Cumulative Transcutaneous Spinal Stimulation with Locomotor Training Safely Improves Trunk Control in Children with Spinal Cord Injury: Pilot Study

Abstract

Background/objectives: Non-invasive spinal cord transcutaneous stimulation (scTS) has expanded the therapeutic landscape of spinal cord injury (SCI) rehabilitation, offering potential benefits beyond compensatory approaches to paralysis. Children with SCI are particularly susceptible to developing neuromuscular scoliosis due to trunk muscle paralysis and ongoing skeletal growth, making targeted interventions crucial. As demonstrated in adults and pediatrics with SCI, the ability of scTS to acutely and safely enable an upright posture and trunk control could be leveraged as a therapeutic adjunct. Activity-based locomotor training (AB-LT) alone significantly improves trunk control in children with SCIs; combining it with scTS may enhance outcomes. This pilot study evaluated the safety, feasibility, and cumulative effects of AB-LT combined with scTS on trunk control in children with SCI.

Methods: Three children with SCI completed 19 to 64 sessions of combined AB-LT and scTS. Adverse effects were monitored session to session, and trunk control was assessed pre- and post-intervention.

Results: Across 130 interventions in three participants, 88.5% of sessions were free from adverse effects. Reported adverse events included autonomic dysreflexia (5.4%), skin redness at electrode sites (4.6%), and headaches (1.5%). No significant impact of scTS on fatigue or central hemodynamic parameters was observed. Post-intervention, all participants demonstrated improved trunk control during quiet and perturbed sitting.

Conclusions: These findings provide the first evidence supporting the safety and feasibility of this combinatorial approach in pediatric SCI rehabilitation while emphasizing the importance of monitoring skin integrity and signs of autonomic dysreflexia. This intervention shows potential synergistic benefits, warranting further research to confirm efficacy and optimize therapeutic protocols.

Keywords: activity-based locomotor training; pediatrics; spinal cord injury; spinal cord transcutaneous stimulation; trunk control.

Figure 1

Study design and intervention session. (a). Scheme of the experiment. Three of the eight participants from the previous Aim [12] were enrolled in the current study. Participants completed a trunk control assessment pre- and post-AB-LT with scTS intervention. (b). AB-LT with scTS intervention. The timeline shows the approximate timing of different locomotor training conditions. After a period of optimization of the stimulation parameters (sitting, scTS), AB-LT on the treadmill (scTS, then with no stimulation) and on the overground (no stimulation, then scTS). BP, HR, fatigue, and pain were measured at the end of each condition (grey arrows). (c). Electrode application. Cathodes were placed on the skin between the T10/T11 and T12/L1 vertebrae, and anodes were placed over bilateral ASIS. (d). Current modulation. The 10 kHz current was packaged in 15, 30, or 60 Hz packs. (e). Treadmill condition. Activity-based Locomotor Training was performed by therapists and activity-based technicians to promote load-bearing, stepping, and standing with partial body weight support, manual facilitation of pelvic rotation, and limb stepping kinematics with upright trunk and coordinated arm swing. Trunk control was also challenged during standing activities with reaching overhead and fine motor tasks [19]. (f). Overground condition. Therapists and trainers challenged trunk control during sitting, standing, and stepping overground via activities, e.g., batting balloons, catching and throwing balls, sit-to-stand with the upright trunk. Abbreviations: EMG—Electromyography, AB-LT- activity-based locomotor training, scTS—spinal cord transcutaneous stimulation, BP—blood pressure, HR—heart rate, ASIS—anterior superior iliac spine, ms—milliseconds, Hz—hertz, CoP—center of pressure.

Figure 2

Fatigue levels and correlation scTS intensity. (a–c). Fatigue levels averaged across all AB-LT sessions for each participant, as measured by the Modified Borg Rate of Perceived Exertion scale. Data points represent individual fatigue scores at each training time point: sitting and sitting + scTS (circles), treadmill and treadmill + scTS (squares), overground and overground + scTS (triangles). Gray marker indicates the no-stimulated condition, while red markers indicate scTS condition; (d–f). Correlation between fatigue scores and the scTS intensity of each channel. Red filled dots (circles, squares, and triangles) and black trend line—stimulation intensity for T10–T11 stimulation site; red unfilled dots and red trend line—stimulation intensity for T12-L1; circles—sitting, squares—treadmill, triangles—overground. Statistical analysis: mixed linear model including a random intercept for each training point: sitting, sitting + scTS, treadmill + scTS, treadmill, overground, overground +scTS. The correlation between fatigue and mean scTS intensity was used with a mixed linear model #—vs. sitting, *—vs. scTS condition, p < 0.05. Correlation between fatigue and stimulation intensity was calculated with repeated measure correlation which adjusts for multiple measures on stimulation channels (channels 1 and 2) using the rmcorr package [35]. Abbreviations: scTS—spinal cord transcutaneous stimulation, mA-milliampere.

Figure 3

Central hemodynamic parameters averaged over all AB-LT. (a–c). SBP for each time point training for P1, P14, and P23; (d–f). DBP; (g–i). HR. Data are represented as violin plots with median (bold line) and quartiles (dashed line). Gray shades—no-stimulated condition, red color—scTS. Hatching plane—50th to 90th sex- and age-matched normal range for typical developing children. Statistical analysis: mixed linear model including a random intercept for each training point: sitting, sitting + scTS, treadmill + scTS, treadmill, overground, overground +scTS. #—vs. sitting, *—vs. scTS condition, p < 0.05. Abbreviations: SBP—systolic blood pressure, DBP—diastolic blood pressure, HR—heart rate, AB-LT—activity-based locomotor training, scTS—spinal cord transcutaneous stimulation.

Figure 4

Trunk control in P1 pre- and post-39 AB-LT with scTS interventions. (a). Trunk angular excursion during anterior and posterior leaning task for head-T8 and T8-pelvis segments for three trials; (b). Trunk angular excursion during left and right leaning task for head-T8 and T8-pelvis segments; (c). Pre-post CoP parameters during “right arm up” self-perturbating task; (d). CoP trace for “right arm up” self-perturbating task over three trials; (e). RMS of EMG of rector spinae at T10 and (f). RMS of EMG of upper trapezius. (g). Study timeline showing AB-LT with scTS interventions (red rectangles) and trunk control testing (triangles). Data are represented as mean for three-four trials with SEM. Statistics: mixed linear models were used with timepoint (pre-, post-intervention) as a fixed factor and trial nested in timepoint as a random factor (three trials). *—vs. pre-intervention, p < 0.05. Abbreviations: AB-LT—activity-based locomotor training, AP—anteroposterior, ML—mediolateral, CoP—Center of Pressure, EMG—electromyogram, RMS—root-mean-square envelope of the EMG, T10—tenth thoracic vertebrata, ES—erector spinae, UT—upper trapezius.

Figure 5

Trunk control in P14 pre- and post-19 AB-LT with scTS interventions. (a). Trunk angular excursion during anterior and posterior leaning task for head-T8 and T8-pelvis segments for three trials; (b). Trunk angular excursion during left and right leaning task for head-T8 and T8-pelvis segments; (c). Pre-post CoP parameters during “right arm up” self-perturbating task; (d). CoP trace for “right arm up” self-perturbating task over three trials; (e). RMS of EMG of rector spinae at T10 and (f). RMS of EMG of upper trapezius. (g). Study timeline showing AB-LT with scTS interventions (red rectangles) and trunk control testing (triangles). Data are represented as mean for three trials with SEM. Statistics: mixed linear models were used with timepoint (pre-, post-intervention) as a fixed factor and trial nested in timepoint as a random factor (three trials). *—vs. pre-intervention, p < 0.05. Abbreviations: AB-LT—activity-based locomotor training, AP—anteroposterior, ML—mediolateral, CoP—Center of Pressure, EMG—electromyogram, RMS—root-mean-square envelope of the EMG, T10—tenth thoracic vertebrata, ES—erector spinae, UT—upper trapezius.

Figure 6

Trunk control in P23 before and after AB-LT with scTS sessions. (a). Independent sitting time (unaided) in a relaxed posture and while attempting to sit upright. Red crosses indicate attempts in which P23 lost balance, and black circles indicate the participant successfully maintained balance and completed the trial; (b). Anteroposterior and mediolateral excursion of head-T8 and T8-pelvis during an attempt to sit upright; (c). RMS of EMG of the upper trapezius; (d). erector spinae at Th-10; (e). and erector spinae at L-5 while quiet and upright sitting; (f). Study timeline showing AB-LT with scTS interventions (red rectangles) and trunk control testing (triangles). Data are represented as mean for three trials with SEM. Statistics: mixed linear models were used with timepoint (pre-, post-intervention) as a fixed factor and trial nested in timepoint as a random factor (three trials). *—vs. pre-intervention, p < 0.05. Abbreviations: BB—before break due to COVID-19, AB—after break due to COVID-19, AP—anteroposterior, ML—mediolateral, EMG—electromyogram, RMS—root-mean-square envelope of the EMG, UT—upper trapezius, ES—Erector spinae, T10—tenth thoracic vertebrata, L5—fifth lumbar vertebrata.