VNS-REHAB
Vagus Nerve Stimulation Paired with Rehabilitation for Upper Limb Motor Function After Ischaemic Stroke (VNS-REHAB): A Randomised, Blinded, Pivotal, Device Trial
Clinical Question
In patients with moderate to severe chronic arm weakness (at least 9 months post-ischaemic stroke), does Vagus Nerve Stimulation (VNS) paired with rehabilitation improve upper limb motor function compared to rehabilitation paired with sham stimulation?
Study Overview
Objective
To evaluate whether vagus nerve stimulation (VNS) paired with rehabilitation improves upper limb motor function after chronic ischemic stroke.
Study Summary
VNS paired with rehabilitation led to significantly greater improvement in upper limb function compared to rehabilitation with sham stimulation in patients with moderate-to-severe arm weakness following chronic ischemic stroke.
Intervention
Patients with chronic stroke and moderate-severe upper limb impairment were randomized to receive either implanted VNS paired with task-specific rehab or sham stimulation with rehab, over 6 weeks of in-clinic therapy followed by home-based therapy.
Patients per Arm
VNS: 53, Sham: 55
Bottom Line
Vagus Nerve Stimulation (VNS) paired with intensive rehabilitation resulted in clinically meaningful and statistically significant improvements in arm impairment and function for individuals with chronic ischemic stroke compared to rehabilitation with a sham VNS device. The rate of clinically meaningful response was approximately doubled with active VNS.
Major Points
- Pivotal registration trial for VNS paired with rehabilitation in chronic stroke: 108 patients with moderate-severe arm weakness (≥9 months post-ischemic stroke) across 19 UK/US centers.
- First rigorous triple-blind, sham-controlled neuromodulation trial in stroke rehabilitation — all patients received VNS implant; active (0.8 mA, or 0.7 and 0.6 mA in two participants) vs sham (0 mA) stimulation during rehab movements.
- Primary outcome: FMA-UE score improved significantly more with active VNS (+5.0 vs +2.4 points, P=0.001) — a 2.6-point between-group difference immediately after 6 weeks of in-clinic therapy.
- Clinically meaningful response (≥6 point FMA-UE improvement) nearly doubled at 90 days: 47% VNS vs 24% sham (P=0.01) — NNT of ~4 for a meaningful motor recovery response.
- WMFT functional score significantly better with VNS at 90 days (P<0.0001) — translating impairment-level gains into real-world functional improvement.
- Favorable safety profile: 334 adverse events (163 VNS, 171 Control) in 85 (78%) participants, majority mild; no serious adverse device events; one case of vocal cord palsy (surgery-related) in the control group that resolved in 5 weeks.
- Conducted under FDA Investigational Device Exemption (#G170031) and UK MHRA (#CI/2015/0011) — the first adequately powered pivotal RCT of VNS-paired rehabilitation for chronic stroke (FDA approval status not addressed in the primary paper).
- Mechanism: VNS during movement is thought to release neuromodulators (acetylcholine, norepinephrine) that enhance synaptic plasticity, pairing neural activity with rehabilitative movements.
- All participants in the chronic phase with a mean time from stroke of over three years (VNS 3.1 ± 2.3 years, Control 3.3 ± 2.6 years) — demonstrating that neuroplasticity can be enhanced even years after stroke, challenging the conventional 'recovery plateau' concept.
- Gains sustained through 90-day follow-up (FMA-UE change VNS +5.8 vs Control +2.8, P=0.008); the paper explicitly notes long-term data from this study are not yet available.
Design
Study Type: Pivotal, randomised, triple-blind, sham-controlled device trial
Randomization: 1
Blinding: Triple-blind (participants, outcomes assessors, and treating therapists)
Enrollment Period: October 2, 2017 to September 12, 2019
Follow-up Duration: 90 days after completion of the 6-week in-clinic therapy
Centers: 19
Countries: United Kingdom, United States
Sample Size: 108
Analysis: Intention-to-treat analysis was used for all outcomes. The primary outcome was analyzed using an ANCOVA model (with treatment arm, region, and treatment-by-region interaction as factors; age and baseline FMA-UE as covariates). Secondary outcomes used logistic regression (for the responder analysis) and ANCOVA models in a hierarchical testing manner.
Inclusion Criteria
- Age ≥22 and ≤80 years
- History of unilateral supratentorial ischaemic stroke that occurred 9 months to 10 years prior to enrollment
- Moderate to severe arm impairment, defined as a Fugl-Meyer Assessment-Upper Extremity (FMA-UE) score between 20 and 50
Exclusion Criteria
- FMA-UE score outside the 20–50 range
- Insufficient voluntary finger/wrist movement to perform the required rehabilitation tasks
- Severe upper-limb sensory loss
- Non-qualifying stroke type/location (hemorrhagic, non-supratentorial, or otherwise not meeting the stroke criterion)
- Severe spasticity
- Severe depression (high Beck Depression Inventory score)
- Note: the paper states that the full inclusion and exclusion criteria are provided in the study supplement; additional exclusions listed on ClinicalTrials.gov (NCT03131960) are not detailed in the primary paper.
Arms
| Field | Control | VNS (Active VNS + Rehabilitation) |
|---|---|---|
| Intervention | All participants were implanted with a VNS device. The control group received 6 weeks of intensive in-clinic rehabilitation (3x/week for 18 sessions) paired with sham (0 mA) stimulation delivered during each movement repetition. This was followed by a 90-day home exercise program with sham stimulation. | The active group received 6 weeks of identical in-clinic rehabilitation paired with active VNS (0.8 mA — reduced to 0.7 mA and 0.6 mA in two participants for tolerability; 100 µs, 30 Hz, 0.5 sec pulses) delivered during each movement repetition. This was followed by a 90-day home exercise program with active stimulation. |
| Duration | 6 weeks in-clinic therapy + 90 days home therapy | 6 weeks in-clinic therapy + 90 days home therapy |
Outcomes
| Outcome | Type | Control | Intervention | HR / OR / RR | P-value |
|---|---|---|---|---|---|
| Change in Fugl-Meyer Assessment-Upper Extremity (FMA-UE) score from baseline to the first day after completion of the 6-week in-clinic therapy. | Primary | +2.4 points (SD 3.8) | +5.0 points (SD 4.4) | 0.001 | |
| Clinically meaningful response on FMA-UE score (≥6 point improvement) at Day 90 | Secondary | 24% | 47% | Between-group difference 24% (95% CI 6 to 41%); NNT 4.3 | 0.01 |
| Change in Wolf Motor Function Test (WMFT)-Functional Score at Day 90 | Secondary | +0.16 (SD 0.30) | +0.46 (SD 0.40) | Between-group difference 0.30 (95% CI 0.16 to 0.43) | <0.0001 |
| Change in FMA-UE score at Day 90 | Secondary | +2.8 (SD 5.2) | +5.8 (SD 6.0) | Between-group difference 2.96 (95% CI 0.83 to 5.08) | 0.008 |
| Post-hoc: WMFT-Functional clinically meaningful response (≥0.4 point change) at Day 90 | Secondary | 22% | 57% | NNT 2.8 | 0.01 |
| Total adverse events (all) | Adverse | 171 events | 163 events | 334 total in 85 (78%) participants; majority (n=242) mild | |
| Participants with implantation-related AE (possibly/probably/definitely related) | Adverse | 24/55 (55%) | 21/53 (40%) | Mostly post-operative pain | |
| Participants with device-use-related AE (possibly/probably/definitely related) | Adverse | 9/55 | 13/53 | ||
| Serious adverse event related to surgery | Adverse | 1 case of vocal cord palsy (resolved after 5 weeks) | 0 | No serious adverse device events reported in either group |
Criticisms
- Narrow eligibility (FMA-UE 20–50, no severe spasticity/sensory loss) excludes the majority of chronic stroke patients with arm weakness — the most impaired patients who need help most are excluded.
- Durability beyond 90 days not established by this trial — the paper explicitly notes long-term data from this study are not yet available.
- Small sample size (n=108) limits subgroup analyses and precision of treatment effect estimates; authors acknowledge subgroup effects cannot be reliably assessed.
- Predominantly male population (~64–65%) — results may not fully represent the female stroke population.
- Industry-sponsored by MicroTransponder Inc. (VNS device manufacturer) — inherent conflict of interest in a pivotal registration trial.
- Requires surgical implantation of a VNS device — invasive procedure with inherent risks (vocal cord palsy, infection, hardware complications) not applicable to all patients.
- Cost of VNS device, surgical implantation, and intensive 6-week rehabilitation program creates significant access barriers — likely limited to well-resourced centers.
- The sham control group still received 6 weeks of intensive rehabilitation (18 sessions) and improved (+2.4 FMA-UE points) — the incremental VNS benefit (+2.6 points) may not justify the cost and surgical risk for all patients.
- Ischemic stroke only — excludes hemorrhagic stroke and non-supratentorial stroke patients who may also benefit from neuromodulation-enhanced rehabilitation.
Funding
MicroTransponder Inc.
Based on: VNS-REHAB (The Lancet, 2021)
Authors: Prof. Jesse Dawson, MD; Prof. Charles Y. Liu, PhD; Prof. Gerard E. Francisco, ..., PhD; Prof. Teresa J Kimberley
Citation: Lancet. 2021 April 24; 397(10284): 1545-1553.
Content summarized and formatted by NeuroTrials.ai.