We have previously shown the safety and feasibility of vagus nerve stimulation (VNS) paired with upper-limb rehabilitation after ischemic stroke. In this exploratory study, we assessed whether clinical and brain MRI variables predict response to treatment. We used data from two completed randomised and blinded clinical trials (N = 35). All participants had moderate to severe upper-limb weakness and were randomised to 6-weeks intensive physiotherapy with or without VNS. Participants had 3 T brain MRI at baseline. The primary outcome was change in Fugl-Meyer Assessment, upper-extremity score (FMA-UE) from baseline to the first day after therapy completion. We used general linear regression to identify clinical and brain MRI predictors of change in FMA-UE. VNS-treated participants had greater improvement in FMA-UE at day-1 post therapy than controls (8.63 ± 5.02 versus 3.79 ± 5.04 points, t = 2.83, Cohen’s d = 0.96, P = 0.008). Higher cerebrospinal fluid volume was associated with less improvement in FMA-UE in the control but not VNS group. This was also true for white matter hyperintensity volume but not after removal of an outlying participant from the control group. Responders in the VNS group had more severe arm impairment at baseline than responders to control. A phase III trial is now underway to formally determine whether VNS improves outcomes and will explore whether these differ in people with more severe baseline upper-limb disability and cerebrovascular disease.
Vagus nerve stimulation (VNS) paired with upper-limb rehabilitation is a potential novel treatment for arm weakness after stroke. VNS triggers release of plasticity promoting neuromodulators, such as acetylcholine and norepinephrine, throughout the cortex1. Timing this with motor training drives task-specific plasticity in the motor cortex2 and VNS paired with rehabilitative training has been shown to improve recovery in different preclinical models of stroke, both in comparison to VNS alone and rehabilitation alone3,4. These improvements were associated with synaptic reorganization of cortical motor networks and recruitment of residual motor neurons controlling the impaired forelimb5. Two clinical studies comparing VNS paired with upper-limb rehabilitation with upper-limb rehabilitation alone have shown it to be acceptably safe and feasible and that it may improve arm weakness after ischemic stroke6,7.
Arm weakness is the most common symptom of stroke and approximately half of stroke survivors with arm weakness have prolonged disability, which is associated with reduced quality of life8,9. Restoration of arm function after stroke is a priority for many stroke survivors10. However, recovery of motor function after stroke varies, so identifying factors that help predict response is important to aid patient selection and identify those most likely to respond. This is particularly true where therapies are invasive (involving surgery) and/or time consuming; VNS requires implantation of a nerve stimulator which is costly and associated with risks of anaesthesia, a small risk of infection and small risk of vocal cord palsy. There are several clinical and brain imaging markers that predict cognitive and functional recovery after stroke including age, level of impairment, white matter hyperintensity (WMH) volume, stroke lesion volume, corticospinal tract damage and blood pressure level11,12,13.
In the present study, we combined clinical and brain magnetic resonance imaging (MRI) data from our two previous randomised trials of VNS paired with rehabilitation for the upper-limb after ischemic stroke6,7. We performed exploratory analyses to assess predictors of response to VNS paired with upper-limb rehabilitation. Our goal was to identify predictive factors for further study that may help with patient selection for this promising novel therapy. […]