Upper limb rehabilitation after stroke: constraint versus intensive training. A longitudinal case-control study correlating motor performance with fMRI data

Background The reproducible beneficial effect of constraint-induced movement therapy (CIMT) in hemiparetic stroke patients makes it a good model to study brain plasticity during rehabilitation procedures. Objective Assess the functional brain reorganization induced by each of the two components of CIMT: (i) non-affected upper-limb constraint and (ii) intensive training of the paretic arm. Methods Brain activity of a right hemiparetic chronic stroke patient and of 10 healthy controls was recorded with a functional magnetic resonance imaging (fMRI) during a finger opposition task. For the patient, a total of 8 assessments were performed, before and after each component of CIMT. At each time point, brain activity during movement was compared with rest. Patient’s results were first compared to the control group and then correlated to motor performance across sessions. Results Constraint-therapy-related improvement was correlated with a decrease of cerebral activity in sensory-motor regions of both the affected and the non-affected hemispheres. Intensive-therapy-related improvement was correlated with the recruitment of pre-motor cortices and cerebellum in both hemispheres. Conclusions Two different patterns of brain activity underlie the effects of intensive training and constraint which could account for the respective effect of each component of the therapy.