Abstract 3090: Transcranial Magnetic Stimulation (TMS) reveals Correlates of Response to Intensive Robotic Rehabilitation in Chronic Stroke

Introduction: Intensive task practice assisted by robots is a promising treatment for hemiparesis after stroke. However, not all respond to the treatment, so a better patient selection method is needed. Previous studies have shown that motor cortical responsiveness to transcranial magnetic stimulation (TMS) may predict gains from practice and may change with those gains. Hypotheses: In the first study, we hypothesized that more intact motor cortical pathways would predict better response to treatment, and that motor cortical excitability would increase when motor function improved. In the second study we had the same hypotheses, and: Motor cortical excitability would increase during a single session of intensive therapy, and that this increase would itself predict gains in motor function. Methods: Study 1: participants in trials of upper extremity robotic rehabilitation in chronic stroke were recruited to have TMS measures for extensor digitorum communis (EDC) and biceps muscles before and after the therapeutic intervention, which consisted of 6-12 wks. of intensive rehabilitation for 3 hr. each wk., using either robotics or more conventional methods. TMS measures included motor maps, threshold, recruitment curves, and ipsilateral silent period. Study 2: all participants received 12 wks. of a newly optimized robotic training, with or without functional task practice. In addition to TMS measures before and after the intervention, participants had the same measures before and after one of the first sessions of therapy using a particular robot. Results: Study 1: (N=13) TMS thresholds in the EDC was lowered in 6/8 and recruitment curve threshold increased in 5 /8 with measurable responses after treatment, particularly in one participant with an 18 point increase in upper extremity Fugl-Meyer (FM) score. MEP presence showed a non-significant trend towards prediction of functional response (3.3 higher FM gain) even when baseline FM was included in the analysis. Study 2: While this is an ongoing study and outcome data are not available, motor threshold fell in the EDC after 1 hr. of robotic therapy for one participant and an MEP was present only after a therapy session for another. Conclusion: TMS measures in multiple muscles in moderately impaired stroke patients are feasible for rehabilitation trials and may be valuable both for prediction of response to therapy as well as demonstrating mechanisms of motor function improvement.