Upper limb motor rehabilitation impacts white matter microstructure in multiple sclerosis

Upper limb impairments can occur in patients with multiple sclerosis, affecting daily living activities; however there is at present no definite agreement on the best rehabilitation treatment strategy to pursue. Moreover, motor training has been shown to induce changes in white matter architecture in healthy subjects. This study aimed at evaluating the motor behavioral and white matter microstructural changes following a 2-month upper limb motor rehabilitation treatment based on task-oriented exercises in patients with multiple sclerosis. Thirty patients (18 females and 12 males; age = 43.3 +/- 8.7 years) in a stable phase of the disease presenting with mild or moderate upper limb sensorimotor deficits were randomized into two groups of 15 patients each. Both groups underwent twenty 1-hour treatment sessions, three times a week. The "treatment group" received an active motor rehabilitation treatment, based on voluntary exercises including task-oriented exercises, while the "control group" underwent passive mobilization of the shoulder, elbow, wrist and fingers. Before and after the rehabilitation protocols, motor performance was evaluated in all patients with standard tests. Additionally, finger motor performance accuracy was assessed by an engineered glove. In the same sessions, every patient underwent diffusion tensor imaging to obtain parametric maps of fractional anisotropy, mean diffusivity, axial diffusivity, and radial diffusivity. The mean value of each parameter was separately calculated within regions of interest including the fiber bundles connecting brain areas involved in voluntary movement control: the corpus callosum, the corticospinal tracts and the superior longitudinal fasciculi. The two rehabilitation protocols induced similar effects on unimanual motor performance, but the bimanual coordination task revealed that the residual coordination abilities were maintained in the treated patients while they significantly worsened in the control group (p = 0.002). Further, in the treatment group white matter integrity in the corpus callosum and corticospinal tracts was preserved while a microstructural integrity worsening was found in the control group (fractional anisotropy of the corpus callosum and corticospinal tracts: p = 0.033 and p = 0.022; radial diffusivity of the corpus callosum and corticospinal tracts: p = 0.004 and p = 0.008). Conversely, a significant increase of radial diffusivity was observed in the superior longitudinal fasciculi in both groups (p = 0.02), indicating lack of treatment effects on this structure, showing damage progression likely due to a demyelination process. All these findings indicate the importance of administering, when possible, a rehabilitation treatment consisting of voluntary movements. We also demonstrated that the beneficial effects of a rehabilitation treatment are task-dependent and selective in their target; this becomes crucial towards the implementation of tailored rehabilitative approaches.