Vagus nerve stimulation paired with rehabilitation shows uncertain effects on upper extremity function after stroke

Rationale After a stroke, many people experience persistent upper extremity (UE) motor impairments (that is, deficits in strength, coordination, and muscle tone), which can limit activities such as reaching, grasping, and manipulation. Despite conventional rehabilitation, recovery of UE function often remains limited. Vagus nerve stimulation (VNS), applied invasively or non‐invasively (transcutaneously), has been proposed as an adjunct to rehabilitation to enhance neuroplasticity and improve motor outcomes after stroke. While early trials have reported mixed findings, the overall effectiveness and safety of VNS interventions in this context are uncertain. Objectives To assess the potential benefits and harms of vagus nerve stimulation (VNS) as an add‐on treatment to rehabilitate people who have post‐stroke UE motor function impairments and activity limitations. Search methods We searched for published trials in the Cochrane Library, MEDLINE, Embase, Scopus, PsycINFO, CINAHL and PEDro. We also handsearched for reference lists and looked for other relevant studies on Google Scholar. We performed the searches up to May 2025. Eligibility criteria We included randomised controlled trials (RCTs) that assessed the effect of VNS paired with conventional rehabilitation, compared to conventional rehabilitation alone (or paired with sham VNS) in adults (≥ 18 years) with a first‐ever ischaemic or haemorrhagic stroke, at any post‐stroke phase. Outcomes Our critical outcomes were UE motor function and the number of participants with at least one serious adverse event (SAE). Our main important outcomes were UE activity and quality of life. Risk of bias Two review authors independently screened references, selected studies based on eligibility criteria, extracted data, and assessed the risk of bias in each included study using the Cochrane RoB 2 tool. Synthesis methods When the studies were sufficiently similar, we synthesised VNS benefits and harms using a fixed‐effects model meta‐analysis for dichotomous outcomes and a random‐effects model for continuous outcomes. Included studies We included 10 studies with a total of 547 participants that met our inclusion criteria. The included studies took place in China, the UK, the USA, and Italy. All studies were published in peer‐reviewed journals. Nine were written in English, and one in Chinese. Three studies applied VNS invasively and seven non‐invasively (transcutaneously). All interventions were paired with rehabilitation (motor training) and compared with rehabilitation alone. All studies included both men and women. Most treatments were given in outpatient hospital services or health centres. All studies measured outcomes at short‐term (six to 12 weeks after treatment), three studies at medium‐term (six months after treatment), and one study at long‐term (12 months after treatment). We also identified 23 ongoing studies and 14 studies are awaiting classification. Synthesis of results Compared with conventional rehabilitation alone, the evidence is very uncertain about the effect of VNS paired with rehabilitation on UE motor function in the short‐term (SMD 1.22, 95% CI 0.68 to 1.77; 10 studies, 499 participants; very low‐certainty evidence). VNS paired with rehabilitation may result in little to no increased risk of SAEs compared to rehabilitation alone (RR 2.38, 95% CI 0.77 to 7.30; 8 studies, 416 participants; low‐certainty evidence). The evidence is very uncertain about the effect of VNS plus rehabilitation on UE activity (SMD 0.88, 95% CI ‐0.09 to 1.86; 6 studies; 186 participants; very low‐certainty of evidence) and quality of life (SMD 0.04, 95% CI ‐0.30 to 0.37; 3 studies, 180 participants; very low‐certainty of evidence) compared to rehabilitation alone in the short‐term. For the critical and main important outcomes, most studies had an overall high risk of bias. Few included studies took long‐term follow‐up measurements. Authors' conclusions Comparing VNS paired with rehabilitation to rehabilitation alone, we found very uncertain evidence for the effects on UE motor function, UE activity, and quality of life in the short‐term. VNS may result in little to no increased risk of SAEs. The body of evidence on the benefits and harms of VNS is limited by the risk of bias in the included studies, the small number of included studies, and the relatively small sample sizes. Funding This Cochrane review did not receive any specific funding. Registration Protocol (2024) DOI: 10.1002/14651858.CD015859 PICOs PICOs Population Intervention Comparison Outcome