Important

The information on this page is for educational purposes only. It is not medical advice and is not a substitute for evaluation, diagnosis, or treatment by a qualified healthcare professional. No physician-patient relationship is established through this website. If you are experiencing a medical or psychiatric emergency, call 911 or go to the nearest emergency room. If you are having thoughts of suicide or self-harm, call or text 988 to reach the Suicide and Crisis Lifeline.

VNS · For Patients and Families

Vagus nerve stimulation: a patient's guide

This page is for people who have tried many treatments for depression without lasting relief, and for the family members supporting them. It explains what vagus nerve stimulation (VNS) is, who it is meant for, what the surgery and the months afterward are like, and what the research honestly shows, in plain language and without a sales pitch.

Adam Hart, MD · Section Chief, Neuromodulation Program · Greenville, SC

Are you a clinician, or do you want the full detail? The companion page, Vagus Nerve Stimulation for Clinicians, has the staging framework, the complete trial results, the coverage history and every reference behind the numbers on this page.

What is VNS?

VNS stands for vagus nerve stimulation. It is a small device, similar in size to a heart pacemaker, placed under the skin of the chest. A thin wire runs under the skin to the vagus nerve in the left side of the neck. Every few minutes, around the clock, the device sends a short, gentle electrical pulse up the nerve into the brain.

The vagus nerve is one of the body's main communication lines between the organs and the brain. Most of its fibers carry information upward, and VNS uses that. The pulses first reach a relay station in the brainstem, the part of the brain that connects to the spinal cord. From there they reach the parts of the brain that regulate mood. These include the areas that make the chemical messengers norepinephrine and serotonin, the same systems that antidepressant medicines act on.

The essentials: One operation to start, usually one to two hours, and most people go home the same day. The battery is replaced in a smaller procedure every several years. Between visits the device runs on its own: nothing to charge and no daily routine, but regular check-ups so we can adjust the settings. It works slowly, over months rather than weeks, and when it works the benefit tends to build over one to two years. It is added to your other treatment, not a replacement for it. VNS has been FDA-approved for hard-to-treat depression since 2005.
R patient's right L patient's left Brain: mood circuits prefrontal cortex, amygdala, hippocampus (via the relays below) Nucleus tractus solitarius (NTS) first relay, in the medulla; passes the signal to locus coeruleus and raphe Larynx (voice box) supplied by the recurrent laryngeal branch of the vagus (dashed); stimulation contracts the left vocal fold, hence voice change Heart the right vagus supplies the heart's pacemaker (SA node), so the left nerve is the one stimulated Direction of signal about 80% of vagal fibers are sensory, so stimulation travels mostly up to the brainstem (arrows on the nerve) Jugular foramen the vagus nerve leaves the skull here Carotid sheath carotid artery (red, medial), vagus nerve (pink), jugular vein (blue, lateral) Helical electrodes coiled around the nerve at the level of the thyroid cartilage Neck incision in a skin crease on the left side Strain-relief loop and lead the lead is tunneled under the skin, over the collarbone, to the generator Chest incision 2–3 cm below the left collarbone Pulse generator in a pocket under the skin and fat of the chest wall; battery lasts several years Front view, as on a chest X-ray: the patient's left is on the viewer's right. Not to scale.

Front view of the implanted system (the patient's left is on the viewer's right, as on a chest X-ray). The surgeon makes one small cut (an incision) in a skin crease of the left neck and coils the tips of the wire, called the electrodes, around the nerve. A second small cut below the left collarbone makes room for the generator, the battery-powered unit that produces the pulses. The left nerve is used because the right one helps control the heart's rhythm. The most common side effect, a change in the voice while the device is pulsing, happens because a branch of the same nerve runs to the voice box.

The implanted system

Medical illustration of a person with a vagus nerve stimulator: a small pulse generator in the upper chest connected by a thin lead that runs up the left side of the neck toward the brainstem

Illustration of the generator in the upper chest and the wire running up the left neck. Image: "Vagus nerve stimulation" by Manu5 / Scientific Animations, via Wikimedia Commons, CC BY-SA 4.0. Unmodified.

What it looks like on X-ray

Two chest radiograph views showing an implanted vagus nerve stimulator: the pulse generator below the left collarbone and the lead coiling up the neck to the electrodes on the vagus nerve

A real implanted system on X-ray: the generator below the left collarbone and the wire coiling up to the neck. The extra loop of wire is deliberate slack so the wire is not pulled when you turn your head. This patient was treated for epilepsy; the hardware is the same one used for depression. Image: "Vagal nerve stimulator VNS bei Epilepsie 56W – CR – 001" by Hellerhoff, via Wikimedia Commons, CC BY-SA 4.0. Unmodified.

How is it different from TMS, ECT or ketamine?

You may already have had some of these. TMS (transcranial magnetic stimulation) uses a magnet held against the head in daily clinic sessions over several weeks. ECT (electroconvulsive therapy) triggers a brief, controlled seizure while you are asleep under anesthesia. Ketamine is a medicine given by infusion, and esketamine is a related nasal spray; both can act within hours to days. VNS is different in four ways:

  • It is slow. VNS usually takes three months to a year to show benefit, and often keeps improving into the second year.
  • It runs in the background. The device pulses on a fixed cycle around the clock, so there is no course of sessions to finish. You still come in for check-ups, every few months at first, so we can adjust the settings.
  • It is added on. People keep taking their medicines. In the studies, VNS was always given alongside usual treatment, not instead of it.
  • It involves surgery. One operation, usually one to two hours, rather than a series of clinic visits.

Is VNS right for me?

VNS was studied in, and is approved for, people whose depression has lasted a long time and has not responded to at least four adequate treatments. That usually means someone who has been ill for years, has tried medicines from several different families, and has often already had TMS, ketamine or ECT.

VNS may be worth discussing if:

  • Your depression has lasted two years or more, or keeps coming back
  • You have had four or more treatments at a proper dose for long enough, and they did not help enough
  • ECT helped you, but staying well requires ongoing ECT sessions that are hard to keep up
  • You can wait months for a benefit and keep coming to follow-up visits while you wait

VNS is probably not the right next step if:

  • You need to feel better quickly, for example because of a crisis or suicidal thoughts. ECT or ketamine work much faster and are better choices in that situation.
  • You have not yet tried TMS. It is non-surgical, works faster, and is usually tried first.
  • You would find it hard to attend check-ups every few months for the first year or two

VNS is not an option, or has not been studied, if:

  • The left vagus nerve was cut in a previous operation (a "vagotomy")
  • You need heat-based physical therapy treatments known as diathermy (short-wave, microwave or therapeutic ultrasound). Ordinary diagnostic ultrasound scans are fine.
  • You are under 18, or in a suicidal crisis right now
  • You have certain heart-rhythm or nervous-system conditions, or schizophrenia. The device's safety and benefit have not been established in these situations.

This list is from the manufacturer's patient safety information (linked under Helpful resources). Your evaluation covers all of it in detail.

Why "four treatments"?

Four is the line the FDA wrote into VNS's approval in 2005, and Medicare's 2019 rules and the RECOVER trial (described under Does it actually work?) use the same line. A large government-funded study called STAR*D explains why doctors count failed treatments at all. It followed thousands of people through up to four rounds of antidepressant treatment. About one in three reached remission (their symptoms mostly went away) on the first medicine, and about the same share on the second. After that, the odds fell to roughly one in seven for the third and fourth rounds, and people who needed more rounds were more likely to relapse afterward. Once two well-run treatments have failed, many specialists now call the depression "treatment-resistant" and think it is time to talk about treatments that work differently, such as TMS, esketamine and ECT. VNS sits further along that path: it was studied and approved for people with four or more failed treatments and a long-standing or recurring illness.

If ECT is keeping you well

One group is often a good fit for VNS: people who respond to ECT but need it again and again to stay well. Maintenance ECT works, but it means repeated anesthesia, memory side effects and a lot of logistics. In the five-year study described below, people who had responded to ECT in the past did especially well: 71% of them responded to VNS at some point during the study, compared with 57% of similar people who did not get VNS. ("Responded" means their depression score fell by at least half.) People whose ECT had not worked also did better with VNS (60% versus 34%), so a poor ECT result does not rule VNS out. Those are five-year figures from an observational study funded by the device maker, in which people were not assigned to groups by chance, so the comparison is suggestive rather than proof. Small case series also describe patients who were able to space out or stop maintenance ECT after getting VNS. If that is your situation, it is worth a conversation.

Getting the device

Here is what the process looks like from the first conversation to the day the device is switched on. One thing to know up front: our program does not implant VNS devices. We evaluate whether VNS is a reasonable next step, assemble the treatment history an implanting center will need, help you understand coverage, refer you to a center that implants (see Helpful resources), and can manage the device afterward.

  1. Consultation and records. We review your treatment history in detail. Because VNS is for people who have tried at least four adequate treatments, the implanting center will need documentation of what you have taken, at what dose, for how long, and what happened. Bring records if you have them; if not, we can request them.
  2. Coverage and referral. Coverage for VNS in depression is limited (see Cost and insurance). For people on Medicare, the only covered route at the moment is a research study called RECOVER, explained under Does it actually work? and Cost and insurance; whether it is taking new participants changes over time. The implanting center confirms coverage before anything is scheduled; we help you get there.
  3. The operation. At the implanting center, a surgeon trained in the procedure implants the device while you are asleep under general anesthesia. It usually takes one to two hours. There are two small incisions. One is in a skin crease on the left side of the neck, where the electrode coils are placed around the nerve. The other is about an inch below the left collarbone, where the generator sits in a pocket under the skin. Almost everyone goes home the same day; occasionally the surgeon will want you to stay overnight.
  4. Recovery. Expect some soreness at the incisions for the first few days. Your surgeon will tell you what to avoid, usually heavy lifting and strenuous exercise, while the incisions heal, typically for a week or two.
  5. Switching it on. About two to four weeks after surgery, once the incisions have healed, the device is turned on at a low setting in the clinic. Over the following visits the strength is increased gradually to a level you tolerate well. This can be done by the implanting center or, if you prefer, by our program.

What to expect afterward

The timeline

  • First weeks Getting used to the sensation in the throat during each pulse. Settings adjusted for comfort.
  • 3 to 12 months When VNS helps, the change usually starts to show somewhere in this window, and it comes on slowly. In the RECOVER trial most of the gains appeared in the last three months of the first year.
  • 1 to 2 years Benefit often keeps building. In RECOVER, roughly 3 in 10 people who had no meaningful benefit at one year had gained it by the end of the second year.
  • Long term Whoever manages your device sees you often while the settings are being adjusted in the first year or two, then about once a year once things are stable. The battery lasts several years and is replaced in a shorter operation, because the wire on the nerve stays in place.

Side effects and risks

Most side effects of the stimulation happen only while the device is actually pulsing (for example, 30 seconds every five minutes) and tend to fade for most people as they get used to it or as the settings are adjusted:

  • Voice change or hoarseness during stimulation. The most common one. A branch of the vagus nerve runs to the voice box, and the pulse makes the left vocal cord tighten briefly.
  • Cough, a tight feeling in the throat, a sore throat, or trouble swallowing during stimulation.
  • Shortness of breath with exertion during stimulation. In the RECOVER trial this was reported more often with the device on than off. It is a known VNS side effect, and the trial found no new safety problems.
  • Neck or throat discomfort during stimulation. Usually relieved by lowering the settings.

The operation itself carries risks like any surgery:

  • Pain at the incisions for the first days to weeks.
  • Infection at an incision. This is the main surgical risk. It is uncommon, and when it happens it usually shows up in the weeks after the operation; a serious infection can require the device to be removed.
  • Bleeding, or injury to nearby nerves or blood vessels. Rarely, the nerve branch to the voice box is injured during surgery, which can leave longer-lasting hoarseness or weakness of the left vocal cord.

If a side effect is bothersome, the settings can usually be turned down, and the stimulation can be switched off by the team managing your device at any time. The full list of warnings is in the manufacturer's patient safety information, linked under Helpful resources.

Watch for worsening. Depression can get worse at any time, including after a change in your medicines or in the device settings. The manufacturer's labeling asks that you be watched closely for worsening mood or thoughts of suicide, especially around those changes. If that happens, contact your treating clinician the same day, or call or text 988, or go to the nearest emergency room.

Check-ups and MRI scans

At each visit the clinician managing your device reads it with a wand held over the skin. It takes a few minutes and does not hurt. They review how you are doing and adjust the settings if needed. MRI scans are usually possible, but only under conditions that depend on your device model, the scanner and the body part being scanned, and some scans cannot be done with the wire in place. Call whoever manages your device before any MRI is scheduled: the device has to be programmed off for the scan and back on afterward, the imaging center needs your model number, and if you were given a hand-held magnet it must never go into the scanner room. The manufacturer's patient safety information lists the few other things to avoid, such as diathermy; ask us about anything you are unsure of.

Does it actually work?

This is the question that matters most. The honest answer: for some people with long-standing, hard-to-treat depression, VNS can bring a slow, lasting improvement that other treatments have not. It is not a cure, it does not work for everyone, and some of the research results fell short of the goals the researchers set. You should know about that before deciding.

The five-year study

The largest long-term study enrolled 795 people with long-standing, moderate-to-severe depression and followed them for up to five years (about half were still being seen at the five-year visit). All had failed at least four treatments; on average they had failed about eight. Roughly two-thirds got VNS added to their usual care and one-third continued usual care alone. The chart shows the share of people who were in "response" (their depression score had dropped by at least half) at each check-up:

Share of people in response at each visit over five years

0% 10% 20% 30% 40% 50% 60% 12 24 36 48 60 months after baseline VNS + usual care: 49.6% Usual care alone: 25.9% 24% 8.8% VNS + treatment as usual (n=454 at 3 mo, 250 at 60 mo) Treatment as usual (n=249 → 116)

By the end of five years, about half of the people with VNS were in response at their visit, compared with about a quarter of those on usual care alone. Counting everyone who responded at any point, 68% of the VNS group responded at some time versus 41% of the usual-care group. The two groups separated within the first three months, and the gap never closed; it was widest in the final two years. Source: results posted on ClinicalTrials.gov for the Treatment-Resistant Depression Registry (NCT00320372) and Aaronson 2017. This was an observational study: people were not assigned to the two groups by chance, so the groups may have differed in ways that affect the results. It was also funded by the device maker. Both are real limitations.

The randomized trial (RECOVER)

Medicare would not accept the five-year study on its own, so a randomized trial was run. In a randomized trial people are assigned to groups by chance, which is the fairest test of whether a treatment works. RECOVER enrolled 493 adults, and all but 15 had the device implanted. By chance, half had it switched on for a year and half had it left off. Neither the patients nor the clinicians scoring them knew who was which. These were profoundly treatment-resistant patients. On average they had failed 13 treatments (the middle value was 11). They had been in their current bout of depression for about 18 years on average (the middle value was 12). Three in four were not working.

The trial's main measurement did not show a clear difference. People with the device on were in full response (their depression score had fallen by at least half) for about 19% of the time between months 3 and 12, compared with about 16% with it off. That gap was small enough that it could have been due to chance. It was a real disappointment. But the trial also measured other things, and on several of those (its "secondary" measurements) the device-on group did better. Their own doctors rated them as at least partly improved for more of the time. Independent raters who did not know their group saw the same. On most, though not all, of the quality-of-life and daily-activity questionnaires they did better; a disability questionnaire and a self-rated overall-health scale showed no difference between the groups.

RECOVER: share of months 3 to 12 spent in response or meaningful benefit, device on vs. off

0%20%40%60% MADRS response primary endpoint, ≥50% CGI-I partial response treating clinician, ≥30% QIDS-C partial response offsite blinded rater, ≥30% Daily activities (WPAI item 6) meaningful benefit Quality of life meaningful benefit 18.9%16.3% not significant (p=0.137) 53.8%39.8% p<0.001 39.6%30.7% p=0.006 43.3%37.5% significant, Rush 2025 45.4%37.1% significant, Rush 2025 Active VNS (n=249) Sham (n=244)

The first pair of bars is the trial's main measurement. The difference there was small enough that it could have been due to chance (in research terms, it was not statistically significant). The differences in the other pairs were large enough that chance is unlikely. "Partial response" means at least a 30% drop in depression scores; in a group this ill, that is a meaningful change. The abbreviations on the chart (MADRS, CGI-I, QIDS-C, WPAI) are simply the names of the questionnaires used. Source: Conway 2025 and Rush 2025 (Brain Stimulation), with percentages as reported in Conway 2026 (American Journal of Psychiatry) and Philip & Brunoni 2025. Full detail and every p-value are on the clinician page.

Does it last?

The people whose device was on were followed for a second year, during which everyone knew the device was on, so the second-year numbers are less rigorous than the first-year ones. Of those who had a meaningful benefit at one year, about 8 in 10 still had it at two years. And roughly 3 in 10 of those who had not benefited by one year had a meaningful benefit by two years. That slow, cumulative pattern is the single most distinctive thing about VNS.

What we can say honestly about VNS

In the long-standing, severely treatment-resistant depression it is meant for, VNS gives a slow, lasting benefit to a meaningful share of people, and the benefit tends to grow rather than fade. It does not help everyone. The largest randomized trial missed its main target while meeting several secondary ones, and the studies were paid for by the company that makes the device. Full remission, meaning the symptoms mostly go away, is hard to reach in this group. In the randomized trial it was rare in both groups. In the five-year study, roughly a quarter of the people with VNS were in remission at any given visit by year five, and about four in ten reached it at least once. What the evidence supports is a realistic hope of feeling substantially better and functioning better over one to two years, for people who have run out of faster options.

Cost and insurance

Published estimates for the device and the operation range from about $25,000 to $50,000 or more, before surgeon, anesthesia and follow-up visit charges, and the generator will need paid replacement every several years. What is actually billed depends on the implanting hospital and your plan; ask the implanting center for a written estimate before anything is scheduled. Getting insurance to pay is usually the hardest practical part, because the device is expensive and most plans still do not cover it for depression:

  • Medicare pays for a new VNS implant for depression only when it is done inside a Medicare-approved research study. Right now the only such study is RECOVER (see Helpful resources). Its main group finished enrolling in 2023; a bipolar-depression group and a follow-on phase may still be enrolling, so many Medicare patients with regular (unipolar) depression currently have no covered route to a new implant. The implanting center or study site can tell you the current status. Medicare's national policy does provide for replacing the generator in people who already have a VNS device for depression when the battery runs out or the device fails; whether a specific claim is paid depends on your plan and the documentation. The device maker asked Medicare in June 2025 to reconsider the rule; as of September 2026 the rule has not changed and we are not aware of a decision.
  • Private insurance varies by plan. Most large insurers still class VNS for depression as experimental and do not cover it. Approval is sometimes won case by case. Your plan will require prior authorization, meaning written approval from the insurer before surgery can be scheduled; the implanting center handles that paperwork with you.
  • The manufacturer (LivaNova) has care coordinators who check insurance, and a Patient Access Program that helps with paperwork and prior authorization for people with commercial insurance or Tricare. It does not advertise help with the cost itself; check the company's current terms.

Questions people often ask

These are questions people commonly ask about VNS. If yours is not here, ask us. Remember that our program evaluates, refers and manages devices; the implant itself is done at a center that offers it.

Will I feel the device working?

Usually yes, at first. During each pulse most people notice a tightness in the throat or a change in their voice. For most people it is uncomfortable rather than painful, and the incisions will be sore for the first days to weeks. For most people the sensation becomes less noticeable over time, though some notice it for as long as the device is on; the settings can be adjusted if it stays bothersome. If stimulation ever hurts, tell whoever manages your device; that usually means the settings need changing.

Do I have to stop my medications?

No. VNS is approved and studied as an add-on. In the trials, VNS was added on top of people's ongoing treatment, which continued (and could be adjusted) throughout, and expert guidance is to continue medication for at least the first year after the implant.

How long before I know whether it is helping?

Longer than with medicines, TMS or ECT. Improvement, when it comes, usually appears gradually over the first year: in the five-year study the share of people in response rose from about one in four at three months to nearly four in ten at one year, and benefit often keeps building into the second year. Whoever manages your device should track your scores at each visit so that small changes are not missed; we do.

What if it does not work for me?

Depending on how benefit is measured, roughly a third to a half of people in the RECOVER trial had no meaningful improvement after one to two years of stimulation, and in the five-year study about one in three never reached a 50% improvement. If we are managing your device, we will say so plainly if that is where things stand after a fair trial of one to two years. Your other treatments continue throughout, and the remaining options should be talked through with you. The stimulation can be turned off at any time. If you want the hardware out, the generator can be removed in an outpatient operation; the electrode coils on the nerve are usually left in place, because removing them is a more delicate operation, and expert advice is that switching the device off is usually better than another surgery.

Is this the same as ECT or TMS?

No. ECT uses a brief, controlled seizure under anesthesia and works within weeks. TMS uses magnetic pulses on the outside of the head in daily clinic sessions. VNS is an implanted device that sends short pulses to a nerve in the neck every few minutes, around the clock, for years. Many people who get VNS have already had one or both of the others.

Can I have an MRI?

Usually yes, but only under conditions that depend on your device model, the scanner and the body part being scanned; some scans cannot be done with the wire in place. Call whoever manages your device before any MRI is scheduled. The device has to be programmed off for the scan and back on afterward, the imaging center needs your model number, and if you were given a hand-held magnet it must never go into the scanner room.

Will my voice change permanently?

Usually not. For most people the voice change happens only while the device is pulsing, because a branch of the vagus nerve supplies the voice box. Between pulses the voice is normal, and lowering the settings reduces the effect. Rarely, the surgery itself can injure that branch and leave lasting hoarseness or weakness of the left vocal cord; your surgeon will discuss that risk with you before the operation.

How long does the battery last, and what happens then?

Several years. Published estimates range from about six years for older models to more than ten for current ones, depending on the settings used. The battery is checked at every visit so it is replaced before it runs out. Replacing it is a shorter operation than the original implant, because the wire to the nerve stays in place and only the generator is swapped.

I already have a VNS device from another program. Can you manage it?

Usually, yes, if the device was implanted for depression. Managing VNS devices is a standard part of neuromodulation psychiatry, and we work with the manufacturer's clinical support team for adjustments. Device complications may need a surgeon. Bring your programming history if you have it; if not, we can request it from your previous provider or the manufacturer.

Helpful resources

Reputable places to learn more. Every link below was opened and checked in September 2026. Pages published by the device maker are labeled as such; they are useful, and they are also selling something, so read them with that in mind. None of these organizations pays us to be listed here, and we receive nothing for these links. Our program does not implant VNS devices; when we manage a device for a patient who has one, we use the manufacturer's field clinical support for programming, as every program does.

Patient stories reported by journalists

Programs that implant VNS for depression, and how to find others

The study through which Medicare covers VNS

  • RECOVER (NCT03887715) on ClinicalTrials.gov Government registryThe only Medicare-approved study. Listed as recruiting as of June 2026; the main randomized group finished enrolling in 2023, and a bipolar-depression group and a follow-on phase continue, so ask the site what it is currently enrolling. Entry requires: age 18 or over; a depressive episode lasting two years or more, or at least four episodes; at least four treatments that did not help enough; and other criteria the site will review. Sites within driving distance of Greenville include MUSC in Charleston (the only South Carolina site) and Emory University in Atlanta; the full list is on ClinicalTrials.gov.
  • Medicare's own page on VNS coverage GovernmentThe official statement of what Medicare covers, and under what conditions.

Official device information, manufacturer pages and support

About patient stories: we link to independently reported accounts rather than posting testimonials from our own patients. Federal advertising rules and the AMA's ethics code treat a physician's testimonial as a claim about what patients generally achieve, so it has to be accompanied by the typical result, which is what the charts above show; and privacy law requires a patient's written permission before their story is used in marketing. The clinician page explains this in detail.

Where these numbers come from

The main studies behind this page. The clinician page lists the full reference list behind both pages, including the older trials, the coverage documents and the anatomy sources.

  1. Rush AJ, et al. Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. Am J Psychiatry. 2006;163:1905–1917. PubMed. (The "why four treatments" numbers. A 2023 reanalysis argued that the remission rates were lower than reported, both at each step and overall (35% rather than 67%); the original investigators disagreed. Both analyses show the odds falling sharply after the second treatment.)
  2. Conway CR, George MS, Sackeim HA. Toward an evidence-based, operational definition of treatment-resistant depression: when enough is enough. JAMA Psychiatry. 2017;74:9–10. PubMed.
  3. Aaronson ST, et al. A 5-year observational study of patients with treatment-resistant depression treated with vagus nerve stimulation or treatment as usual: comparison of response, remission, and suicidality. Am J Psychiatry. 2017;174:640–648 (erratum 174:907). PubMed. Visit-by-visit data: ClinicalTrials.gov NCT00320372.
  4. Conway CR, et al. Clinical characteristics and treatment exposure of patients with marked treatment-resistant unipolar major depressive disorder: a RECOVER trial report. Brain Stimul. 2024;17:448–459. PubMed. (How ill the RECOVER patients were.)
  5. Conway CR, et al. Vagus nerve stimulation in treatment-resistant depression: a one-year, randomized, sham-controlled trial. Brain Stimul. 2025;18:676–689. PubMed.
  6. Rush AJ, et al. Effects of vagus nerve stimulation on daily function and quality of life in markedly treatment-resistant major depression: findings from a one-year, randomized, sham-controlled trial. Brain Stimul. 2025;18:690–700. PubMed.
  7. Conway CR, et al. Durability of the benefit of vagus nerve stimulation in markedly treatment-resistant major depression: a RECOVER trial report. Int J Neuropsychopharmacol. 2026;29(1):pyaf080. PubMed. (The two-year results; the second year was open-label.)
  8. Conway CR, et al. The RECOVER trial of vagus nerve stimulation in markedly treatment-resistant depression: critical findings, lessons learned, and future directions. Am J Psychiatry. 2026; online August 26. PubMed.
  9. McAllister-Williams RH, et al. The use of vagus nerve stimulation (VNS) in the management of patients with difficult-to-treat major depressive disorder: an expert consensus statement. Neuropsychiatr Dis Treat. 2026;22:554007. PubMed. (The "keep taking medication for at least a year" guidance and the advice that switching off is usually better than removal. Funded by the device maker.)
  10. Ardesch JJ, et al. Vagus nerve stimulation for epilepsy activates the vocal folds maximally at therapeutic levels. Epilepsy Res. 2010;89:227–231. PubMed. (Why the voice changes.)
  11. Mandalaneni K, Rayi A. Vagus Nerve Stimulator. StatPearls, updated August 2023. NCBI Bookshelf. (The surgery and the anatomy.)
  12. LivaNova. VNS Therapy for depression: brief summary of safety information for patients, February 2025. livanova.com. (Who should not have the device, and the warnings.)
  13. Centers for Medicare & Medicaid Services. Decision memo CAG-00313R2, February 15, 2019, and National Coverage Determination 160.18. cms.gov. (What Medicare covers.)

Last updated: September 2026 · Information on this page is educational and does not substitute for a conversation with your clinician. Treatment decisions should be made individually, based on your history and circumstances.