Beta activity reflects change in upper limb activity rather than impairment following high-dose high-intensity upper limb neurorehabilitation in chronic stroke

Background: High-dose high-intensity upper limb neurorehabilitation can lead to meaningful clinical gains even in chronic stroke, yet substantial variability in recovery remains unexplained. Identifying neurophysiological markers linked to neuroplasticity and recovery could provide mechanistic insights and guide personalised rehabilitation. Objective: To characterise stroke-related alterations in {beta}-activity during movement and neural activity at rest and explore associations between brain activity and changes in upper limb clinical outcomes in chronic stroke survivors undergoing three-week high-dose rehabilitation. Methods: Electroencephalography (EEG) was recorded during the three-week rehabilitation programme in 40 chronic stroke survivors participating in the Queen Square Upper Limb (QSUL) Programme, as well as in 26 healthy controls. Recordings were taken during passive movement of the affected and unaffected index fingers (~70 movements per hand) and at rest (~7 min). Clinical assessments included the Fugl-Meyer Upper Limb Assessment (FM-UE), reflecting impairment-level deficits, and the Chedoke Arm and Hand Activity Inventory (CAHAI-13), capturing real-world upper limb activity, to examine their differential relationships with movement-related {beta}-activity. Results: Stroke survivors showed significant improvements in FM-UE and CAHAI scores following the rehabilitation programme (Mean {Delta}: FM-UE = 7.5, CAHAI = 7.4), exceeding minimum clinically important differences. Compared to controls, stroke survivors exhibited less strong {beta}-event-related desynchronization/synchronization ({beta}-ERD/ERS) during passive movement of the affected and unaffected index finger, with effects lateralised to the lesioned hemisphere. No significant differences at rest were observed between stroke participants and healthy controls. Only improvements in CAHAI, but not FM-UE, were associated with stronger {beta}-ERD (more negative) and stronger {beta}-ERS (more positive) responses during passive movement. Conclusions: Stronger movement-related {beta}-activity is associated with improvements in upper limb activity following high-dose high-intensity neurorehabilitation, suggesting {beta}-activity as a potential marker of neuroplasticity.