Percutaneous spinal stimulation to enhance gait and alleviate functional impairments in multiple sclerosis: A pilot study

Abstract

Objective

To evaluate the feasibility and potential efficacy of percutaneous spinal stimulation (epidural stimulation, ES) combined with task-specific training to reduce spasticity and improve gait and balance in individuals with progressive multiple sclerosis (MS).

Methods

Two men with progressive MS (EDSS 6.5) underwent ES lead implantation targeting the lower spinal cord, followed by one month of rehabilitation involving 12 ES-assisted training sessions. Assessments included instrumented gait analysis, Modified Ashworth Scale (MAS), pendulum test, and static balance testing performed at baseline, and at end of the study with ES-Off and ES-On conditions.

Results

Participant 1, with spastic hemiparetic gait, demonstrated improved lower extremity joint kinematics and muscle activation during gait, reduced knee extensor spasticity, and enhanced functional movement patterns with ES-On. Participant 2, with significant paraparesis and minimal spasticity, showed limited gait changes but experienced marked improvements in static balance, particularly under eyes-closed conditions. No adverse events were reported.

Conclusion

This pilot study demonstrates the feasibility and tolerability of ES paired with task-specific training in progressive MS. Participant-specific responses were observed, including improvements in gait kinematics, neuromuscular activation, spasticity, or balance; however, spatiotemporal gait parameters did not improve. These preliminary findings highlight response heterogeneity and the need for individualized ES parameter tuning.

Trial registry name and URL: https://clinicaltrials.gov/study/NCT06019611

Keywords

Epidural spinal cord stimulation balance gait neuromodulation spasticity and volitional motor function

Introduction

Progressive motor impairment, specifically myelopathy, is the most common presentation of progressive multiple sclerosis (MS).^1,2 Mounting evidence suggests that cortical spinal tract involvement with MS lesions is responsible for the development of motor disability, primarily characterized by gait dysfunction.^3–6 Gait issues, often associated with spasticity,^7,8 weakness,⁹ and impaired balance,¹⁰ are among the most debilitating MS symptoms, affecting daily function and quality of life.¹¹

Up to 75% of patients with MS develop progressive disease, with no current treatment to halt or reverse its disability. Pathologically, progressive MS is characterized by gradual axonal loss in the absence of overt inflammatory disease. Current treatment options are limited to addressing the inflammatory component of the disease; they do not halt or reverse the neuronal death and axonal loss that contribute to declining motor function and walking ability. Progressive motor dysfunction in MS is most common in patients who have spinal cord lesions or lesions in the brainstem involving the motor tracts.^5,12,13 Complete loss of axons in any particular tract, even in patients with progressive MS, is unusual, suggesting the presence of residual but dysfunctional neural pathways. Spinal cord stimulation, which is believed to activate remaining neuronal pathways, has shown potential for facilitating volitional motor function, as demonstrated in individuals with spinal cord injury by altering spinal network excitability rather than restoring damaged axons.^14–18 However, its potential as a promising intervention to address motor dysfunction in MS patients is understudied.

Spinal cord stimulation, particularly epidural spinal cord stimulation (ES), was initially developed in the 1960s to treat chronic pain by suppressing noxious input.¹⁹ Unexpectedly, ES was discovered to enable motor function in an MS patient, where it was originally intended to relieve pain.²⁰ Despite these early findings, the potential of ES to enhance motor function in MS patients remained largely unexplored for many years. In the following decades, ES research focused on determining effective stimulation parameters to improve motor function, specifically in spinal cord injury,²¹ where ES has been shown to enhance volitional motor function and improve seated posture, reaching, standing, and stepping.^{14–18,22,23} Emerging evidence indicates that ES can similarly suppress spasticity and clonus in individuals with MS,^24–27 supporting a shared spinal mechanism despite differing primary pathologies. Collectively, these observations suggest that targeting spinal circuit dysfunction with ES may represent a promising, yet understudied, approach for addressing motor impairment in individuals with progressive MS.

The rationale for investigating the use of ES in patients with MS is that we can potentially positively impact on both spasticity and weakness by modulating the excitability of spinal circuits and residual descending pathways, without the systemic side effects of oral anti spasticity medications or the weakness associated with intrathecal baclofen. Motor deficits related to MS can be highly individualized, and one of the key benefits of ES over traditional approaches to improve gait function is that it can be tailored to an individual's disability to address specific impairments, such as weakness or spasticity. This highlights the need for further investigation into the effects of ES on progressive MS to strengthen the evidence base for clinical application. The purpose of this study was to examine the feasibility of implementing temporary ES in individuals with progressive MS and to explore preliminary within-participant changes in spasticity, balance, and gait. We hypothesized that temporary ES would be feasible and well tolerated in individuals with progressive MS, and that spasticity, balance, and gait may improve within participants.

Methods

Study participants

Following FDA investigational device exemption and Mayo Clinic Institutional Review Board approval, two male participants with progressive MS met inclusion criteria and were enrolled in this pilot study (NCT06019611). Inclusion criteria included no clinical or radiologic MS relapses for over 5 years, an Expanded Disability Status Scale (EDSS) score of 6.5, the ability to ambulate 10 feet independently or with an assistive device, and no changes to spasticity medications or dalfampridine in the last 3 months. Exclusion criteria included Grade 4 spasticity measured bilaterally in four muscle groups using the Modified Ashworth Scale (MAS), active participation in any other interventional clinical trial, any active implanted medical device, and any illness or condition that would compromise the participant's ability to comply with the protocol or their safety.

Study design

As outlined in Figure 1, baseline assessments, including gait and balance analysis (Figure 1A), were obtained prior to ES lead implantation, described as No ES. One day after surgical lead implantation, lead location and potential motor activation was determined by performing spatiotemporal electrophysiological mapping of the lumbosacral spinal cord, iterating through multiple electrode combinations while adjusting stimulation amplitude (mA) and frequency (Hz).²⁸ Subsequently, the participants completed one month of rehabilitation, which included ES and task-specific training across 12, two-hour sessions. Finally, prior to explanting the leads, each participant underwent post-rehabilitation assessments, including spasticity assessment, and gait and balance analysis, with stimulation off (ES-Off) and on (ES-On), (Figure 1F). The methodological approach is outlined in Appendix A, Figure A1.

Figure 1.

Study design. A. Baseline tests. B. Fluoroscopic image and illustration of ES lead placement (created with BioRender). C. Motor recruitment curves to enable motor activation of the lower extremities. D. Frequency curves protocol where stimulation amplitude is increased until whole leg muscle activation is reached followed by stepwise progression in frequency. E. ES and task-specific training, and F. End of study test with ES-On and ES-Off.

Implantation surgery

The present study employed temporary percutaneous cylinder leads. Percutaneous implantation can be performed under local anesthesia, is comparable in magnitude to a lumbar puncture, and allows for simple outpatient lead removal. In contrast, paddle electrode implantation requires laminectomy under general anesthesia and a similar surgical procedure for removal. In pain medicine, it is standard practice to conduct a trial of spinal cord stimulation prior to permanent implantation to determine efficacy. Because spinal cord stimulation for motor indications is not FDA approved, we followed this established pain medicine trial-based approach to assess motor outcomes while minimizing surgical risk, reducing cost, and addressing ethical concerns related to permanent hardware and potential device abandonment.

The procedure was performed in an outpatient setting, under sedation, with anesthesiology monitoring. Two temporary percutaneous epidural stimulation leads (Abbott Neuromodulation, Plano, TX) were placed parallel to one another along the dorsal epidural surface of the lumbosacral enlargement (Figure 1B). The externalized ends of the ES electrode leads were connected to a neurostimulation system (Abbott programmer; Abbott, Plano, TX) for delivering electrical pulse waveforms to the epidural surface. To refine electrode array alignment, intraoperative myography was used to record ES-evoked motor potentials from LE muscles bilaterally.^{29, 30}

Spatiotemporal electrophysiological mapping of the lumbosacral spinal cord

To confirm lead location and determine the general muscle activation patterns of the LE's, baseline motor recruitment curves were collected on day one of testing following surgical implantation (Figure 1C). Motor recruitment curves were generated with a train of five charge-balanced, biphasic paired pulses each with a pulse width of 250 µs and separated by 50 ms, with different electrode configurations. Amplitude was increased stepwise (0.25–15.0 mA) at a frequency of 0.5 Hz to enable motor activation of the lower extremities (Ripple Neuromed, Salt Lake City, UT, USA). To measure evoked electromyography (EMG) responses specific to lower-limb muscles, electrodes were connected to a 16-channel biosignal amplifier (g.BSamp, g.tec medical engineering GmbH, Schiedlberg, Oberösterreich, Austria). Each channel was configured with a gain of 100 μV/V. The resulting surface EMG signals were amplified and digitized at a sampling rate of 4 kHz using a PowerLab 16/35 system (AD Instruments, Dunedin, New Zealand). Motor activation thresholds were defined as LE EMG recordings exceeding 50µv and three standard deviations beyond resting motor activity while in the supine position.²⁸ On day 2, frequency curves were collected at the lowest amplitude of whole leg muscle activation identified through EMG, with similar stepwise progression from 16 to 40 Hz (Figure 1D). Recruitment and frequency curves were collected in a passive, supine position. The LE muscles studied to determine proximal versus distal activation patterns include the rectus femoris, vastus lateralis, medial hamstring, gastrocnemius, soleus, and tibial anterior muscle groups. Representative recruitment and frequency curve results for each participant are included in Appendix B.

ES and task-specific training

The participants completed one month of physical therapy, which included ES paired with task-specific training across 12 two-hour sessions (Figure 1E). ES was applied using information obtained during the spatiotemporal electrophysiological mapping including muscle activation specificity based on electrode location (proximal vs. distal). Stimulation parameters were fine-tuned at each session to enhance motor performance during task-specific training and were used as the effective ES parameters for the end-of-study tests. A task-specific training, a form of rehabilitation, focused on improving performance of functional tasks through goal-directed practice and repetition, was utilized during this trial.³¹ Functional tasks consisted of repeated gait activities on both the treadmill and overground, with or without body weight support, upright standing and balance exercises, pre-gait activities, and isolated strength training of the LE muscles. All task-specific training activities were done with stimulation on ES-On and stimulation was only available when the participant was in the laboratory. Rest breaks from both the task-specific training and stimulation were provided intermittently based on performance and/or the participant's request.

Data collection and processing

Modified Ashworth Scale (MAS)

Spasticity was assessed using the MAS with participants positioned in supine with a 10-inch therapy wedge supporting their trunk while their legs hung off the edge of the bed. Knee flexion/extension and ankle dorsiflexion/plantar flexion were tested by applying a high-velocity stretch, with resistance graded on a 0–4 scale per MAS protocol.³²

Pendulum test

The pendulum test evaluates spasticity by observing the passive swinging of the lower limb under gravity.³³ The motion resembles a smooth, damped pendulum, while spasticity reduces oscillations and alters the pattern. The pendulum test has good predictive value for detecting the presence of knee extensor muscle spasticity.³⁴ In this test, the participants were positioned in the same manner as the MAS; the examiner lifted the relaxed leg to a horizontal position, extended the knee, and released it to swing freely. Three trials per side, with 60-second rests, were recorded using inertial measurement unit (IMU) sensors placed on the shank and thigh. Knee joint angles were derived from shank sensor data, with the first swing angle (FSA; Appendix A, Figure A2) as the primary outcome, correlating with knee extensor spasticity.³⁵

Gait analysis

Participants had 41 retro-reflective markers affixed to their extremities, pelvis, trunk, and head. Surface EMG was collected (Motion Lab Systems, Baton Rouge, LA) from LE muscles. Participants walked approximately 10 m at a self-selected, preferred walking speed using a front-wheeled walker (Figure 1F). Marker trajectories of 3–6 strides were recorded by 19 cameras (120 Hz, Raptor-12, Motion Analysis Corporation, Santa Rosa, CA). Kinematic variables were calculated using commercial software (C-Motion Inc., Rockville, MD), with the 3D coordinates of the markers as inputs. Embedded right-hand Cartesian coordinate systems were used to describe the position and orientation of the LE rigid body segments.

Three trials for each of the walking conditions were averaged and used in data analysis for each participant. The period of a gait cycle (stride) was defined from the initial contact of one lower extremity to the next initial contact of that same extremity. All gait variables including the temporospatial parameters (velocity, cadence, step and stride length, double and single support, and step width) and joint kinematics were expressed as a percentage of the gait cycle.

The raw EMG signals were digitized at 4200 Hz, then were full-wave rectified and smoothed using a root mean square (RMS) algorithm with a time-averaging period of 25 ms using a custom program in MATLAB (Mathworks, Natick, MA). The signal magnitude from the RMS envelope of the EMG data was used to determine muscle activity during each gait cycle.

Postural control (static balance) assessment

Postural control and static balance were evaluated while standing on an instrumented pressure mat (Tekscan Inc., South Boston, MA), with eyes open (EO) and closed (EC), at baseline with No ES and at the end of study under both ES-Off and ES-On conditions (Figure 1). The participant was asked to stand on the instrumented pressure mat in a quiet stance with arms at their sides and feet positioned approximately hip-width apart. This was repeated in two conditions: EO and EC, with a 60-second resting interval between the tests. The Tekscan system and software were utilized to record the trajectory of the center of force during the balance tests using the instrumented pressure mat and to calculate the balance variables including, sway range in the anterior-posterior and medial-lateral directions (cm), and sway area (cm²). Due to technical issues, static balance tests were performed only on Participant 2.

Data analysis

Descriptive statistics including plots and tables for all outcomes were reported for the baseline assessments and end of study assessments under ES-On and ES-Off conditions.

Results

The two individuals who were enrolled to participate were in their 50 s and presented with very different gait impairments. These impairments resemble differences seen across patients in the clinical setting. Participant 1 (male, height: 180 cm, weight: 76 kg) had a primarily spastic hemiparetic gait pattern affecting the right leg greater than the left and presented with a hip circumduction gait pattern. Participant 2 (male, height: 172 cm, weight: 66 kg) had minimal spasticity and had significant paraparesis and presented with a crouched (flexed hips, knees, and ankles) gait pattern, affecting the left leg greater than the right. Both participants utilized a front wheeled walker, could walk 10–20 meters independently with an EDSS of 6.5. Both participants attended all planned study visits and completed all study-related activities without adverse events or early termination of any visit, demonstrating overall feasibility and tolerance of the study design. Study procedures, including assessments and physical therapy sessions involving task-specific training with ES, were well tolerated, with no serious adverse events or limitations in participation. Participant 2 occasionally reported discomfort at higher stimulation amplitudes, described as sensations of “buzzing,” “tingling,” “pressure,” and general discomfort; therefore, stimulation amplitudes were initiated at lower levels at the beginning of sessions and gradually increased as acclimation occurred. Overall, participants with progressive MS were able to fully participate in and tolerate ES paired with task-specific training.

The fine-tuned stimulation parameters that enhanced performance during task-specific training were subsequently used for the end of study assessments with ES-On, were for Participant 1, a current amplitude of 4.4 mA with a pulse frequency of 150 Hz; and for Participant 2, a current amplitude of 9.5 mA with a pulse frequency of 130 Hz.

Baseline assessments

Gait analysis data during the baseline assessment for Participant 1 indicated a reduced gait speed (0.25 m/s) and decreased range of motion in the LE joints, compared to normative data,³⁶ with right knee hyperextension and no ankle dorsiflexion. The data for participant 2 indicated a reduced gait speed (0.38 m/s) and increased ankle dorsiflexion during terminal stance caused by a crouched gait. Table 1 and Figure 2 indicate the spatiotemporal parameters and LE joint kinematics during the baseline assessment.

Figure 2.

LE joint kinematics in the sagittal plane for participant 1 (A), and Participant 2 (B), during baseline assessment without epidural stimulation (NO ES). Joint angles for the hip and knee are expressed as extension–flexion (Ext–Flx), and for the ankle as plantarflexion–dorsiflexion (PF–DF). Each plot displays joint angle trajectories over the gait cycle (%GC; 0–100%). Vertical lines represent the transition from stance to swing. Solid black lines represent lab normative data (unpublished), with shaded regions showing ±1 standard deviation. Normative data were collected from 20 healthy adults (10 females; mean (SD): age = 26.7 (4.8) years, height = 173.2 (9.0) cm, mass = 70.1 (12.2) kg, BMI = 23.2 (2.1) kg/m²).

Table 1.

Spatiotemporal parameters during the baseline and end of study assessments.

	Participant 1			Participant 2			Normative data
Spatiotemporal parameters	Baseline (No Stim)	End of Study, ES-Off	End of Study, ES-On	Baseline (No Stim)	End of Study, ES-Off	End of Study, ES-On	P₅	Mean	P₉₅
Velocity (cm/s)	25	17	18	38	31	39	112	138	163
Cadence (step/min)	44	29	31	53	50	53	104	120	136
Stride (cm)	70	73	74	87	76	88	120	138	156
Step Width (cm)	13	11	13	10	12	14	5	10	16
Total Support Time (% of cycle)
Right	73	77	75	78	80	79	61	63	65
Left	78	83	85	80	81	80
Single Support Time (% of cycle)
Right	21	17	15	20	19	21	35	37	39
Left	28	24	25	22	20	22
Step Length (cm)
Right	39	37	37	44	40	45	60	69	78
Left	31	37	38	43	36	44

Static balance tests at baseline (No ES) with Participant 2 indicated that postural sway was greater during the EC condition compared to the EO condition, particularly in the medial-lateral direction (Figure 3).

Figure 3.

Static balance tests with eyes open (A, EO) and eyes closed (D; EC), and the results for the center of force (COF) trajectory sway in the anterior-posterior (AP) and medial-lateral (ML) directions (B) and the postural sway area (C) at the baseline visit without epidural stimulation (NO ES) in Participant 2.

End of study assessments

Both participants tolerated ES lead implantation, task-specific training with ES, and lead explant well with no signs or symptoms of fatigue or MS flares during the clinical trial.

In Participant 1 (specifically the most affected leg), analysis of gait data from the end-of-study assessments with ES-Off revealed notable improvements, including increased hip extension, knee flexion, and ankle dorsiflexion compared to the baseline visit (Figure 4A). The end of study comparison between ES-Off and ES-On conditions (Figures 4B and 5) showed that ES resulted in marked increases in right knee maximum flexion during the swing phase (39%) and hamstring peak muscle activity (30%), a decrease in rectus femoris peak muscle activity (14%), and the restoration of a more natural walking pattern by enabling the foot to go through its typical rolling motion (three rockers of the foot) during each step during the ES-On condition (Figure 5). The gait speed decreased during end of study tests with ES-Off and ES-On in comparison to baseline tests (Table 1). In Participant 2, gait parameters and joint kinematics did not change notably between the ES-On and ES-Off conditions compared to baseline (Figure 4B and C and Table 1). Overall, no notable changes in spatiotemporal parameters were observed between ES-Off and ES-On at the end of the study for either participant (Table 1).

Figure 4.

Lower extremity joint kinematics in the sagittal plane are illustrated for Participant 1 (right leg) during baseline and ES-off condition at the end of the study (A), as well as with ES-Off and ES-On conditions at the end of the study (B). Stimulation was delivered using a current amplitude of 4.4 mA, pulse frequency of 150 Hz, and an asymmetric charge-balanced waveform with a pulse width of 250 μs, as indicated by the vertical bar on the left side of panel B. Similarly, data for Participant 2 (left leg) are shown baseline and ES-Off condition at the end of the study (C), as well as with ES-Off and ES-On conditions at the end of the study (D). using a current amplitude of 9.5 mA, pulse frequency of 130 Hz, and an asymmetric charge-balanced waveform with a pulse width of 250 μs, as indicated by the vertical bar on the left side of panel D. Joint angles for the hip and knee are expressed as extension–flexion (Ext–Flx), and for the ankle as plantarflexion–dorsiflexion (PF–DF). Each plot displays joint angle trajectories over the gait cycle (%GC; 0–100%). Vertical lines indicate the transition from stance to swing phase. Solid black lines represent lab normative data, with shaded regions showing ±1 standard deviation.

Figure 5.

Lower extremity (LE) joint kinematics and Root Mean Square (RMS) (time averaging, period of 0.25 s) electromyography (EMG) of hamstring and rectus femoris for a representative gait cycle during post-rehabilitation assessments with ES-Off and ES-On (right limb). The top panel shows hip, knee, and ankle joint angles, with the y-axis representing joint angle in degrees (extension–flexion for hip and knee; dorsiflexion–plantarflexion for the ankle). The bottom panel shows RMS EMG signals (normalized to the maximum value across all gait cycles). Vertical lines represent the transition from stance to swing. Highlighted boxes represent the time that kinematic and EMG data were processed.

The MAS score for the right side (Participant 1, most affected leg) knee extensors decreased from 2 during ES-Off to $1^{+}$ during ES-On, and on the left side (participant 2, most affected leg) decreased from $1^{+}$ during ES-Off to $1$ during ES-On (Table 2). Data from the pendulum test indicated that the first swing angle increased by 6%, and rectus femoris muscle activity decreased by 58% during passive swing of the lower limb with ES-On compared to ES-Off in Participant 1 (Appendix A, Figure A3); however, the pendulum test data were not different between the two conditions for Participant 2 (Table 2).

Table 2.

Modified Ashworth Scale (MAS) scores for the knee extensors (Ext), knee flexors (Flex), ankle dorsiflexors (Dorsi), and ankle plantarflexors (Plantar), as well as mean (SD) of first swing angle (FSA) during the pendulum test.

	Participant 1								Participant 2
	End of Study, ES-Off				End of Study, ES-On				End of Study, ES-Off				End of Study, ES-On
Modified Ashworth Scale
	Knee Ext	Knee Flex	Ankle Plantar	Ankle Dorsi	Knee Ext	Knee Flex	Ankle Plantar	Ankle Dorsi	Knee Ext	Knee Flex	Ankle Plantar	Ankle Dorsi	Knee Ext	Knee Flex	Ankle Plantar	Ankle Dorsi
Right	2	0	1+	0	1+	0	1	0	1+	0	1	0	1+	0	1	0
Left	2	0	1+	0	1	0	0	0	1+	0	0	0	1	0	0	0
Pendulum Test, FSA (Deg)
Right	78 (2)				83 (6)				68 (3)				76 (5)
Left	90 (5)				99 (3)				77 (2)				85 (4)

In Participant 2, postural sway decreased notably during EC (91%) and EO (50%) conditions at the end of study (ES-Off) compared to baseline. Post-rehabilitation comparison between ES-Off and ES-On conditions demonstrated that ES led to a notable decrease in postural sway during EC (56%) and EO (30%) conditions (Figure 6).

Figure 6.

Static balance tests with eyes open (EO) and eyes closed (EC) in Participant 2: The center of force (COF) trajectory sway in the anterior-posterior (AP) and medial-lateral (ML) directions (A & B) and the postural sway area (C) at the end of the study.

Discussion

We present two patients with progressive multiple sclerosis with EDSS of 6.5 who had improvement in gait and balance parameters in response to ES combined with task-specific training. The response to stimulation was different in the two patients, likely due to differences in gait impairments, which are not captured by the EDSS. In this pilot study, ES paired with task-specific training was well tolerated, with no serious adverse events or study withdrawals. Participants with progressive MS were able to complete all study procedures, supporting the feasibility of this intervention in this population.

Gait and spasticity

Participant 1 exhibited a spastic hemiparetic gait pattern characterized by hip and knee flexor weakness and knee extensor spasticity, resulting in a circumduction gait. ES use in this participant led to increased hamstring strength and reduced quadriceps spasticity during walker-assisted gait (Figure 5). These improvements included greater hip extension, knee flexion, and ankle dorsiflexion during the end-of-study assessment (both ES-On and ES-Off) compared to baseline (Figure 4). This suggests a potential synergistic effect of one month of ES combined with task-specific training on walking ability. The marked increase in right knee maximum flexion during the swing phase, along with increased hamstring peak muscle activity, suggests that ES may enhance volitional motor function during walking. Specifically, it improved hamstring muscle activity while reducing co-contraction of the knee extensors. Additionally, the notable decrease in rectus femoris peak muscle activity may indicate reduced spasticity of the knee extensors during late stance as well as initial and mid-swing (Figure 5). The results from MAS and the pendulum test also support the possible impact of ES on reducing knee extensor muscle spasticity (Table 2). However, despite these neuromuscular and kinematic improvements, no meaningful differences were observed in spatiotemporal gait parameters between ES-Off and ES-On conditions at the end of the study (Table 1).

Participant 2 had greater deficits in strength overall but very little spasticity. ES use in this participant did not result in notable gait improvement. Addressing such leg weakness would require broader and higher-intensity ES, but he was unable to tolerate it due to discomfort.

Balance

ES use in Participant 2 led to an improvement in static balance with EO and EC (Figure 6). The maintenance of postural control requires the integration of multiple sensorimotor processes (visual, vestibular, and proprioceptive) to generate coordinated movements that maintain the center of mass within the limits of stability. Any adverse alteration in these processes or motor impairments, such as spasticity, results in deficits in postural control and balance. Postural control partly depends on the ability to modulate ankle stiffness,^37,38 and individuals with spasticity struggle to adjust their ankle stiffness due to increased reflex modulation, leading to poorer postural control. Sosnoff et al. showed that spasticity significantly contributes to the postural control abnormalities in individuals with MS.³⁹ Additionally, ankle joint proprioception is crucial for postural control, as it provides key information about standing postural sway.⁴⁰ In MS, proprioception is impaired due to cerebral or dorsal column involvement,⁴¹ which may explain the greater postural sway with eyes closed compared to a healthy individual.⁴² The results of the static balance tests in this study indicate a notable impact of ES on improving static balance and postural control in Participant 2 who had significant paraparesis. This improvement may be attributed to the reduction in ankle spasticity facilitated by ES; however, the results from MAS did not show any notable changes in ankle spasticity. This might be due to subjectivity of MAS and its poor inter-session reliability.⁴³ The pronounced effect of ES observed during the EC condition may be due to its role in enhancing proprioception.

Limitations and future directions

The primary limitations of this study include the small sample size and the absence of a rigorous method for determining effective ES parameters and objectively assessing lower-extremity spasticity. In addition, ES-On and ES-Off conditions were evaluated only after completion of the task-specific training period due to feasibility constraints and to minimize participant burden following implantation. Despite these limitations, this study provides valuable preliminary insights and lays the groundwork for future investigations. Further research should focus on evaluating the impact of ES in a larger sample of individuals with progressive MS to determine effective ES parameters for enhancing gait function.

Conclusion

The preliminary findings from this pilot study suggest that ES, when combined with task-specific training, is feasible and well tolerated in individuals with progressive MS and may have a positive impact on gait and motor function outcomes. These promising results highlight the need for further research with larger sample sizes to confirm and refine these outcomes, as well as to individualize the application of ES for individuals with MS.

Supplemental Material

sj-docx-1-mso-10.1177_20552173261432301 - Supplemental material for Percutaneous spinal stimulation to enhance gait and alleviate functional impairments in multiple sclerosis: A pilot study

Supplemental material, sj-docx-1-mso-10.1177_20552173261432301 for Percutaneous spinal stimulation to enhance gait and alleviate functional impairments in multiple sclerosis: A pilot study by Omid Jahanian, Anders J Asp, Megan L Gill, Daniel D Veith, Katrina A Fernandez, Cecilia A Hogen, Candee J Mills, Andrew R Thoreson, Ryan Solinsky, Kenton R Kaufman, Peter J Grahn, Kristin D Zhao and W Oliver Tobin in Multiple Sclerosis Journal – Experimental, Translational and Clinical

Supplemental Material

sj-docx-2-mso-10.1177_20552173261432301 - Supplemental material for Percutaneous spinal stimulation to enhance gait and alleviate functional impairments in multiple sclerosis: A pilot study

Supplemental material, sj-docx-2-mso-10.1177_20552173261432301 for Percutaneous spinal stimulation to enhance gait and alleviate functional impairments in multiple sclerosis: A pilot study by Omid Jahanian, Anders J Asp, Megan L Gill, Daniel D Veith, Katrina A Fernandez, Cecilia A Hogen, Candee J Mills, Andrew R Thoreson, Ryan Solinsky, Kenton R Kaufman, Peter J Grahn, Kristin D Zhao and W Oliver Tobin in Multiple Sclerosis Journal – Experimental, Translational and Clinical

Supplemental Material

sj-docx-3-mso-10.1177_20552173261432301 - Supplemental material for Percutaneous spinal stimulation to enhance gait and alleviate functional impairments in multiple sclerosis: A pilot study

Supplemental material, sj-docx-3-mso-10.1177_20552173261432301 for Percutaneous spinal stimulation to enhance gait and alleviate functional impairments in multiple sclerosis: A pilot study by Omid Jahanian, Anders J Asp, Megan L Gill, Daniel D Veith, Katrina A Fernandez, Cecilia A Hogen, Candee J Mills, Andrew R Thoreson, Ryan Solinsky, Kenton R Kaufman, Peter J Grahn, Kristin D Zhao and W Oliver Tobin in Multiple Sclerosis Journal – Experimental, Translational and Clinical

Footnotes

Acknowledgments

The authors thank Mrs. Lea Dacy for proofreading and formatting assistance

Author contributions

Conceptualization: OJ, AJA, MLG, PJG, and WOT; data curation: OJ; formal analysis: OJ; funding acquisition: KDZ, PJG, and WOT; investigation: OJ, AJA, MLG, DDV, KAF, CAH, and CJM; methodology: OJ, AJA, MLG, PJG, and WOT; resources: KRK, PJG, and KDZ; supervision: WOT; visualization: DDV; writing – original draft preparation: OJ; and writing – review & editing: AJA, MLG, DDV, KAF, CAH, CJM, ART, RS, KRK, PJG, KDZ, and WOT

Consent to participate

The two participants with progressive MS provided informed written consent to share their medical information.

Consent to publish

The participant pictured in provided a written, signed consent for photograph publication.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declaration of conflicting interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: PJG reports research support from the National Institutes of Health under Award Number R01NS115877, which partially supported this study. WOT reports receiving research funding from the National Institutes of Health, Mayo Clinic Center for Multiple Sclerosis and Autoimmune Neurology and Merck. He receives royalties from the publication of “Mayo Clinic Cases in Neuroimmunology” (OUP). The other authors declare that they have no competing interests.

Ethical approval and informed consent

The Mayo Clinic Institutional Review Board approved this pilot study (NCT06019611). The U. S. Food and Drug Administration provided an investigational device exemption. The two participants with progressive MS provided informed written consent to share their medical information. The participant pictured in provided a written, signed consent for photograph publication.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the Mayo Clinic Center for Multiple Sclerosis and Autoimmune Neurology. This work was also supported by the NINDS of the NIH under Award Number R01NS115877 (awarded to PJG). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

ORCID iDs

Omid Jahanian

W Oliver Tobin

Supplemental material

Supplemental material for this article is available online.

References

Bashir

Whitaker

. Clinical and laboratory features of primary progressive and secondary progressive MS. Neurology 1999; 53: 765–765.

Stevenson

Miller

Rovaris

, et al. Primary and transitional progressive MS: a clinical and MRI cross-sectional study. Neurology 1999; 52: 839–839.

Jackson-Tarlton

Keegan

Fereidan-Esfahani

, et al. Spinal cord and brain corticospinal tract lesions are associated with motor progression in tumefactive multiple sclerosis. Mult Scler Relat Disord 2023; 73: 104614.

Keegan

Kaufmann

Weinshenker

, et al. Progressive solitary sclerosis: gradual motor impairment from a single CNS demyelinating lesion. Neurology 2016; 87: 1713–1719.

Keegan

Kaufmann

Weinshenker

, et al. Progressive motor impairment from a critically located lesion in highly restricted CNS-demyelinating disease. Mul Scler J 2018; 24: 1445–1452.

Zeydan

Atkinson

, et al. Cervical spinal cord atrophy: an early marker of progressive MS onset. Neurol Neuroimmunol Neuroinflammation 2018; 5: e435.

Pau

Coghe

Corona

, et al. Effect of spasticity on kinematics of gait and muscular activation in people with multiple sclerosis. J Neurol Sci 2015; 358: 339–344.

Rizzo

Hadjimichael

Preiningerova

, et al. Prevalence and treatment of spasticity reported by multiple sclerosis patients. Mult Scler J 2004; 10: 589–595.

Behrens

Pfüller

Mansow-Model

, et al. Using perceptive computing in multiple sclerosis-the short Maximum speed walk test. J Neuroeng Rehabil 2014; 11: 1–10.

10.

Martin

Phillips

Kilpatrick

, et al. Gait and balance impairment in early multiple sclerosis in the absence of clinical disability. Mult Scler J 2006; 12: 620–628.

11.

Sutliff

. Contribution of impaired mobility to patient burden in multiple sclerosis. Curr Med Res Opin 2010; 26: 109–119.

12.

Kassa

Sechi

Flanagan

, et al. Onset of progressive motor impairment in patients with critical central nervous system demyelinating lesions. Mult Scler J 2021; 27: 895–902.

13.

Sechi

Keegan

Kaufmann

, et al. Unilateral motor progression in MS: association with a critical corticospinal tract lesion. Neurology 2019; 93: e628–e634.

14.

Angeli

Boakye

Morton

, et al. Recovery of over-ground walking after chronic motor complete spinal cord injury. N Engl J Med 2018; 379: 1244–1250.

15.

Gill

Grahn

Calvert

, et al. Neuromodulation of lumbosacral spinal networks enables independent stepping after complete paraplegia. Nat Med 2018; 24: 1677–1682.

16.

Gill

Linde

Hale

, et al. Alterations of spinal epidural stimulation-enabled stepping by descending intentional motor commands and proprioceptive inputs in humans with spinal cord injury. Front Syst Neurosci 2021; 14: 590231.

17.

Grahn

Lavrov

Sayenko

, et al. Enabling task-specific volitional motor functions via spinal cord neuromodulation in a human with paraplegia. In: Mayo Clinic Proceedings, 2017, pp.544–554. Elsevier.

18.

Wagner

Mignardot

J-B

Goff-Mignardot

, , et al. Targeted neurotechnology restores walking in humans with spinal cord injury. Nature 2018; 563: 65–71.

19.

Shealy

Mortimer

Reswick

. Electrical inhibition of pain by stimulation of the dorsal columns: preliminary clinical report. Anesth Analg 1967; 46: 489–491.

20.

Cook

. Electrical stimulation in multiple sclerosis. Hosp Pract 1976; 11: 51–58.

21.

Pinter

Gerstenbrand

Dimitrijevic

. Epidural electrical stimulation of posterior structures of the human lumbosacral cord: 3. Control of spasticity. Spinal Cord 2000; 38: 524–531.

22.

Gorgey

Trainer

Khalil

, et al. Epidural stimulation and resistance training (REST-SCI) for overground locomotion after spinal cord injury: randomized clinical trial protocol. J Clin Med 2025; 14: 1829.

23.

Gorgey

Trainer

Sutor

, et al. A case study of percutaneous epidural stimulation to enable motor control in two men after spinal cord injury. Nat Commun 2023; 14: 2064.

24.

Asp

Gill

Veith

, et al. Percutaneous spinal cord stimulator enables novel clonus biomarker and management in persons with spinal cord injury and multiple sclerosis: an exploratory study. Top Spinal Cord Inj Rehabil 2025; 31: 114–124.

25.

Asp

Jahanian

Gill

, et al. Percutaneous epidural spinal cord stimulation immediately reduces clonus in an individual with multiple sclerosis. Mult Scler J 2026; 32: 144–147.

26.

D’hondt

Marcial

Mittal

, et al. A scoping review of epidural spinal cord stimulation for improving motor and voiding function following spinal cord injury. Top Spinal Cord Inj Rehabil 2023; 29: 12–30.

27.

Romeni

Losanno

Emedoli

, et al. High-frequency epidural electrical stimulation reduces spasticity and facilitates walking recovery in patients with spinal cord injury. Sci Transl Med 2025; 17: eadp9607.

28.

Veith

Asp

Gill

, et al. Prevalence of autonomic dysreflexia during spinal cord stimulation after spinal cord injury. J Neurophysiol 2024; 132: 1371–1375.

29.

Calvert

Grahn

Strommen

, et al. Electrophysiological guidance of epidural electrode array implantation over the human lumbosacral spinal cord to enable motor function after chronic paralysis. J Neurotrauma 2019; 36: 1451–1460.

30.

Veith

Gill

Beck

, et al. Functional outcomes and participants’ perspectives during short-term application of spinal stimulation in individuals with spinal cord injury. J Spinal Cord Med 2025; 48: 978–989.

31.

Hubbard

Parsons

Neilson

, et al. Task-specific training: evidence for and translation to clinical practice. Occup Ther Int 2009; 16: 175–189.

32.

Bohannon

Smith

. Interrater reliability of a modified Ashworth scale of muscle spasticity. Phys Ther 1987; 67: 206–207.

33.

Bajd

Vodovnik

. Pendulum testing of spasticity. J Biomed Eng 1984; 6: 9–16.

34.

Whelan

Sexton

Jones

, et al. Predictive value of the pendulum test for assessing knee extensor spasticity. J Neuroeng Rehabil 2018; 15: 68.

35.

Sangari

Perez

. Prevalence of spasticity in humans with spinal cord injury with different injury severity. J Neurophysiol 2022; 128: 470–479.

36.

Lencioni

Carpinella

Rabuffetti

, et al. Human kinematic, kinetic and EMG data during different walking and stair ascending and descending tasks. Sci Data 2019; 6: 309.

37.

Angulo-Kinzler

Mynark

Koceja

. Soleus H-reflex gain in elderly and young adults: modulation due to body position. J Gerontol A Biol Sci Med Sci 1998; 53: M120–M125.

38.

Gatev

Thomas

Kepple

, et al. Feedforward ankle strategy of balance during quiet stance in adults. J Physiol 1999; 514: 915–928.

39.

Sosnoff

Shin

Motl

. Multiple sclerosis and postural control: the role of spasticity. Arch Phys Med Rehabil 2010; 91: 93–99.

40.

Fitzpatrick

Rogers

McCloskey

. Stable human standing with lower-limb muscle afferents providing the only sensory input. J Physiol 1994; 480: 395–403.

41.

Fling

Dutta

Schlueter

, et al. Associations between proprioceptive neural pathway structural connectivity and balance in people with multiple sclerosis. Front Hum Neurosci 2014; 8: 814.

42.

Daley

Swank

. Quantitative posturography: use in multiple sclerosis. IEEE Trans Biomed Eng 2007; 7: 668–671.

43.

Craven

Morris

. Modified ashworth scale reliability for measurement of lower extremity spasticity among patients with SCI. Spinal Cord 2010; 48: 207–213.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

2.56 MB

0.03 MB

0.00 MB

0.79 MB