Neural network plasticity in Parkinson's disease: the impact of exercise-driven modulation

Parkinson's disease (PD) is characterized by progressive motor impairment and large-scale network dysfunction that extends beyond dopaminergic degeneration. Although rehabilitative exercise improves clinical outcomes, its capacity to induce structural brain plasticity remains incompletely understood. Here, we tested the hypothesis that forced (high-intensity) upper-limb exercise induces white matter microstructural remodeling in PD.

Twenty-three patients with idiopathic PD (mean age 69.1 ± 6.5 years) were allocated to a forced exercise group (FE; n = 13) or a voluntary exercise group (VE; n = 10) and underwent an 8-week supervised upper-limb training protocol. Groups were matched for age, sex, and clinical severity. Assessment included diffusion tensor imaging (DTI) of cerebellar and thalamocortical tracts, computerized dynamic posturography (CDP), the motor section of the Unified Parkinson's Disease Rating Scale (UPDRS-III) and the Hoehn and Yahr scale.

Significant Group  ×  Time interactions were identified in the inferior cerebellar peduncle mean diffusivity (MD), Axial Diffusivity, Radial Diffusivity (RD); all p < 0.001 and middle cerebellar peduncle (MDP) p = 0.021), confirming that microstructural changes were significantly larger in the FE group. Results for the anterior corticothalamic radiation represent exploratory trends (p ≈ 0.06 for the interaction) and should be interpreted with caution. Findings were accompanied by significant improvements in postural stability in the FE group.

These findings identify exercise-induced plasticity within distributed motor-vestibular networks, supporting a systems-level compensatory mechanism. Our results provide in vivo evidence that targeted rehabilitation can reshape structural connectivity in PD, with implications for developing disease-modifying interventions.