacy of the different types of VNS​ for stroke rehabilitation (Malakouti et al., 2024). All the above-mentioned review papers combined results of non-implantable (transcutaneous auricular) and implantable VNS, misrepresenting the exact effects implantable VNS itself. 

Another systematic review concluded that there are risks related to device implantation to deliver invasive VNS compared to non-invasive VNS (Korupolu et al., 2024). Future human comparison studies are required to study and quantify the efficacy vs. risks of paired VNS delivered via different methods with rehabilitation. Invasive VNS studies reported severe adverse events such as vocal cord palsy, dysphagia, surgical site infection, and hoarseness of voice, which were found to be related to surgery (Korupolu et al., 2024). 

The purpose of non-implantable tVNS is to noninvasively apply low-voltage electrical currents to stimulate the cervical branch of the vagus nerve. nVNS has been tested primarily in the setting of headache. nVNS has been proposed as an intervention to relieve pain in acute attacks of cluster or migraine headaches as an alternative to standard care and to reduce the frequency of attacks for both cluster headaches and migraine as an adjunct to standard care. Proposed uses have been tested in other neurologic, psychiatric, or metabolic disorders as well. tVNS is being investigated as a noninvasive alternative to surgery for numerous indications.

 

PROPHYLAXIS AND TREATMENT OF CLUSTER AND MIGRAINE HEADACHES

In an open-label, single-arm, pilot study, Goadsby et al. (2014) evaluated a novel nVNS for acute treatment of migraine. Participants with migraine (with or without aura) were eligible for this study. Up to four migraine attacks were treated with two 90-second doses, at 15-minute intervals delivered to the right cervical branch of the vagus nerve within a 6-week time period. Subjects were asked to self-treat at moderate or severe pain, or after 20 minutes of mild pain. Of 30 enrolled individuals (25 females, 5 males, median age of 39 years), two treated no attacks, and one treated aura only, leaving a full analysis set of 27 treating 80 attacks with pain. An adverse event was reported in 13 individuals, notably: neck twitching (n=1), raspy voice (n=1) and redness at the device site (n=1). No unanticipated, serious, or severe adverse events were reported. The pain-free rate at 2 hours was 4 of 19 (21%) for the first treated attack with a moderate or severe headache at baseline. For all moderate or severe attacks at baseline, the pain-free rate was 12/54 (22%). The authors concluded that nVNS may be an effective and well-tolerated acute treatment for migraine in certain individuals. These preliminary findings need to be validated by well-designed studies.

In 2015, Kinfe et al. enrolled 20 individuals with treatment-refractory migraine in this 3-month, open-label, prospective, noncomparative observational study of preventive nVNS. The number of headache days per month decreased from 14.7 to 8.9 (P<0.01) between baseline and end of treatment (3 months). The migraine disability assessment score improved from 26 to 15 (P<0.01).

Nesbitt et al. (2015) reported their initial experience with a novel device, designed to provide portable, noninvasive, transcutaneous stimulation of the vagus nerve, both acutely and preventively, as a treatment for cluster headache. Individuals with cluster headache (11 chronic, eight episodic), from two centers, including seven who were refractory to drug treatment, had sufficient data available for analysis in this open-label observational cohort study. The device, known as the gammaCore, was used acutely to treat individual attacks as well as to provide prevention. Patient-estimated efficacy data were collected by systematic inquiry during follow-up appointments up to a period of 52 weeks of continuous use. Fifteen individuals reported an overall improvement in their condition, with four reporting no change, providing a mean overall estimated improvement of 48%. Of all attacks treated, 47% were aborted within an average of 11±1 minutes of commencing stimulation; 10 individuals reduced their acute use of high-flow oxygen by 55%, with nine reducing use of triptan by 48%. Prophylactic use of the device resulted in a substantial reduction in estimated mean attack frequency from 4.5/24 hours to 2.6/24 hours (P<0.0005) posttreatment. The authors concluded that these data suggested that noninvasive VNS may be practical and effective as an acute and preventive treatment in chronic cluster headaches. They stated that further evaluation of this treatment using randomized sham-controlled trials is thus warranted. This study provided Class IV evidence that for individuals with cluster headache, transcutaneous stimulation of the vagus nerve aborts acute attacks and reduces the frequency of attacks.

Puledda and Goadsby (2016) stated that neuromodulation is a promising, novel approach for the treatment of primary headache disorders. Neuromodulation offers a new dimension in the treatment that is both easily reversible and tends to be very well-tolerated. The autonomic nervous system is a logical target given the neurobiology of common primary headache disorders, such as migraine and the trigeminal autonomic cephalalgias (TACs). These investigators reviewed new encouraging results of studies from the most recent literature on neuromodulation as acute and preventive treatment in primary headache disorders, and discussed some possible underlying mechanisms. These researchers focused on vagal nerve stimulation (VNS) and sphenopalatine ganglion stimulation (SPGS) since they have targeted autonomic pathways that are cranial and can modulate relevant pathophysiological mechanisms. The initial data suggested that these approaches will find an important role in headache disorder management going forward. The authors concluded that the armamentarium for the treatment of migraine and the TACs is rapidly expanding thanks to neuromodulation techniques. The newer methods appear much better tolerated and offer important therapeutic benefits. Equally attractive in many ways is that bench-based understanding is being applied to neuromodulation to yield bedside advances in treatment. They stated that clinicians can look forward to the results of a number of ongoing studies and the real possibility to add these exciting methods to their practice.

In 2016, Silberstein et al. published the results from the EVENT trial, which was a feasibility study of prevention with a sample size of 59 individuals. It was not powered to detect differences in efficacy outcomes. For the outcome of response, defined as 50% or more reduction in the number of headache days, 10% of the individuals in the nVNS group versus 0% in the sham group were responders; statistical testing was not performed.

In a prospective, open-label, randomized study, Gaul et al. (2016) compared adjunctive prophylactic nVNS (n=48) with standard of care (SoC) alone (control; n=49) for the acute treatment of chronic cluster headache (cCH). A 2-week baseline phase was followed by a 4-week randomized phase (SoC plus nVNS vs control) and a 4-week extension phase (SoC plus nVNS). The primary endpoint was the reduction in the mean number of cluster headache (CH) attacks per week. Response rate, abortive medication use and safety/tolerability were also assessed. During the randomized phase, individuals in the intent-to-treat population treated with SoC plus nVNS (n=45) had a significantly greater reduction in the number of attacks per week versus controls (n=48) (−5.9 vs −2.1, respectively) for a mean therapeutic gain of 3.9 fewer attacks per week (95% CI, 0.5– 7.2; P=0.02). Higher (≥ 50%) response rates were also observed with SoC plus nVNS (40% [18/45]) versus controls (8.3% [4/48]; P<0.001). No serious treatment-related adverse events occurred. The authors concluded that adjunctive prophylactic nVNS is a well-tolerated novel treatment for chronic CH, offering clinical benefits beyond those with SoC.

The authors stated that study limitations included the lack of a placebo/sham device, an open-label study design, the short treatment duration and the use of patient-reported outcomes. No placebo arm was incorporated into the study because a suitable placebo/sham device had not yet been designed. Instead of a placebo/sham arm, SoC was deemed the most appropriate control treatment that was reflective of a real-world clinical scenario. The open-label study design and short treatment duration may have contributed to a placebo effect in both treatment groups. The 16.7% response rate in the control group during the extension phase may partially reflect a placebo response to nVNS. The initial response experienced in the control group during the randomized phase may have also impacted the capacity for a meaningful response to nVNS during the extension phase. Furthermore, fewer individuals in the control arm (50%) than in the nVNS arm (64.4%) were highly adherent (≥80%) to prophylactic nVNS, which may have further confounded response rates and reductions in abortive medication use in this group. Only individuals with chronic, treatment-refractory CH were included because of their stable CH attack frequency and intensity. A 2.5-month study duration was deemed sufficient to observe a treatment effect. Treatment response in favor of nVNS was consistent across intent-to-treat (ITT), modified ITT (mITT) and per-protocol populations (per-protocol population was defined as participants in the mITT population who had ≥ 12 days of observation in the randomized phase and no major protocol violation). Because no CH-specific QOL instruments exist, the EQ-5D-3L and HIT-6 measures were considered most appropriate, and nVNS prophylaxis resulted in meaningful improvements for both measures. The apparent lack of effect of acute nVNS therapy on CH duration or severity was consistent with findings in the chronic CH population that were reported in a recent study of acute nVNS therapy for CH. The nVNS adherence rates in this study (50%–64%) were consistent with those reported for prophylactic noninvasive neuromodulation in migraine and were considered meaningful given that twice-daily nVNS requires more effort and participation than a conventional oral medication regimen.

In a randomized, double-blind, sham-controlled study (ACT1), Silberstein et al. (2016) evaluated nVNS as an acute CH treatment. A total of 150 subjects were enrolled and randomly assigned (1:1) to receive nVNS or sham treatment for less than or equal to 1 month during a double-blind phase; completers could enter a 3-month nVNS open-label phase. The primary endpoint was response rate, defined as the proportion of subjects who achieved pain relief (pain intensity of 0 or 1) at 15 minutes after treatment initiation for the first CH attack without rescue medication use through 60 minutes; secondary endpoints included the sustained response rate (15–60 minutes). Subanalyses of eCH and cCH cohorts were pre-specified. The intent-to-treat population comprised 133 subjects: 60 nVNS-treated (eCH, n=38; cCH, n=22) and 73 sham-treated (eCH, n=47; cCH, n=26). A response was achieved in 26.7% of nVNS-treated subjects and 15.1% of sham-treated subjects (P=0.1). Response rates were significantly higher with nVNS than with sham for the eCH cohort (nVNS, 34.2%; sham, 10.6%; P=0.008) but not the cCH cohort (nVNS, 13.6%; sham, 23.1%; P=0.48). Sustained response rates were significantly higher with nVNS for the eCH cohort (P=0.008) and total population (P=0.04). Adverse device effects (ADEs) were reported by 35 of 150 (nVNS, 11; sham, 24) subjects in the double-blind phase and 18 of 128 subjects in the open-label phase. No serious ADEs occurred. The authors concluded that in one of the largest randomized sham-controlled studies for acute CH treatment, the response rate was not significantly different (versus sham) for the total population; nVNS provided significant, clinically meaningful, rapid, and sustained benefits for eCH but not for cCH, which affected results in the total population. They stated that this safe and well-tolerated treatment represents a novel and promising option for eCH.

The authors noted that the drawbacks of this study included the analysis of the cCH cohort as part of the primary endpoint, the need for careful interpretation of subanalyses results, challenges with blinding inherent in medical device studies, and the time to first measurement of response used to define the primary efficacy endpoint. Primary endpoint results were significant for the eCH cohort but were diminished overall by the cCH cohort results. When subanalyses results were interpreted, the lack of statistical powering and the potential for type 1 and type 2 errors (in the eCH and cCH cohorts, respectively) should be considered. The difference in AE descriptions provided by subjects treated with the nVNS (e.g., drooping/pulling of the lip/face) and sham (e.g., burning, soreness, stinging) devices may help to explain results of the blinding analyses, which were similar to those observed in previous sham-controlled trials. The burning sensation and other pain-related AEs reported by the sham-treated group in ACT1 may have led to a placebo effect based on impressions that the subjects were receiving active treatment. Sham device–associated pain may have also produced a diffuse noxious inhibitory control (DNIC) effect, a phenomenon in which the application of a noxious electrical stimulus to remote body regions inhibits dorsal horn activity and attenuates the original pain. Potential placebo and DNIC effects in the sham group may have reduced the magnitude of the therapeutic benefit associated with nVNS treatment. Another drawback was that the time-point used to define the ACT1 primary end-point was 15 minutes after treatment initiation, which has been used in other CH studies, rather than after treatment completion. In ACT1, this 15-minute interval comprised an 8‐minute nVNS stimulation period followed by only a 7-minute period that appeared to be sufficient for significant treatment effects to become evident in the eCH cohort but not in the cCH cohort or total population. The 15-minute assessment time-point may have also contributed to the nonsignificant difference in average pain intensities between the nVNS and sham groups; other potential contributing factors include the combined statistical influence of the responders and nonresponders as well as the assessment after all attacks (rather than after the first attack). Thus, methodological implications in ACT1 regarding distinct effects among the eCH and cCH cohorts, the painful sham stimulation, and the use of a longer time-point to first measurement of response such as 30 minutes, as used in CH studies of other therapies, should be considered for future RCTs.

In 2016, Grazzi et al. reported on the use of preventive nVNS in an open-label, prospective, noncomparative study of 56 women with menstrual migraine. The treatment period was 12 weeks. At the end of treatment, the mean number of headache days per month was reduced from baseline (7.2–4.7; P<0.01). Twenty individuals (39%; 95% CI, 26%–54%) had a greater than or equal to 50% reduction in headache days.

Lainez and Guillamon (2017) summarized CH pathophysiology and the effectiveness of various neuromodulating techniques. In individuals with cCH, VNS with a portable device used in conjunction with SoC in CH individuals resulted in a reduction in the number of attacks. The authors concluded that new recent noninvasive approaches such as nVNS have shown effectiveness in a few trials and could be an interesting alternative in the management of CH, but require more testing and positive RCTs.

An UpToDate review on “Cluster headache: Treatment and prognosis" (May 2017) states that “When chronic cluster headache is unresponsive to medical treatments, various surgical interventions and neurostimulation techniques are potential treatment options, though none are clearly established as effective. In such cases, it is particularly important to exclude potential causes of secondary cluster headache. Neurostimulation techniques, including sphenopalatine ganglion stimulation and VNS, appear promising but remain investigational. Destructive surgical procedures are unproven and should be viewed with great caution."

Yuan and Silberstein (2017) stated that neuromodulation is an emerging area in headache management. Through neurostimulation, multiple brain areas can be modulated to alleviate pain, hence reducing the pharmacological need. These investigators discussed the recent development of the VNS for headache management. Early case series from epilepsy and depression cohorts using invasive VNS showed a serendipitous reduction in headache frequency and/or severity. Noninvasive VNS, which stimulates the carotid vagus nerve with the use of a personal handheld device, also demonstrated efficacy for acute migraine or CH attacks. Long-term use of nVNS appeared to exert a prophylactic effect for both chronic migraine and cCH. In animal studies, nVNS modulated multiple pain pathways and even lessen cortical spreading depression. Progression in nVNS clinical efficacy over time suggests an underlying disease-modifying neuromodulation. The authors concluded that nVNS appears to be as effective as the invasive counterpart for many indications. They noted that with an enormous potential therapeutic gain and a high safety profile, further development and application of nVNS is promising.

Miller and Matharu (2017) noted that there is growing interest in neuromodulation for primary headache conditions. Invasive modalities such as orbital nerve stimulation (ONS), deep brain stimulation (DBS) and SPGS are reserved for the most severe and intractable individuals. Noninvasive options such as VNS, supraorbital nerve stimulation, and transcranial magnetic nerve stimulation (TMS) have all emerged as potentially useful headache treatments. These researchers examined the evidence base for noninvasive neuromodulation in TACs and migraine. Although a number of open-label series of noninvasive neuromodulation devices have been published, there is very little controlled evidence for their use in any headache condition. Open-label evidence suggested that VNS may have a role in the prophylactic treatment of CH and there is limited evidence to suggest it may be useful in the acute treatment of cluster and potentially migraine attacks. There is limited controlled evidence to suggest a role for SONS in the prophylactic treatment of episodic migraine; however, there is no evidence to support its use in CH. TMS may be effective in the acute treatment of episodic migraine; but there is no controlled evidence to support its use as a preventative in any headache condition. The authors concluded that noninvasive neuromodulation techniques are an attractive treatment option with excellent safety profiles; however, their use is not yet supported by high-quality RCTs.

Mwamburi et al. (2017a) noted that CH is a debilitating disease characterized by excruciatingly painful attacks that affects 0.15% to 0.4% of the US population. Episodic cluster headache manifests as circadian and circannual seasonal bouts of attacks, each lasting 15 to 180 minutes, with periods of remission. In cCH, the attacks occur throughout the year with no periods of remission. Although existing treatments are effective for some individuals, many individuals continue to suffer. There are only two FDA-approved medications for eCH in the United States; others, such as high-flow oxygen, are used off-label. Episodic CH is associated with co-morbidities and affects work, productivity, and daily functioning. The economic burden of eCH is considerable, costing more than twice that of nonheadache individuals. These researchers stated that gammaCore adjunct to SoC was found to have superior efficacy in treatment of acute eCH compared with sham-gammaCore used with SoC in ACT1 and ACT2 trials. However, the economic impact has not been characterized for this indication. These investigators conducted a cost-effectiveness analysis of gammaCore adjunct to SoC compared with SoC alone for the treatment of acute pain associated with eCH attacks. The model structure was based on treatment of acute attacks with three outcomes: (i) failures, (ii) nonresponders, and (iii) responders. The time horizon of the model is 1 year using a payer perspective with uncertainty incorporated. Parameter inputs were derived from primary data from the RCTs for gammaCore. The mean annual costs associated with the gammaCore-plus-SoC arm was $9510, and mean costs for the SoC-alone arm was $10,040. The mean quality-adjusted life years for gammaCore-plus-SoC arm were 0.83, and for the SoC-alone arm, they were 0.74. The gammaCore-plus-SoC arm was dominant over SoC alone. All one-way and multiway sensitivity analyses were cost-effective using a threshold of $20,000. The authors concluded that gammaCore dominance, representing savings, was driven by superior efficacy, improvement in QOL, and reduction in costs associated with successful and consistent abortion of episodic attacks. They stated that these findings serve as additional economic evidence to support coverage for gammaCore. Moreover, they stated that additional real-world data are needed to characterize the long-term impact of gammaCore on co-morbidities, utilization, QOL, daily functioning, productivity, and social engagement of these individuals, and for other indications.

 

Mwamburi et al. (2017b) stated that the FDA has cleared gammaCore (nVNS) for the treatment of eCH. With the exception of subcutaneous sumatriptan, all other treatments are used off-label and have many limitations. The FDA approval process for devices differs from that of drugs. These researchers performed a review of the literature to evaluate new evidence on various aspects of gammaCore treatment and impact. The ACute Treatment of Cluster Headache Studies (ACT1 and ACT2), both double-blind sham-controlled randomized trials, did not meet the primary end-points of the trials; but each demonstrated significant superiority of gammaCore among individuals with eCH. In ACT1, gammaCore resulted in a higher response rate (RR) (RR, 3.2; 95% CI, 1.6–8.2; P=0.014), higher pain-free rate for greater than 50% of attacks (RR, 2.3; 95% CI, 1.1–5.2; P=0.045), and shorter duration of attacks (MD, −30 minutes; P<0.01) compared with the sham group. In ACT2, gammaCore resulted in higher odds of achieving pain-free attacks in 15 minutes (OR, 9.8; 95% CI, 2.2–44.1; P=0.01), lower pain intensity in 15 minutes (MD, −1.1; P<0.01), and higher rate of achieving responder status at 15 minutes for greater than or equal to 50% of treated attacks (RR, 2.8; 95% CI, 1.0–8.1; P=0.058) compared with the sham group. The PREVention and Acute Treatment of Chronic Cluster Headache (PREVA) study also demonstrated that gammaCore plus SoC was superior to SoC alone in individuals with cCH. Medical costs, pharmacy refills, and pharmacy costs were higher in individuals coded for CH in claims data compared with controls with non-headache codes. These researchers stated that gammaCore is easy to use, practical, and safe; delivery cannot be wasted; and individuals prefer using gammaCore compared with SoC. The treatment improved symptoms and reduces the need for CH rescue medications. They stated that current US reimbursement policies, which predated nVNS and are based on expensive, surgically implanted, and permanent implanted VNS (iVNS), need to be modified to distinguish nVNS from iVNS. The authors concluded that there is sufficient evidence to support the need to modify current reimbursement policies to include coverage for gammaCore (nVNS) for eCH. Moreover, they stated that one drawback was that the information available from publications that contribute to a review was as reported. This review, however, added new information to the body of evidence, particularly in comparison with previous reviews. While authors of previous reviews had identified gammaCore as a beneficial intervention for individuals with CH, they also pointed to a gap and a need for clinical trials to provide further evidence on its safety and effectiveness. They stated that the recommended future path is to collect real-world data that are specific to individuals suffering from eCH and CH with regard to use or non-use of gammaCore via a registry to monitor usage and performance measurement. Additionally, stakeholders should periodically review data from claims databases to evaluate long-term outcomes related to symptoms, utilization, cost, and reimbursement burden and the impact on co-morbidities and all-cause healthcare utilization, to better understand the value associated with gammaCore use beyond symptom relief. Also needed are continued research efforts, using RCTs, to characterize the benefits of gammaCore in other indications, including migraine, specific inflammatory illnesses, cardiac diseases, and psychiatric disorders.

Silberstein et al. (2017) noted that a panel of nine experts, including neurologists, other headache specialists, and medical and pharmacy directors, from four health plans (one integrated delivery network and three plans with commercial, Medicare, and Medicaid lines of business), convened to discuss CH. Topics covered included the current treatment landscape, treatment challenges, economic impact of disease, and gaps in care for individuals with CH. One major challenge in the management of CH is that it is under-recognized and frequently misdiagnosed, leading to delays in and suboptimal treatment for individuals who suffer from this painful and disabling condition. The management of CH is challenging due to the lack of a robust evidence base for preventive treatment, the AEs associated with conventional preventive treatments, the variability of response to acute treatments, and the challenging reimbursement landscape for well-accepted treatments (e.g., oxygen). The lack of effective prevention for many individuals may lead to the excessive use of acute therapies, often multiple times each day, which drives the cost of illness up significantly. The goal of the panel discussion was to discuss the role of gammaCore, the recently released first nVNS therapy in the acute treatment of individuals with eCH, in the management of CH. The panel reviewed current practices and formulated recommendations on incorporating a newly released therapy into CH management. The panel explored the role of traditional management strategies as well as that of gammaCore in the acute treatment of individuals with eCH. The panel agreed that the treatment guidelines should be updated to reflect the role of gammaCore as a first-line, acute therapeutic option for individuals with eCH and that payers should offer coverage of gammaCore to their members who have a diagnosis of eCH. Healthcare providers, including headache specialists and neurologists, and payers are encouraged to remain up-to-date regarding the results of ongoing clinical trials evaluating the use of gammaCore for the acute and/or preventive treatment of migraine to ensure that individuals are being appropriately treated for these conditions and that they have access to treatment through their insurers. Moreover, the panel noted that additional studies need to be conducted in the United States to verify the role of gammaCore in the preventive therapy of eCH and cCH.

Simon and Blake (2017) noted that stimulation of the cervical vagus nerve with iVNS has been used clinically for more than 20 years to treat individuals with epilepsy. More recently, gammaCore, a nVNS, was developed, which has been purported to also stimulate the vagus nerve without the cost and morbidity associated with an iVNS system. The gammaCore device has been used to acutely treat various types of primary headaches, including migraine and CH, and for the prevention of episodic, chronic, and menstrual migraines and CH. The gammaCore device was cleared by the FDA for the acute treatment of pain in individuals with eCH. These investigators summarized the clinical work that has been published in the use of gammaCore for treating primary headache disorders, presented an overview of studies demonstrating that nVNS does indeed stimulate similar vagus nerve fibers as the implantable VNS system, and then presented several animal headache-related studies that address the mechanism of action (MOA) of nVNS. The authors concluded that preliminary clinical studies in various primary headache disorders are encouraging. Human studies and modeling have demonstrated that nVNS activates vagus nerve fibers similar to those implicated in the clinical benefits of iVNS. They stated that continuing human and animal research is needed to further elucidate the MOA and to help define optimal signal parameters and treatment paradigms for headache and other disorders.

Gaul et al. (2017) stated that although the PREVA trial did not examine the effects of nVNS in individuals with eCH, the rapid beneficial effects on attack frequency observed within 2 weeks of treatment initiation in this cCH analysis, combined with the established safety profile of nVNS, suggested that a trial in eCH would be clinically reasonable.

Ho et al. (2017) noted that SPG is the largest collection of neurons in the calvarium outside of the brain. During the past century, it has been a target for interventional treatment of head and facial pain due to its ease of access. Block, radiofrequency ablation (RFA), and neurostimulation have all been applied to treat a myriad of painful syndromes. Despite the routine use of these interventions, the literature supporting their use has not been systematically summarized. These investigators summarized the level of evidence supporting the use of SPG block, RFA, and neurostimulation. Medline, Google Scholar, and the Cochrane Central Register of Controlled Trials (CENTRAL) databases were reviewed for studies on SPG block, RFA, and neurostimulation. Studies included in this review were compiled and analyzed for their treated medical conditions, study design, outcomes and procedural details. Studies were graded using Oxford Center for Evidence-Based Medicine for level of evidence. Based on the level of evidence, grades of recommendations are provided for each intervention and its associated medical conditions. A total of 83 publications were included in this review, of which 60 were studies on SPG block, 15 were on RFA, and eight were on neuro-stimulation. Of all the studies, 23 had evidence level above case series. Of the 23 studies, 19 were on SPG block, one study on RFA, and three studies on neurostimulation. The rest of the available literature was case reports and case series. The strongest evidence lied in using SPG block, RFA, and neurostimulation for CH; SPG block also had evidence in treating trigeminal neuralgia, migraines, reducing the needs of analgesics after endoscopic sinus surgery and reducing pain associated with nasal packing removal after nasal operations. The authors concluded that SPG is a promising target for treating CH using blocks, RFA, and neurostimulation; SPG block also had some evidence supporting its use in a few other conditions. Moreover, they stated that most of the controlled studies were small and without replications; further controlled studies are needed to replicate and expand on these previous findings.

Goadsby et al. (2018) reported on the results of a randomized, double-blind, sham-controlled study (ACT2) for the treatment of acute cluster headache attacks. Ninety-two individuals with cluster headaches were randomly assigned to tVNS (described in this response as nVNS) or sham treatment. Individuals were further identified as having eCHs or cChs and randomly assigned at approximately 1:1 to the tVNS and sham treatment groups. The primary efficacy endpoint was the ability to achieve pain-free status within 15 minutes of initiation of treatment without use of rescue treatment. There was no difference between tVNS-treated and sham-treated individuals in the overall cluster headache study population. Subgroup analysis of the cCh population showed no differences between tVNS-treated and sham-treated individuals. For the episodic cluster headaches subgroup, tVNS demonstrated a 48% response rate compared with 6% response rate for sham-treated (P<0.01). The interaction P value for the subgroup analysis was statistically significant ( P=0.04).

An UpToDate review on “Cluster headache: Treatment and prognosis" (2018) states that “There are several promising but unproven methods using neurostimulation to treat medically refractory cluster headache, including sphenopalatine ganglion stimulation, occipital nerve stimulation, noninvasive VNS, and deep brain stimulation. All are investigational and require further study to confirm long-term benefit and safety."

Sanchez-Gomez et al. (2018) evaluated the safety and effectiveness of peripheral neurostimulation of the SPG in the treatment of refractory cCH. Various medical databases were used to perform a systematic review of the scientific literature. The search for articles continued until October 31, 2016, and included clinical trials, systematic reviews and/or meta-analyses, health technology assessment reports, and clinical practice guidelines that included measurements of efficiency/effectiveness or adverse effects associated with the treatment. The review excluded cohort studies, case-control studies, case series, literature reviews, letters to the editor, opinion pieces, editorials, and studies that had been duplicated or outdated by later publications from the same institution. Regarding effectiveness, these investigators found that SPG stimulation had positive results for pain relief, attack frequency, medication use, and individuals' QOL. In the results regarding safety, these researchers found a significant number of AEs in the first 30 days following the intervention. Removal of the device was necessary in some individuals. Little follow-up data, and no long-term data, were available. The authors concluded that these findings are promising, despite the limited evidence available. They considered it essential for research to continue into the safety and efficacy of SPG stimulation for individuals with refractory cCH. In cases where this intervention may be indicated, treatment should be closely monitored.

de Coo et al. (2019) combined the data from ACT1 and ACT2 meta-analytically for the two primary outcomes reported in the two studies. The authors reported an interaction between treatment group and CH subtype in the pooled analysis (P<0.05 for both outcomes).

In 2018, the results of a multicenter, double-blind, randomized, sham-controlled trial (the PRESTO trial) of acute treatment of migraine with nVNS in 248 individuals with episodic migraine with/without aura were reported in three separate publications (Tassorelli et al., 2018; Martelletti et al., 2018; Grazzi et al. , 2018). The primary efficacy outcome was the proportion of participants who were pain-free without using rescue medication at 120 minutes. There was not a statistically significant difference in the primary outcome (30% vs. 20%; P=0.07) although it favored the nVNS group. The nVNS group had a higher proportion of individuals with decrease in pain from moderate or severe to mild or no pain at 120 minutes (41% vs. 28%; P=0.03) and a higher proportion of individuals who were pain-free at 120 for 50% or more of their attacks (32% vs. 18%; P=0.02). PRESTO results did not include QOL or functional outcomes and the double-blind treatment and follow-up period was 4 weeks. In the additional 4 weeks of acute nVNS in the open-label period, rates of pain-free response after the first treated attack (28%) and pain relief (43.4%) were similar to the rates in the double-blind period.

In 2018, Trimboli et al. reported on the preventive and acute treatment of nVNS in 41 consecutive individuals with refractory primary chronic headaches (n=23 with chronic migraine) in an open-label, prospective, noncomparative clinical audit. Response was defined as at least 30% reduction in headache days/episodes after 3 months of treatment. Two of 23 (9%) individuals with chronic migraine met the definition for responder.

In 2019, Diener et al. published the results from a phase 3, multicenter, sham-controlled RCT (the PREMIUM trial) conducted in several European countries including individuals who experienced five to 12 migraine days per month. The study included a 4-week run-in period during which no treatment was administered; 477 participants entered the run-in. The criteria to remain eligible after run-in were not described in the publication. After run-in, 341 participants were randomly assigned (nVNS, n=169 or sham, n=172) to a 12-week double-blind treatment period followed by a 24-week open-label period of nVNS. Individuals administered two 120-second stimulations bilaterally to the neck with gammaCore, three times daily. nVNS was not statistically significantly superior to sham with respect to the outcomes of reduction of at least 50% in migraine days from baseline to the last 4 weeks, reduction in number of migraine days from baseline to the last 4 weeks or acute medication days in the intention-to-treat population.

In 2020, Lai et al. analyzed the effectiveness and safety of peripheral nerve stimulation of the cervical branch of the vagal nerve for primary headache disorders. A systematic review and meta-analysis of the literature was carried out on RCTs of nVNS for treating headaches. A random-effects model was used to report all outcomes. The primary outcomes were a reduction in headache days or attacks and pain-free status within 30 minutes. Secondary outcomes were the pain-relief status within 30 minutes, the pain-relief status at 60 minutes, abortive medication use, greater than or equal to 50% responder rate, pain-free status in greater than or equal to 50% of treated attacks, adverse events, and satisfaction. In total, 983 individuals were included from six trials. The authors found that nVNS was effective in achieving a pain-free status within 30 min (odds ratio [OR], 2.27; 95% confidence interval [CI], 1.16~4.44; P=0.02), pain-relief status within 30 min (OR, 1.8; 95% CI, 1.17~2.78; P=0.007), pain-relief status at 60 minutes (OR, 1.93; 95% CI, 1.2~3.1; P=0.006), a reduction in abortive medication use (OR, 0.61; 95% CI, 0.41~0.92; P=0.02), and pain-free status in greater than or equal to 50% of treated attacks (OR, 2.15; 95% CI, 1.27~3.66; P=0.005) compared to sham-device treatment. There were no significant differences in decreased headache days (standardized mean difference [SMD], −0.159; 95% CI, -0.357~0.04; P=0.117), adverse events (OR, 1.084; 95% CI, 0.559~2.104; P=0.811), or satisfaction (OR, 1.45; 95% CI, 0.97~2.17; P=0.07) between nVNS and sham-device treatment. The ≥50% responder rate could not be determined (OR, 3.34; 95% CI, 0.83~13.33; P=0.09; I 2=73%). The authors concluded that "cervical nVNS is effective for acute pain relief for migraine and cluster headache."

OTHER CONDITIONS

The t-VNS system uses a noninvasive approach that combines a stimulation unit and ear electrode to stimulate the auricular branch of the vagus nerve (which supplies the skin over the concha of the ear) for the treatment of pharmacoresistant epilepsy. Individuals self-administer electric stimulation for several hours per day. Currently, the device does not have FDA approval.

In 2020, Wu et al. reported the results of a systematic review and meta-analysis of 3 RCTs (Aihua et al. 2014; Bauer et al. 2016; Rong et al. 2014) of tVNS for the treatment of drug-resistant epilepsy. All treatment groups underwent a cymba conchae stimulus at a frequency of 20 to 30 Hz. The control groups received various kinds of sham stimulation at a frequency of 1 Hz, the same frequency stimulation as treatment but at the nonauricular vagus nerve area or no stimulation. Meta-analysis of the RCTs found that seizure frequency was significantly reduced with tVNS (Mean Difference [MD]=−3.29; 95% CI, −6.31 to −0.27). However, meta-analysis of two RCTs that reported responder rates (undefined) did not find a significant difference between the tVNS and control groups (N=238; OR=0.47; 95% CI, 0.54−4.02]. All three RCTs assessed QOL using the Quality of Life in Epilepsy Inventory (QOLIE)-31 scale, but found no significant differences between treatment and control groups. Important limitations of the RCTs include imprecision, risk of confounding due to potentially imbalanced use of important nonprotocol interventions (i.e., concomitant antiepileptic drugs), and unacceptable flaws in outcome assessment (i.e., unspecified definition of response, between-group differences in measurement timing, lack of EEG data).

 

In 2016, Bauer et al. published the results of a randomized, double-blind controlled trial (cMPsE02) assessing the efficacy and safety of tVNS versus control stimulation in individuals with drug-resistant epilepsy. The primary objective was to demonstrate superiority of add-on therapy with tVNS (stimulation frequency 25 Hz, n=39) versus active control (1 Hz, n=37) in reducing seizure frequency over 20 weeks. Secondary objectives comprised reduction in seizure frequency from baseline to end of treatment, subgroup analyses and safety evaluation. Treatment adherence was 84% in the 1-Hz group and 88% in the 25-Hz group, respectively. Stimulation intensity significantly differed between the 1-Hz group (1.02 ± 0.83 mA) and the 25 Hz group (0.50 ± 0.47 mA; P=0.006). Mean seizure reduction per 28 days at end of treatment was −2.9% in the 1-Hz group and 23.4% in the 25-Hz group (P=0.146). In contrast to controls, there was a significant reduction in seizure frequency in individuals of the 25-Hz group who completed the full treatment period (20 weeks; n=26, 34.2%, P=0.034). Responder rates (25%, 50%) were similar in both groups. Subgroup analyses for seizure type and baseline seizure frequency revealed no significant differences. Adverse events were usually mild or moderate and comprised headache, ear pain, application site erythema, vertigo, fatigue, and nausea. Four serious adverse events were reported including one sudden unexplained death in individuals with epilepsy (SUDEP) in the 1-Hz group which was assessed as not treatment-related. The authors concluded that "tVNS had a high treatment adherence and was well tolerated. Superiority of 25 Hz tVNS over 1 Hz tVNS could not be proven in this relatively small study, which might be attributed to the higher stimulation intensity in the control group. Efficacy data revealed results that justify further trials with larger patient numbers and longer observation periods."

In 2014, Rong et al. evaluated the safety and effectiveness of transcutaneous auricular VNS (ta-VNS) for individuals with refractory epilepsy. A total of 144 individuals were randomly assigned to a ta-VNS group (n=98) or transcutaneous auricular non-VNS (tn-VNS) control group (n=46). Treatment was conducted twice per day for 24 weeks. After 8, 16, and 24 weeks of treatment, the individuals were evaluated according to the Modified Engel Scale (four classes). After 8 weeks, according to the medical ethic design, individuals in tn-VNS group were switched into ta-VNS group uniquely. After 8 weeks' treatment, 41.0% and 27.5% of individuals from ta-VNS and tn-VNS groups, respectively, experienced reduction in seizure frequency that reached I, II, and III levels according to the standards of the Modified Engel Scale compared with the baselines, indicating significant difference in seizure reduction between the two groups. After 24 weeks of treatment, individuals with ta-VNS had a 47.7% reduction, and tn-VNS, with an additional 16 weeks of treatment, reached 47.5% in reduction. After 8 weeks' treatment, the percentages of average seizure frequency in ta-VNS and tn-VNS were reduced by 42.6% and 11.5%, respectively, providing a statistically significant difference in the results between the two groups ( P<0.05). In addition, there were significant improvements in EEG and the quality of daily life of the individuals after treatment. The authors concluded that the "results show that this ta-VNS treatment can effectively reduce the frequency of seizures and improve the patient's QOL. This may be an effective treatment for refractory epilepsy. At the same time, it is also safe, economic, and widely applicable."

In 2014, Aihua et al. reported on results from a series of 60 individuals with pharmaco-resistant epilepsy treated with a tVNS device, who were randomized to stimulation over the earlobe (control group) or the Ramsay-Hunt zone (treatment group), which includes the external auditory canal and the conchal cavity and is considered to be the somatic sensory territory of the vagus nerve. Thirty individuals were randomized to each group; 4 subjects from the treatment group were excluded from analysis due to loss to follow-up (n=3) or adverse events (n=1), while 9 subjects from the control group were excluded from analysis due to loss to follow-up (n=2) or increase or lack of decrease in seizures or other reasons (n=7). In the treatment group, compared with baseline, the median monthly seizure frequency was significantly reduced after 6 months (5.5 months vs. 6.0 months; P<0.001) and 12 months (4.0 months vs. 6.0 months; P<0.001) of tVNS therapy. At 12-month follow-up, tVNS group subjects had a significantly lower median monthly seizure frequency compared with the control group (4.0 months vs. 8.0 months; P<0.001).

 

Two small case series identified used a tVNS device for treatment of medication-refractory seizures. In a small case series of 10 individuals with treatment-resistant epilepsy, Stefan et al. (2012) reported that three individuals withdrew from the study, while 5 of 7 individuals reported a reduction in seizure frequency. In another small case series, He et al. (2013) reported that among 14 pediatric individuals with intractable epilepsy who were treated with bilateral tVNS, of the 13 individuals who completed follow-up, the mean reduction in self-reported seizure frequency was 31.8% after 8 weeks, 54.1% from week 9 to 16, and 54.2% from week 17 to 24.

In 2020, Kutlu et al. reported the results of an RCT that compared a home-based exercise treatment program with or without auricular VNS in 60 female individuals in Turkey with fibromyalgia syndrome (auricular VNS n=30 and no auricular VNS n=30). The VNS was delivered in 30-minute sessions on weekdays for 4 weeks. The home-based exercise program consisted of strengthening, stretching, isometric, and posture exercises that targeted the body and upper and lower extremities. When added to exercise, auricular VNS did not significantly improve mean scores on the Fibromyalgia Impact Questionnaire (37.27 vs. 41.93; P=0.378) or on any 36-Item Short Form Health Survey subscales (e.g., Physical Function: 80.00 vs. 85.00; P=0.167). An important limitation of this RCT is the lack of a sham control group.

In 2014, Huang et al. (2014) reported on results of a pilot RCT of a tVNS device that provides stimulation to the auricle for the treatment of impaired glucose tolerance. The trial included 70 individuals with impaired glucose tolerance who were randomized to active or sham tVNS, along with 30 controls who received no tVNS treatment. After 12 weeks of treatment, individuals who received active tVNS were reported to have significantly lower 2-hour glucose tolerance test results than those who received sham tVNS (7.5 mmol/L vs. 8 mmol/L; P=0.004).

In 2013, Hein et al. reported on results of 2 pilot RCTs of a tVNS device for the treatment of depression, 1 of which included 22 subjects and another assessed 15 subjects. In the first study, 11 subjects were randomized to active or sham tVNS. At 2-week follow-up, Beck Depression Inventory (BDI) self-rating scores in the active stimulation group decreased from 27.0 to 14.0 points (P<0.001), while the sham-stimulated individuals did not show significant reductions in BDI scores (31.0 to 25.8 points). In the second study, 7 individuals were randomized to active tVNS, and eight individuals were randomized to sham tVNS. In this study, BDI self-rating scores in the active stimulation group decreased from 29.4 to 17.4 points (​P<0.05) after 2 weeks, while the sham-stimulated individuals did not show a significant change in BDI scores (28.6 to 25.4 points). The authors did not report direct comparisons in BDI change scores between the sham- and active-stimulation groups.

In 2014, Shiozawa et al. conducted a systematic review of studies evaluating the evidence related to transcutaneous stimulation of the trigeminal or vagus nerve for psychiatric disorders. Overall, the studies assessed were limited by small size and poor generalizability.

 

In 2015, Hasan et al. reported on a randomized trial of tVNS for the treatment of schizophrenia. Twenty individuals were assigned to active tVNS or sham treatment for 12 weeks. There was no statistically significant difference in the improvement of schizophrenia status during the observation period.

In 2020, Wu et al. reported the results of a randomized, pilot sham-controlled RCT in 21 individuals (nVNS=10 and sham nVNS, n=11) with upper limb motor function impairment following subacute ischemic stroke. The mean Fugl-Meyer assessment–upper extremity scores increased by 6.90 with nVNS versus 3.18 points with sham after 15 days of intervention (Difference= −3.72 points; 95% CI, −5.12 to -2.32; p≤.001). The improvement in the mean Fugl-Meyer assessment–upper extremity scores remained significantly higher at both the 4-week (+7.70 vs. +3.36; P≤0.001) and the 12-week (+7.40 vs. +4.18; P=0.038) follow-ups. There was only 1 adverse event noted, which was that one individual in the nVNS group developed skin redness at an electrode point of contact.

AMERICAN ACADEMY OF NEUROLOGY

In 1999, the American Academy of Neurology released a consensus statement on the use of VNS in adults, which stated: “VNS is indicated for adults and adolescents over 12 years of age with medically intractable partial seizures who are not candidates for potentially curative surgical resections, such as lesionectomies or mesial temporal lobectomies." The Academy updated these guidelines in 2013, stating: “VNS may be considered for seizures in children, for LGS [Lennox-Gastaut syndrome]-associated seizures, and for improving mood in adults with epilepsy (Level C). VNS may be considered to have improved efficacy over time (Level C)."

AMERICAN PSYCHIATRIC ASSOCIATION

Updated in 2010, the American Psychiatric Association guidelines for the treatment of major depressive disorder in adults included the following statement on the use of VNS: “Vagus nerve stimulation (VNS) may be an additional option for individuals who have not responded to at least four adequate trials of antidepressant treatment, including ECT [electroconvulsive therapy]," with a level of evidence III (may be recommended on the basis of individual circumstances).

NATIONAL INSTITUTE FOR HEALTH AND CARE EXCELLENCE

National Institute for Health and Care Excellence (NICE) interventional procedures guidance on “Transcutaneous stimulation of the cervical branch of the vagus nerve for cluster headache and migraine" (2016) stated that “Current evidence on the safety of transcutaneous stimulation of the cervical branch of the vagus nerve for cluster headache and migraine raises no major concerns. The evidence on efficacy is limited in quantity and quality. Therefore, this procedure should only be used with special arrangements for clinical governance, consent and audit or research … Clinicians wishing to do transcutaneous stimulation of the cervical branch of the vagus nerve for cluster headache and migraine should ensure that individuals understand the uncertainty about the procedure's efficacy and provide them with clear written information … NICE encourages further research on transcutaneous stimulation of the cervical branch of the vagus nerve for cluster headache and migraine. Studies should describe whether the procedure is used for treatment or prevention, and whether it is used for cluster headache or migraine. Clinicians should clearly document details of patient selection and the treatment regimen. Outcome measures should include changes in the number and severity of cluster headache or migraine episodes, medication use, QOL in the short and long term, side effects, acceptability, and device durability".

The NICE published in 2018 a Medtech innovation briefing on nVNS for cluster headache (MIB162). The briefing states that the "intended place in therapy would be as well as standard care, most likely where standard treatments for cluster headache are ineffective, not tolerated or contraindicated" and that key uncertainties around the evidence are that "people with episodic and chronic cluster headaches respond differently to treatment with gammaCore. The optimal use of gammaCore in the different populations is unclear." In 2019, the National Institute for Health and Care Excellence published a Medical Technologies Guidance [MTG46] on gammaCore for cluster headache in December 2019. The recommendations state that evidence supports using gammaCore to treat cluster headache and that gammaCore is not effective in everyone with cluster headache.

The NICE published in 2020 an Interventional Procedure Overview on implanted VNS for TRD (IPG679). The guidance states: "Evidence on the safety of implanted vagus nerve stimulation for TRD raises no major safety concerns, but there are frequent, well-recognized side effects. Evidence on its efficacy is limited in quality. Therefore, this procedure should only be used with special arrangements for clinical governance, consent, and audit or research." The guidance further states that "NICE encourages further research into implanted vagus nerve stimulation for TRD, in the form of randomized controlled trials with a placebo or sham stimulation arm. Studies should report details of patient selection. Outcomes should include validated depression rating scales, patient-reported QOL, time to onset of effect and duration of effect, and any changes in concurrent treatment."​