Practice guide

This page is intended for UK healthcare professionals. It is the operational anchor document for the clinician zone, written for the clinician who has decided to consider tVNS for a specific patient and needs the practical detail to proceed safely.

What this guide is

This guide covers the operational side of tVNS across the four EU MDR approved indications: epilepsy, depression, migraine, and Prader-Willi syndrome. Where parameters or protocols differ meaningfully between indications, the differences are flagged inline. Where they are essentially the same, they are stated once.

The guide does not duplicate the overview, which covers the mechanism, the anatomy, and the framework for thinking about vagal afferent engagement. Read the overview first if you have not done so. This document assumes that framework.

Candidate selection

Three principles cut across all four indications and frame the decision before the indication-specific criteria are applied.

The patient must have an indication that the trial evidence supports. tVNS is an adjunctive treatment with the strongest evidence in drug-resistant epilepsy. The depression, migraine, and Prader-Willi indications are regulated, but the trial evidence varies in strength. Selection should reflect the actual quality of the evidence, not the regulatory status alone.

The patient must be capable of consistent daily use. The therapeutic effect is cumulative over months. A patient who cannot reliably use the device for 2 to 4 hours daily over 20 weeks is unlikely to benefit, no matter how appropriate the indication. This makes adherence assessment a real part of selection, particularly in patients with cognitive impairment, severe depression, chaotic life circumstances, or significant carer dependency.

The patient must have realistic expectations. Response rates in the strongest trials sit at under half of participants. A patient who expects the device to replace medication, or to produce a rapid effect, will discontinue early and discontinue dissatisfied. The expectation conversation should happen before the trial starts.

Indication-specific selection

The detail of who is and is not a candidate within each indication is on the indication pages: epilepsy, depression, migraine, and Prader-Willi syndrome. The summary criteria are below.

Contraindications and cautions

The following apply across all indications.

Absolute contraindications (per the Instructions for Use)

• Pregnancy. The IFU lists pregnancy as a contraindication; do not use during pregnancy.
• Active implanted electronic device — cardiac pacemaker, implanted defibrillator, implanted vagus nerve stimulator, or cochlear implant. The IFU lists active implants as a contraindication (theoretical risk of interaction with the implant electronics).
• Cerebral shunt (for example, for hydrocephalus). Listed as a contraindication in the IFU.
• Sore or diseased skin at the cymba conchae site. Do not apply the electrode to broken or diseased skin (eczema, psoriasis, recent injury, active dermatitis); resolve before initiating.

Cautions (consult before use)

• Cardiac arrhythmia. The IFU advises that patients with cardiac arrhythmias consult their doctor to confirm suitability before use; it is a warning rather than an absolute contraindication.
• Bilateral vagotomy or other surgical alteration of vagal anatomy. Not addressed in the IFU; engage cautiously and seek specialist input.

Note: "active peptic ulcer" appears in some earlier tVNS materials as a caution based on animal data; it is not listed as a contraindication in the current tVNS E Instructions for Use.

Cautions, not contraindications

• Recent myocardial infarction within 3 months. Wait until cardiology clearance.
• Severe orthostatic hypotension. Monitor blood pressure during initial sessions.
• Severe psychiatric instability. Stabilise before beginning a 20-week trial.

Pre-treatment assessment

Before starting tVNS, a small set of baseline measures is worth taking. These are practical, not investigational, and most should already exist in the patient's record.

1. A baseline diary appropriate to the indication. Seizure diary in epilepsy (ideally 8 weeks). Headache diary in migraine. Validated mood scale (PHQ-9, MADRS, or HADS) in depression. Behaviour log in Prader-Willi syndrome. Without a baseline, the trial cannot be evaluated.
2. A medication review. tVNS is adjunctive. The existing pharmacological regimen should be stable for at least 6 to 8 weeks before starting, so that any change attributed to tVNS is not confounded by recent dose changes.
3. A cardiac history and resting ECG in patients with any cardiac history or in the elderly. Routine ECG is not required in young patients with no cardiac risk factors.
4. Confirmation that the cymba conchae site is free of skin disease.
5. A documented expectation conversation. What the patient hopes for, what the evidence supports, and the realistic time course.

Stimulation parameters

The convention has converged across the strongest trials and is consistent across the four indications, with minor variations.

Standard parameters

What "maximum tolerated, sub-pain threshold" means in practice

The intensity is titrated until the patient reports a clear, comfortable, non-painful tingling at the electrode site. The end-point is one notch below the level at which the sensation becomes uncomfortable. Patients vary substantially in tolerance, and the appropriate intensity for one patient may be uncomfortable for another. Re-titration is often needed during the first two weeks as the skin acclimatises and the perception threshold shifts.

A practical rule: if the patient cannot complete a 2-hour session at the current intensity without removing the device early, the intensity is too high. If the patient reports no perception of stimulation at all, the intensity is too low or the electrode contact is poor.

Initiation protocol

The first sessions matter for adherence. A well-managed initial week produces a patient who continues; a poorly managed one produces a patient who stops at week three.

1. First session, in clinic. Fit the electrode. Confirm correct cymba conchae placement. Demonstrate the device. Titrate intensity slowly to the comfortable suprathreshold range. Stay through one 30-minute observed session. Confirm the patient can fit the electrode independently before they leave.
2. First two weeks, at home, gentle ramp. Start with shorter sessions (1 hour, twice daily) for the first 7 days, building to the target 2 to 4 hours daily by the end of week 2. This reduces electrode-site irritation and gives the patient a tolerable on-ramp.
3. Re-titrate intensity at the week 2 review. Skin acclimatises. The intensity that felt strong on day 1 often feels mild by day 14. Re-titration is normal, not a problem.
4. Week 4 phone or email check. Adherence, tolerability, side effects, fit issues. Adjustments to side, intensity, or session structure as needed. This is the highest-risk drop-off window; a brief check-in keeps patients engaged.
5. Week 8 review. Diary review. Brief side-effect screen. No primary outcome judgement at this point.
6. Week 12 review. Some indication-specific endpoints (migraine attack frequency) are evaluable here. Epilepsy and depression are typically not yet evaluable.
7. Week 20 review. Primary outcome judgement. Decision to continue, modify, or stop.

Monitoring during the trial

Three things are worth monitoring routinely.

The diary. The single most informative document. Without it, response is guesswork. The patient should keep it; the clinician should review it at every contact.

Adherence. The companion app on the standard tVNS® device records session start, duration, and intensity. Reviewing the app log alongside the diary tells you whether a non-response is a true non-response or an adherence non-response.

Side effects. Common, mild, transient effects (electrode-site tingling, redness, occasional headache) generally resolve in the first 2 weeks and do not need intervention beyond reassurance. Persistent or escalating side effects warrant a phone call rather than a wait-and-see at the next scheduled review.

The objective biomarkers (HRV, pupillometry, EEG signatures) are useful in research and in selected complex cases. They are not required in routine practice.

Judging response

The criteria for response differ by indication and are detailed on the indication pages. The framework below is shared.

Responder. A patient who meets the indication-specific responder criterion (typically ≥50% reduction in seizure frequency for epilepsy, ≥50% reduction in headache days for migraine, a meaningful change on a validated mood scale for depression, or a documented reduction in temper-outburst frequency or severity for Prader-Willi syndrome).

Partial responder. Improvement that is meaningful to the patient and clinician but does not reach the formal responder threshold. In a regimen where tVNS is adjunctive to existing care, partial response is often clinically valuable and warrants continuation.

Non-responder. No change at the appropriate evaluation point (20 weeks for most indications), with adherence confirmed at or near 100% of prescribed daily duration. Non-response in a confirmed-adherent patient is a genuine non-response, and discontinuation is appropriate.

Adherence non-responder. No change at the appropriate evaluation point, with documented adherence below approximately 70%. This is not a treatment failure; it is a use failure. The choice is between supporting the patient toward better adherence and stopping the trial without a meaningful answer.

Continuation and de-escalation

Patients who respond at the 20-week mark are typically continued indefinitely on the same regimen. The therapeutic effect appears to require ongoing exposure; intermittent or holiday-style use has not been studied and we do not currently recommend it.

After 12 months of stable response, some clinicians taper to 2 hours daily from 4 hours, particularly where adherence is becoming a burden. There is no formal trial evidence for this approach, and it should be done with diary monitoring so that any deterioration can be identified early and the original duration restored.

For non-responders, discontinuation is straightforward. There is no withdrawal phenomenon, no taper requirement, and no documented rebound effect. Patients can stop and resume freely.

Adverse events and what to do

Serious adverse events are rare: across the published trials and real-world data cited here, the events that occur are predominantly local, mild, and transient, and a systematic review of 1,322 participants recorded only three serious events judged possibly related to tVNS (Redgrave et al., 2018). Note, however, that the Instructions for Use list a broader set of possible effects (including, uncommonly, arrhythmia, dyspnoea, and a preliminary increase in seizures or low mood), which should be reflected in patient counselling.

Common (5 to 20% of patients). Electrode-site tingling, redness, mild itching. Generally settle within 2 weeks; emollient between sessions or repositioning the electrode usually resolves them. Mild headache and occasional sleep disturbance in the first week are also common.

Uncommon (under 5%). Tinnitus, dizziness, nausea, transient sinus bradycardia. The 150-patient Yang RCT reported one episode of self-resolving sinus bradycardia in the active arm. Cease the session, allow recovery, and discuss with the patient before resuming. Most patients can continue at a slightly lower intensity.

Rare. Skin breakdown at the cymba conchae site requiring a treatment break. Ear infections that the patient associates with the device but that are not clearly device-related.

Reporting. Any device-related adverse event should be reported to Anatomical Concepts UK and submitted to the MHRA Yellow Card scheme. We will support submissions and can provide the device-specific information needed for the report.

Where this guide does not go

This document covers the framework that applies across the four primary indications. It does not cover:

• The detail of the trial evidence for each indication. That is on the indication pages and in the evidence library.
• Off-label use in stroke rehabilitation, functional gastrointestinal disease, spinal cord injury, or chronic pain. We are happy to discuss the literature on request.
• Closed-loop or research-only configurations of the device, which are not yet clinically available.
• Paediatric use. The tVNS E Instructions for Use specify an intended population of adults aged 18 and over; use in children and adolescents falls outside the device's intended use and is not covered by this guide.

Where to next

Browse the indication-specific pages

Open the evidence library

See the resources page (downloads and templates)

Book a demo or call

tVNS is a Class IIa medical device manufactured by tVNS Technologies GmbH, Germany. Distributed in the UK by Anatomical Concepts UK Ltd.