Vagus Nerve Stimulation for Migraine: From Trigeminal Pathways to Clinical Evidence
Introduction: Migraine and the Need for Non-Pharmacological Options
Migraine is the most common disabling neurological disorder worldwide, affecting over 1.16 billion people globally (Global Burden of Disease Study, 2021). It ranks as the second leading cause of years lived with disability among all human diseases and the most disabling neurological condition among children, adolescents, and adults under 60 (Prieto Peres et al., 2024). The economic burden is substantial — in the United States alone, migraine is estimated to cost over $36 billion annually in direct healthcare and lost productivity.
Current treatments include acute therapies (triptans, NSAIDs, gepants) and preventive medications (beta-blockers, antiepileptics, CGRP monoclonal antibodies). While these represent important advances, a significant proportion of patients remain inadequately treated. Approximately 40% of migraineurs who could benefit from preventive therapy do not receive it, and among those who do, adherence at one year is less than 30% — often due to side effects, inadequate efficacy, or drug interactions.
This treatment landscape has created substantial interest in neuromodulation — non-pharmacological approaches that modulate neural activity without the systemic effects of drugs. Among these, non-invasive vagus nerve stimulation (nVNS) has emerged as one of the most extensively studied, with a robust mechanistic rationale grounded in the anatomy of the vagal-trigeminal system and clinical evidence from multiple randomised controlled trials.
The Vagal-Trigeminal Connection: Why the Vagus Nerve Matters for Migraine
The vagus nerve and the trigeminal nerve — the primary pain-transmitting nerve in migraine — converge at critical brainstem nuclei. Understanding this anatomical relationship is essential to understanding how VNS may treat migraine.
Brainstem Convergence
The nucleus tractus solitarius (NTS) in the brainstem is the primary central relay station for vagal afferent fibres. It receives sensory information from the vagus nerve and distributes it to multiple brain regions involved in pain processing. The NTS sits in close anatomical proximity to the spinal trigeminal nucleus and the trigeminocervical complex (TCC) — the key relay for headache pain.
These nuclei are not merely neighbours. Direct and indirect connections between the vagal and trigeminal systems mean that modulating vagal input can influence trigeminal pain processing. The NTS receives convergent input from both vagal afferents and trigeminal nociceptors innervating the dura and meningeal blood vessels — the structures from which migraine pain originates.
Descending Pain Modulation
From the NTS, projections reach two critical neuromodulatory centres:
1. The locus coeruleus (LC) — releasing norepinephrine
2. The dorsal raphe nucleus (DRN) — releasing serotonin
Both centres send descending inhibitory projections to the trigeminocervical complex and the cerebral cortex. When VNS activates this pathway, it engages the brain's endogenous pain modulation system — the same system targeted, through different mechanisms, by triptans and other migraine therapies.
Mechanism of Action: Multiple Pathways to Migraine Relief
Suppression of Trigeminocervical Nociception
The most direct evidence for how VNS treats migraine comes from the work of Akerman, Goadsby, and colleagues. In a series of preclinical studies, they demonstrated that VNS dose-dependently inhibits acute nociceptive activation of intracranial dural-vascular trigeminal neurons — the neurons that transmit migraine pain (Akerman et al., 2017).
A single two-minute dose of VNS suppressed dural-evoked trigeminocervical neuronal firing. Two doses suppressed superior salivatory nucleus-evoked responses for up to 2.5 hours. Remarkably, the effect was independent of laterality — VNS on one side of the neck suppressed trigeminal activation on both sides — suggesting a central rather than peripheral mechanism.
In an episodic migraine model, nVNS and the triptan sumatriptan inhibited trigeminal nerve activation to a similar extent, with nVNS increasing inhibitory effects through both GABAergic and serotonergic descending pathways (Akerman et al., 2020).
Cortical Spreading Depression Suppression
Cortical spreading depression (CSD) — a slowly propagating wave of neuronal depolarisation followed by suppression — is the electrophysiological substrate of migraine aura and is thought to activate trigeminal nociceptive pathways even in migraine without aura. Suppressing CSD is therefore a rational therapeutic target.
Chen et al. (2016) provided landmark evidence that VNS significantly suppresses CSD susceptibility:
- Electrical CSD threshold was elevated by more than twofold
- Frequency of CSD during continuous potassium chloride application was reduced by approximately 40%
- CSD propagation speed was reduced by approximately 15%
- The effect was mediated exclusively through central vagal afferents — proximal vagotomy abolished the effect while distal vagotomy did not
A 2024 mechanistic study by Liu et al. elucidated the molecular pathway: VNS increases TrkB phosphorylation in the NTS through a glutamate receptor-mediated mechanism, which in turn modulates serotonergic and noradrenergic innervation to suppress CSD and cortical inflammation. This mechanism is particularly relevant to migraine with aura and helps explain why clinical trials have found preferential benefit in aura patients.
Neurotransmitter Modulation
VNS influences multiple neurotransmitter systems relevant to migraine:
- Glutamate inhibition: VNS reduces glutamate levels in the trigeminal nucleus caudalis and enhances GABA release, shifting the excitatory/inhibitory balance in the trigeminocervical complex
- Serotonin enhancement: Activation of the dorsal raphe nucleus facilitates descending serotonergic inhibition — the same pathway through which triptans exert their effects
- Norepinephrine release: Locus coeruleus activation provides noradrenergic modulation of cortical excitability and pain processing
- Anti-inflammatory effects: Through the cholinergic anti-inflammatory pathway, VNS may reduce neurogenic inflammation in the meninges
The CGRP Connection
Calcitonin gene-related peptide (CGRP) is a key mediator of migraine pathophysiology, released from trigeminal nerve endings during attacks and targeted by the newest class of migraine preventive therapies. VNS may modulate CGRP signalling indirectly through its suppression of trigeminocervical complex activation and CSD — both of which trigger CGRP release from trigeminal nerve fibres.
Clinical Evidence: The Major Trials
The gammaCore Device
gammaCore (electroCore, Inc.) is a handheld, non-invasive vagus nerve stimulator applied to the cervical (neck) region. It delivers a proprietary signal consisting of a 1 ms pulse through the skin to the cervical branch of the vagus nerve. Treatments are self-administered as two consecutive 120-second stimulations.
gammaCore is currently the only FDA-cleared non-invasive VNS device for both acute treatment and prevention of migraine in adults and adolescents.
PRESTO Trial: Acute Migraine Treatment
The PRESTO trial (Tassorelli et al., 2018) was a prospective, multicentre, randomised, double-blind, sham-controlled study of nVNS for acute migraine treatment in 248 participants with episodic migraine.
Treatment was applied within 20 minutes of pain onset, with repeat stimulation if no improvement at 15 minutes.
Results:
- Pain freedom at 120 minutes (primary endpoint): 30.4% nVNS vs. 19.7% sham (p = 0.067 — a trend but not statistically significant)
- Pain freedom at 30 minutes: 12.7% nVNS vs. 4.2% sham (p = 0.012)
- Pain freedom at 60 minutes: 21.0% nVNS vs. 10.0% sham (p = 0.023)
- Post hoc repeated-measures analysis through 30, 60, and 120 minutes: OR 2.3 (p = 0.012)
The study provided Class I evidence that nVNS increases the probability of achieving pain freedom in the first two hours of a migraine attack, with the strongest effects observed at early time points.
EVENT Trial: Chronic Migraine Prevention
The EVENT trial (Silberstein et al., 2016) was the first sham-controlled pilot study of nVNS for chronic migraine prevention, enrolling 59 participants averaging 21.5 headache days per month.
The randomised phase did not reach statistical significance (mean headache day change: -1.4 nVNS vs. -0.2 sham; p = 0.56). However, among 15 completers from the nVNS group who continued in the six-month open-label extension, mean headache days decreased by 7.9 days — an encouraging signal that justified further investigation.
PREMIUM Trial: Episodic Migraine Prevention
The PREMIUM trial (Diener et al., 2019) was the largest sham-controlled study to date, enrolling 477 participants with episodic migraine for a 12-week double-blind treatment period.
The primary endpoint was not met in the intention-to-treat population (migraine day reduction: -2.26 nVNS vs. -1.80 sham; p = 0.15). However, a critical methodological issue emerged: the sham device inadvertently stimulated the vagus nerve, producing measurable parasympathetic activity and inhibiting the trigeminal autonomic reflex. This active sham likely compressed the treatment difference.
Post hoc analysis of participants with at least 67% adherence showed significant benefits of nVNS for migraine days, headache days, and acute medication use — suggesting that the true treatment effect was masked by the compromised sham control.
PREMIUM II Trial: Refined Design
The PREMIUM II trial (Najib et al., 2022) addressed the limitations of its predecessor with a modified inactive sham device, ipsilateral stimulation (matching the predominant pain side), and an expanded population including both episodic and chronic migraine.
The trial was terminated early due to the COVID-19 pandemic, reaching approximately 60% of its planned enrolment — 336 enrolled, with 113 compliant completers.
Results:
- Primary endpoint (migraine day reduction): -3.12 nVNS vs. -2.29 sham (p = 0.23 — not significant)
- 50% responder rate: 44.87% nVNS vs. 26.81% sham (p = 0.048 — significant)
- Migraine with aura subgroup: headache days decreased by 5.5 days (nVNS) vs. 2.7 days (sham) — a therapeutic gain exceeding 100% over sham
The finding that patients with migraine with aura respond preferentially is mechanistically coherent: it aligns directly with the demonstrated ability of VNS to suppress cortical spreading depression, the electrophysiological basis of aura.
VNS for Cluster Headache
While not the primary focus of this article, the evidence for nVNS in cluster headache provides important context. The ACT1 (Silberstein et al., 2016) and ACT2 (Goadsby et al., 2018) trials demonstrated that nVNS is significantly effective for episodic cluster headache — with ACT2 showing 48% pain freedom at 15 minutes versus 6% for sham (p < 0.01) — but not for chronic cluster headache.
A pooled meta-analysis of both trials (de Coo et al., 2019) confirmed a significant interaction between treatment and cluster headache subtype, establishing nVNS as an evidence-based acute treatment for episodic cluster headache.
The Sham Control Challenge
A recurring methodological issue across nVNS trials deserves explicit discussion. Designing an effective sham (placebo) control for a device that produces a perceptible electrical sensation is inherently difficult. The PREMIUM trial demonstrated this concretely: the sham device was found to have actual parasympathetic activity, potentially generating a therapeutic effect of its own.
PREMIUM II addressed this with a modified inactive sham, but the challenge persists. The International Headache Society (IHS) has identified sham control validity as a significant limitation of the current evidence base and a factor contributing to the relatively modest treatment differences observed in primary endpoints.
This context is important for interpreting the evidence: several trials showed consistent trends favouring nVNS without reaching statistical significance — a pattern that may partly reflect an inadequately inert sham condition rather than true lack of efficacy.
Recent Guidelines and Real-World Evidence
IHS 2025 Guidelines
The International Headache Society published evidence-based guidelines in 2025 on non-invasive neuromodulation for migraine (Yuan et al., 2025). After reviewing 15 studies from six databases, the IHS issued a weak recommendation for gammaCore Sapphire for preventive migraine treatment, noting that "these devices are safe and generally well tolerated and devoid of drug interactions."
The weak (rather than strong) recommendation reflects the evidence limitations discussed above — not safety concerns or a negative assessment of efficacy.
Long-Term Real-World Data
Fernandes, Ashraf, and Goadsby (2025) published an 11-year real-world experience with nVNS at a tertiary headache centre, including 108 patients with various headache diagnoses. Median device use was 47 months, and most patients considered nVNS useful, particularly for prevention. This provides reassurance regarding long-term tolerability and patient acceptance in clinical practice.
Auricular VNS (taVNS) for Migraine
While the cervical approach (gammaCore) has the largest evidence base, transcutaneous auricular VNS (taVNS) is an emerging area of investigation.
A meta-analysis by Song et al. (2023) comparing cervical and auricular approaches found that:
- Cervical nVNS significantly impacted the 50% responder rate but did not significantly reduce migraine days
- Auricular nVNS significantly reduced migraine days (mean difference: -1.8 days) and headache intensity
Straube et al. (2015) conducted the key taVNS trial in chronic migraine, finding that 1 Hz stimulation of the cymba conchae was effective for reducing headache days over three months. However, the evidence base for taVNS in migraine remains smaller and less well-controlled than for cervical nVNS.
Comparison with CGRP Monoclonal Antibodies
The emergence of CGRP-targeting therapies (erenumab, fremanezumab, galcanezumab, eptinezumab) has transformed migraine prevention. How does nVNS compare?
| Feature | nVNS (gammaCore) | CGRP monoclonal antibodies |
|---------|------------------|---------------------------|
| Mechanism | Central: inhibits TCC, suppresses CSD, modulates NTS/LC/DRN | Peripheral: blocks CGRP or CGRP receptor |
| Administration | Self-administered, on-demand | Monthly subcutaneous injection or quarterly IV infusion |
| Acute treatment | Yes (FDA-cleared) | No (gepants serve this role) |
| 50% responder rate | ~45% in adherent patients | 40–60% across trials |
| Drug interactions | None | Minimal |
| Safety profile | Excellent; no serious device-related adverse events | Good; injection site reactions, constipation |
The key advantage of nVNS is its dual utility as both an acute and preventive treatment, its complete absence of drug interactions, and its suitability for patients who are contraindicated for or have failed pharmacological treatments. VNS is also being investigated for other chronic pain conditions, where similar anti-nociceptive mechanisms appear to be at work.
Limitations and Future Directions
Current Limitations
Mixed primary endpoint results: No migraine prevention trial has met its primary intention-to-treat endpoint (EVENT, PREMIUM, PREMIUM II all missed). While methodological factors (sham control issues, COVID-19 truncation) partially explain this, it remains a limitation.
Evidence strength: The IHS 2025 guidelines issued only "weak recommendations" due to evidence limitations, though this applies to most neuromodulation devices, not nVNS specifically.
Chronic cluster headache: nVNS shows no benefit for chronic cluster headache, despite strong efficacy in episodic cluster headache — an unexplained dissociation.
Limited head-to-head comparisons: No trials have directly compared nVNS with pharmacological migraine treatments, making it difficult to position nVNS in the treatment algorithm.
Emerging Directions
- Biomarker-guided patient selection — particularly aura status, which appears to predict preferential response
- taVNS vs. tcVNS comparative studies — to determine whether the more accessible auricular approach can match cervical efficacy
- Combination therapy — nVNS with CGRP-targeting agents, leveraging their complementary central and peripheral mechanisms
- Paediatric and adolescent populations — gammaCore is already FDA-cleared for adolescent migraine prevention
- Vestibular migraine and medication overuse headache — IHS expert consensus recommendations were developed for these under-studied populations
Summary
The scientific rationale for VNS in migraine rests on solid anatomical and physiological foundations: the vagal-trigeminal convergence at brainstem nuclei, the demonstrated ability of VNS to suppress trigeminocervical nociception and cortical spreading depression, and the engagement of descending serotonergic and noradrenergic pain modulation pathways.
Clinical evidence from multiple randomised trials demonstrates that nVNS provides meaningful acute migraine relief (particularly at early time points) and shows signals of preventive efficacy — with the strongest benefit observed in patients with migraine with aura. The responder rate in adherent patients approaches 45%, comparable to many established preventive therapies.
The evidence base has been complicated by methodological challenges — particularly sham control validity and a truncated pivotal prevention trial — that make it difficult to demonstrate the full magnitude of the treatment effect. As these issues are addressed in future studies, the role of VNS in the migraine treatment armamentarium will become increasingly well-defined.
For patients seeking a drug-free, non-invasive option with no systemic side effects and dual acute-preventive utility, nVNS represents a scientifically grounded and clinically validated approach.
---
References
Akerman, S. et al. (2017). Vagus nerve stimulation suppresses acute noxious activation of trigeminocervical neurons in animal models of primary headache. Neurobiology of Disease, 102, 96–104.
Akerman, S. et al. (2020). Inhibition of trigeminal nociception by non-invasive vagus nerve stimulation: investigating GABAergic and serotonergic pathways in a model of episodic migraine. Frontiers in Neurology, 11, 146.
Chen, S.P. et al. (2016). Vagus nerve stimulation inhibits cortical spreading depression. Pain, 157(4), 797–805.
de Coo, I.F. et al. (2019). Differential efficacy of nVNS for the acute treatment of episodic and chronic cluster headache: a meta-analysis. Cephalalgia, 39(8), 967–977.
Diener, H.C. et al. (2019). Non-invasive vagus nerve stimulation (nVNS) for the preventive treatment of episodic migraine: the PREMIUM trial. Cephalalgia, 39(12), 1475–1487.
Fernandes, C.S. et al. (2025). Neuromodulation in trigeminal autonomic cephalalgias: 11-year experience of nVNS. Cephalalgia, 45(3).
Goadsby, P.J. et al. (2018). Non-invasive vagus nerve stimulation for the acute treatment of episodic and chronic cluster headache: the ACT2 study. Cephalalgia, 38(5), 959–969.
Liu, T.T. et al. (2024). Vagus nerve stimulation inhibits cortical spreading depression via glutamate-dependent TrkB activation in the NTS. Cephalalgia, 44(2).
Najib, U. et al. (2022). Non-invasive vagus nerve stimulation for prevention of migraine: the PREMIUM II trial. Cephalalgia, 42(7), 560–569.
Prieto Peres, M.F. et al. (2024). Global burden of migraine. Cephalalgia, 44(6).
Silberstein, S.D. et al. (2016). Chronic migraine headache prevention with nVNS: the EVENT study. Neurology, 87(5), 529–538.
Silberstein, S.D. et al. (2016). Non-invasive VNS for acute treatment of cluster headache: the ACT1 study. Headache, 56(8), 1317–1332.
Song, Y. et al. (2023). Noninvasive vagus nerve stimulation for migraine: a systematic review and meta-analysis. Frontiers in Neurology, 14, 1190062.
Straube, A. et al. (2015). Treatment of chronic migraine with transcutaneous stimulation of the auricular branch of the vagal nerve. Journal of Headache and Pain, 16, 543.
Tassorelli, C. et al. (2018). Noninvasive vagus nerve stimulation as acute therapy for migraine: the PRESTO study. Neurology, 91(4), e364–e373.
Yuan, H. et al. (2025). IHS evidence-based guidelines on non-invasive neuromodulation devices for migraine. Cephalalgia, 45(5).
