Context: Reorganization in central motor networks occurs during early recovery from hemiparetic stroke. In chronic stroke survivors, specific rehabilitation therapy can improve upper extremity function. Abstract:Objective: To test the hypothesis that in patients who have chronic motor impairment following stroke, specific rehabilitation therapy that improves arm function is associated with reorganization of cortical networks. Design, Setting, and Patients: A randomized controlled clinical trial conducted in a US ambulatory rehabilitation program with 21 patients (median [IQR], 50.3 [34.8- 77.3] months after unilateral stroke). Data were collected between 2001 and 2004. Interventions: Patients were randomly assigned to bilateral arm training with rhythmic auditory cueing (BATRAC) (n=9) or standardized dose- matched therapeutic exercises (DMTE) (n=12). Both were conducted for 1 hour, 3 times a week, for 6 weeks. Main Outcome Measures: Within 2 weeks before and after the intervention, brain activation during elbow movement assessed by functional magnetic resonance imaging (fMRI) and functional outcome assessed using arm function scores. Results: Patients in the BATRAC group but not in the DMTE group increased hemispheric activation during paretic arm movement (P = .03). Changes in activation were observed in the contralesional cerebrum and ipsilesional cerebellum (P=.009). BATRAC was associated with significant increases in activation in precentral (P < .001) and postcentral gyri (P=.03) and the cerebellum (P < .001), although 3 BATRAC patients showed no fMRI changes. Considering all patients, there were no differences in functional outcome between groups. When only BATRAC patients with fMRI response were included (n=6), BATRAC improved arm function more than DMTE did (P=.02). Conclusions: These preliminary findings suggest that BATRAC induces reorganization in contralesional motor networks and provide biological plausibility for repetitive bilateral training as a potential therapy for upper extremity rehabilitation in hemiparetic stroke. Context: Reorganization in central motor networks occurs during early recovery from hemiparetic stroke. In chronic stroke survivors, specific rehabilitation therapy can improve upper extremity function. Abstract: Objective: To test the hypothesis that in patients who have chronic motor impairment following stroke, specific rehabilitation therapy that improves arm function is associated with reorganization of cortical networks. Design, Setting, and Patients: A randomized controlled clinical trial conducted in a US ambulatory rehabilitation program with 21 patients (median [IQR], 50.3 [34.8-77.3] months after unilateral stroke). Data were collected between 2001 and 2004. Interventions: Patients were randomly assigned to bilateral arm training with rhythmic auditory cueing (BATRAC) (n = 9) or standardized dose- matched therapeutic exercises (DMTE) (n = 12). Both were conducted for 1 hour, 3 times a week, for 6 weeks. Main Outcome Measures: Within 2 weeks before and after the intervention, bra in activation during elbow movement assessed by functional magnetic resonance imaging (fMRI) and functional outcome assessed using arm function scores. Results: Patients in the BATRAC group but not in the DMTE group increased hemispheric activation during paretic arm movement (P = .03). Changes in activation were observed in the contralesional cerebrum and ipsilesional cerebellum (P = .009). BATRAC was associated with significant increases in activation in precentral (P <.001) and postcentral gyri (P = .03) and the cerebellum (P < .001), although 3 BATRAC patients showed no fMRI changes. Considering all patients, there were no differences in functional outcome between groups. When only BATRAC patients with fMRI response were included (n = 6), BATRAC improved arm function more than DMTE did (P = .02). Conclusions: These preliminary findings suggest that BATRAC induces reorganization in contralesional motor networks and provide biological plausibility for repetitive bilateral training as a potential therapy for upper extremity rehabilitation in hemiparetic stroke.