Transcutaneous spinal stimulation alters cortical and subcortical activation patterns during mimicked-standing: A proof-of-concept fMRI study

Abstract

Transcutaneous spinal stimulation (TSS) is a non-invasive neuromodulation technique that has been used to facilitate the performance of voluntary motor functions such as trunk control and self-assisted standing in individuals with spinal cord injury. Although it is hypothesized that TSS amplifies signals from supraspinal motor control networks, the effect of TSS on supraspinal activation patterns is presently unknown. The purpose of this study was to investigate TSS-induced activity in supraspinal sensorimotor regions during a lower-limb motor task. Functional magnetic resonance imaging (fMRI) was used to assess changes in neural activation patterns as eleven participants performed mimicked-standing movements in the scanner. Movements were performed without stimulation, as well as in the presence of (1) TSS, (2) stimulation applied to the back muscle, (3) paresthesia stimulation, and (4) neuromuscular electrical stimulation. TSS was associated with greater activation in subcortical and cortical sensorimotor regions involved in relay and processing of movement-related somatosensory information (e.g., thalamus, caudate, pallidum, putamen), as compared to the other stimulation paradigms. TSS also resulted in deactivation in both nucleus accumbens and posterior parietal cortex, suggesting a shift toward somatosensory feedback-based mechanisms and more reflexive motor control. Together, these findings demonstrate that spinal stimulation can alter the activity within supraspinal sensorimotor networks and promote the use of somatosensory feedback, thus providing a plausible neural mechanism for the stimulation-induced improvements of sensorimotor function observed in participants with neurological injuries and disorders.

Keywords: Brain-Spinal Connectome; Neuroimaging; Neuromodulation; Sensorimotor network; Spinal Cord; Standing.

Figures

Figure 1.

Electrode configurations for each stimulation paradigm: (A) transcutaneous spinal stimulation (TSS), (B) sham stimulation (SHAM), (C) paresthesia stimulation (PARA), (D) neuromuscular electrical stimulation (NMES). Black circles indicate the position of the cathode and red circles indicate the position of the anode electrodes.

Figure 2.

View of (A) experimental apparatus and (B) the paradigm used for the fMRI protocol. Participants were positioned on the scanner bed with both legs placed in the Exolab apparatus, and viewed an LCD monitor through a mirror attached to the head coil. Each fMRI scan was implemented in a block design. Specifically, participants started each scanning run by viewing a fixation cross for 24 s (“fixation, no-movement period”). After this, an on-screen instruction to “stand” was presented to the participant for 24 s (“movement block”). A fixation period then followed and was displayed for either 21, 24 or 27 s. There were 10 stimulation blocks in total.

Figure 3.

Patterns of activation and deactivation in ROIs located in the cerebellum and brainstem. BOLD signal changes are plotted as mean percent signal change (%PSC) from baseline for each condition. The strength of differences outputted by the BML for the comparisons between NO-STIM and TSS are indicated by the asterisks above the lines, whereas the asterisks below the lines indicate the strength of the differences outputted for comparisons between TSS and the NMES, PARA, and SHAM. Error bars indicate 95% confidence interval.

Figure 4:

Patterns of activation and deactivation in subcortical ROIs. BOLD signal changes are plotted as mean percent signal change (%PSC) from baseline for each condition. The strength of differences outputted by the BML for the comparisons between NO-STIM and TSS are indicated by the asterixis above the lines, whereas the asterixis below the lines indicate the strength of the differences between TSS and the NMES, PARA, and SHAM. Error bars indicate 95% confidence interval.

Figure 5:

Patterns of activation and deactivation in cortical ROIs. BOLD signal changes are plotted as mean percent signal change (%PSC) from baseline for each condition. The strength of differences outputted by the BML for the comparisons between NO-STIM and TSS are indicated by the asterixis above the lines, whereas the asterixis below the lines indicate the strength of the differences between TSS and the NMES, PARA, and SHAM. Error bars indicate 95% confidence interval.

Figure 6:

Visual summary of changes in supraspinal activation during exposure to TSS. Axial contrast maps outputted from a mixed-effect model analysis (using 3dMEMA AFNI program; p