Instrumented selective control assessment of the lower extremity to identify neural constraints in muscle co-activation during treadmill walking after stroke

• Published in: Gait & posture Vol. 106; pp. S261 - S262
• Main Authors: Buurke, Tom, Febrer-Nafría, Míriam, Verheyden, Geert, De Groote, Friedl
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.09.2023
• ISSN: 0966-6362; 1879-2219
• DOI: 10.1016/j.gaitpost.2023.07.041

Deficits in muscle coordination during walking after stroke are characterized by reduced selective control, i.e. the inability to perform isolated joint movements. Reduced selective control is often quantified by the number of muscle synergies during walking [1]. However, muscle synergies could capture both muscle coordination patterns to perform the task (e.g. walking) and muscle coordination patterns due to constraints on a neural level (e.g. dysfunctional co-activation), while only the neural constraints can be considered a contributing factor to muscle coordination deficits. Here, we aim to find muscle co-activation patterns that reflect neural constraints but not task constraints. Therefore, we assess the relationship between muscle co-activation patterns during instrumented selective control assessment of the lower extremity (SCALE) [2] and treadmill walking in people post-stroke. Based on previously observed muscle synergies post-stroke [1] we hypothesize that the amount of co-activation between the knee extensor muscles and either 1) knee flexor muscles or 2) ankle plantarflexor muscles during the SCALE relates to the amount of muscle co-activation between the same muscle during treadmill walking. 18 people with chronic stroke performed the SCALE [2] and the knee task was further analyzed. During the SCALE knee task, participants were instructed to perform isolated knee extension-flexion-extension movements with the paretic leg, while sitting in a chair with both legs hanging freely. Then, participants walked on an instrumented treadmill for three minutes, of which the last minute was further analyzed. Surface-electromyography was measured during SCALE and treadmill walking from the rectus femoris (RF), vastus lateralis (VL), biceps femoris (BF), semitendinosus (ST), soleus (SOL) and gastrocnemius medialis (GM). To find generalized muscle co-activation patterns, we first quantified co-activation as the correlation coefficient between EMG data of the hypothesized muscle pairs for each individual participant within each task. Then, Pearson’s correlation coefficients were calculated on group level between co-activation during SCALE and co-activation during treadmill walking, with alpha=0.05. Significant correlations were found between the co-activation during SCALE and treadmill walking for the knee extensors and 1) knee flexors (RF-ST: r=0.56, p=0.016) and 2) ankle plantarflexors (RF-GM: r=0.49, p=0.039; VL-GM: r=0.49, p=0.038; Fig. 1). Fig. 1. Scatter plots of muscle co-activation during the SCALE isolated knee movement and treadmill walking (left) and two typical participants with respectively high and low co-activation (right). Colored dots indicate individual participants, asterisks indicate significant correlations (p<0.05). [Display omitted] Co-activation of knee extensors and knee flexors/ankle plantarflexors during walking post-stroke could reflect neural constraints in muscle coordination rather than a compensation strategy for other impairments. The ability to distinguish task constraints from neural constraints in muscle co-activation might aid the identification of therapeutic treatment targets and design of assistive devices. We will perform similar analyses in children with cerebral palsy, for which data collection is currently ongoing.