TABLE 1.

Details of key studies examining the mechanisms of electrical stimulation.

References	Study details	Key outcome	Stimulation parameters	Acute/immediate functional outcomes	Functional outcomes at conclusion	Adverse events	Analysis methods for mechanisms	Mechanism explored
Epidural electrical stimulation
Angeli et al. (2014)	Clinical study: –4 Males 23.8–32.8 years old. –2.3–4.2 years post-SCI. –Injury levels: C7, C6–C7, T5, and T5–T6. –AIS grade: –A (2) and B (2).	Motor function	16 electrode array at L1 – S2 –25 Hz or 30 Hz. –0.8–9 V. –Participant specific. –450 μs pulse width. –Continuous stimulation. –1 h daily.	3/4 participants could voluntarily execute movements within 4–11 days of stimulation. Final participant showed voluntary movement after 7 months of stimulation but was not assessed earlier.	Voluntary motor function continued to improve in all participants with long-term rehabilitative training and EES.	None reported	Electrophysiology	Descending neuroplasticity: –Modulation of motor output in response to visual and auditory cues.
Angeli et al. (2018)	Clinical study: –4 participants 22–32 years old. –2.5–3.3 years post-SCI. –Injury levels: C5, T1, and T4 (2). –AIS grade: –A (2) and B (2)	Motor function	16 electrode array at L1 – S2 –20–50 Hz. –1–5.7 V. –Participant specific. –450 μs pulse width. –1–2 training sessions daily with 1 h of stimulation.	Not reported	Long term training and stimulation facilitated intentional overground walking in 2/4 participants. All four participants achieved independent standing and trunk stability. –Execution of walking only with active EES and participant’s intention to walk.	Drainage of surgery site (1). Ankle oedema (1). Hip fracture during training (1).	Motor and sensory exam.	Descending neuroplasticity: –Walking only occurred with intent of participant.
Kathe et al. (2022)	Clinical study: –9 males 23–53 years old. –1 year, 3 months – 14 years, and 3 months post-SCI. –Injury levels: T3, T4, T6, and unclear for 6 participants. –AIS grade: A (2), B (1), C (5), and D (1). –Preclinical study: –Male or female mice severe contusion (95 kdyne) T8/T9 SCI. –Endpoint at 30 min or after 4 weeks rehabilitation	Motor function. –Mechanisms (preclinical).	Clinical study: 16 electrode array: specify 5-6-5 medtronic paddle lead or a custom design (ONWARD medical) –parameters personalized. –Preclinical study: Electrodes at L2 and S1 Continuous stimulation: –40 Hz, 0.2 ms pulses, and 50–300μA. High frequency burst stimulation: –10 ms busts of 0.2 ms pulses. –50–300 μA. –600 Hz. –30 Hz modulating frequency.	All participants immediately improved/regained walking with robotic interface support. Participants exerted volitional control over stepping amplitude.	5 months of EES rehab improved weight-bearing capacities, outdoor walking with EES and assistive device.	None reported	Neuroimaging (PET). Electrophysiology. Tract tracing. Immunohistology. Single nucleus RNA sequencing. Spatial Transcriptomics.	Descending neuroplasticity: –Decreased neuronal activity within lumbar spinal cord. –Identified neuronal population necessary for walking with EES.
Kinfe et al. (2017)	Clinical study: –12 people with Failed Back Surgery Syndrome (6 females and 5 males; 44–76 years old). –Blood samples obtained at baseline and at 3 months of stimulation.	Pain	16 contact paddle lead at Th8-9 –40 Hz burst rate. –500 Hz intra-burst rate. –1 ms pulse width. –0.15–1.6 mA intensity (participant specific). –Three months of stimulation.	Not reported	3 months of burst stimulation significantly decreased back and leg pain intensity (visual analog scale).	No serious adverse events reported. –Temporary skin irritation at implanted pulse generator site (3).	Enzyme-linked immunosorbent assays (ELISA)	Neuroinflammation: –Elevated serum IL-10 expression.
Asboth et al. (2018)	Preclinical study: –Female rat severe contusion (255.5 kdyne) T8/9 SCI. –Male or female C57BL/6 mice severe contusion (90 kdyne) T8-9 SCI. –Endpoint at 12 weeks post-SCI.	Motor function	Electrodes at L2 and S1 –Continuous stimulation at 40 Hz. –0.2 ms pulse width. –100–300 μA. –Training 6 days per week for 40 min. –Combinatorial treatment: EES paired with agonists to serotonergic and dopaminergic receptors (electrochemical neuromodulation).	Electrochemical neuromodulation immediately restored involuntary locomotion on treadmill but not voluntary.	100% of rats with electrochemical neuromodulation and 88% of rats with EES only regained weight-bearing locomotion. Trained rats adapt limb kinematics for different tasks.	None reported	Behavioral testing. Electrophysiology. Tract tracing. Optogenetics. Immunohistology.	Descending neuroplasticity: –Cortico-reticulo-spinal networks facilitate descending control.
Ghorbani et al. (2020)	Preclinical study: –Male rat moderate contusion (150 kdyne) T10 SCI. –Endpoint at 15 days post-SCI.	Mechanisms	Electrodes at upper injury site (T10) –Subthreshold stimulation at 100 Hz, 0.1 ms. 0.3–0.6 mA. –1 h stimulation daily for 14 days.	Functional outcomes not assessed.	Functional outcomes not assessed.	None reported.	Immunoblotting. Immunohistology	Neurotrophic factors: –Elevated BDNF in homogenized spinal cord samples.
Li et al. (2020)	Preclinical study: –Female rat contusion (25 mm 20 g weight drop) T10 SCI. –Endpoints at 7 and 28 days post-SCI.	Motor function	Custom Array at T10 – T13 –Stimulation at 90% motor threshold. –50 Hz. –200 μs pulse width, 0.045 mA. –30 min stimulation per day for 1 week.	Significantly improved BBB score at 7 days post-SCI.	Significantly improved BBB scores at 14, 21, and 28 days post-SCI.	None reported.	Semi quantitative RT-PCR. Immunoblotting. Immunofluorescence. Histology.	Glial cells: –Reduced oligodendrocyte and myelin loss.
Sivanesan et al. (2019)	Preclinical study: –Paclitaxel-induced peripheral neuropathy study in male rats. –Endpoint at day 30.	Pain	Quadripolar medtronic SCS electrode. –T13 to L1 spinal cord level. –50 Hz, 0.2 ms, constant current, and 80% motor threshold. –6–8 h daily for 14 days.	Stimulation before and during paclitaxel administration alleviated the development of neuropathic pain.	Neuropathic pain relief extended for at least 2 weeks after stimulation.	Spinal cord injury from implant. Poor lead placement Damaged electrodes	RNA-seq	Neuroinflammation: –Upregulating of genes involved in inflammatory processes, particularly astrocytes and microglia.
Stephens et al. (2018)	Preclinical study: –Chronic constriction injury of sciatic nerve in male and female rats. –Endpoint at 39 days post injury.	Pain	Quadripolar medtronic SCS electrode. –T13 to L1 spinal cord level. –50 Hz, 0.2 ms, constant current, and 80% motor threshold. –120 min/session, twice per day for 3 days.	Peak neuropathic pain relief within 60–90 min of commencing stimulation. Withdrawal responses returned to baseline within 30 min of ceasing stimulation.	Long term outcomes not assessed.	Impaired motor function after implantation (3).	RNA-seq.	Neuroinflammation: –Upregulating of genes involved in inflammatory processes.
Thornton et al. (2018)	Preclinical study: –Female rat T7/T8 complete transection injury. –Endpoint at 6–7 months post-SCI.	Motor. –Mechanisms.	Electrodes at L2 and S1 –40 Hz at 95% of threshold. –Training for 20 mins per day, 3 days per week for 6 months. –Combinatorial treatment: cellular transplant of OECs or FBs.	Short term outcomes not assessed.	No difference in function between OEC and FB implanted animals either with or without EES.	None reported	Immunohistology.	Axonal regeneration: –Greater presence of NF and 5-HT positive axons into lesion.
Vallejo et al. (2020)	Preclinical study: –Male rat spared nerve injury model. –Endpoint after 48 h of stimulation.	Pain	Four electrode SCS lead (Heraeus medical). –Low rate stimulation: –50 Hz, 150 μs pulse width, 0.03–0.09mA (70% motor threshold). High rate stimulation: –1,200 Hz, 50 μs pulse width, 0.02–0.010 mA (70% motor threshold). DTMP stimulation: –Multiplexed charges in 20–1,200 Hz range, 500 μs max pulse width, 0.03–0.10 mA (70% motor threshold).	Significantly improved mechanical hypersensitivity with all stimulation forms. DTMP significantly improved hot and cold thermal hypersensitivity.	Long term outcomes not assessed.	None reported	RNA-seq	Neuroinflammation –Upregulating of genes involved in inflammatory processes. –Decreased GFAP, C1qa, Casp1, and Tal1 expression.
Peripheral nerve stimulation
Perez et al. (2003)	Clinical study: –Twenty health volunteers (13 men and 7 women). Mean age 29.4 years old.	Motor function.	30 min of common peroneal nerve Patterned stimulation: –10 pulses at 100 hz every 1.5 s. –1 ms pulse width. Uniform stimulation: –10,000 pulses total. –150 ms even spacing of pulses.	Patterned stimulation enhanced reciprocal inhibition for at least 5 min after treatment. Effect was temporary and returned to baseline by 20 min.	Long term outcomes not assessed.	None reported.	Electrophysiology.	Local neuroplasticity: –Improved strength of reciprocal inhibition.
Ayanwuyi et al. (2022)	Preclinical study: –Male rat T12 unilateral dorsal column focal demyelination. –Endpoint at 7 or 14 days post-demyelination.	Mechanisms	Sciatic nerve stimulation –20 Hz continuous stimulation for 1 h. –100 ms pulse width, 3 V.	Functional outcomes not reported.	Functional outcomes not reported.	None reported.	Immuno-fluorescence.	Glial cells: –Increased myelin basic protein expression, increased oligodendrocyte numbers. –Neuroinflammation: –Microglia/macrophage polarization to pro-repair phenotype.
Gajewska-Woźniak et al. (2016)	Preclinical study: Healthy male rats	Mechanisms	Tibial nerve –Continuous burst of 3 pulses every 25 ms. –200 μs pulse width. –4 ms inter-pulse interval. –4 min × 20 min daily for 7 days.	Functional outcomes not reported.	Functional outcomes not reported.	None reported.	Retrograde labeling. Electrophysiology Immuno-fluorescence	Local neuroplasticity: –Increased glutaminergic and cholinergic inputs to alpha motoneurons.
Goganau et al. (2018)	Preclinical study: –Female rat C4 dorsal column wire knife lesion. –Endpoint at 4 weeks post-SCI.	Mechanisms. Pain.	Sciatic nerve –20 Hz stimulation. –0.2 ms pulse duration. –1 h stimulation post-SCI.	Electrode placement was not associated with neuropathic pain.	Fine touch did not recover with stimulation.	No adverse effects of pain.	Immunohistology. Tract tracing.	Axonal regeneration: –Increased neurite length after C4 dorsal column injury.
Hahm et al. (2014)	Preclinical study: –Male rat severe contusion (50 mm 10 g weight drop) T12 SCI. –Endpoint at 5 weeks post-SCI.	Spasticity	Bilateral hindlimb stimulation: transcutaneous stimulation at 4 Hz or 100 Hz. –250 μs pulse width. –50% or 90% motor threshold. –30 min single session 5 weeks post-SCI.	High frequency stimulation alleviated spasticity from 20 to 50 min after application.	Long term effects not assessed.	None reported.	Immunohistology.	Neuroinflammation: –Reduced activated microglia expression.
Hayashi et al. (2019)	Preclinical study: –Male rat severe contusion (25 mm 20 g weight drop) T9 SCI. –Endpoint at 1 or 4 weeks post-SCI.	Motor function.	Bilateral hindlimb stimulation –2 Hz at 10 mA. –10 min of stimulation 5 days per week, 4 weeks total.	No improvement to BBB score at 1 week post-SCI.	BBB score significantly higher from 2 to 4 weeks post-SCI for stimulation animals. –Significantly higher inclined plane angle at 4 weeks post-SCI for stimulated animals.	None reported	Histology. TUNEL staining. Immunoblotting. Immunohistology. ELISA.	Neurotrophic factors: –Increased levels of BDNF in spinal cord.
Matsuo et al. (2014)	Preclinical study: –Spared nerve injury in male jcl:ICR mice. –Endpoint 8 days post-injury.	Pain.	Electrodes over L1 to L6 dorsal rami. –Transcutaneous stimulation at 100 Hz. –Sub-motor threshold Intensity for 30 min.	Early stimulation significantly reduced mechanical hyperalgesia from days 3–7 of stimulation. –Significantly reduced thermal hyperalgesia at days 6 and 7.	Late stimulation (1 or 2 weeks post-injury) did not alleviate neuropathic pain.	None reported.	Immunohistology. Immunoblotting Flow cytometry	Neuroinflammation: –Decreased microglia and astrocyte activation and pro-inflammatory cytokines.
Udina et al. (2008)	Preclinical study: –Female rat T8 dorsal funiculus lesion. –Endpoint at 15 weeks post-SCI.	Mechanisms.	Sciatic nerve –20 Hz at 0.02 ms pulse width or 200 Hz stimulation for 25 ms every 250 ms. –72,000 pulses per hour. –2× motor threshold. –0.02 ms pulse width. – 1 h stimulation post-SCI.	Functional outcomes not reported.	Functional outcomes not reported.	None reported.	Tract tracing. ELISA. Immunohistology.	Axonal regeneration: –Increased axonal outgrowth into T8 dorsal column injury. –Neurotrophic factors: –Increase in cAMP in lumbar dorsal root ganglion.
Wenjin et al. (2011)	Preclinical study: –Female rat sciatic nerve transection model. –Endpoint at 1 week post-injury.	Mechanisms.	Sciatic nerve –Continuous 20 Hz stimulation. –100 μs pulse width and 3–5 V. – 1 h of total stimulation.	Functional outcomes not reported.	Functional outcomes not reported.	None reported.	Immuno-fluorescence.	Neurotrophic factors: –Increased levels of BDNF in spinal cord neurons.
Wong et al. (2022)	Preclinical study: –Male rat L5 nerve root ligation. –Endpoint at 7 days after injury.	Pain.	Sciatic nerve stimulation –2, 20, or 60 Hz. –200 μs square wave pulses, pulse trains separated by 8 s. –1–10 mA intensity. –1 h single stimulation session (L5 nerve root ligation model).	2 and 20 Hz stimulation alleviated mechanical and thermal hypersensitivity from day 1 to 7 post-injury.	Long term outcomes not assessed.	None reported.	Immuno-fluorescence. Immunoblotting.	Neuroinflammation: –Decreased microglia and astrocyte density, decreased inflammatory cytokines.
Functional electrical stimulation
Bakkum et al. (2015)	Clinical study: –FES cycling: 9 males 40–64 years old. –10–34 years since SCI. –Injury levels: C3, C6, C7 (2), T1, T6, T8, T9, and T10. –AIS grade: –A (6), B (1), and C (2).	Mechanisms. General health outcome.	FES cycling –Surface electrodes over quadriceps, gluteal and hamstring muscles. –50 Hz stimulation for all muscles. –Adjustable intensity with maximum of 140 mA. –16 weeks exercise program, 18–32 mins cycling time.	Short term outcomes not assessed.	Both FES cycling and control hand cycling improved inflammatory status, visceral adiposity, and metabolic syndrome symptoms.	None reported, but high dropout rate (46%) possible due to time intensive training program.	ELISA	Neuroinflammation: –Reduced CRP, IL-6, and IL-6/IL-10 ratio.
Griffin et al. (2009)	Clinical study: –18 participants (13 male and 5 female) aged 27–56 years old (mean = 40). –1–53 years post-injury (mean = 11). –Injury levels: C4 (5), C5, C8 (2), T2, T3, T4 (3), T5 (2), T6, and T7(2). –5 complete injuries, 13 incomplete.	Motor and sensory score	FES cycling –Surface electrodes over quadriceps, gluteal and hamstring muscles. –50 Hz, <140 mA. –2–3 times weekly, 10 weeks.	Short term outcomes not assessed.	Lower extremity ASIA scores and motor and sensory components of the ASIA test improved with FES training.	None reported.	ELISA	Neuroinflammation: –Reduced CRP, IL-6, and TNF-α expression.
Piazza et al. (2017)	Clinical study: –10 participants (6 males and 4 females) aged 32–72 (mean = 48.1). –Time since injury: 12 to 43 months (mean = 32). –Injury levels: C4 (9) and C6. –AIS grades: C (6) and D (4).	Motor function.	10 min FES cycling –200 Hz stimulation. –1 ms pulse width. –Intensity just below generation of visible muscle contractions.	Increase in H-reflex excitability after stimulation.	Long term outcomes not assessed.	None reported.	Electrophysiology.	Local neuroplasticity: –Improved H reflex.
Becker et al. (2010)	Preclinical study: –Female rat suction-ablation T9 SCI. –Endpoint at 36 or 43 days post-SCI.	Mechanisms	Common peroneal nerve –20 Hz stimulation. –200 μs pulse width and 3V. –Alternated 1 s on/off for each leg. –Stimulation 3 times a day for 1 h	Functional outcomes not assessed.	Functional outcomes not assessed.	None reported.	Immunohistology.	Glial cells: –Increased progenitor cell birth and differentiation into oligodendrocytes.