Effects of prolonged robot-assisted training on upper limb motor recovery in subacute stroke

BACKGROUND: High intensity and early initiation of rehabilitation have been extensively demonstrated to enhance neural plasticity and motor recovery after stroke. However, the optimal duration of rehabilitation programs in order to have the highest impact on motor outcomes has not been established. OBJECTIVES: To evaluate motor outcomes in subacute stroke survivors with moderate to severe upper limb paresis over an extended period of rehabilitation consisting of usual care augmented with a large number of upper limb robot-assisted sessions (54 +/- 13 sessions). METHODS: Retrospective study in 10 inpatients. RESULTS: The results showed a gradual decrease in motor and functional impairments throughout the training period with a clinically meaningful increase in the Fugl-Meyer Assessment scores and in the Motor Status Scores (1st vs 37th day of the training, FMA,+ 48%, p = 0.018, MSS,+ 64%, p = 0.012; 37th vs. 79th day, FMA,+ 23%, p = 0.012, MSS,+ 30%, p = 0.017). In addition, there were improvements in hand kinematics recorded by the robot during a pointing task, with quantitative improvement (1st vs 40th day, movement efficacy,+ 97%, p = 0.0499; hand velocity,+ 335%, p = 0.013) prior to qualitative improvement (1st vs 80th day, number of hand trajectory reversals, -50%, p = 0.028; root mean square error of the trajectory/linear displacement, -52%, p = 0.059). CONCLUSIONS: Although this study was a retrospective analysis of a small sample of patients, the results suggested that a prolonged period of intensive upper limb rehabilitation, including robot-assisted training incorporated into a multidisciplinary program throughout the subacute phase after stroke resulted in significant improvements in patients with moderate to severe motor impairments.