Intensive non-paretic arm training in chronic stroke patients with severe paresis improves functional independence without compromising paretic arm function

We previously demonstrated functionally limiting hemisphere specific motor deficits in the non-paretic, ipsilesional arm of chronic stroke patients. In a small pilot study in patients with severe paresis, we showed that non-paretic arm deficits can improve with non-paretic arm training. We now extend this study to a larger two-track cross-over design that includes both non-paretic arm training and sham-training. We ask whether non-paretic arm training can improve functional independence, without detriment to the paretic arm, and we explore the durability of these effects. This report includes only one track of our study during year 1, for which we have collected data in stroke survivors with moderate to severe paresis over a 12-week interval with 5 testing sessions that assessed non-paretic arm function, functional independence, and paretic arm impairment. After the initial test (Test 1), participants were retested (Test 2) after 3-weeks to confirm test re-test reliability and stability in baseline performance. During the following 3 weeks, participants engage in intense ipsilesional arm training for three 1.5-hour sessions per week. During training, patients engage in virtual reality (VR) games that required rapid and accurate motions of the non-paretic arm for 45 minutes. Following VR activities, the patients engage in real-life activities involving resistive exercise, and challenging use of the non-paretic arm. After a post-test (Test 3), participants engage in 3 weeks of sham training involving playing computer and board games to control for non-specific effects. They are again tested following the sham training (Test 4), and again after 3 weeks to assess durability of training (Test 5). For the non-paretic arm, our primary dependent measures are: 1) Jebsen-Taylor Hand Function Test (JTHFT), 2) The motor subscale of the Functional Independence Measure (FIM), and 3) Hand dynamometry. For the paretic arm, our primary dependent measure is the Fugl-Meyer Assessment. Our preliminary results indicate substantial improvements in response to non-paretic arm training in non-paretic arm performance (JTHFT) and functional independence (FIM), but not in general strength (dynamometry). This suggests that improvements are in coordination. Importantly, the paretic arm shows a modest, but significant reduction in impairment. Our results suggest that training of the ipsilesional arm in stroke survivors can improve non-paretic arm performance, which generalizes to improve functional independence. These improvements are durable over time, and this training is not detrimental to paretic arm function, and may slightly decrease paretic arm impairment.