Vagus Nerve Stimulation for Tinnitus: Reversing the Brain's Phantom Sound
Introduction: The Phantom Sound
Tinnitus — the perception of sound in the absence of an external acoustic source — affects more than 740 million people worldwide, with approximately 14% of adults experiencing the condition at some point. For roughly 2% of the global population — over 120 million people — tinnitus is severe enough to substantially interfere with daily life, affecting concentration, sleep, emotional wellbeing, and social functioning (Jarach et al., 2022).
Despite its enormous prevalence, no FDA-approved pharmacological treatment exists specifically for tinnitus. Current management strategies include cognitive behavioural therapy (CBT), tinnitus retraining therapy (TRT), sound therapy, and hearing aids. While these approaches can reduce the distress associated with tinnitus, they do not address the underlying neural pathology — they manage the condition rather than reverse it.
This distinction between management and reversal is central to understanding the scientific excitement surrounding vagus nerve stimulation (VNS) paired with sound therapy for tinnitus. Unlike existing approaches, VNS-paired tone therapy is designed to target and reverse the specific brain changes that generate the phantom sound — using the same neural plasticity mechanisms that created the problem to undo it.
The Neurobiology of Tinnitus: What Goes Wrong in the Brain
Maladaptive Auditory Cortex Reorganisation
The prevailing model of tinnitus centres on maladaptive plasticity in the central auditory system. When hearing loss occurs — typically from noise exposure, ageing, or ototoxicity — peripheral auditory neurons are damaged or destroyed, reducing input to the brain from specific frequency ranges.
The brain responds to this reduced input through a compensatory process: it increases synaptic gain in the auditory pathway, attempting to restore normal levels of evoked neural activity. However, this gain increase has an unintended consequence — it also raises spontaneous firing rates in the absence of sound. The brain, in essence, turns up the volume on silence, and the result is the phantom sound of tinnitus.
Three Pathological Hallmarks
Research by Seki, Eggermont, and Norena has identified three key pathological changes in the auditory cortex of tinnitus patients:
1. Tonotopic map reorganisation: The frequency-specific organisation of auditory cortex becomes distorted. Cortical regions that previously represented the lost frequencies become "colonised" by neighbouring frequencies, expanding their representation
2. Increased neural synchrony: Correlated firing across auditory cortex neurons increases by approximately 40% in the reorganised regions. This hypersynchrony — neurons firing together that should fire independently — creates a coherent neural signal that the brain interprets as sound
3. Increased spontaneous firing rates: Neurons in deafferented regions fire more frequently even without acoustic input, providing the "raw material" that the brain interprets as the tinnitus percept
It is important to note that not all tinnitus requires large-scale tonotopic reorganisation. In cases of normal or mild hearing loss, enhanced synchronicity across the high-frequency region alone may be sufficient to generate a tinnitus percept (Eggermont, 2012). The common thread is aberrant neural activity — activity that the auditory system misinterprets as an external sound.
The Innovation: VNS-Paired Tone Therapy
Turning Plasticity Against Itself
The insight that drives VNS-paired tone therapy is that the same neural plasticity mechanisms that create tinnitus can be harnessed to reverse it. If maladaptive plasticity reorganises the auditory cortex in a pathological direction, then targeted plasticity can reorganise it back.
The approach exploits a fundamental property of cortical plasticity: the brain preferentially strengthens neural representations of stimuli that co-occur with neuromodulatory signals — particularly norepinephrine and acetylcholine. These neuromodulators act as a biological "stamp of importance," telling the brain which sensory experiences deserve to be encoded into lasting circuit changes.
The Treatment Protocol
VNS-paired tone therapy works as follows:
1. The patient's tinnitus frequency is identified through audiometric assessment
2. A series of tones spanning the hearing range is presented, excluding the tinnitus frequency
3. Each tone is paired with a brief burst of vagus nerve stimulation
4. The concurrent release of norepinephrine and acetylcholine — triggered by VNS — "stamps in" the auditory representation of each presented frequency, expanding its cortical territory
Over repeated sessions, this process:
- Strengthens auditory neuron populations representing non-tinnitus frequencies
- Asynchronously activates neurons surrounding the hyperactive tinnitus region
- Reduces the over-expanded, synchronous representation of the tinnitus frequency
- Effectively normalises the tonotopic map
The elegance of this approach is that it does not attempt to suppress neural activity (which could impair hearing further) or mask the tinnitus signal. Instead, it reorganises the auditory cortex, restoring the balanced frequency representation that existed before the maladaptive plasticity occurred.
The Neuromodulatory Pathway
When VNS is delivered, afferent signals travel to the nucleus tractus solitarius (NTS) in the brainstem, which projects to:
1. The locus coeruleus (LC) — releasing norepinephrine
2. The nucleus basalis of Meynert (NBM) — releasing acetylcholine
Brief bursts of these neuromodulators enhance plasticity associated with coincident sensory events — in this case, the paired tones. VNS provides a safe and precise means to trigger this release with high temporal control. Both neuromodulators are independently required: noradrenergic depletion prevents VNS-driven cortical plasticity, and serotonergic depletion also blocks the effect (as demonstrated in motor cortex studies that established the mechanistic framework).
The Preclinical Breakthrough: Engineer et al. (2011)
The foundational evidence for VNS-paired tone therapy was published in Nature in January 2011 by Navzer Engineer, Michael Kilgard, and colleagues at the University of Texas at Dallas. This study remains one of the most important papers in tinnitus neuroscience.
Study Design
Rats were exposed to loud noise under anaesthesia to induce tinnitus. The presence of tinnitus was documented using a gap detection model — a behavioural paradigm exploiting the fact that tinnitus-affected animals show impaired detection of quiet gaps in a tone at their tinnitus frequency, but not at other frequencies.
VNS was then paired with interleaved multiple tones spanning the hearing range, excluding the tinnitus frequency.
Key Findings
The results were striking:
1. Noise exposure degraded the frequency tuning of auditory cortex neurons and increased cortical synchronisation — the pathological hallmarks of tinnitus
2. VNS paired with multiple tones completely eliminated both the behavioural and physiological correlates of tinnitus
3. VNS paired with a single tone was sufficient to generate specific and long-lasting changes in the cortical map
4. Improvements persisted for weeks after the end of therapy
5. The changes occurred not only in primary auditory cortex but across multiple areas within the auditory pathway
This was the first demonstration that targeted plasticity therapy could reverse the neural and behavioural hallmarks of tinnitus. It provided the preclinical foundation for all subsequent human trials and established a new therapeutic paradigm: rather than treating tinnitus as a symptom to be managed, it could be treated as a neural reorganisation problem to be reversed (Engineer et al., 2011).
Clinical Evidence: Implanted VNS Trials
De Ridder et al. (2014): First Human Case Series
Dirk De Ridder and colleagues in Belgium, working with Kilgard's group, conducted the first human study of VNS paired with tones for tinnitus. Ten patients with severe chronic tinnitus received surgically implanted VNS electrodes and underwent 2.5 hours per day of therapy for 20 days.
Four of ten patients (40%) showed clinically meaningful improvements on both the Tinnitus Handicap Inventory (THI) and minimum masking level, with improvements remaining stable for more than two months after therapy (De Ridder et al., 2014).
A critical observation emerged from this study: all five non-responders were taking medications that interfere with acetylcholine, norepinephrine, or GABA signalling — including muscarinic antagonists, norepinephrine agonists, and GABA agonists. All four responders were medication-free. This finding strongly suggests that drugs interfering with the neuromodulatory pathways through which VNS drives plasticity can block its therapeutic effect — a finding with important implications for patient selection and medication management.
Tyler et al. (2017): Randomised Controlled Pilot
Richard Tyler and colleagues conducted the first randomised, double-blind, controlled pilot study using the MicroTransponder Serenity System — a purpose-built implantable VNS device for tinnitus.
Thirty participants were randomised to receive VNS paired with tones (n=16) or VNS alone as an active control (n=14) for six weeks of home therapy. After six weeks, all participants received the paired therapy.
Results:
- After six weeks: 50% of the paired VNS group showed more than 20% improvement on the THI versus 28% of controls
- At one year: 50% of all participants had a clinically meaningful response
- Using the Tinnitus Handicap Questionnaire: 70% of patients experienced significant reduction in tinnitus severity (Tyler et al., 2017)
Vanneste et al. (2017): Neuroimaging Evidence
Sven Vanneste and De Ridder provided the first electrophysiological evidence in humans that VNS-tone pairing modulates the specific neural networks implicated in tinnitus. Using EEG, they demonstrated that VNS-tone pairing:
- Reduced gamma-band activity in the left auditory cortex — with the reduction correlating to the degree of loudness improvement
- Increased alpha synchronisation in auditory cortex, consistent with normalisation of pathological alpha/gamma activity
- Reduced phase coherence between auditory cortex and brain areas associated with tinnitus distress, including the cingulate cortex
These electrophysiological changes map directly onto the pathological neural signatures of tinnitus, providing objective evidence that VNS-tone pairing reverses the abnormal brain activity that underlies the condition (Vanneste et al., 2017).
Non-Invasive Approaches: taVNS for Tinnitus
The invasiveness and cost of implanted VNS have driven substantial interest in transcutaneous auricular VNS (taVNS) as a more accessible alternative.
Lehtimaki et al. (2013): First taVNS Pilot
The first pilot study of taVNS combined with sound therapy for tinnitus enrolled 10 subjects. Stimulation at the tragus was combined with personalised sound therapy. The study reported significant reductions in THI and Mini Tinnitus Questionnaire scores, with magnetoencephalography (MEG) showing decreased amplitude of auditory N1m responses — indicating reduced cortical excitability in the auditory cortex (Lehtimaki et al., 2013).
Shim et al. (2015): taVNS with Notched Music
Shim and colleagues explored an innovative approach: pairing taVNS with notched music therapy — music from which the tinnitus frequency had been removed, creating a spectral notch that preferentially stimulates non-tinnitus frequencies. Thirty patients received taVNS at the auricular concha during notched music listening.
After 10 sessions, 50% of patients reported symptom relief, with significant improvements in subjective tinnitus loudness (from 6.32 to 5.16 on a 10-point scale) and tinnitus awareness (from 82.4% to 65.6% of waking hours) (Shim et al., 2015).
Yakunina et al. (2018): fMRI Confirmation
Using functional MRI in 36 tinnitus patients, Yakunina and colleagues confirmed that taVNS at the inner tragus and cymba conchae activated the locus coeruleus and nucleus of the solitary tract — the same brainstem structures engaged by invasive VNS. Intriguingly, taVNS also activated the cochlear nuclei in tinnitus patients — an effect not observed in healthy subjects — suggesting tinnitus-specific brainstem engagement that may relate to the compensatory gain mechanisms underlying the condition (Yakunina et al., 2018).
Ylikoski et al. (2017, 2020): Autonomic Effects
A large retrospective study by Ylikoski et al. (2017) of 97 tinnitus patients showed that taVNS significantly shifted autonomic balance from sympathetic dominance toward parasympathetic tone, as measured by heart rate variability. At one-year follow-up in a subsequent study of 171 patients, tinnitus annoyance decreased in 72% and tinnitus-triggered stress in 82% of patients (Ylikoski et al., 2020).
These findings highlight that taVNS may benefit tinnitus patients through both targeted auditory cortex plasticity (when paired with tones) and autonomic rebalancing — the latter being relevant because tinnitus is strongly associated with sympathetic overdrive and chronic stress.
Raj-Koziak et al. (2024): Mixed Results
A controlled trial by Raj-Koziak et al. enrolled 29 patients for 12 weeks of taVNS paired with sounds. While subjective tinnitus measures did not improve significantly versus control, theta-band EEG activity increased in the taVNS group — indicating cerebral effects of stimulation even in the absence of symptomatic improvement. This dissociation between neural effects and symptomatic benefit underscores the need for optimised protocols and longer treatment durations.
Systematic Reviews and Meta-Analyses
Current Evidence Assessment
A 2021 systematic review by Stegeman et al. identified nine studies of VNS for tinnitus and concluded that, while results were heterogeneous, important methodological limitations — including considerable risk of bias, low sample sizes, and inconsistent protocols — prevented firm conclusions (Stegeman et al., 2021).
A 2023 meta-analysis by Perez et al. of non-invasive VNS trials found a significant positive effect on tinnitus disability (THI scores), with effect sizes of 0.51–0.69. However, no effect on loudness intensity was observed, and the authors noted that the clinical relevance of the effects was low given available data (Perez et al., 2023).
The scoping review by Yakunina and Nam (2021) — the most comprehensive assessment to date — concluded that while animal research has convincingly demonstrated that VNS paired with tones can reverse tinnitus, human clinical results remain mixed, and transcutaneous approaches have received increasing attention as safer, cheaper alternatives to implanted VNS.
How to Succeed — and Fail — with Paired VNS Therapy
A critical 2023 paper by Kilgard et al. — "How to Fail with Paired VNS Therapy" — catalogued the conditions under which VNS-paired therapy fails, providing essential guidance for clinical practice:
Stimulation intensity follows an inverted-U curve: Low intensities fail to sufficiently activate the neuromodulatory system, moderate intensities produce optimal effects, and high intensities overactivate and desensitise the system, paradoxically reducing efficacy.
Pharmaceutical interference is a real concern: Drugs that interfere with norepinephrine, acetylcholine, or GABA signalling can block VNS-driven plasticity — directly explaining the non-response observed in medicated patients in the De Ridder (2014) case series.
Timing between VNS-tone pairings determines efficacy: Suboptimal pairing intervals can reduce or eliminate cortical map plasticity. The temporal precision of VNS delivery relative to tone presentation is not a minor detail — it is a primary determinant of therapeutic outcome (Kilgard et al., 2023).
Comparison with Other Tinnitus Treatments
| Approach | Mechanism | Targets Neural Pathology? | Evidence Strength |
|----------|-----------|--------------------------|-------------------|
| VNS-paired tones | Reverses auditory cortex reorganisation | Yes — directly | Moderate (strong preclinical, mixed clinical) |
| CBT | Changes cognitive/emotional response | No | Strong (multiple meta-analyses) |
| TRT | Promotes habituation to tinnitus signal | Partially | Moderate |
| rTMS | Modulates auditory cortex excitability | Partially | Mixed (meta-analyses show variable results) |
| Bimodal (Lenire) | Sound + trigeminal nerve stimulation | Partially | FDA-approved; large trials |
| Sound therapy | Masking / partial masking | No | Moderate |
The distinguishing feature of VNS-paired tone therapy is its specificity: it is the only approach designed to reverse the precise neural reorganisation that generates tinnitus, leveraging endogenous neuromodulatory systems to drive controlled, targeted cortical plasticity. This same principle of targeted plasticity therapy has been applied successfully in stroke rehabilitation, where VNS paired with movement practice drives motor cortex reorganisation.
Limitations and Future Directions
Current Limitations
Small sample sizes: The largest VNS-tinnitus trial to date enrolled only 30 participants (Tyler et al., 2017). Large-scale confirmatory RCTs are urgently needed.
Invasiveness of the most effective approach: The strongest clinical evidence comes from implanted VNS, which requires surgery, carries risks (cough, hoarseness, paraesthesia), and is costly. This limits clinical adoption.
No standardised protocol: There is no consensus on optimal stimulation parameters for taVNS in tinnitus — pulse width, frequency, intensity, electrode placement, session duration, and number of sessions all remain open questions.
Medication interference: The requirement that patients be free of medications affecting norepinephrine, acetylcholine, and GABA signalling significantly limits the eligible patient population.
Tinnitus heterogeneity: Tinnitus encompasses multiple subtypes with potentially different underlying mechanisms. Not all patients may have the type of maladaptive cortical reorganisation that VNS-paired tone therapy is designed to reverse.
Future Directions
- Large-scale RCTs for both invasive VNS-tone pairing and optimised taVNS protocols
- Non-invasive translation — determining whether taVNS can match invasive VNS efficacy when properly optimised
- Patient stratification — identifying audiometric, neuroimaging, or clinical features that predict response
- Medication screening protocols — systematic management of drugs that may block plasticity
- Combination approaches — integrating VNS with CBT or sound therapy to address both neural pathology and psychological distress
- Home-based therapy — development of user-friendly taVNS-tone pairing systems for extended treatment
- Closed-loop systems — future devices that adapt stimulation parameters based on real-time neural feedback
Summary
Tinnitus represents a compelling case study in the power — and limits — of neural plasticity. The same capacity for cortical reorganisation that allows the brain to adapt to changing sensory input can, when misdirected, create a persistent phantom sound that degrades quality of life for hundreds of millions of people.
VNS-paired tone therapy attacks this problem at its source. By delivering controlled neuromodulatory signals — norepinephrine and acetylcholine — precisely timed to coincide with specific acoustic inputs, the therapy drives the auditory cortex to reorganise in a therapeutic direction, reversing the maladaptive map changes and neural synchrony that generate the tinnitus percept.
The preclinical evidence is compelling: in the foundational Nature study, VNS-paired tones completely eliminated both the behavioural and physiological correlates of tinnitus in rats. The translation to humans has been more challenging — small clinical trials show meaningful improvement in a subset of patients, but optimal protocols remain to be defined, and the field has yet to produce a large-scale confirmatory RCT.
The scientific logic is sound, the mechanistic rationale is among the strongest for any tinnitus intervention, and the early clinical results are encouraging. The critical next step is rigorous optimisation of stimulation parameters and treatment protocols — particularly for non-invasive approaches — followed by adequately powered trials. If the preclinical promise can be fully realised in clinical practice, VNS-paired tone therapy has the potential to transform tinnitus from a condition that can only be managed into one that can be reversed.
---
References
De Ridder, D. et al. (2014). Safety and efficacy of vagus nerve stimulation paired with tones for the treatment of tinnitus: a case series. Neuromodulation, 17(2), 170–179.
Eggermont, J.J. (2012). Tinnitus does not require macroscopic tonotopic map reorganization. Frontiers in Systems Neuroscience, 6, 2.
Engineer, N.D. et al. (2011). Reversing pathological neural activity using targeted plasticity. Nature, 470, 101–104.
Jarach, C.M. et al. (2022). Global prevalence and incidence of tinnitus: a systematic review and meta-analysis. JAMA Neurology, 79(9), 888–900.
Kilgard, M.P. et al. (2023). How to fail with paired VNS therapy. Brain Stimulation, 16(5), 1424–1428.
Lehtimaki, J. et al. (2013). Transcutaneous vagus nerve stimulation in tinnitus: a pilot study. Acta Otolaryngologica, 133(4), 378–382.
Perez, C.R. et al. (2023). Effects of non-invasive neuromodulation of the vagus nerve for tinnitus: a systematic review with meta-analysis. Journal of Clinical Medicine, 12(11), 3673.
Raj-Koziak, D. et al. (2024). Effectiveness of transcutaneous vagus nerve stimulation for tinnitus treatment. International Journal of Audiology, 63(4), 260–269.
Shim, H.J. et al. (2015). Feasibility and safety of transcutaneous vagus nerve stimulation paired with notched music therapy for tinnitus. Journal of Audiology and Otology, 19(3), 159–167.
Stegeman, I. et al. (2021). Tinnitus treatment by vagus nerve stimulation: a systematic review. PLoS ONE, 16(3), e0247221.
Tyler, R. et al. (2017). Vagus nerve stimulation paired with tones for tinnitus: a prospective randomised double-blind controlled pilot study. Scientific Reports, 7, 11960.
Vanneste, S. et al. (2017). Pairing sound with vagus nerve stimulation modulates cortical synchrony and phase coherence in tinnitus. Scientific Reports, 7, 17345.
Yakunina, N. et al. (2018). BOLD fMRI effects of taVNS in patients with chronic tinnitus. PLoS ONE, 13(11), e0207281.
Yakunina, N. & Nam, E.C. (2021). Direct and transcutaneous vagus nerve stimulation for treatment of tinnitus: a scoping review. Frontiers in Neuroscience, 15, 680590.
Ylikoski, J. et al. (2017). Non-invasive vagus nerve stimulation reduces sympathetic preponderance in patients with tinnitus. Acta Otolaryngologica, 137(4), 426–431.
Ylikoski, J. et al. (2020). Stress and tinnitus: transcutaneous auricular vagal nerve stimulation attenuates tinnitus-triggered stress reaction. Frontiers in Psychology, 11, 570196.
