Percutaneous Spinal Cord Stimulator Enables Novel Clonus Biomarker and Management in Persons With Spinal Cord Injury and Multiple Sclerosis: An Exploratory Study

Abstract

Background:

Spasticity is common in spinal cord injury (SCI) and multiple sclerosis (MS), and it can manifest as repeated, rhythmic muscle contractions called clonus. The spontaneous nature of clonus can disrupt independent performance of activities of daily living and negatively impact overall health and quality of life.

Objectives:

To quantify biomarkers of clonus and explore management of clonus in individuals with SCI or progressive MS using epidural spinal cord stimulation (ES) or dorsal root stimulation (DRS).

Methods:

Four male participants were included in this case series study: 3 with SCI and 1 with MS. All participants underwent temporary percutaneous ES lead placement over the dorsolateral thoracolumbar region of the spinal cord. Participants with SCI were also implanted with DRS leads at the L4 dorsal root. Clonus was elicited mechanically at the ankle while recording electromyography of the soleus muscle synchronized to direct spinal cord field potential recordings from ES and DRS percutaneous leads.

Results:

Clonus was evident as a prominent band (5-8 Hz) in recordings from ES leads, DRS leads, soleus muscle, and accelerometry. In 4 participants, percutaneous spinal stimulation reduced median clonus duration and cycle count. Clonus was immediately suppressed upon activation of spinal cord stimulation, and the suppression persisted even when clonus was reinitiated after turning off the stimulation.

Conclusion:

The use of objective biomarkers, including spinal cord potentials, to quantify clonus in real time combined with the immediate and reversible effects of stimulation highlight the potential of neuromodulation as a therapeutic tool for managing clonus. These data demonstrate preliminary efficacy of ES and DRS for clonus monitoring and treatment in 4 participants.

Damage to the brain or spinal cord, whether from traumatic spinal cord injury (SCI), progressive neurological conditions like multiple sclerosis (MS), or stroke, leads to similar symptoms across various neurological disorders. In cases of myelopathy due to either SCI or MS, individuals often report nonvoluntary muscle activity that leads to undesirable limb movement called spasticity. This occurs in 65% of patients with SCI,¹ 80% with MS,² and 21% with stroke.³ Features of spasticity include resistance to passive movement, clonus, cocontractions of agonist/antagonist muscle groups, poor motor control, and poor coordination with slowed movement.⁴ Spasticity and clonus occur due to upper motor neuron damage, reducing quality of life through gait dysfunction and undesirable nonvoluntary movement.^1,5

Clonus involves involuntary rhythmic contractions in any joint, but it often occurs in the plantar flexor muscles (5-8 Hz) due to hyperactive stretch reflexes.⁶ Clonus is tested by rapidly dorsiflexing the ankle and observing for rhythmic oscillations between dorsiflexion and plantar flexion. In healthy individuals, no reaction typically occurs, but those with upper motor neuron lesions exhibit rhythmic plantarflexion.⁷ Previous research has primarily characterized clonus through clinical assessments such as the Spinal Cord Assessment Tool for Spastic Reflexes (SCATS) test,⁸ kinematic assessments through smartphones,⁹ motion capture equipment,¹⁰ goniometers,¹¹ kinetic assessments such as foot reaction forces^10,12–14 or accelerometry,⁹ and electromyography (EMG).^{10–13,15,16} However, these existing biomarkers rely on clinical expertise alongside specialized kinematic and electrophysiological equipment that is time-consuming to implement and can suffer from low user compliance, especially when used outside of a laboratory setting, which limits their practicality in everyday contexts.¹⁷ Moreover, collecting EMG, kinetic, and kinematic data during daily activities— such as sleep or community engagement—poses significant challenges, highlighting the urgent need for more accessible and reliable biomarkers that can effectively guide therapeutic interventions in real-world scenarios.

Current treatments for spasticity include physical therapy, oral medications (e.g., baclofen, tizanidine), intramuscular Botulinum toxin injections, and intrathecal baclofen pumps. These options often have side effects, are short-lived, are effort intensive, and offer limited benefits for individuals with SCI and MS.^2,18 Clonus treatment options are similarly limited. Botulinum toxin type A may help, but it shows variable effects and potential adverse events.¹⁹ Antispasticity drugs have limited efficacy, with drawbacks like sedation and weakness. While functional electrical stimulation may reduce spasticity,²⁰ its impact on clonus is unclear, and stretching may harm locomotor function.²¹ Surgical procedures such as neurectomy²² may effectively reduce clonus, but this comes at the undesirable cost of irreversibly severing neural connections. Conversely, emerging treatments like cryoneurolysis²³ are surgical procedures with temporary effects, requiring repeated treatments at unpredictable intervals, and may also lead to undesirable weakness. Crucially, although clonus biomarkers have been studied in humans, no current interventions leverage clonus biomarkers to optimize real-time treatment.

Spinal cord stimulation (SCS) to restore lost functions following SCI has gained substantial traction in the field following reports of unprecedented improvements in motor function.^24,25 These reports have led to the resurgence of transcutaneous SCS (TS), epidural SCS (ES), and dorsal root ganglion stimulation (DRS) for other clinical signs associated with upper motor neuron damage, such as spasticity and/or clonus. TS, a neuromodulation tool that utilizes skin surface electrodes to enable dormant spinal circuitries below the level of injury, was shown to temporarily decrease spasticity and clonus in individuals with SCI.²⁶ However, due to the limitations of applying skin surface electrodes to a specific location, risks of improper electrode adhesion and subsequent skin breakdown, poor user compliance, and the relatively short effects of carryover, an implanted stimulating system may be more efficacious for precise clonus mitigation.

In contrast to traditional noninvasive recording and stimulation methods to monitor and treat clonus, our approach leverages percutaneously implanted SCS electrodes, providing a more reliable, less cumbersome, and consistent means of quantifying and treating clonus. This innovative strategy not only enhances the understanding of clonus characteristics through the analysis of time and frequency domain features across various recording modalities, but it also lays the groundwork for improved clinical management. In this prospective crossover case series study, we demonstrate that both ES and DRS effectively alleviate clonic activity in the lower extremities in participants with SCI or MS. Furthermore, to overcome the limitations of existing noninvasive clonus biomarkers in real-world scenarios, we utilize spinal cord electrophysiological recordings from stimulation leads along with EMG of the soleus muscle and accelerometry data as confirmatory biomarkers for clonus.

Methods

Participants

Participant characteristics are described in Table 1. Four male participants were enrolled, with 3 participants diagnosed with chronic SCI and 1 participant with secondary progressive MS. Participants over 21 years old were screened for disease diagnosis for greater than 1 year prior, intact spinal reflexes, and no active fractures or infections. Participants were included in this exploratory analysis if they exhibited clonus, as the study aimed to investigate the incidental finding of spinal cord stimulation's effect on clonus. Of the 11 participants enrolled following this discovery, a subset of 4 participants exhibiting clonus were included in the analysis. Those without clonus were excluded, as the intervention was not applicable to individuals without clonus. All participants underwent temporary percutaneous ES lead placement bilaterally over the lumbosacral region (T11-L1). Additionally, the participants with SCI underwent dorsal root ganglion stimulation (DRS) lead placement at the bilateral L4 dorsal roots.

Table 1.

Participant summary

Participant		Participant characteristics					SCS parameters
		Sex	Age	Years since diagnosis	NLI	AIS	Days of SCS	Lead location	Frequency, Hz	Amplitude, mA	Pulse width, μs
	SCI01	M	20s	13	C8	B	10	L4	80	1.1-2.2	260
	SCI02	M	30s	10	T7	C	10	T12	80	0.7	250
	SCI03	M	40s	3	T4	B	10	T11	80	2.2	250
	MS01	M	50s	10	EDSS 6.5	N/A	12	T11	80	6.0-14.0	250

Note: Colors correspond to participant-specific results in Figures 3 and 4. AIS = American Spinal Injury Impairment Scale; EDSS = Expanded Disability Status Scale; N/A = not applicable; NLI = neurologic level of injury; SCS = spinal cord stimulation.

Study design

Temporary percutaneous lead placement was conducted as part of 2 parent studies (ClinicalTrials. gov identifiers: NCT06019611, NCT04736849) aimed at enhancing reaching, standing, and stepping motor function in individuals with SCI and MS, respectively. Following one day of postsurgery recovery, participants completed task-specific motor rehabilitation training for 2 weeks (SCI) and 4 weeks (MS), respectively. Next, participants underwent 2 days of somatotopic mapping of stimulation parameters, followed by multiple days of reaching and weight-bearing activities, including standing and walking. Clonus was assessed 9 to 16 days post implantation using a prospective crossover study design.

Clonus was triggered manually in a controlled, reproducible manner by rapid, passive dorsiflexion of the foot. In a seated position, this was achieved by dropping the foot onto a ledge from an approximately 6-inch height (Figure 1B). Clonus duration and cycle count were assessed using surface EMG from the soleus. Two to 3 repeated trials were conducted under each of the 3 conditions and averaged for analysis. First, a baseline clonus assessment was performed with stimulation off. Second, ES or DRS was applied at submotor thresholds, as defined by the root mean square (RMS) signal below twice the baseline RMS. A pilot experiment on one participant was conducted to determine the minimum effective stimulation frequency. This pilot experiment resulted in clonus suppression at 80 Hz (eFigure 1). For all participants in this study, stimulation was provided at a frequency of 80 Hz, submotor threshold current amplitude (0.7-14.0 mA), and 250-260 μs pulse width with Abbott Octrode (ES) and Proclaim (DRS) trial leads (Abbott, Plano, TX) (Figure 1A). After ES or DRS was applied, a clonus assessment was performed with stimulation on. Third, a baseline clonus assessment retest was repeated with stimulation off. Three repeated trials were performed for each condition in a crossover study design with a rest period of at least 20 seconds between trials and between conditions. Stimulation was applied 5 to 30 seconds before the first “stimulation on” condition trial. Each trial lasted for a maximum of 60 seconds.

Figure 1.

Design and timeline of study-related activities. Clonus trials were completed in a seated position with EMG recordings of the soleus. (A) Intraoperative fluoroscopy and a schematic of epidural and dorsal root stimulation lead implantation (BioRender). (B) Initial clonus trials were completed without stimulation, followed by clonus trials with stimulation, and ending with retesting clonus trials without stimulation. (C) The foot was dropped from a short height onto a ledge to create rapid dorsiflexion of the ankle. Recordings were taken from the soleus EMG, inertial measurement unit placed mid-shank, and local field potentials from ES and DRG percutaneous leads.

Electromyography

The skin surface was shaved and prepped with 70% isopropanol pads (Covidien, Mansfield, MA). Then, bipolar pairs of Ag/AgCl skin surface electrodes (Cardinal Health, Dublin, OH) were placed 1 to 3 cm apart parallel to the muscle fiber direction over the soleus muscle. An Ag/AgCl ground electrode was placed on the skin surface on the anterior tibial crest approximately mid-tibia. Electrodes were connected to a 16-channel biosignal amplifier (g.tec medical engineering GmbH, Schiedlberg, Austria). The gain was set to 100 μV/V for all channels. Amplified skin surface EMG data was sampled at 4 kHz (AD Instruments, Dunedin, New Zealand).

Spinal cord field potentials

Field potentials were recorded from the Abbott Octrode (ES) and Proclaim (DRS) trial leads (Abbott, Plano, TX) using a Nomad Macro+Stim front end (Ripple Neuro, Salt Lake City, UT) in conjunction with the Trellis software package. Electrode potentials were recorded from the most rostral-medial contact with the most caudal-lateral contact of the ES and DRS electrodes serving as a reference, all of which shared a common ground with the EMG recording system. Data were collected at 7.5 kHz and digitized using a 16-bit analog-todigital converter with a 0.25 μV resolution.

Accelerometry

The inertial measurement unit (IMU) sensor (Opal, APDM Inc., Portland, OR) was placed on the lateral side of the shank, distal to the knee joint, at the head of the fibula, using a soft Velcro strap with the sensor's x-axis aligned with the long axis of the fibula. Data were collected at 128 Hz. The vector magnitude of the acceleration data was used for the quantification of clonus in this study.

Data processing

Recorded data were exported for post hoc analysis using MATLAB R2023a (The MathWorks Inc., Natick, MA). The EMG, ES, and DRS data were low pass filtered (500 Hz, 10th order Infinite Impulse Response, 0.1 passband ripple). An additional high pass filter was applied to the ES and DRS datasets to remove excessive low-frequency content noise (2.5 Hz, 10th order Infinite Impulse Response, 0.1 passband ripple). A 100-point (25 ms) RMS window was calculated across the EMG signal. Clonus events were identified as peaks in the RMS-windowed EMG signal exceeding 4X the standard deviation of a pretrigger quiescent period. A short-time Fourier transform was performed to visualize time series EMG, ES, DRS, and accelerometry data in the frequency domain. Spectrograms were produced by segmenting time series EMG, ES, DRS, and accelerometry data into partially overlapping intervals with a frequency resolution of 2 Hz and a time resolution of 1.28 seconds, applying a Kaiser window, computing the short-time Fourier transform, and graphically representing the resulting matrix. The signal-tonoise ratio (SNR) was calculated as 10 times the log of the ratio of 5 to 8 Hz band power²⁷ during clonus to that of the 10 seconds before clonus onset.

Statistical analysis

Differences in dependent variables (clonus duration, cycle count, EMG RMS amplitude, and cycle frequency outcome measures) between stimulation conditions (no stimulation and stimulation) were analyzed using separate Wilcoxon signed-rank tests. For each outcome measure, test statistics (Z), significance (P), and effect size (r) were reported. This nonparametric test was selected due to the small sample size and its conservative approach to detecting differences. Effect size was calculated as the Z value divided by the square root of the total number of observations (number of subjects × 2). Trial data were reported for all outcome measures for all participants except SCI01, for whom data from the no stimulation retest condition were not collected (eTable 1). This was because SCI01 was the first participant in whom this phenomenon was observed, and a standardized evaluation protocol, including the retest condition, had not yet been established at that time. Differences were considered statistically significant with P < .05. Statistical analyses were performed using SPSS Statistics software (SPSS Inc., Chicago, IL).

Results

Our goal was to characterize field potentials from ES and DRS electrodes to assess similarities with soleus EMG and accelerometry and their potential as clonus-specific biomarkers. All 4 participants experienced clonus when rapid dorsiflexion was mechanically evoked at baseline with no stimulation. Figure 2 presents the data from one participant (SCI03). This instance of dorsiflexion-evoked clonus lasted for 22 seconds and was observed in the time series data from the EMG, ES, DRS, and accelerometry recordings (Figure 2A). Dorsiflexion-evoked clonus was also associated with a notable spike in the frequency domain at 5.1 Hz (Figure 2B) and an increase from baseline in the band power between 5 and 8 Hz in biopotentials recorded from soleus EMG (SNR of 20.55dB), ES leads (SNR of 7.07dB), DRS leads (SNR of 10.07 dB), and accelerometer affixed to the leg (SNR of 35.19dB) (Figure 2C).

Figure 2.

Clonus biomarkers of participant 3 (SCI) for biosignals recorded from the soleus muscle, dorsal root ganglion (DRG) electrodes, epidural stimulation (ES) electrodes, and inertial measurement unit (IMU) accelerometry magnitude. Data are represented in the (A) time domain and (B) frequency domain and as a (C) time-frequency spectrogram. Inset horizontal scale bars in panel A represent 200 ms while vertical scale bars in A represent 100 mV for soleus EMG, 10 µV for DRS, 10 µV for ES, and 2 m/s² for accelerometry magnitude data.

SCS reduced the median mechanically evoked clonus duration from 40.77 seconds (range 10.5659.79) at baseline without stimulation to 1.57 seconds (range 0.73-2.57) with stimulation (Z = -1.826, P = .068, r = 0.65; Figure 3A). Similarly, clonus cycle count was reduced from a median of 208.75 cycles (range 51.43-329.21) at baseline to a median of 7.25 cycles (range 4.33-13.00) with stimulation (Z = -1.826, P = .068, r = 0.65; Figure 3B).

Figure 3.

Clonus (A) duration and (B) cycle count during baseline (no stimulation), stimulation, and baseline retest (no stimulation retest) conditions. The box plot central lines represent the median. Colors represent individual subjects. Orange squares represent the participant with multiple sclerosis (MS) while all other colors represent participants with spinal cord injury (SCI).

SCS also lowered dorsiflexion-evoked clonus soleus EMG RMS amplitude, reducing it from 3650.51 μV (range 815.02-6786.03) without stimulation to 189.31 μV (range 36.20-541.01) with stimulation (Z = -1.826, P = .068, r = 0.65; Figure 4A). However, cycle frequency remained unchanged, with medians of 5.09 Hz (range 4.815.56) without stimulation and 4.90 Hz (range 4.566.09) with stimulation (Z = -0.365, P = .715, r = 0.13; Figure 4B).

Figure 4.

Clonus (A) RMS EMG amplitude and (B) frequency during baseline (no stimulation) and stimulation conditions. The box plot central lines represent the median. Colors represent individual participants as indicated in Table 1.

Additionally, a baseline dorsiflexion retest without stimulation demonstrated sustained clonus suppression (eFigure 2, eTable 1). The median clonus duration during the retest was 1.79 seconds (range 0.588-4.958), with a median cycle count of 9.67 (range 3.67-22.00) and a median RMS amplitude of 106.99 μV (range 31.05-734.57), similar to those observed during stimulation. Clonus frequency remained consistent across baseline, stimulation, and retest conditions (eFigure 2).

Discussion

The findings reported herein demonstrate that ES and DRS leads can record novel invasive biomarkers of clonus (Figure 2). Additionally, we report preliminary evidence that ES and DRS reduce clonus duration and severity in individuals with SCI or progressive MS when stimulation is applied (Figures 3 and 4). Current treatments for clonus are often limited in their efficacy and can pose significant side effects.^16,18,28 Thus, the exploration of neuromodulation as a therapeutic tool offers a promising alternative for managing this condition. In this study, ES and DRS were shown to reduce median values clonus duration and cycle count, providing an immediate and durable effect on clonic activity. The reduction of clonus was consistent across all 4 participants, with decreases in median clonus duration (Figure 3), cycle count (Figure 3), and EMG RMS amplitude (Figure 4), accompanied by a moderate effect size suggesting a notable impact of stimulation. Importantly, these effects were sustained, as demonstrated by the baseline retest condition, which suggests some degree of carryover immediately after stimulation cessation (eFigure 2).

Mechanistic insights and comparison with prior studies

A widely accepted hypothesis for the mechanism of clonus is that hyperactive stretch reflexes result from self-excitation, which creates a feedback loop with a harmonic frequency influenced by the propagation rate of information through 1A sensory afferents, synapses in the spinal cord, and alpha motor neuron efferents to activate muscles.⁶ An alternative explanation is that clonus is driven by central generator activity within the lumbosacral enlargement,²⁹ which arises from specific peripheral events and leads to rhythmic stimulation of the lower motor neurons.⁶ The reduction in clonus achieved through ES and DRS supports the hypothesis of hyperactive stretch reflexes. Stimulation at T12 may influence clonus by modulating ascending and descending pathways, thereby disrupting hyperactive reflex arcs. While the soleus muscle is innervated by tibial nerve fibers that originate from the L4-S3 spinal segments, effects of high-frequency stimulation at the T12 vertebral level likely propagate across spinal structures and in turn influence nearby neural networks that reside within the conus medullaris. Despite ES and DRS acting in different anatomical locations, with ES implanted at T12 and DRS implanted at L4, both were effective at reducing clonus duration, cycle count, and magnitude. These findings align with the “information lesion” theory in neuromodulation, where high-frequency stimulation (80-130 Hz) disrupts pathological neural signals to provide symptom relief.³⁰ Although this study did not systematically investigate it, frequencies below 80 Hz were ineffective in reducing clonus (eFigure 1). Future studies should explore a wider range of frequency and amplitude parameters. In contrast to previous reports showing that intrathecal baclofen reduces clonus frequency, this study found that ES and DRS did not affect clonus frequency (Figure 4B). This difference may be attributed to their distinct mechanisms of action: Baclofen may act partially by disrupting the central pattern generator to reduce frequency, while SCS likely modulates clonus by suppressing transmission of sensory signals that activate pathologically imbalanced spinal networks to produce spastic activity.

The present findings add to the growing body of literature supporting the therapeutic potential of neuromodulation in individuals with SCI and MS. Studies utilizing TS have also demonstrated temporary reductions in spasticity and clonus, with improvements in motor function and quality of life in SCI patients.²⁶ However, TS poses challenges, such as poor electrode adhesion, skin breakdown, and short-lasting effects. The percutaneous, implanted approach used in this study mitigates these limitations by providing a safe, consistent, and “hands-off ” user experience. Although invasive procedures like ES and DRS provide targeted neuromodulation, they carry inherent risks, including infection, lead migration, device-related complications, and procedural discomfort. These risks should be weighed against the potential therapeutic benefits when considering this approach for clinical application.

Roles of biomarkers and SCS in clonus management

Leveraging spinal cord stimulators to record spinal cord potentials as a clonus biomarker offers several advantages over standalone external systems. By integrating both diagnostic and therapeutic capabilities into a single implanted platform, we eliminate the need for bulky external electrophysiological equipment, which is often cumbersome and restricted to controlled laboratory settings. Additionally, real-time biomarker monitoring through implanted leads facilitates the development of adaptive, closed-loop systems that can automate spinal cord stimulation in response to detected clonus events. This innovation has the potential to reduce the need for constant clinical oversight and address the limitations of current clonus assessment methods, paving the way for broader clinical adoption and streamlined management of clonus.

A key strength of this study is the use of multiple biomarkers, including surface EMG, accelerometry, and novel local field potentials from the ES and DRS leads, to objectively and consistently quantify clonus. These objective biomarkers offer rigorous real-time clonus quantification compared to clinical measures based on subjective observation or patient reports. In contrast to wearable technologies, which can be cumbersome and difficult to integrate into everyday activities, implanted devices such as those used in this study provide a more pragmatic option. The ability to continuously monitor spinal cord biomarkers and apply neuromodulation in real-world settings could be integrated into adaptive closed-loop clonus suppression systems. Such systems would continuously monitor clonus activity and adjust stimulation parameters automatically, reducing the need for manual intervention and enhancing clinical efficiency. Spinal cord recordings may be curated as a library of patient data for identifying future biomarkers and optimizing stimulation parameters.³¹ This approach could significantly improve the quality of life for individuals with SCI and MS by reducing the impact of clonus on overall health and activities of daily living.

ES and DRS offer immediate clonus relief without the side effects of pharmaceuticals.³ Additionally, the sustained effects observed in the no stimulation retest condition (eFigure 2 and eTable 1) suggest the potential for longer term benefits, which may be further explored in future studies. There are several challenges to the widespread clinical implementation of ES and DRS for clonus management. First, the optimal stimulation settings remain unclear, though the biomarkers studied here may help refine these parameters. Developing an implantable closed-loop clonus suppression system also presents difficulties, such as SCS artifacts producing harmonics that interfere with the 5 to 8 Hz clonus frequency range.²⁷ This issue could be addressed by limiting SCS duration or applying advanced signal processing methods like empirical mode decomposition.³²

Limitations

This study has several limitations that impact the interpretation of its findings. First, the small sample size of 4 participants must be considered when generalizing the results. Although statistical significance was not reached with the Wilcoxon signed-rank test, moderate effect sizes were evident when comparing stimulation to baseline conditions across clonus duration, cycle count, and RMS amplitude outcome measures. Given these moderate effect sizes, it is likely that increasing the sample size would achieve statistical significance. A larger, controlled study with a diverse cohort (e.g., SCI, MS, stroke) and a preimplantation baseline timepoint is needed to validate biomarker reliability and intervention efficacy across conditions with clonus, ensuring the observed outcomes are attributable to stimulation to support broader clinical applicability. Moreover, synchronized EMG, ES, DRS, and accelerometry data were collected from only 1 of the 4 participants. Second, the study is neither blinded nor randomized, as it is not feasible to control for the unavoidable stimulation-induced paresthesias with reasonable sham interventions.³³ While the 2 participants with the motor complete American Spinal Injury Association Impairment Scale (AIS) B classification have no preserved motor function below the neurological level of injury³⁴ and are likely unable to introduce bias into the data reported, this is not the case with the AIS C or MS participants.

Although the acute effects are promising, their long-term sustainability is uncertain. Variability in clonus induction and potential selection bias may limit broader applicability. Repeated testing alone can reduce spasticity, though this was partially addressed in our design, as no consistent trial order effect was observed in the baseline no stimulation condition (Table S1). EMG, local field potentials, and accelerometry were used to measure clonus, but these techniques may not capture the full complexity of the neuromuscular activity involved. The source of spinal field cord potentials remains unclear, as they may reflect a mix of stretch reflexes, motor efferents, motion artifacts, paraspinal EMG, or distant EMG signals. Detailed signal phase analysis could potentially clarify their origins, though this is beyond the scope of this article. Lastly, the study did not include comprehensive assessments of functional implications, underscoring the need for larger, controlled studies with extended follow-up and thorough functional evaluations in real-world environments to validate these preliminary findings. Addressing these limitations will be crucial for advancing toward a reliable, implantable system for clonus management.

Conclusion

This case series study presents evidence that both ES and DRS are effective in reducing clonus in individuals with SCI and MS. The use of objective biomarkers, including spinal cord potentials, to quantify clonus in real time, combined with the immediate and reversible effects of stimulation, highlights the potential of neuromodulation as a therapeutic tool for managing clonus. Future research should aim to refine stimulation parameters, explore long-term outcomes, and develop closed-loop systems for more personalized and effective clonus management. Ultimately, these advancements could lead to an improved quality of life for individuals with debilitating clonic spasticity.

Funding Statement

Financial Support Support for this article was provided by the National Institute of Neurological Disorders and Stroke and the Mayo Clinic Center for Multiple Sclerosis and Autoimmune Neurology. Abbott provided devices and technical support.