Differential Neuronal Network Remodeling Induced by Passive and Assistive Lower-Limb Exoskeleton Robot Training in Stroke: A Randomized Controlled Trial.

OBJECTIVES

To investigate differences in neuronal network remodeling induced by passive and assistive therapies of lower-limb exoskeleton robots training and their respective impacts on clinical outcomes in stroke rehabilitation.

DESIGN

A double-blind randomized controlled trial.

SETTING

Rehabilitation medicine department in a tertiary hospital.

PARTICIPANTS

Fifty-three (N=53) adult patients with lower-limb motor dysfunction after the first stroke.

INTERVENTIONS: Patients were randomly allocated to either a passive therapy group (PG, n=24) or assistive therapy group (AG, n=29). Both groups received lower-limb exoskeleton robot training for 20 minutes, twice daily, over a 14-day intervention period.

MAIN OUTCOME MEASURES: Primary outcomes included wavelet phase coherence (WPCO) and information flow (IF), both derived from wavelet transform of hemodynamic signals measured by functional near-infrared spectroscopy, to characterize neuro-specific changes in brain network connectivity. Secondary outcomes comprised the Fugl-Meyer Assessment for the Lower Extremity, Modified Barthel Index, Berg Balance Scale, and Functional Ambulation Category Scale.

RESULTS

Passive therapy significantly increased WPCO value in ipsilesional primary motor cortex (P=.02) and reduced intrahemispheric IF. In contrast, assistive therapy significantly increased WPCO in occipital cortex (P=.047) and enhanced IF within the contralesional hemisphere. Correlation analysis indicated that the 14-day improvement in Fugl-Meyer Assessment for the Lower Extremity scores were positively correlated with WPCO changes in the PG (r=.560; P=.016), but negatively correlated in the AG (r=-.542; P=.008).

CONCLUSIONS

These findings provide evidence that passive and assistive therapies produce different neurovascular coupling responses that directly affect clinical outcomes in patients with stroke. Direct assistive therapy may induce compensatory neural activity, which could influence the effectiveness of rehabilitation. In contrast, passive therapy promotes interhemispheric motor network balanced and should be considered a foundational intervention before the application of assistive therapy.