The vagus nerve is the longest nerve of the parasympathetic nervous system, the part of the body that runs rest and recovery, and it carries a steady stream of signals from the organs back up to the brain. It is also part of the inflammatory reflex, the loop through which the nervous system can turn the immune response up or down. Stimulating it deliberately covers a family of methods with very different evidence behind them. An implanted device is an established treatment for drug-resistant epilepsy, for treatment-resistant depression, and, paired with therapy, for arm recovery after stroke.
A non-invasive ear clip is a fast-growing research field with many small trials and open questions. And the everyday practices that raise vagal tone, slow breathing most of all, have measurable effects, though the claims made for humming and gargling run well ahead of what has been shown. The cautions that matter are gathered in one place lower down.
Findings & Outcomes
What It Is
The vagus nerve is the tenth cranial nerve and the main nerve of the parasympathetic nervous system, the part of the autonomic nervous system that slows the heart, settles digestion and governs rest and recovery. The name comes from the Latin for wandering, because it leaves the brainstem and travels down through the neck and chest to reach the heart, the lungs and much of the gut. Most of its fibers run the other way, upward, carrying a constant report from the organs back to the brain. Vagus nerve stimulation means putting a small electrical signal onto that nerve on purpose. It began as an implanted device for epilepsy and has since branched into several forms with very different evidence behind them.
The inflammatory reflex is why vagus stimulation is studied beyond the brain. Work led by Kevin Tracey showed that the vagus nerve is part of a loop the body uses to sense inflammation and lower it: signals traveling up the nerve register rising inflammatory molecules, and signals traveling down can lower the output of those same molecules from immune cells. This direct link from the nervous system to the immune response is why vagus stimulation is now tested in conditions outside epilepsy, including rheumatoid arthritis.
The Anatomy Of The Vagus Nerve
1From The Brainstem
The vagus begins in the medulla, low in the brainstem, where a hub called the nucleus tractus solitarius receives the sensory traffic coming up from the body. From there the nerve descends through an opening in the base of the skull and runs down the neck alongside the great vessels.
2Through The Neck And Chest
In the neck it gives off the branch that supplies the voice box, which is why stimulating the nerve can change the voice. It continues into the chest, sending fibers to the heart, where it is the main slowing influence on heart rate, and to the lungs and airways.
3To The Organs And Back
Below the chest it reaches the stomach, gut, liver and other organs. Roughly four in five of its fibers are sensory, carrying information from these organs up to the brain, so the vagus carries more information up from the organs than commands down to them, and that sensory traffic is part of the inflammatory reflex.
How It Works
An electrical pulse on the vagus does two things at once, because the nerve runs in both directions. Downward, on the parasympathetic side, it can slow the heart and shift the body toward its rest-and-recover setting. Upward, along the larger sensory bundle, it reaches the nucleus tractus solitarius in the brainstem, which relays onward to the regions that regulate mood, arousal and the release of the brain's own signaling chemicals. The antiseizure and antidepressant effects are thought to work mainly through this upward route rather than through the heart.
The immune effect runs through the inflammatory reflex. Downward vagal signaling, relayed through the spleen, prompts the release of acetylcholine, which binds receptors on immune cells and lowers their output of inflammatory molecules such as tumor necrosis factor. This is the cholinergic anti-inflammatory pathway, and it is the mechanism behind testing vagus stimulation in an inflammatory disease like rheumatoid arthritis.
There is also a route to the nerve from the surface of the ear. A small branch of the vagus supplies part of the outer ear, so a clip or electrode there can reach the same central projections without surgery. Brain imaging in people has shown that stimulating this spot activates the brainstem hub and its onward targets, which is the basis of the non-invasive ear-clip form. The everyday, gentler way into the same system is the breath: slow breathing at about six breaths a minute is the most direct way most people can raise vagal tone, the resting influence of the vagus on the heart that is read out as heart rate variability.
What Changed
For its first years vagus nerve stimulation was a single, specialized procedure: an implanted pulse generator under the collarbone with a lead coiled around the nerve in the neck, cleared in the United States in 1997 for epilepsy that had not responded to medication. A depression approval followed in 2005. Two developments widened the field. The first was the inflammatory reflex, which recast the vagus as an influence on the immune system as well as the brain. The second was the discovery that the ear branch offers a way in without surgery, which turned a hospital procedure into something a small clinic, or a consumer device maker, could study. More recently, pairing brief stimulation with physical therapy was cleared for arm recovery after stroke.
One phrase now covers an established surgical treatment for epilepsy, an active research frontier at the ear, and a range of consumer devices, and the evidence behind them is very different.
The Forms And Their Evidence
Implanted VNS For Epilepsy And Depression
The implanted device is the established form. In drug-resistant epilepsy it reduces seizure frequency by a modest amount: in its pivotal trial, high stimulation cut seizures by about 28 percent over three months against about 15 percent on low stimulation, and it is a standard option when medication and surgery are not enough. For treatment-resistant depression the evidence splits by time frame. The short controlled trial that supported approval did not clearly beat a sham device over ten weeks, which is a mark against it. The stronger signal is long-term: over years, people with the implant showed higher response and remission rates than those on usual care in a large observational comparison.
Paired VNS For Stroke Rehabilitation
A newer use pairs a brief pulse of vagus stimulation with each movement during rehabilitation, so the stimulation lands during the period when the brain forms new connections most readily. In the pivotal blinded trial, VNS-REHAB, people who received paired stimulation during arm therapy gained more movement than those who received the therapy alone, and were about twice as likely to reach a meaningful improvement. That trial was funded by MicroTransponder, the maker of the device. This use is cleared for arm recovery after ischemic stroke and sits at a moderate level of evidence, resting on that one pivotal trial.
Transcutaneous Auricular VNS, The Ear Clip
The non-invasive form, transcutaneous auricular vagus nerve stimulation or taVNS, delivers current through the skin of the ear. It is where most of the current interest sits, and where the evidence is thinnest. There are many small trials in depression, anxiety, inflammation and atrial fibrillation, and several are encouraging. Three limits run through them:
- they are small;
- they are hard to blind convincingly, because a person can feel the ear clip;
- they vary in how and where on the ear they stimulate.
In depression, an ear-clip protocol reduced symptoms in early trials, including one that found it comparable to a standard antidepressant, though these are small and the comparison was not against a dummy device. In atrial fibrillation, low-level ear stimulation lowered the burden of the arrhythmia in small randomized studies. This is a promising early field, and the evidence sits at an emerging stage.
Everyday Vagal Tone Practices
Outside any device, several practices raise vagal tone. Slow breathing near six breaths a minute is the best supported: it drives the widest, smoothest heart-rate swings and its mechanism is well understood, which is covered in full on the breathwork and HRV page. Cold exposure and the diving response also engage the vagus, and humming, chanting and gargling are often promoted on the grounds that they involve the throat muscles the vagus supplies. The breathing effect on vagal tone is well established. Going from there to humming or gargling treating a disease is where the popular claims outrun the evidence.
The Research & Studies
Everything here is based on the research we have collected and checked, sorted into groups and ordered with the strongest evidence first. Click any claim to open the studies behind it.
How it works
Signals down the vagus tell immune cells to make less TNF, the inflammatory reflex
The vagus nerve is wired into the immune system. Signals traveling down it lower how many inflammatory molecules immune cells make, which is why a nerve is being studied for inflammatory disease.
The inflammatory reflex, characterized in work led by Kevin Tracey, is a physiological loop in which the vagus nerve both senses rising inflammation through its afferent fibers and dampens it through its efferent, cholinergic anti-inflammatory pathway. Efferent vagal activity, relayed via the splenic nerve, drives release of acetylcholine that acts on alpha-7 nicotinic receptors on macrophages to reduce production of tumor necrosis factor and other cytokines. This is a foundational mechanism paper and a review of the concept rather than a clinical trial, so it establishes the pathway, not the size of any treatment effect.
The study · 1
Tracey, The inflammatory reflex · Nature 2002;420(6917):853-9
Ear-skin stimulation reaches the same brainstem vagus hub on brain scans
A branch of the vagus reaches the outer ear, so a clip there can send a signal to the same brainstem hub the surgical device targets, without any operation. Brain scans confirm the ear route activates that hub.
Functional MRI in healthy people showed that transcutaneous stimulation of the auricular concha, supplied by the auricular branch of the vagus, produced significant activation in the nucleus tractus solitarius and connected regions including the locus coeruleus, matching the central projection pattern expected from the vagus nerve. This is the anatomical and imaging basis for the non-invasive ear-clip form. It demonstrates central engagement of the pathway, which is the necessary first step, but not any clinical outcome.
The study · 1
Frangos, non-invasive access to vagus central projections via the external ear, fMRI evidence · Brain Stimul 2015;8(3):624-36
Ear-clip stimulation shifts the depression-linked default mode network on brain scans
Brain scans show the ear clip changes activity in a network that behaves abnormally in depression, which is a plausible mechanism for the mood effect. It shows the signal reaches relevant circuits, not that symptoms improve.
A functional MRI study in patients with major depressive disorder found that transcutaneous vagus nerve stimulation modulated connectivity of the default mode network, including its coupling with regions implicated in mood regulation, moving it in a direction away from the pattern typically seen in depression. This provides a mechanistic bridge between the ear route and the mood findings. As an imaging study it describes central effects on brain networks; it does not itself measure clinical improvement.
The study · 1
Fang et al., transcutaneous vagus nerve stimulation modulates default mode network in major depressive disorder · Biol Psychiatry 2016;79(4):266-73
Seizure Control
Implanted stimulation cut drug-resistant seizures about 28% versus 15% on low stimulation in its pivotal trial
For epilepsy that medication cannot control, the implanted device is an established, approved treatment. In its pivotal trial it cut seizures about 28 percent against 15 percent on low stimulation, and the benefit tends to grow over the first year or two.
Implanted vagus nerve stimulation was the first form of the therapy, cleared in the United States in 1997 as an adjunctive treatment for drug-resistant (medication-refractory) focal, or partial-onset, epilepsy. In the pivotal multicenter active-control trial, patients on high stimulation had an average 28 percent reduction in seizure frequency over three months compared with 15 percent on low stimulation (p=0.04). A 2022 systematic review and meta-analysis from the International League Against Epilepsy pooled long-term data and reported a mean seizure reduction of about 35 percent (34.7 percent), with responder rates that improve over years, and concluded that neurostimulation is an effective option for drug-resistant epilepsy. The device reduces how often seizures occur and is used alongside medication.
Who this may not transfer to:The trials enrolled adults and adolescents with focal epilepsy that had not responded to medication, so the figures describe that refractory population and the implanted device, not the non-invasive ear form.
The studies · 2
Touma et al., neurostimulation in people with drug-resistant epilepsy, systematic review and meta-analysis (ILAE Surgical Therapies Commission) · Epilepsia 2022;63(6):1314-1329
Handforth et al., VNS therapy for partial-onset seizures, a randomised active-control trial · Neurology 1998;51(1):48-55
Mood & stress
Over 10 weeks the implant did not beat a sham for depression, about 15% versus 10% response
Over its first ten weeks, the implanted device did not clearly beat a switched-off dummy device for treatment-resistant depression. This is a mark against it, and it is why the short-term evidence is weaker than the long-term picture.
The acute-phase, double-blind randomized controlled trial of implanted vagus nerve stimulation in treatment-resistant depression compared active stimulation against a sham (implanted but not delivering therapy) over 10 weeks in 235 participants. Response rates were roughly 15 percent for active and 10 percent for sham, and the difference on the primary outcome was not statistically significant. The disappointing acute result is the core of the long-running debate over the device's depression approval, which rests more on longer-term open data than on this controlled trial.
The study · 1
Rush et al., VNS for treatment-resistant depression, randomised controlled acute phase trial · Biol Psychiatry 2005;58(5):347-54
Over 5 years, 68% responded to the implant versus 41% on usual care
Over five years, people with the implant did better than those on usual care: about 68 percent responded against 41 percent, with more reaching remission. This long-term signal is the main reason the depression use stands, despite the weak short-term trial.
A large 5-year observational registry compared patients with treatment-resistant depression treated with adjunctive implanted vagus nerve stimulation against those receiving treatment as usual (roughly 795 patients across the two arms). The stimulation group showed a higher cumulative response rate (about 68 versus 41 percent) and higher remission over the follow-up. Because assignment was not randomized, the groups may have differed in ways that also affect outcome, so the size of the true effect is uncertain even though the direction is consistent.
The study · 1
Aaronson et al., 5-year observational study of VNS versus treatment as usual in treatment-resistant depression · Am J Psychiatry 2017;174(7):640-8
Ear-clip stimulation eased depression on par with the antidepressant citalopram in a 107-patient trial
The ear-clip version lowered depression scores in small early studies, and in one trial worked about as well as a common antidepressant. The results are promising but the trials are small and hard to blind.
Early trials of transcutaneous auricular vagus nerve stimulation (taVNS) in major depressive disorder reported reductions in symptom scores, first in a non-randomized controlled pilot and later in randomized work. In a 107-patient randomized comparative-effectiveness trial, taVNS given for eight weeks produced symptom improvement similar to the antidepressant citalopram, with no significant difference between the two groups; both arms improved, so a placebo contribution cannot be excluded. These are small studies, and blinding an ear-clip intervention convincingly is difficult, so expectancy is a live concern and the citalopram comparison used an active drug in place of a dummy device. The evidence is at an emerging stage, encouraging but not established.
The studies · 2
Rong et al., effect of taVNS on major depressive disorder, non-randomised controlled pilot · J Affect Disord 2016;195:172-9
Comparative effectiveness of taVNS versus citalopram for major depressive disorder, a randomised trial · Neuromodulation 2022;25(3):450-60
Heart And Vascular
Ear stimulation lowered atrial fibrillation burden about 85% over six months in a 53-person trial
Gentle stimulation of the ear cut the amount of atrial fibrillation people had in small trials, both in the moment and over six months. The effect is clear in these studies but they are small and single-center.
In a randomized acute study, low-level transcutaneous stimulation of the tragus suppressed induced atrial fibrillation and lowered inflammatory markers compared with sham. The follow-up TREAT AF randomized clinical trial (about 53 participants) delivered daily ear stimulation over six months and found the median atrial fibrillation burden about 85 percent lower than with sham (p=0.011). These are small, mostly single-center trials in paroxysmal atrial fibrillation, so the direction is encouraging while the magnitude and durability remain uncertain and unconfirmed in large multicenter work.
The studies · 2
Stavrakis et al., low-level transcutaneous electrical vagus nerve stimulation suppresses atrial fibrillation · J Am Coll Cardiol 2015;65(9):867-75
Stavrakis et al., TREAT AF, transcutaneous electrical vagus nerve stimulation to suppress atrial fibrillation, a randomised clinical trial · JACC Clin Electrophysiol 2020;6(3):282-91
Inflammatory Arthritis
In 17 patients, vagus stimulation lowered TNF and eased rheumatoid arthritis, worse when switched off
In a tiny early study, an implant that stimulated the vagus lowered inflammation and eased rheumatoid arthritis, and the arthritis worsened again when it was turned off. This is a first demonstration, not a treatment yet.
A first-in-human open-label study implanted vagus nerve stimulators in a small group of patients with rheumatoid arthritis (about 17 participants) and found that active stimulation reduced whole-blood production of tumor necrosis factor and lowered clinical disease-activity scores, with worsening during a period when the device was switched off. The study was funded by SetPoint Medical, the company developing the implant, a further reason to read a small open-label result with care. This is the clinical demonstration of the inflammatory reflex in an autoimmune disease. With no control group, a small sample and an open-label design, it is an early demonstration that motivates larger controlled trials rather than an established therapy.
The study · 1
Koopman et al., VNS inhibits cytokine production and attenuates disease severity in rheumatoid arthritis · Proc Natl Acad Sci USA 2016;113(29):8284-9
Go Deeper
- Breathwork and HRV: the free, best-evidenced way to raise vagal tone, with the protocol and the mechanism.
- Heart rate variability: the number that reads vagal tone, what it can and cannot tell you.
- Resonance and entrainment: why breathing near six breaths a minute couples to the heart the way it does.
- Depression: where implanted and ear-clip stimulation sit among the things studied for low mood.
The Chinese Medicine View
Chinese medicine has no vagus nerve, because it does not describe the body through nerves at all. What it does describe, in detail, is the territory the vagus governs, and the overlap reads best as a lens laid over the same body, not a proof of either account. The Lung governs Qi and controls respiration, 肺主氣，司呼吸, and the breath is where the practice of settling the body begins. The instruction 氣沉丹田, sink the Qi to the Dan Tian, directs the breath low into the abdomen rather than high in the chest, which is the same postural move a modern teacher gives for the slow breathing that raises vagal tone, arrived at from a completely different account of why.
The tradition holds that healthy Qi descends: Lung Qi and Stomach Qi are meant to move downward, and when they rebel upward the result is cough, breathlessness, nausea or a restless, ungrounded feeling. That descending, settling direction sits close to what the parasympathetic side of the nervous system does when the vagus quiets the heart and turns the body toward rest. The Heart is said to house the Shen, the mind and spirit, and a calm, even pulse is read as a settled Shen, a picture that sits close to what a steady, responsive heart rhythm reflects in modern terms and is drawn out on the heart rate variability page. These are two descriptive languages laid over the same body, developed independently, and neither one certifies the other. The modern anatomy does not prove the classical model literally correct, and the classical model did not predict the nerve.
Cautions For This Practice
Everything to be aware of is here, in one place. This practice suits most healthy people; a few situations call for real care.
The ear-clip form is generally well tolerated, mostly mild skin or tingling effects
A systematic review of the safety and tolerability of transcutaneous vagus nerve stimulation in humans gathered adverse events across the published studies and found the intervention generally well tolerated. The most frequently reported effects were mild and local, chiefly skin irritation or redness at the stimulation site, tingling around the ear or face, and occasional headache or dizziness, with serious events rare. Reporting quality varied between the source studies, so the review characterizes a favorable safety picture rather than a precise event rate.Redgrave et al., safety and tolerability of transcutaneous vagus nerve stimulation in humans, a systematic review
The implanted device commonly causes a hoarse voice, cough and throat sensations while stimulating
The randomized active-control epilepsy trial (199 implanted, 196 analyzed) documented the side-effect profile of the implanted device. The effects the trial itemized were voice alteration or hoarseness and shortness of breath, arising chiefly during active stimulation; cough and throat or neck sensations are also recognized effects, because the vagus supplies the larynx. They were generally mild, and voice effects in particular tended to lessen over time and could be reduced by adjusting the stimulation parameters. Serious device-related complications were uncommon in the trial.VNS for partial-onset seizures, a randomised active-control trial (Handforth et al.)
The implanted device is surgery
Placing an implanted stimulator is a minor operation and carries the ordinary risks of one, including infection and, rarely, injury to nearby structures. Once it is running, the most common effects come from the stimulation itself: a hoarse or altered voice, a cough, throat tightness and shortness of breath, usually during the seconds the device is active. These tend to be mild and often ease over the first months, and the settings can be adjusted. This is information to weigh with the specialist managing the device, not a decision to make alone.
The ear clip and the heart
For someone with a cardiac pacemaker, an implanted defibrillator or a known heart-rhythm disorder, ear-clip stimulation is worth discussing with a cardiologist first, because it acts on the same autonomic pathways that influence heart rate and rhythm. Skin irritation or a little redness at the clip is the common minor effect for everyone else.
Not a stand-in for treating serious depression
For someone in a depressive episode, and especially with any thoughts of self-harm, an ear clip bought online is not a substitute for proper care. The implanted form is used under psychiatric supervision for depression that has resisted other treatments, which is a very different setting from a consumer device. If low mood is severe or persistent, the right first step is a clinician.
Consumer device quality varies widely
The ear-clip devices sold direct to the public differ a great deal in how well they deliver and target current, and most of them have not been tested in the trials described above. A promising result from a research protocol does not automatically transfer to a given gadget, so treat marketing claims of specific benefits with care.
Start slow, be smart, read the research, and consult a professional if you have any concerns. This is here to inform your choice, not make it for you.
Common Questions
What does the vagus nerve actually do?
It runs the parasympathetic, rest-and-recover side of the nervous system, and it is a two-way line. Going down from the brainstem it slows the heart, calms the airways and helps digestion move. Going up, which is most of its fibers, it carries a steady report from the heart, lungs and gut to the brain. It is also part of the inflammatory reflex, the loop the body uses to sense inflammation and turn it down through the release of acetylcholine onto immune cells. That combination, quieting arousal and lowering inflammation, is why one nerve is studied across epilepsy, depression, stroke recovery and inflammatory disease.
Is vagus nerve stimulation FDA approved?
The implanted form is, for specific uses. It has been cleared in the United States for drug-resistant epilepsy since 1997, for treatment-resistant depression since 2005, and, paired with rehabilitation, for arm recovery after ischemic stroke since 2021. These are the established uses, delivered by a surgically placed device under specialist care. The ear-clip and other non-invasive forms are mostly research tools and consumer products rather than approved treatments for those conditions, and the everyday breathing and cold practices are not medical devices at all.
Do the ear-clip vagus devices work?
The research is early and mixed. Small trials of ear stimulation in depression, anxiety, inflammation and atrial fibrillation have produced some encouraging results, and brain imaging confirms that stimulating the ear does reach the same brainstem hub the implanted device targets. The limits are clear too: the trials are small, difficult to blind convincingly, and inconsistent in exactly where and how they stimulate, so the field sits at an emerging stage. Consumer devices vary widely in quality, and a good result from a careful research protocol does not guarantee that a particular gadget delivers the same.
Can breathing or cold exposure stimulate the vagus nerve?
Yes, measurably, and slow breathing is the clearest example. Breathing at about six breaths a minute drives the largest, smoothest swings in heart rate, which reflects a rise in vagal tone, and the mechanism behind it is well understood. Cold exposure and the diving response also engage the vagus. The reasonable claim stops there, at effects on tone and calm. Popular advice that humming or gargling treats a disease by toning the vagus is not established. The breathing effect on vagal tone is well established; the disease claims for humming and gargling are not.
Is vagus nerve stimulation safe?
It depends on the form. The implanted device carries the risks of a minor surgery plus stimulation effects such as a hoarse voice, cough and throat sensations, which are usually mild and often settle over time. The non-invasive ear clip is low risk for most people, with skin irritation the common minor complaint, and a systematic review of its use in people found it generally well tolerated. The main concern is the heart: anyone with a pacemaker, an implanted defibrillator or a rhythm disorder should check with a cardiologist before trying ear stimulation, since it acts on the pathways that influence heart rate. The full cautions are gathered in the section above.
Explore Related
Other pages this one connects to, by the evidence they share, the outcomes they touch, and the ground they cover.
All 14 sources on this page independently checked and cross-referenced.
Thomas Dehli, Founder & Editor, Sacred Lotus
Sacred Lotus has published Chinese medicine reference material since 2001. Integrative pages are held to the same standard as the herb and formula library: cite the source, grade the claim at its real strength, and say where the research has not looked. This page is educational and it is not medical advice. Last reviewed and updated August 9, 2026.
Evidence strength
How confidently the research supports a claim. Strength describes the evidence, not our endorsement.