Comparison of exercise training effect with different robotic devices for upper limb rehabilitation: a retrospective study

BACKGROUND: Several robotic devices have been proposed for upper limb rehabilitation, but they differ in terms of application fields and the technical solutions implemented. AIM: The aim of this study was to compare the effectiveness of three different robotic devices for shoulder-elbow rehabilitation in reducing motor impairment and improving motor performance in post-stroke patients. DESIGN: Retrospective multi-center study. SETTING: Inpatient rehabilitation hospital. POPULATION: Eighty-seven chronic and subacute post-stroke patients, aged 48-85 years. METHODS: Data were obtained through a retrospective analysis of patients who underwent a 3-week rehabilitation program including robot-assisted therapy of the upper limb and conventional physical therapy. Patients were divided into three groups according to the robot device used for exercise training: 'Braccio di Ferro" (BdF), InMotion2 (IMT), and MEchatronic system for MOtor recovery after Stroke (MEMOS). They were evaluated at the beginning and end of treatment using the Fugl-Meyer (FM) and Modified Ashworth (MAS) clinical scales and by a set of robot measured kinematic parameters. RESULTS: The three groups were homogeneous for age, level of impairment, time since the acute event, and spasticity level. A significant effect of time (P<0.001) was evident on FM and kinematic parameters across all groups. The average change in the FM score was 9.5, 7.3 and 7.1 points, respectively, for BdF, IMT and MEMOS. No significant between-group differences were observed at the MAS pre-vs. post-treatment. A significant interaction between time and groups resulted for the mean velocity (MV, P<0.005) and movement smoothness parameters (nPK, P<0.001 and SM, P<0.02). The effect size (ES) was large for the FM score and MV parameter, independently of the type of robot device used. Further, the ES ranged from moderate to large for the remaining kinematic parameters except for the movement accuracy (mean distance, MD), which exhibited a small ES in the BdF and MEMOS groups. CONCLUSIONS: The motor function gains obtained during robot-assisted therapy of stroke patients seem to be independent of the type of robot device used for the training program. All devices tested in this study were effective in improving the level of impairment and motor performance. CLINICAL REHABILITATION IMPACT: This study could help rehabilitation professionals to set-up comparative studies involving rehabilitation technologies.