A robotic system for quantitative assessment and poststroke training of forelimb retraction in mice.

Background: Neurorehabilitation protocols based on the use of robotic devices have recently shown to provide promising clinical results. However, their efficacy is still limited because of the poor comprehension of the mechanisms at the basis of functional enhancements.
Objective: To increase basic understanding of robot-mediated neurorehabilitation by performing experiments on a rodent model of stroke.
Methods: Mice were trained to pull back a handle on a robotic platform and their performances in the task were evaluated before and after a focal cortical ischemic stroke. The platform was designed for the quantitative assessment of forelimb function via a series of parameters (time needed to complete the task, t-target; average force; number of sub-movements).
Results: The animals rapidly learned the retraction task and reached asymptotic performance by the fifth session of training. Within 2 to 6 days after a small, endothelin-1-induced lesion in the caudal forelimb area, mice showed an increase in t-target and number of sub-movements and a corresponding decrease in the average force exerted. These parameters returned to baseline, pre-lesion values with continued platform training (10-14 days after stroke).
Conclusions: These results highlight the utility of the devised platform for characterizing post-infarct deficits and improvements of forelimb performance. Further research is warranted to widen the understanding of device-dependent rehabilitation effects.