Improvement of voluntary movement following task-specific training with subthreshold lumbosacral epidural stimulation in individuals with severe spinal cord injury

Recovery of voluntary motor function several years after clinical diagnosis of motor complete spinal cord injury (SCI) is rare. Subthreshold network targeted lumbosacral epidural stimulation (scES) has been previously shown to restore voluntary movement in individuals with motor complete SCI (Angeli et al. 2014, Grahn et al. 2017, Darrow et al. 2022). The aim of this study is to determine the effects of task-specific training with scES on voluntary movement ability in individuals with severe SCI.Individuals with chronic motor complete SCI (n=27) were implanted at L1-S1 spinal cord level with an electrode array and neurostimulator. These individuals (age: 35.5 ± 10.6 yrs; time post-injury: 10.1 ± 8.1 yrs; 37% female, 15 AIS A, 11 AIS B, 1 AIS C) were unable to move their legs voluntarily or stand independently. Voluntary movement mapping was performed in all participants while asking individuals to attempt flexion movements of the first toe, ankle and hip. Further changes in parameters were dependent on the observed EMG modulation and joint movements. Individuals were randomized to a voluntary training group (Vol-scES) or a cardiovascular group (CV-scES). Individuals used scES for 6 hours a day practicing voluntary movement (Vol-scES) or for blood pressure regulation (CV-scES). We performed torque and EMG assessments in a dynamometer pre and post training with voluntary specific stimulation.All individuals were able to modulate EMG activity in the presence of scES and were able to move at least one joint. There was no difference between groups prior to the start of the training intervention. Individuals randomized to the Vol-scES group showed an improvement in maximum torque generation and ability to perform multiple consecutive repetitions of hip flexion when compared to their baseline (post-implant, pre-intervention values). While individuals randomized to the CV-scES group showed a slight decrease in both outcomes. When asked to perform ankle dorsiflexion the median RMS value of TA showed a slight increase for the VoL-scES group with a decrease for the CV-scES group. Torque generation during dorsiflexion increased in the VoL-scES post training while there was a decrease in the CV-scES. These results provide evidence that task-specific training with scES results in greater voluntary movement ability. Further, scES modulates network excitability of the injured spinal cord to allow for the integration of afferent and supraspinal descending input to generate movement below the injury after severe SCI.