Vagus nerve stimulation: An update of currently registered clinical trials on ClinicalTrials.gov.

Vagus nerve stimulation (VNS) has emerged as a promising therapeutic modality, eliciting significant interest across various medical disciplines due to its profound effects on physiological and psychological processes.[ 27 , 40 , 55 , 56 ] VNS is the stimulation of the vagus nerve, a key component of the parasympathetic nervous system that regulates various physiologic functions. Conventional VNS involves the surgical implantation of a device that delivers electrical impulses to the vagus nerve that travel to the brainstem and subsequently diffuse into various brain regions, although non-invasive methods have become more prominent.[ 28 , 49 , 63 ] The precise mechanisms underlying the therapeutic effects of VNS are multifaceted and not yet fully understood. However, it is believed to involve the activation of neural circuits associated with mood regulation, autonomic function, and inflammatory response modulation.[ 61 ] In addition, VNS has been shown to promote the release of neurotransmitters such as serotonin and norepinephrine, which play pivotal roles in mood and pain modulation.[ 35 ] Its effects on the nucleus tractus solitarius and downstream autonomic circuits underlie its cardiovascular and anti-inflammatory effects.[ 24 ] Overall, the complex interplay of neural pathways and neurotransmitter systems engaged by VNS underscores its potential as a versatile therapeutic approach for various conditions. Beyond its initial use in refractory epilepsy and stroke, VNS holds significant promise for an expanding array of medical indications. Emerging research has indicated its potential efficacy in addressing treatment-resistant depression (TRD), migraines, fibromyalgia, neurodegenerative disorders such as Alzheimer’s disease, rheumatoid arthritis, and inflammatory bowel disease, among other pathologies.[ 7 , 11 , 12 , 15 , 36 , 37 , 39 , 54 , 60 ] As research continues to unveil the diverse mechanisms and therapeutic potential of VNS, its application is poised to extend beyond its current indications, revolutionizing the landscape of neuromodulation-based therapies. While basic science advancements have deepened our understanding of vagal nerve physiology and the mechanisms underlying VNS, the translation of these discoveries into widespread clinical applications faces significant regulatory hurdles. Research has highlighted the potential of VNS as a therapy for a wide range of conditions. Despite this progress, its the United States food and drug administration (FDA)-approved uses remain limited to epilepsy management, TRD, rheumatoid arthritis (SetPoint Device), and stroke rehabilitation, reflecting the complexities of moving from early clinical investigation to widespread clinical use. Other trials, such as gammaCore for migraines and emergency authorization for COVID-19, have been approved indications for non-invasive VNS. Clinical trials are critical for bridging this gap, validating the safety and efficacy of VNS while shedding light on the mechanisms of VNS in diverse pathophysiological contexts. To date, hundreds of trials have been conducted or are ongoing globally. However, challenges such as the variability in trial designs, diverse patient populations, and the need for long-term follow-up have limited the integration of VNS into broader clinical practice. Therefore, this study aims to evaluate the current state of VNS clinical trials by performing a retrospective review of ClinicalTrials.gov, a nationwide publicly available database for clinical trials. Unlike traditional systematic reviews, which focus on published outcomes from completed studies, our registry-based approach offers real-time insight into ongoing and emerging research efforts. This type of surveillance is critical for identifying future therapeutic directions, regulatory trends, and areas of clinical uncertainty. While we report available outcomes where applicable, we refrain from making subjective assessments of clinical significance, allowing readers to draw their own conclusions based on the data. The primary objective of this manuscript is to provide an update on all clinical trials performed to date, as well as highlight new disorders being assessed by VNS in current clinical trials.

The search of ClinicalTrials.gov initially yielded 440 clinical trials before screening. Of these initial trials, 346 met our inclusion criteria and 94 were excluded [ Figure 1 ]. Out of the 346 included trials, 147 (42%) were completed, 127 (37%) were in the recruitment process, 8 (2.31%) were recruiting by invitation, 18 (5%) were active but not yet recruiting, and 46 (13%) had yet to begin recruiting [ Table 1 ]. The trials spanned from 2000 to 2024, exhibiting a non-linear exponential growth in VNS clinical trials ( P < 0.001, R 2 = 0.924) [ Figure 2 ]. A significant increase in trials occurred post-2015. A similar pattern was observed for interventional trials ( P < 0.0001, R 2 = 0.924) [ Figure 3 ]. Scatter plot of clinical trials initiated per year. Statistical testing was performed with Pearson correlation coefficient testing ( P < 0.001). Scatter plot of interventional clinical trials initiated per year. Statistical testing was performed with Pearson correlation coefficient testing ( P < 0.001). Majority of the clinical trials were interventional (330, 95.38%), while 16 (4.6%) studies were observational [ Table 1 ]. In addition, most of the trials (86%) were non-invasive, while 14% were invasive. All conditions with completed trials featured interventional study types, except for aging, which comprised only observational trials. Across all conditions, five included both observational and interventional trials: neurodegenerative diseases (20% observational trials), epilepsy/seizure (19%), diabetes (14%), healthy patients (10%), and pain disorders (7%) [ Table 2 ]. Overview of 346 clinical trials by status and study type. The table provides information on 147 completed trials, the corresponding percentage of total completed trials for each condition, the percentage breakdown of study types (interventional or observational) for each condition, and the overall participant count for each condition. The interventions included multiple modalities of VNS, primarily the non-invasive transcutaneous auricular VNS, in which an external device delivers electrical stimulation through the skin. Invasive interventions require the surgical implantation of a device with a lead wire that connects directly to the left vagus nerve. Some interventions included VNS in conjunction with a drug or behavioral intervention. In interventional trials where VNS is the primary variable, sham stimulation was used as a control, or VNS was compared to the current standard of care. Stimulation modality and protocol specifications (e.g., frequency and amplitude) were inconsistently reported; however, representative examples of cervical and auricular VNS trials are noted. Among the 147 completed clinical trials, categories of conditions with the highest number of trials included epilepsy/seizure with 16 trials (11%), pain disorders with 15 trials (10%), cardiovascular diseases with 14 trials (10%), stroke with 11 trials (8%), and gastrointestinal disorders with 11 trials (8%) [ Table 2 ]. Nearly all clinical conditions had completed trials, except for three, which included autoimmune disorders, vagus nerve disorders, and eating disorders [ Table 2 ]. Of those completed, results have been published for 34 trials (10%). Majority of these trials (27, 79%) were conducted in the United States, with the rest being conducted in Europe. The trials with published results all spanned 1–5 years, except for one trial lasting 8 years and another 13 years (median 2.6 years). The participant counts across conditions exhibited a wide range. Among completed trials, migraine/headache had the highest number of participants (1,041), followed by pain disorders (770), epilepsy/seizure (665), and gastrointestinal disorders (511). Other conditions had very few participants, such as kidney disease (7) and endometriosis (18). Of note, although there was only one completed trial for VNS in the setting of cancer, it had a total of 288 participants. In addition, 10 trials involving neuromodulation in healthy individuals were completed, involving a total of 486 participants [ Table 2 ]. The conditions with the highest percentage of trials in the “Active, Not Recruiting” category included diabetes (27%), arthritis (20%), and infant development (20%). However, 21 conditions in this category had no trials in the “Active, Not Recruiting” category. A similar pattern was observed in the “Enrolling by Invitation” category, where 27 conditions did not have any currently enrolling trials. The three medical conditions with the greatest percentage of trials in the “Enrolling by Invitation” category included aging (25%), vagus nerve disorders (20%), and ear disorders (14%) [ Table 3 ]. In the “Not Yet Recruiting” category, the highest representations were within post-operative conditions and spinal cord conditions (both 33%). In the “Recruiting” category, all but one condition had trials that were actively recruiting participants. All three trials for eating disorders were in this category. Information on 346 trials, percentage of total trials, percentage of active trials (not recruiting), percentage completed, percentage enrolling by invitation, percentage not yet recruiting, and percentage currently recruiting. The largest category of active or completed clinical trials involves cardiovascular conditions [ Figure 4 ]. Cardiovascular VNS clinical trials investigated VNS to treat forms of heart failure ( NCT01385176 , NCT03327649 ) and atrial fibrillation ( NCT02548754 ). NCT01385176 showed promising preliminary results with improved peak VO2 and left ventricular ejection fraction in patients who received VNS versus a sham treatment for 18 months; however, these results failed to reach statistical significance. NCT02548754 revealed a statistically significant reduction in atrial fibrillation burden in the VNS group. In a trial investigating patients with heart failure ( NCT03327649 ), the VNS group had improved quality of life and global longitudinal strain compared to sham controls. An additional study, published by Nearing et al ., found that in symptomatic heart failure patients with reduced ejection fraction, VNS led to long-term improvements in autonomic function.[ 44 ] However, a separate clinical trial, INOVATE-HF (Increase of Vagal Tone in Heart Failure), did not find a reduction in the rate of death or heart failure-related adverse events with VNS intervention in chronic heart failure patients.[ 23 ] Overall, peer-reviewed clinical trial results suggest that VNS can improve the condition of heart failure. Number of clinical trials per condition treated by vagus nerve stimulation. Other* :Trials including those for cardiovascular conditions, inflammatory diseases, and COVID-19. Epilepsy, psychological disorders, and pain disorders were among the next most common conditions where VNS is being investigated. Four randomized controlled trials found that VNS treatment reduced seizure activity by at least 50% in up to one-third of individuals in the treatment group.[ 3 , 9 , 21 , 25 ] In addition to improved seizure control, Englot et al . reported patient-rated improved alertness, mood changes, verbal communication, memory, and school/work performance with VNS treatment.[ 22 ] In addition, NCT01325623 found that patient-reported quality of life and seizure severity score (on the National hospital seizure severity scale (NHS)-3 scale) improved following 12 months of VNS treatment, and NCT01846741 found a decreased seizure severity score (NHS-3 scale) in 20 patients treated with VNS. While VNS is approved for TRD, it is being investigated to treat other mood disorders, including posttraumatic stress disorder (PTSD). Trial NCT02992899 reported that VNS may lead to a reversal of neurobiological changes that occur with PTSD. Separately, published results show that VNS has utility in substance use disorder ( NCT04556552 ), with significantly reduced opioid withdrawal symptoms and pain in the VNS treatment group compared to sham controls. Many VNS clinical trials for migraines have shown a decrease in the number of episodes ( NCT01532830 , NCT01667250 ) and intensity of migraines ( NCT02686034 ). VNS treatment led to a significant reduction in the number of migraine attacks with phonophobia in NCT01532830 , whereas NCT01667250 and NCT02686034 showed a decrease in mean headache days and headache pain scales in patients treated. Given the wide range of physiologic functions mediated by the vagus nerve, VNS has been explored for a variety of conditions, even in cases where the role of the vagus nerve is incompletely understood. Published results describe VNS as a treatment for tinnitus ( NCT01962558 ), to increase pancreatic glucose uptake ( NCT01117311 ), and even to help with infant feeding ( NCT04643808 ). VNS was also tested as a treatment for respiratory diseases with limited efficacy. In the treatment of asthma, patients treated with VNS experienced no symptomatic improvement, and one-third of patients experienced side effects. In the treatment of COVID-19, results were mixed. NCT01679314 showed no significant improvement in symptoms; however, NCT00762931 found an improvement in forced expiratory volume in COVID-19 patients receiving VNS treatment, and NCT05608629 found that VNS can improve long COVID-19 symptoms. Search of PubMed for the ClinicalTrials.gov Identifier of the 34 completed trials with results on ClinicalTrials.gov yielded 18 peer-reviewed reports with an approximately 53% rate of publication. These trials were: NCT01701245 ; NCT01958125 ; NCT02983448 ; NCT03327649 ; NCT02548754 ; NCT01846741 ; NCT02385526 ; NCT04157621 ; NCT04643808 ; NCT04556552 ; NCT01667250 ; NCT03592745 ; NCT01325623 ; NCT01792817 ; NCT02243020 ; NCT03131960 ; NCT02686034 ; NCT02992899 [ Table 4 ]. Summary of VNS clinical trials with published results ( n =34).

The vagus nerve controls myriad physiological functions, ranging from heart rate and respiration to digestion. With the nerve’s large responsibility in autonomic function, improper stimulation and signaling could lead to life-threatening conditions. By stimulating the vagus nerve, VNS devices improve outcomes in patients with impaired vagus nerve function. Our review found that VNS treatment yielded symptomatic improvement across a range of conditions. Of the clinical trials with published results, 31 of 34 concluded that VNS is a safe and effective treatment modality that can increase patient quality of life. A large proportion of included VNS clinical trials focused on the treatment of cardiovascular conditions, encompassing nearly 10% of all active clinical trials. VNS interventional research has explored cardiovascular health since the advent of this technology.[ 20 ] Importantly, cardiovascular diseases hold significant clinical relevance as the leading cause of death for all populations in the US, and with an estimated 6.2% of adults living with heart disease.[ 26 ] When active, the vagus nerve slows the heart rate, reduces atrioventricular conduction, and lowers ventricular contractility.[ 16 ] These negative dromotropic effects have been used to treat persistent sinus tachycardia, through recently developed closed-loop and more commonly used open-loop paradigms.[ 6 , 14 , 21 , 45 ] Closed-loop approaches show promise to optimize therapeutic effects while minimizing associated risks.[ 59 ] Another medical adaptation of VNS is commonly used for individuals with medically intractable epilepsy, defined as an uncontrollable seizure following two well-tolerated and appropriately chosen antiseizure drug regimens.[ 32 ] VNS has been shown to be both safe and effective in individuals aged 4 years and older and is FDA-approved for use in epilepsy.[ 46 ] Researchers hypothesize that VNS limits epileptic activity by desynchronizing cortical activity and altering electroencephalogram (EEG) patterns during a seizure; however, the exact method of action is unclear.[ 10 , 62 , 64 ] The most frequently reported psychiatric comorbidity among epilepsy patients is depression, with an estimated 40% of patients also managing depression, leading to lower quality of life and worse outcomes.[ 30 , 31 , 58 ] In 2005, clinical trials led by Rush et al . found that VNS can have antidepressant effects in patients with epilepsy.[ 50 - 52 ] Researchers hypothesized that VNS improves circulation to cortical regions responsible for mood regulation and can increase the production of regulatory neurotransmitters.[ 13 , 19 ] These trial domains align with preclinical models in which VNS has been shown to modulate cytokine levels, affect vagal tone, and suppress seizure activity through thalamocortical desynchronization. VNS can also be used in the treatment of depression in patients without epilepsy. As VNS depression trials started later than epilepsy and cardiovascular VNS trials, with many still recruiting, results from these trials are incomplete. Future reports on the utility of VNS in the treatment of depression are of great importance, as an estimated 29% of the United States population reports receiving a major depressive disorder diagnosis in their lifetime, and 17% reported being actively depressed in 2023.[ 29 ] VNS was FDA-approved in 2005 for TRD, defined as having failed at least two adequate trials of antidepressants during the current depressive episode, but unfortunately, it is no longer covered by public insurance, limiting its use.[ 1 ] Despite this setback, research into the use of VNS for the treatment of TRD has increased in recent years. Aaronson et al . found that patients who underwent VNS treatment had significantly shorter times to remission, and a reduction in suicidal ideation and all-cause mortality.[ 2 ] Further reports show clinical symptom improvement with VNS at long-term follow-up, and Conway et al . found a quality of life improvement in TRD patients who received VNS versus the standard of care.[ 18 , 41 , 53 ] Notably, a current randomized controlled trial, RECOVER, is investigating the use of VNS to modulate baseline depressive symptom severity ( NCT03887715 ). It is postulated that VNS reduces anxiety and depression by increasing parasympathetic tone and decreasing the sympathetic stress response.[ 38 ] Regarding its use in pain management, it is hypothesized that VNS provides an analgesic effect through the stimulation of afferent fibers.[ 47 , 48 ] VNS utility in pain management has been shown in fibromyalgia, chronic pelvic pain, and headaches.[ 8 , 33 , 42 , 57 ] This shows how VNS may be an optimal alternative to current pain management strategies, such as oral analgesics with poor adverse effect profiles or higher risk surgical treatments. Similarly, VNS is beneficial for arthritis, as the method of action for VNS in arthritis is the ability of vagal afferents to block the cytokine storm in response to molecular markers of injury and inflammation in the spleen, contributing to arthritic symptoms.[ 4 ] Given the high prevalence of arthritis in the adult population, clinical trials exploring these neuroimmunological interactions have the largest possible patient population overall; however, due to the novelty of this field, only 3% of included clinical trials focused on this use case. With early success, we can expect the number of clinical trials utilizing VNS to treat arthritis and other chronic pain conditions to increase. Finally, our longitudinal analysis revealed that the number of VNS clinical trials began to increase dramatically since 2015, suggesting that VNS as a treatment is still in its early exploratory phase. Each trial lasts several years, with an average time from bench to bedside being 10–15 years and even longer for widespread clinical adoption.[ 17 ] This registry-based review complements prior indication-specific or outcome-focused analyses by surveying the broader research pipeline for VNS. These data can help guide clinicians, industry partners, and policymakers in understanding the future landscape of VNS applications. Given the breadth of conditions treated using VNS, the number of trials and participants will continue to grow as more studies publish results describing VNS’s utility with positive results. While this review offers a summary of VNS clinical trials, there are several limitations. First, not all clinical trials may be included in the registries consulted, especially those conducted outside the United States, which could result in the omission of relevant studies. In addition, despite efforts to capture all available data, the inclusion of studies is dependent on the accuracy and completeness of information provided in trial registers. In particular, trial demographic reporting often does not follow standardized guidelines, hindering the ability to provide a more granular analysis of the composition of trial participants.[ 5 ] This meant the risk of bias could not be formally assessed due to the variable quality and completeness of registry-reported data. Despite a substantial increase in reporting rates, continued suboptimal adherence to registration and reporting standards by investigators and sponsors means that accurate, complete, and timely information is not provided for all studies. It is important to note that our analysis does not make direct claims regarding the clinical significance of reported outcomes. In many cases, trial data were incomplete or lacked the granularity necessary for such assessment (e.g., no effect size, minimal patient-reported outcome metrics). As a result, we chose a neutral reporting strategy, enabling readers to interpret the potential clinical impact based on their own expertise and contextual knowledge. While we used the Pearson correlation to summarize the relationship between year and trial registration counts, this method was chosen purely for descriptive visualization. We recognize that clinical trial trends are often non-linear and influenced by external factors such as FDA device approvals, insurance reimbursement changes, or global events such as the COVID-19 pandemic. Future analyses incorporating non-linear regression or segmented time-series models may provide deeper insight into growth trajectories or inflection points. In addition, the analysis of VNS clinical trials often relies on self-reported outcomes (i.e., quality of life scores) and interpretations of trial data, which could introduce bias or variability in study quality. Differences in trial design, participant populations, and reporting methodologies further complicate formal quantitative meta-analyses of trial results. Interpretation is limited by the substantial heterogeneity in stimulation protocols, including electrode placement and waveform parameters, which were not uniformly available across trials. In addition, due to the heterogeneity in trial design, endpoints, and limited availability of standardized results, we did not conduct a meta-analysis. Future studies focused on specific indications with harmonized endpoints may enable quantitative synthesis using forest or funnel plots. In particular, these trials do not have an established standard for the control treatment, nor is there a discrete metric to quantify the impact of each study. Given that many trial entries did not report whether blinding or sham controls were used, we were unable to assess the potential impact of placebo effects systematically. Finally, unpublished data or studies with negative outcomes may remain inaccessible, skewing the representation of VNS efficacy and safety across conditions. This may introduce survivorship bias and inflate the perceived success of VNS interventions. Three clinical trials included in this review ( NCT03359902 , NCT01679314 , NCT02388269 ) published with no significant results. In addition, there were 15 clinical trials with completed results but no associated peer-reviewed PubMed publications. Addressing these limitations in future studies, ensuring better adherence to reporting standards, and facilitating the timely publication of results are essential to bring novel treatments to patients suffering from intractable conditions.

VNS has shown substantial utility across a wide range of conditions and has been increasingly studied in clinical trials since 2015. In addition to its currently approved indications, VNS has demonstrated promising results for cardiovascular diseases, psychological disorders, and pain disorders, opening new horizons for the treatment of intractable conditions. Physicians should familiarize themselves with these results and potentially upcoming indications for VNS.

This was a retrospective study that was conducted on March 11, 2024, through a search of the ClinicalTrials. gov database.[ 43 ] A query was performed using the term “Vagus Nerve Stimulation.” Both invasive and non-invasive VNS methods were included for the purpose of obtaining a complete review of the full scope of VNS clinical trials. Two authors (MAH, JR) independently conducted the screening, and the senior author (MEA) was consulted for a final decision in the event of discrepancies. Trials were accepted for inclusion if they were completed, recruiting, not yet recruiting, active but not recruiting, or enrolling by invitation. The exclusion criteria consisted of any trials being classified as having an unknown status, or those that were withdrawn, terminated early, or suspended [ Figure 1 ]. Flow diagram for identification of relevant vagus nerve stimulation clinical trials. For trials that met the inclusion criteria, data on each trial’s ClinicalTrial.gov identifier, title, study status, study type, procedure type, injury type studied, country, year of initiation, sample size, and intervention were retrieved. For completed trials with associated results, a literature search was performed using the clinical trial number provided in the database (NCT ********). This search was performed using the PubMed database to identify peer-reviewed literature related to the clinical trial.[ 34 ] Associations of the number of clinical trials over time were analyzed and fit using Pearson correlation coefficients as a descriptive tool to illustrate the general association between calendar year and the number of trials registered. No inferential modeling was performed. A P < 0.05 was considered statistically significant. Due to incomplete and non-standardized reporting across trial entries, we did not apply formal risk of bias tools such as the Cochrane RoB 2.0. All analysis was performed using RStudio 4.4.0 (R Core Development Team, Vienna, Austria).