Vagus nerve stimulation for tinnitus is still experimental: small implant trials found some people improved, but the randomised pilot found no significant difference between groups, and non-invasive versions have shown limited benefit or none.
It is often mentioned alongside bimodal devices such as Lenire, because both pair sound with a second signal to the nervous system. It is a separate line of research with its own history, and worth understanding on its own terms.
The core idea
The vagus nerve carries signals between the brain and much of the body. Stimulating it through a surgically implanted device is an FDA-approved procedure for epilepsy and depression, and it is thought to trigger the release of chemical messengers in the brain that make it more open to change.
For tinnitus, researchers found in animal studies that pairing brief vagus nerve stimulation with specific tones — deliberately excluding the frequency matching the animal's tinnitus — could drive targeted rewiring in the auditory cortex and reduce the abnormal neural activity associated with tinnitus. The theory is that this "paired" stimulation strengthens the brain's response to non-tinnitus frequencies relative to the overactive tinnitus frequency, gradually correcting the imbalance thought to underlie the phantom sound.
Two cautions come with that account, both from a 2021 scoping review of the field. The animal studies replicating the original finding all come from the same group of authors, and no independent team has reproduced them. And the theory rests on a model of tinnitus — reorganisation of the auditory cortex's frequency map — that the review describes as strongly criticised, partly because tinnitus can begin within seconds of noise exposure and can develop without extensive hearing loss.
Implanted VNS: what the trials found
Implanted (direct) VNS involves surgically placing an electrode on the vagus nerve in the neck, connected to a small implanted pulse generator — the same general approach used for epilepsy.
The first case series. In 2014, ten people with severe chronic tinnitus were implanted and heard tones paired with brief nerve stimulation for two and a half hours a day over 20 days. Four of the ten improved in a clinically meaningful way, on both the Tinnitus Handicap Inventory and a measure of the sound itself, and the improvement lasted more than two months. The five patients taking certain medications that may interfere with the nerve's chemical signalling did not improve.
The randomised pilot. In 2017, 30 implanted people were randomised to paired stimulation (16) or a control version (14) for six weeks of home therapy, after which everyone received paired stimulation.
- The paired group's Tinnitus Handicap Inventory scores improved significantly from where they started; the control group's did not.
- The difference between the two groups, 10.3%, was not statistically significant. That is the comparison a randomised trial exists to make.
- 50% of the paired group had a clinically meaningful improvement at six weeks, against 28% of controls. At one year, after everyone had received paired stimulation, 50% of participants had a clinically meaningful response.
- The benefit was larger for people with tonal tinnitus not caused by blast exposure — a subgroup finding, and a hint that the approach might suit some tinnitus types better than others.
- Two of the 30 had vocal-cord paralysis after the implant surgery, with hoarseness or voice weakness lasting more than 12 weeks, and in one it did not fully recover. One broken lead needed replacement surgery. Other side effects were mild, and the authors describe a safety profile similar to VNS for epilepsy.
- Several authors were employees of, or a paid consultant to, the company making the device.
The authors called it a pilot that provides impetus for a larger pivotal study, which is an accurate description. It is a reason to keep testing the idea, not evidence that the implant works.
Non-invasive VNS: what the trials found
Transcutaneous VNS stimulates a branch of the vagus nerve through the skin, usually at the ear, without surgery. It is the version most likely to be offered outside a research implant programme, and its results so far are modest at best.
- A 2023 meta-analysis found five clinical trials and pooled four. Stimulation produced a significant effect on tinnitus handicap after treatment compared with a comparison group, and no effect on loudness. The authors judged the clinical relevance low and concluded that no firm conclusions are available.
- A 2024 controlled study of 29 adults, not randomised, paired transcutaneous stimulation with sound for 12 weeks and found no overall improvement against controls. The authors' conclusion is plain: it was not effective in reducing tinnitus symptoms in their group.
- A 2026 randomised feasibility trial of 30 people with pure-tone tinnitus tested three ways of timing non-invasive stimulation against tinnitus-matched sound over nine sessions in three weeks. It was safe, but found no significant differences between groups in tinnitus loudness or the Tinnitus Functional Index. Reductions in how unpleasant the sound felt, anxiety and sleep problems in one group were exploratory findings that need a properly powered trial.
A 2026 meta-analysis of neuromodulation trials in The Laryngoscope summarises VNS in one line: modest benefits, with safety concerns limited to the implanted approach. Bimodal stimulation was the modality with the most consistent and durable results in that analysis.
How this relates to Lenire
Lenire does not use vagus nerve stimulation; it pairs sound with electrical stimulation of the tongue. What the two share is the broad idea that combining sound with a second, simultaneous signal to the nervous system can do more than sound alone. They are separate lines of research with separate trials, and a result for one says nothing about the other.
Where this stands today
Implanted VNS for tinnitus remains investigational, and it carries the risks of any surgical implant — a very different risk profile from sound therapy or CBT. The 2021 scoping review went further than most: it found no reliable study of either implanted or non-invasive VNS for tinnitus, because every study it reviewed had major flaws such as no control group, small samples, or no randomisation or blinding. It argued that the non-invasive version should be firmly established before implants are considered for tinnitus at all.
The newer trials have not changed that picture much. This is a research direction with a plausible mechanism and some encouraging individual results, not a treatment to seek out. If you are interested, the route is a registered clinical trial, and how to join a tinnitus clinical trial explains how to find one.