Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Transspinal stimulation increases motoneuron output of multiple segments in human spinal cord injury

  • Lynda M. Murray,

    Roles Data curation, Formal analysis, Investigation, Validation, Writing – original draft, Writing – review & editing

    Affiliation Klab4Recovery Research Laboratory, Department of Physical Therapy, College of Staten Island, The City University of New York, Staten Island, New York, United States of America

    ⨯
  • Maria Knikou

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Writing – original draft, Writing – review & editing

    Maria.Knikou@csi.cuny.edu

    Affiliations Klab4Recovery Research Laboratory, Department of Physical Therapy, College of Staten Island, The City University of New York, Staten Island, New York, United States of America, PhD Program in Biology and Collaborative Neuroscience Program, Graduate Center of The City University of New York, New York, New York, United States of America

    ⨯

Abstract

Targeted neuromodulation strategies that strengthen neuronal activity are in great need for restoring sensorimotor function after chronic spinal cord injury (SCI). In this study, we established changes in the motoneuron output of individuals with and without SCI after repeated noninvasive transspinal stimulation at rest over the thoracolumbar enlargement, the spinal location of leg motor circuits. Cases of motor incomplete and complete SCI were included to delineate potential differences when corticospinal motor drive is minimal. All 10 SCI and 10 healthy control subjects received daily monophasic transspinal stimuli of 1-ms duration at 0.2 Hz at right soleus transspinal evoked potential (TEP) subthreshold and suprathreshold intensities at rest. Before and two days after cessation of transspinal stimulation, we determined changes in TEP recruitment input-output curves, TEP amplitude at stimulation frequencies of 0.1, 0.125, 0.2, 0.33 and 1.0 Hz, and TEP postactivation depression upon transspinal paired stimuli at interstimulus intervals of 60, 100, 300, and 500 ms. TEPs were recorded at rest from bilateral ankle and knee flexor/extensor muscles. Repeated transspinal stimulation increased the motoneuron output over multiple segments. In control and complete SCI subjects, motoneuron output increased for knee muscles, while in motor incomplete SCI subjects motoneuron output increased for both ankle and knee muscles. In control subjects, TEPs homosynaptic and postactivation depression were present at baseline, and were potentiated for the distal ankle or knee flexor muscles. TEPs homosynaptic and postactivation depression at baseline depended on the completeness of the SCI, with minimal changes observed after transspinal stimulation. These results indicate that repeated transspinal stimulation increases spinal motoneuron responsiveness of ankle and knee muscles in the injured human spinal cord, and thus can promote motor recovery. This noninvasive neuromodulation method is a promising modality for promoting functional neuroplasticity after SCI.

Citation: Murray LM, Knikou M (2019) Transspinal stimulation increases motoneuron output of multiple segments in human spinal cord injury. PLoS ONE 14(3): e0213696. https://doi.org/10.1371/journal.pone.0213696

Editor: Antal Nógrádi, Szegedi Tudomanyegyetem, HUNGARY

Received: December 15, 2018; Accepted: February 26, 2019; Published: March 7, 2019

Copyright: © 2019 Murray, Knikou. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript.

Funding: This research was supported by a grant from the Craig H. Neilsen Foundation [Grant No. 339705] awarded to Maria Knikou. The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Targeted neuromodulation strategies that strengthen neuronal activity are in great need for restoring sensorimotor function after chronic spinal cord injury (SCI). Within this concept, several therapeutic approaches promoting neuromodulation and thereby neuroplasticity have been adapted over the last few decades [1–6]. One of these methods includes spinal cord stimulation delivered epidurally or transcutaneously to the lumbar spinal region, the location of the leg motor circuits. Evidence supports for some recovery of standing and walking ability during or after epidural delivery of electrical current to the spinal cord of individuals with complete SCI that received concomitantly a variable number of locomotor training sessions [7–14]. Epidural stimulation also promoted bladder control, cardiovascular function, respiration and cough in human SCI [14–18]. Similarly, transcutaneous spinal cord (termed here transspinal) stimulation at low and high frequencies strengthens corticomotoneuronal connectivity, decreases hyperreflexia, improves bladder function, seated postural control, and autonomic cardiovascular function [19–23]. Step-like movements produced by transspinal stimulation [24], further support its ability to modulate spinal locomotor circuits in persons with paralysis. However, the effect of repeated low-frequency transspinal stimulation on the responsiveness of spinal motoneurons in individuals with SCI remains largely undetermined.

At the neuronal level, transspinal stimulation induces synaptic activity and intrinsic polarization of motoneurons [25], and decreases the magnitude of dendritic persistent inward currents that are related to hyperreflexia and spasms after SCI [26–28]. Further, transspinal stimulation activates receptors required for long-term potentiation-like actions of neurotrophins that mediate plasticity [29,30]. At the circuitry level, single-pulse transspinal stimulation generates transspinal evoked potentials (TEPs) simultaneously in bilateral lower limb muscles in individuals with and without SCI with distinct neurophysiological characteristics regarding their latency, duration, shape, and spinal integration [19,24,31–34]. TEPs exhibit a sigmoid function at increasing stimulation intensities, which coincides with bilateral leg extension, supporting the depolarization of motoneurons across multiple spinal segments that has been linked to stimulation-induced formation and retention of motor memories [35].

Collectively, the objectives of this study were to establish the effects of daily low-frequency transspinal stimulation on spinal motoneuron output and spinal inhibition of multiple spinal segments. We hypothesized that low-frequency transspinal stimulation at levels that produces motoneuron depolarization over multiple segments strengthens spinal connections and thus increases motoneuron output. We further hypothesized that motoneuron responsiveness will be greater in individuals with incomplete SCI compared to that observed in individuals with complete SCI. To test our hypotheses, the TEP recruitment input-output curves, TEP depression at low stimulation frequencies, termed here homosynaptic depression, and TEP depression in response to transspinal paired stimuli, termed here postactivation depression, recorded bilateral from ankle and knee flexor/extensor muscles were assessed at rest before and two days after an average of 13 sessions of transspinal stimulation delivered at 0.2 Hz in people with and without SCI.

Materials and methods

Participants

All experimental and stimulation procedures were conducted in compliance with the Declaration of Helsinki after full Institutional Review Board (IRB) approval by the City University of New York IRB committee (IRB Number: 515055). A written informed consent was obtained from all participants before study enrollment. Participants were instructed to refrain from alcohol and caffeine consumption at least 24 hours before testing, and be cannabis-free for more than one week.

Ten individuals with chronic SCI (Table 1) and 10 individuals without SCI (5 female; 30.9 ± 14 years, mean ± SD) were enrolled in the study. Individuals with SCI were recruited based on the following criteria: age 18–65 years; chronic injury (more than 6 months after injury); and able to tolerate lying supine or sitting upright for at least 1 hour. Individuals with SCI were excluded if they presented with concomitant traumatic brain injury or stroke, cognitive impairment, not medically stable, or had contraindications to spinal stimulation (presence of an implanted object such as a pacemaker or pump). Five individuals with chronic SCI had a neurological deficit grade D on the American Spinal Injury Association Impairment Scale (AIS), 1 had AIS C, 2 had AIS B, and 2 had AIS A. The level of SCI ranged from C4 to T11. Individuals with complete SCI were included in order to assess potential changes when corticomotoneuronal control is minimal. Individuals without SCI had no history of neurological or musculoskeletal disorders.

thumbnail
Download:
Table 1. Characteristics and demographics of individuals with spinal cord injury (SCI).

https://doi.org/10.1371/journal.pone.0213696.t001

Noninvasive transspinal stimulation

The T10 spinous process was identified via palpation and in consolidation with anatomical landmarks. A single cathode electrode (Uni-PatchTM EP84169, 10.2 × 5.1 cm2, MA, USA) was placed along the vertebrae equally between the left and right paravertebral sides. The electrode covered from T10 to L1-L2 vertebral levels. These vertebral levels correspond to L1 and S2 spinal segments and thus to the segmental innervation of the muscles from which action potentials were recorded in this study. A pair of interconnected reusable self-adhered anode electrodes (same type as the cathode), were placed on either side of the umbilicus or bilaterally on the iliac crests depending on the participant’s level of comfort or if the stimulation caused bladder discomfort [31,34]. The cathode and anode electrodes were connected to a constant current stimulator (DS7A, Digitimer, UK) that was triggered by Spike 2 scripts (Cambridge Electronics Design Ltd., UK). Optimal electrode placement was based on presence of TEPs in both proximal and distal lower limb muscles at low stimulation intensities, and presence of TEP depression upon paired transspinal stimuli. Once the optimal location was identified, the electrodes were affixed to the skin via Tegaderm transparent film (3M Healthcare, St Paul, Minnesota, USA). After each session, the stimulated skin area was marked by non-toxic skin pen and covered by Tegaderm film to ensure consistency of stimulation site across sessions.

During the intervention, transspinal stimulation was delivered on weekdays excluding holidays at 0.2 Hz with a single monophasic square-wave pulse of 1 ms duration while the participant lay supine at rest [36]. Knee and hip joints were flexed at 30° and ankles were supported in a neutral position. Individuals with SCI received an average of 16.6 ± 1 sessions for an average of 60 ± 2 min per session. Individuals without SCI received a total of 10.2 ± 0.2 sessions for an average of 40 ± 0.1 min per session. Stimulus intensity was established based on the intensity needed to determine the right soleus (SOL) TEP resting threshold at baseline. At baseline, the SOL TEP resting threshold intensity was 96.9 ± 24 and 28.9 ± 5.7 mA for individuals with and without SCI, respectively. Because the SOL TEP amplitude decreased after 10 to 15 min of continuous stimulation, we alternated the suprathreshold stimulation (15-min for SCI group; 10-min for control group) that evoked bilateral leg extension with subthreshold stimulation (5-min for both groups), similar to a stimulation paradigm we have previously utilized [20]. During each session, intensities ranged from 0.4 ± 0.1 to 4.3 ± 0.9 (2.2 ± 0.4) and 0.7 ± 0.1 to 10.3 ± 2.9 (6.4 ± 1.7) times the SOL TEP resting threshold for individuals with and without SCI, respectively. Transspinal stimulation was well tolerated by all participants, blood pressure remained stable, and no adverse events were encountered during the experiments or intervention.

EMG recordings

Surface electromyographic (EMG) activity was recorded bilaterally by single bipolar differential electrodes (Motion Lab Systems Inc., Baton Rouge, Louisiana, USA) from the SOL, medial gastrocnemius (MG), peroneus longus (PL), tibialis anterior (TA), medial hamstring (MH), lateral hamstring (LH), rectus femoris (RF), and gracilis (GRC) muscles. EMG signals were amplified, filtered (10–1000 Hz), sampled at 2000 Hz via a 1401 plus (Cambridge Electronics Design Ltd., Cambridge, UK), and stored for offline analysis.

Neurophysiological assessments before and after repeated transspinal stimulation

Measurements were performed at rest while lying supine, with transspinal stimulation electrodes placed using the same procedures for the intervention described above. Neurophysiological tests were performed before and two days after cessation of stimulation in individuals with and without SCI. Changes in multisegmental spinal motoneuron output were established based on the TEP recruitment input-output curve that was assembled at increasing intensities from below resting motor threshold until a plateau in the response was reached. Furthermore, TEPs from all muscles evoked at 1.3 ± 0.07 and 1.5 ± 0.16 times the SOL TEP threshold were recorded at 0.1, 0.125, 0.2, 0.33, and 1.0 Hz in order to establish changes in TEP homosynaptic depression in individuals with and without SCI, respectively. TEPs were also recorded upon paired transspinal stimuli randomly at the interstimulus intervals (ISIs) of 60, 100, 300 and 500 ms at 0.2 Hz in order to establish changes in TEPs postactivation depression for both subject groups. In all cases, 15 TEPs were recorded.

Data analysis and statistics

Neurophysiological assessments were grouped separately for AIS C-D, AIS A-B, and healthy control subjects based on time of testing. TEPs recorded from ankle and knee muscles were measured as the area under the full-wave rectified waveform for identical time windows (Spike 2, CED Ltd., U.K.). The onset latency of TEPs recorded at 0.1 Hz upon single pulse at baseline were estimated for each participant based on the cumulative sum technique [37] on the unrectified waveform average and was compared between SCI and control subjects using a Student’s t-test.

TEPs recorded upon single-pulse transspinal stimulation at increasing stimulation intensities were normalized to the associated maximal TEP (TEPmax). The normalized TEPs were plotted against the non-normalized stimulation intensities and a Boltzmann sigmoid function (Eq 1; SigmaPlot 11, Systat Software Inc.) was fitted to the data [38,39]. In Eq 1, m is the slope parameter of the function, S50-TEPmax is the stimulus required to elicit a TEP equivalent to 50% of the TEPmax, and s is the TEP amplitude at a given stimulus value TEP. The TEP slope and stimuli corresponding to TEP threshold, 50% of the TEPmax, and maximal TEP were estimated based on Eqs 2, 3 and 4, respectively. The estimated parameters from the sigmoid fit were compared before and after transspinal stimulation using a Student’s t-test.

Eq 1Eq 2Eq 3Eq 4

The predicted stimulation intensity from the sigmoid fit corresponding to S50-TEPmax obtained before transspinal stimulation was used to normalize the stimulation intensities for TEPs recorded before and after the intervention. Averages of normalized TEPs were calculated in steps of 0.05 from 0.3 up to 1.3 times the S50-TEPmax. The above described offline analysis was performed separately for each TEP input-output curve of each muscle for each participant assembled before and after transspinal stimulation. Repeated measures analysis of variance (rmANOVA) was then performed to the normalized TEPs grouped at multiples of S50-TEPmax before and after transspinal stimulation to establish the main effects of time grouped based on the normalized stimulation intensities. When a main effect was found, the Holm-Sidak t-tests for multiple comparisons were used to test for significant interactions between these factors.

To establish changes in homosynaptic depression, TEPs evoked at 0.125, 0.2, 0.33, and 1.0 Hz were expressed as a percentage of the mean amplitude of the homonymous TEP evoked at 0.1 Hz for each participant and muscle separately. To establish changes in postactivation depression, the TEP evoked by the second transspinal stimulus in the paired paradigm was normalized to the mean amplitude of the homonymous first TEP. This was done for TEPs recorded at ISIs of 60, 100, 300 and 500 ms for each subject and muscle separately. For both neurophysiological measures, rmANOVA was performed to establish the main effects of time. When a significant effect was found, Holm-Sidak t-tests for multiple comparisons were performed to establish interactions between time and ISI or between time and stimulation frequency. All values are presented as mean ± standard error of the mean unless otherwise stated. Values were considered significant with a p < .05.

Results

Characteristics of TEPs

Bilateral TEPs recorded from ankle and knee muscles from representative participants with and without SCI are indicated in Fig 1. TEPs are indicated along with the SOL maximal M-wave and SOL H-reflex for comparison of onset and shape between responses. Note that despite differences in the shape of TEPs across muscles, TEPs recorded from knee muscles have shorter latencies compared to the more distal ankle muscles, a phenomenon that was evident in both individuals with and without SCI. For individuals without SCI, TEPs were both biphasic and triphasic in nature for ankle muscles, whilst mostly triphasic for knee muscles. After incomplete SCI, TEPs were mostly triphasic with some polyphasic waveforms (see left RF in subject R03 and right RF in subject R10), and more turns were present within the waveforms. The latency of the SOL H-reflex and all TEPs recorded bilateral from ankle and knee muscles from all participants (N = 20) upon single pulses at 0.1 Hz are presented in Table 2. Latencies observed in individuals without SCI are consistent to those we have previously reported [31,34,40]. TEPs were present at similar latencies in the left and right leg muscles between subject groups. Note that the SOL TEP latency is approximately two thirds the SOL H-reflex latency for participants with (20.1 ± 0.5 ms; 32 ± 0.4 ms; 63%; based on grouped data) or without SCI (20.1 ± 0.4 ms; 31.5 ± 0.5 ms; 64%). These results suggest that SOL H-reflexes and SOL TEPs appear at similar latencies in individuals with and without SCI.

thumbnail
Download:
Fig 1. Transspinal evoked potentials (TEPs) in individuals with and without spinal cord injury (SCI).

Non-rectified TEP waveform averages recorded (N = 15, elicited at 0.2 Hz) bilaterally from ankle and knee muscles (solid black lines) in a healthy control participant (N-005) and two individuals with SCI presenting with American Spinal Injury Association Impairment Scale (AIS) D (R-003) and AIS A (R-010), respectively. Latencies of the soleus (SOL) maximal M-wave and H-reflex (dashed lines) along with all TEPs across individuals are also presented. MG: medial gastrocnemius; PL: peroneus longus; TA: tibialis anterior; MH: medial hamstring; LH: lateral hamstring; RF: rectus femoris; GRC: gracilis.

https://doi.org/10.1371/journal.pone.0213696.g001

thumbnail
Download:
Table 2. Transspinal evoked potential (TEP) onset latency (ms) in individuals with and without spinal cord injury (SCI).

https://doi.org/10.1371/journal.pone.0213696.t002

Changes in multisegmental motoneuron output after repeated low-frequency transspinal stimulation

The TEP recruitment input-output curves recorded from all muscles along with the associated sigmoid fit for individuals with AIS C-D before and after transspinal stimulation are presented in Fig 2A. TEPs are grouped based on stimulation intensities normalized to the S50-TEPmax obtained at baseline. For clarity purposes, a schematic representation of muscles that changes in the TEP recruitment input-output curves were observed is indicated in Fig 2B. In AIS C-D participants, a significant increase in the TEP input-output curve in right SOL (F1,174 = 14.51, p < .001), right MG (F1,189 = 7.35, p = .007), right PL (F1,131 = 8.19, p = .005), left PL (F1,144 = 4.94, p = .028), right RF (F1,138 = 12.28, p < .001), and left GRC (F1,168 = 5.19, p = .024) muscles was found (Fig 2). These results support for increased responsiveness of ankle and knee extensors motoneurons after repeated transspinal stimulation, functionally related to maintenance of standing posture.

thumbnail
Download:
Fig 2. Transspinal evoked potential (TEP) recruitment curves before and after transspinal stimulation in AIS C-D.

(A) TEPs normalized to the homonymous maximal TEP and plotted against normalized stimulation intensities to the S50-TEPmax obtained before transspinal stimulation training along with the sigmoid function fitted to the data. (B) Schematic representation of changes observed in the TEP recruitment input-output curves where red represents an increase and blue represents a decrease in the corresponding muscle from which the TEP was recorded. SOL: soleus; MG: medial gastrocnemius; TA: tibialis anterior; PL: peroneus longus; MH: medial hamstring; LH: lateral hamstring; RF: rectus femoris; GRC: gracilis. An asterisk indicates a significant difference before and after transspinal stimulation.

https://doi.org/10.1371/journal.pone.0213696.g002

The TEP recruitment input-output curves recorded from all muscles along with the associated sigmoid fit for individuals with AIS A-B before and after transspinal stimulation are presented in Fig 3A. TEPs are grouped based on stimulation intensities normalized to the S50-TEPmax obtained at baseline. For clarity purposes, a schematic representation of muscles that changes in the TEP recruitment input-output curves were observed is indicated in Fig 3B. In participants with AIS A-B, a significant increase in the TEP input-output curve in the right RF (F1,121 = 11.86, p < .001), left RF (F1,100 = 31.73, p < .001), and right GRC (F1,86 = 7.56, p = .007) muscles was found after transspinal stimulation (Fig 3A and 3B). Further, the significant decrease in the right TA TEPs (F1,103 = 98.15, p < .001) coincided with a significant interaction between time and normalized intensities (F24,103 = 6.40, p < .001). These results suggest that in cases of absent volitional motor control, transspinal stimulation increased motoneuron responsiveness mostly for proximal leg muscles.

thumbnail
Download:
Fig 3. Transspinal evoked potential (TEP) recruitment curves before and after transspinal stimulation in AIS A-B.

(A) TEPs normalized to the homonymous maximal TEP and plotted against normalized stimulation intensities to the S50-TEPmax obtained before transspinal stimulation training along with the sigmoid function fitted to the data. (B) Schematic representation of changes observed in the TEP recruitment input-output curves where red represents an increase and blue represents a decrease in the corresponding muscle from which the TEP was recorded. SOL: soleus; MG: medial gastrocnemius; TA: tibialis anterior; PL: peroneus longus; MH: medial hamstring; LH: lateral hamstring; RF: rectus femoris; GRC: gracilis. An asterisk indicates a significant difference before and after transspinal stimulation.

https://doi.org/10.1371/journal.pone.0213696.g003

The TEP recruitment input-output curves recorded from all muscles along with the associated sigmoid fit for healthy control subjects are indicated in Fig 4. TEPs are grouped based on time and stimulation intensities normalized to the S50-TEPmax obtained at baseline. Two-way rmANOVA showed that the left MH (F1,357 = 14.46, p < .001), right MH (F1,394 = 8.56, p = .004), right RF (F1,255 = 23.99, p < .001), and left LH (F1,218 = 7.295, p = .007) TEP recruitment curves and associated maximal values were significantly increased after repeated transspinal stimulation (Fig 4). In contrast, a significant decrease in the right TA (F1,318 = 6.33, p = .012), and left RF (F1,145 = 48.2, p < .001) TEPs was found (Fig 4). A significant interaction between time and intensities was found for the right (F21,255 = 2.17, p = .003) and left (F20,145 = 5.93, p < .001) RF TEPs. These results support for increased motoneuron output of knee flexor and hip extensor muscles after transspinal stimulation, whilst decreasing motoneuron output in ankle and hip flexors, and knee extensor muscles.

thumbnail
Download:
Fig 4. Transspinal evoked potential (TEP) recruitment curves before and after transspinal stimulation in healthy control subjects.

(A) TEPs normalized to the homonymous maximal TEP and plotted against normalized stimulation intensities to the S50-TEPmax obtained before transspinal stimulation training along with the sigmoid function fitted to the data. (B) Schematic representation of changes observed in the TEP recruitment input-output curves where red represents an increase and blue represents a decrease in the corresponding muscle from which the TEP was recorded. SOL: soleus; MG: medial gastrocnemius; TA: tibialis anterior; PL: peroneus longus; MH: medial hamstring; LH: lateral hamstring; RF: rectus femoris; GRC: gracilis. An asterisk indicates a significant difference before and after transspinal stimulation.

https://doi.org/10.1371/journal.pone.0213696.g004

The predicted parameters from the sigmoid function fitted to TEPs plotted against the actual stimulation intensities (recruitment input-output curves) for healthy control subjects only showed a significant decrease in stimulation threshold after repeated transspinal stimulation in the left PL TEP (before: 355.8 ± 73.8 mA, after: 330.5 ± 72.8 mA; p = .016), and therefore are not presented in detail. In contrast, for AIS C-D, the m function decreased (p = .023), the TEP slope increased (p = .017), and stimuli for the maximal TEP increased (p = .036) for the left MG TEP (Table 3). The m function also decreased for the left TA (p = .003), suggesting for small selective increased recruitment gains of ankle flexors and extensors in AIS C-D. For AIS A-B, the TEP slope (p = .043), S50-TEPmax (p = .004), and predicted stimuli at threshold and at maximal intensities decreased (p < .05) for the left GRC TEP. The predicted maximum TEP amplitude also decreased for the right SOL (p = .022), suggesting for small selective decreased recruitment gains of knee flexors and ankle extensors in AIS A-B.

thumbnail
Download:
Table 3. Transspinal evoked potential (TEP) recruitment curve sigmoid function parameters before and after transspinal stimulation.

https://doi.org/10.1371/journal.pone.0213696.t003

Comparisons between subjects groups showed that before transspinal stimulation, the predicted maximum TEP amplitude was increased in SCI compared to healthy control subjects for the right MG (SCI: 24.6 ± 2.9; controls: 9.5 ± 1.6; p < .05), left GRC (SCI: 21.9 ± 5.2; controls: 7.2 ± 1.3; p = .014) and right GRC (SCI: 15.5 ± 3.6; controls: 7.0 ± 1.7; p = .037). In contrast, the right SOL TEP was markedly reduced (SCI: 17.5 ± 1.6; controls: 27.6 ± 4.2; p = .032) for individuals with SCI. After transspinal stimulation, the predicted maximum TEP amplitude was increased in SCI compared to healthy control subjects for the left MG (SCI: 25.6 ± 6.5 ; controls: 10.3 ± 2.7; p = .043), right TA (SCI: 9.4 ± 1.4; controls: 4.3 ± 0.9; p = .008), left RF (SCI: 21.2 ± 3.9; controls: 4.6 ± 1.4; p = .003), left GRC (SCI: 20.2 ± 3.9; controls: 6.1 ± 1.1; p = .001) and right GRC (SCI: 17.6 ± 2.9; controls: 5.5 ± 1.2; p = .002) muscles. All other predicted parameters were not significantly different from controls.

Changes in homosynaptic and postactivation TEP depression after repeated low-frequency transspinal stimulation

The average amplitude of normalized TEPs recorded from the left and right leg muscles at different stimulation frequencies grouped for AIS C-D and AIS A-B before and after repeated transspinal stimulation are presented in Fig 5. One-way rmANOVA in AIS C-D for the left SOL TEP recorded at baseline showed a decrease in amplitude as stimulation frequency decreased (F3,20 = 8.49, p < .001). The same result was also found for TEPs recorded from the left/right TA and left PL muscles, while no significance for the remaining TEPs was found (p > .05). This resulted in 4 out of 16 muscles that homosynaptic TEP depression was evident in AIS C-D. In a similar manner, homosynaptic TEP depression was evident in 4 out of 16 muscles in AIS A-B subjects, with TEPs recorded from the left SOL (F3,10 = 5.51, p = .017), right SOL (F3,10 = 4.04, p = .04), right TA (F3,10 = 6.21, p = .012), and right PL (F3,10 = 4.62, p = .028) muscles to display depression that varied as a function of stimulation frequency at baseline (Fig 5). After transspinal stimulation, homosynaptic TEP depression decreased in the left MG (F1,36 = 9.038, p = .005) and remained unaltered in the remaining TEPs (Fig 5) in AIS C-D subjects, while homosynaptic TEP depression increased in the left SOL TEP (F1,21 = 7.44, p = .013) and right RF TEP (F1,19 = 6.71, p = .018) in AIS A-B subjects.

thumbnail
Download:
Fig 5. Homosynaptic transspinal evoked potential (TEP) depression before and after transspinal stimulation in spinal cord injury (SCI).

Overall percent change of TEPs recorded at 0.125, 0.2, 0.33 and 1.0 Hz from the associated TEP recorded at 0.1 Hz before (black lines) and after (green lines) transspinal stimulation training in individuals with (A) AIS C-D and (B) AIS A-B. SOL: soleus; MG: medial gastrocnemius; TA: tibialis anterior; PL: peroneus longus; MH: medial hamstring; LH: lateral hamstring; RF: rectus femoris; GRC: gracilis. Error bars indicate SE. An asterisk indicates a significant difference before and after transspinal stimulation.

https://doi.org/10.1371/journal.pone.0213696.g005

The average amplitude of TEPs recorded from left and right leg muscles upon paired transspinal stimuli grouped for AIS C-D and AIS A-B before and after transspinal stimulation are presented in Fig 6. One-way rmANOVA in AIS C-D for the right LH TEP recorded at baseline showed that the amount of depression varied significantly as a function of the ISI tested (H3 = 10.38, p = .016), while in the remaining TEPs the amount of depression did not vary significantly as a function of the ISI. Similarly, one-way rmANOVA in AIS A-B showed that the amount of depression upon paired transspinal stimuli at baseline varied significantly as a function of ISI only for the left MH TEP (F3,10 = 4.67, p = .027). After transspinal stimulation, TEP depression upon transspinal paired stimuli remained unaltered in AIS C-D (Fig 6A), and was decreased only in the right GRC TEP (F1,14 = 8.31, p = .012) in AIS A-B (Fig 6B), suggesting for minimal changes of postactivation depression after transspinal stimulation in individuals with SCI.

thumbnail
Download:
Fig 6. Postactivation transspinal evoked potential (TEP) depression upon paired transspinal stimuli before and after transspinal stimulation in spinal cord injury (SCI).

Overall amplitude of the second TEP (TEP2) as a percentage of the first mean homonymous TEP amplitude (TEP1) evoked at interstimulus intervals of 500, 300, 100 and 60 ms at a constant stimulation frequency of 0.2 Hz for participants with (A) AIS C-D and (B) AIS A-B. SOL: soleus; MG: medial gastrocnemius; TA: tibialis anterior; PL: peroneus longus; MH: medial hamstring; LH: lateral hamstring; RF: rectus femoris; GRC: gracilis. Error bars indicate SE. An asterisk indicates a significant difference before and after the transspinal stimulation.

https://doi.org/10.1371/journal.pone.0213696.g006

The group amplitude of normalized TEPs recorded from left and right leg muscles at different stimulation frequencies and upon paired pulses before and after transspinal stimulation in healthy control subjects are indicated in Fig 7. One-way rmANOVA for the left SOL TEP recorded at baseline showed a decrease in amplitude as stimulation frequency decreased (H3 = 16.19, p = .001). The same result was also found for TEPs recorded from the remaining muscles, except for the left TA (H3 = 6.41, p = .09) and left RF (H3 = 3.99, p = .26), suggesting for a generalized susceptibility of TEPs to homosynaptic depression in healthy control subjects. After transspinal stimulation, the amount of homosynaptic depression differed between time (F1,65 = 8.70, p = .004) and ISI (F3,65 = 18.61, p < .001) only for the left LH TEP, while no significant changes in the remaining TEPs were found (Fig 7A).

thumbnail
Download:
Fig 7. Homosynaptic and postactivation transspinal evoked potential (TEP) depression before and after transspinal stimulation in healthy control subjects.

(A) Overall percent change of TEPs recorded at 0.125, 0.2, 0.33 and 1.0 Hz from the associated TEP recorded at 0.1 Hz before (black lines) and after (green lines) transspinal stimulation training. (B) Overall amplitude of the second TEP (TEP2) as a percentage of the first homonymous mean TEP (TEP1) evoked at interstimulus intervals of 500, 300, 100 and 60 ms at a constant stimulation frequency of 0.2 Hz. SOL: soleus; MG: medial gastrocnemius; TA: tibialis anterior; PL: peroneus longus; MH: medial hamstring; LH: lateral hamstring; RF: rectus femoris; GRC: gracilis. Error bars indicate SE. An asterisk indicates a significant difference before and after transspinal stimulation.

https://doi.org/10.1371/journal.pone.0213696.g007

The amount of TEP depression upon paired transspinal stimuli recorded at baseline in healthy control subjects (Fig 7B) did not vary based on the ISI for the left SOL TEP (H3 = 5.39, p = .15). The same result was found for the left/right MG, left/right TA, right PL, left MH/LH, right RF, and left/right GRC TEPs, and thus resulting in TEPs postactivation depression independent of the ISI used. After repeated transspinal stimulation, the amount of postactivation depression decreased in the left TA (F1,55 = 9.997, p = .003), right TA (F1,62 = 6.03, p = .017), and left RF (F1,47 = 6.43, p = .015) TEPs. These results support that repeated transspinal stimulation affects TEPs homosynaptic and postactivation depression differently in healthy control subjects.

Discussion

Neuromodulatory noninvasive therapeutic approaches that can strengthen motoneuron depolarization are in great need to promote recovery of motor function after SCI. Here we show for the first time that repeated low-frequency transspinal stimulation over the thoracolumbar enlargement, the location of leg motor circuits, increases the responsiveness of motoneurons over multiple segments in individuals with chronic motor incomplete and complete SCI. While transspinal stimulation induced mixed effects on homosynaptic and postactivation TEP depression, this intervention may be used to functionally improve the neural control of standing and walking.

TEPs had similar latencies and shapes in left and right legs among all subject groups (Fig 1), but more turns within the shape were present in people with SCI, possibly due to less synchronous firing of motor fibers represented by phase cancellation and temporal dispersion in demyelinated nerves [41]. In individuals with SCI, the orderly sigmoidal TEP excitability altered significantly after repeated transspinal stimulation. TEPs recorded mostly at intermediate and maximal intensities were significantly increased largely for the ankle and knee extensors in AIS C-D (Fig 2), while in AIS A-B the increased TEP excitability of knee extensors coincided with decreased excitability of ankle flexors (Fig 3). An increased motoneuron output was also evident for the GRC muscle in the left leg of AIS C-D and right leg in AIS A-B. In the latter case, a significant decrease in the slope of the recruitment curve was evident (Table 3) suggesting altered motoneuronal gain after repeated transspinal stimulation.

While we do not know whether maximal TEPs correspond to the recruitment of the whole motoneuron pool, repeated transspinal stimulation increased the responsiveness of motoneurons that are critical for standing and maintenance of upright posture. This potentiation may have resulted from an increased excitability state enabling alpha motoneurons to produce a more synchronized depolarization. It is possible that the discharge zone of alpha motoneurons residing within the subliminal fringe was expanded after repetitive stimulation of afferents over multiple segments, making motoneurons more easily depolarized to a given input. This is supported by the well-described reversal of subliminal fringe of Renshaw cells upon convergent excitatory inputs [42]. While decreases in threshold intensities based on the sigmoid function were present only in the left GRC TEP in AIS A-B subjects (Table 3), a shift of the TEP recruitment curve to the left was present for the left PL/SOL and right/left MH in AIC C-D (Fig 2A), and right RF and left MH in AIS A-B (Fig 2B). The shift of TEP excitability to the left suggests that repeated low-frequency transspinal stimulation delivered at intensities that motoneuronal depolarization results in bilateral leg extensions followed by subthreshold intensities that may excite interneurons and synapses without direct efferent activity renders the spinal neuronal networks more excitable. A possible mechanism underlying these changes is facilitation of excitatory synapses resulting in long-term potentiation (LTP)-like mechanisms [43]. Because of anatomical and physiological cytoarchitecture differences between the brain and spinal cord, especially after SCI [44], we should be cautious with the general assumption that LTP-mediated plasticity for the spinal cord requires high-frequency stimulation [45,46]. This mechanism is supported by the LTP-induced following low frequency (0.2 Hz) stimulation [47], similar to the frequency we used in this study. Additional evidence supports for LTP-like facilitation mechanisms. For example, transspinal stimulation in midthoracic lateral hemisected rats for 50 min at 0.2 Hz increased the response amplitude of lateral white matter or dorsal corticospinal tracts, and this increase required the activation of N-methyl-D-aspartate (NMDA) receptors [29]. Activity of NMDA receptors is known to affect AMPA/kainate receptors which play a crucial role in LTP induction [48–50]. It is possible that, similar to the role described for the silent synapses in LTP [48], repeated transspinal stimulation could have affected the number of motoneurons failing to discharge by raising the firing threshold and thus increasing the size of motoneuronal depolarization [51,52]. Because transspinal stimulation can excite several types of afferents and efferents over multiple segments, it is possible that presynaptic activity may coincide with postsynaptic excitation, which is required for Hebbian LTP [49]. In conclusion, the increased motoneuron output in individuals with chronic motor incomplete and complete SCI after repeated low-frequency transspinal stimulation may be the result of strengthening of spinal synapses via LTP-like mechanisms or via changes in the intrinsic properties of motoneurons. Regardless the underlying mechanism, it is apparent that functionally the increased responsiveness from multiple motoneuron pools can contribute to recovery of motor function, and especially improve standing and walking ability, as reported after epidural stimulation [8,12].

In SCI, a notable contradiction is that although hyperreflexia can produce significant muscle force output it coincides with decreased motoneuronal excitability and volitional motor control [53]. The lack of neuromodulation prevents motoneurons from firing at a frequency sufficient to generate volitional muscle contractions at adequate forces. These results thus strongly support our overall hypothesis that transspinal stimulation alters motoneuron excitability over multiple segments by bringing motoneurons closer to threshold, which is a pre-requisite for functioning descending and local inputs. Additionally, in several cases of complete or incomplete SCI, a single session of transspinal stimulation decreased spasticity, hyperreflexia, and ankle clonus [19,54]. Collectively, repeated transspinal stimulation may be an optimal therapeutic neuromodulation intervention to concomitantly increase motoneuron responsiveness and decrease spasticity. However, spasticity and hyperreflexia over the course of transspinal stimulation requires to be established by neuromechanical studies. Lastly, we should note that repeated transspinal stimulation may be utilized as an adjunct to physical therapy. This proposal is consistent with the reduced spasticity reported after combined interventions including locomotor training and transcranial or transspinal direct current stimulation [55,56].

In individuals with chronic motor complete and incomplete SCI, only 4 out of 16 TEPs portrayed homosynaptic depression (Fig 5) in contrast to the 14 out of 16 TEPs that were susceptible to homosynaptic depression in healthy control subjects (Fig 7A). Based on these results, we suggest that the susceptibility of TEPs to homosynaptic depression depends on presence of volitional motor activity. The soleus H-reflex homosynaptic depression is mediated by consequent activation of the same muscle spindle group Ia afferents, and occurs at the same Ia/motoneuron synapses with absent contribution from spinal inhibitory interneurons or motor axons [57,58]. In the case of TEPs, the decrease of response amplitude as stimulation frequency decreases likely involves spinal inhibitory interneurons spanning multiple segments since TEPs coincide with bilateral muscle contractions resulting in excitation of various afferent types and motor axons [59]. Furthermore, paired transspinal stimuli at ISIs ranging from 60 to 500 ms produced a significant TEP depression. This postactivation depression was similar between subject groups, and was present at baseline in nearly all of TEPs in individuals with chronic motor complete and incomplete SCI (Fig 6) and in healthy control subjects (Fig 7). Repeated low-frequency transspinal stimulation resulted in mixed changes on TEP homosynaptic depression, with both decreases (left MG TEP in AIS C-D; Fig 5A) and increases (left SOL TEP in AIS A-B, Fig 5A and left LH TEP in controls, Fig 7A) being present. Similarly, mixed changes were observed in the TEP postactivation depression, which remained unaltered in AIS C-D (Fig 6A), increased only for one muscle in AIS A-B (right GRC TEP; Fig 6B), and decreased for two muscles in healthy control subjects (left and right TA TEP; Fig 7B). Postactivation depression affects synaptic actions of group I and group II afferents, and low threshold cutaneous afferents synapsing onto intermediate spinal interneurons differently [60]. Based on the mixed changes we observed, different susceptibility of afferents to postactivation depression, and multiple transynaptic actions on spinal neurons as a result of transspinal stimulation, we theorize that in order to postulate changes of homosynaptic and postactivation depression in TEPs recorded bilaterally from knee and ankle muscles, a widespread synaptic plasticity is required, likely related to the number of stimulation sessions.

Limitations

Transspinal stimulation was delivered with a single, 1-ms square pulse at 0.2 Hz based on previous studies [31,34,36]. We did not use medium or high stimulation frequencies, as we were particularly interested to establish neuroplasticity with low-frequency single pulse stimulation based on its powerful effects on cortical and subcortical neuronal circuits [20,61–63]. Further, we did not perform neurophysiological tests at multiple time points after cessation of stimulation. Thus we are unable to comment on the sustainability of neuroplasticity beyond two days. Lastly, given our sample size and the heterogeneity of SCI [64] we cannot conclude that repeated low-frequency transspinal stimulation is effective for increasing motoneuron responsiveness in all patients. Future studies are warranted to examine different stimulation parameters and time course of neuroplasticity.

Conclusions

This study provides evidence that repeated transspinal stimulation increases the responsiveness of motoneuron pools over multiple spinal segments, and produces mixed effects in spinal inhibitory mechanisms after motor paralysis or paresis. From a functional perspective, increased motoneuron output may contribute to better motor control and improved tasks such as standing and walking. We propose that noninvasive low-frequency transspinal stimulation can be used to strengthen spinal synapses and depolarization of alpha motoneurons in individuals with chronic motor paralysis or paresis.

Acknowledgments

The authors would like to thank all research participants and their families for their unwavering dedication, adherence and support that made this study possible.

References

  1. 1. Baptiste DC, Fehlings MG. Pharmacological approaches to repair the injured spinal cord. J Neurotrauma. 2006;23(3–4): 318–334. pmid:16629619
  2. 2. Alam M, Rodrigues W, Pham BN, Thakor N V. Brain-machine interface facilitated neurorehabilitation via spinal stimulation after spinal cord injury: recent progress and future perspectives. Brain Res. 2016;1646: 25–33. pmid:27216571
  3. 3. Knikou M. Neural control of locomotion and training-induced plasticity after spinal and cerebral lesions. Clin Neurophysiol. 2010;121(10): 1655–1668. pmid:20427232
  4. 4. James ND, McMahon SB, Field-Fote EC, Bradbury EJ. Neuromodulation in the restoration of function after spinal cord injury. Lancet Neurol. 2018;17(10): 905–917. pmid:30264729
  5. 5. Smith AC, Knikou M. A review on locomotor training after spinal cord injury: reorganization of spinal neuronal circuits and recovery of motor function. Neural Plast. 2016;2016: 1216258. pmid:27293901
  6. 6. Hofstoetter US, Knikou M, Guertin PA, Minassian K. Probing the human spinal locomotor circuits by phasic step-induced feedback and by tonic electrical and pharmacological neuromodulation. Curr Pharm Des. 2017;23(12): 1805–1820. pmid:27981912
  7. 7. Courtine G, Gerasimenko Y, van den Brand R, Yew A, Musienko P, Zhong H, et al. Transformation of nonfunctional spinal circuits into functional states after the loss of brain input. Nat Neurosci. 2009;12(10): 1333–1342. pmid:19767747
  8. 8. Harkema S, Gerasimenko Y, Hodes J, Burdick J, Angeli C, Chen Y, et al. Effect of epidural stimulation of the lumbosacral spinal cord on voluntary movement, standing, and assisted stepping after motor complete paraplegia: a case study. Lancet. 2011;377(9781): 1938–1947. pmid:21601270
  9. 9. Angeli CA, Edgerton VR, Gerasimenko YP, Harkema SJ. Altering spinal cord excitability enables voluntary movements after chronic complete paralysis in humans. Brain. 2014;137(5): 1394–1409. pmid:24713270
  10. 10. Angeli CA, Boakye M, Morton RA, Vogt J, Benton K, Chen Y, et al. Recovery of over-ground walking after chronic motor complete spinal cord injury. N Engl J Med. 2018;379(13): 1244–1250. pmid:30247091
  11. 11. Gerasimenko YP, Lu DC, Modaber M, Zdunowski S, Gad P, Sayenko DG, et al. Noninvasive reactivation of motor descending control after paralysis. J Neurotrauma. 2015;32(24): 1968–1980. pmid:26077679
  12. 12. Rejc E, Angeli CA, Bryant N, Harkema SJ. Effects of stand and step training with epidural stimulation on motor function for standing in chronic complete paraplegics. J Neurotrauma. 2017;34(9): 1787–1802. pmid:27566051
  13. 13. Rejc E, Angeli C, Harkema S. Effects of lumbosacral spinal cord epidural stimulation for standing after chronic complete paralysis in humans. PLoS One. 2015;10(7):e0133998. pmid:26207623
  14. 14. Gill ML, Grahn PJ, Calvert JS, Linde MB, Lavrov IA, Strommen JA, et al. Neuromodulation of lumbosacral spinal networks enables independent stepping after complete paraplegia. Nat Med. 2018;24(11): 1677–1682. pmid:30250140
  15. 15. Harkema SJ, Wang S, Angeli CA, Chen Y, Boakye M, Ugiliweneza B, et al. Normalization of blood pressure with spinal cord epidural stimulation after severe spinal cord injury. Front Hum Neurosci. 2018;12:83. pmid:29568266
  16. 16. Aslan SC, Legg Ditterline BE, Park MC, Angeli CA, Rejc E, Chen Y, et al. Epidural spinal cord stimulation of lumbosacral networks modulates arterial blood pressure in individuals with spinal cord injury-induced cardiovascular deficits. Front Physiol. 2018;9:565. pmid:29867586
  17. 17. DiMarco AF, Kowalski KE, Geertman RT, Hromyak DR, Frost FS, Creasey GH, et al. Lower thoracic spinal cord stimulation to restore cough in patients with spinal cord injury: results of a National Institutes of Health–Sponsored Clinical Trial. Part II: clinical outcomes. Arch Phys Med Rehabil. 2009;90(5): 726–732. pmid:19406290
  18. 18. DiMarco AF, Kowalski KE, Geertman RT, Hromyak DR. Lower thoracic spinal cord stimulation to restore cough in patients with spinal cord injury: results of a National Institutes of Health-Sponsored Clinical Trial. Part I: methodology and effectiveness of expiratory muscle activation. Arch Phys Med Rehabil. 2009;90(5): 717–725. pmid:19406289
  19. 19. Hofstoetter US, McKay WB, Tansey KE, Mayr W, Kern H, Minassian K. Modification of spasticity by transcutaneous spinal cord stimulation in individuals with incomplete spinal cord injury. J Spinal Cord Med. 2014;37(2): 202–211. pmid:24090290
  20. 20. Murray LM, Knikou M. Remodeling brain activity by repetitive cervicothoracic transspinal stimulation after human spinal cord injury. Front Neurol. 2017;8:50. pmid:28265259
  21. 21. Gad PN, Kreydin E, Zhong H, Latack K, Edgerton VR. Non-invasive neuromodulation of spinal cord restores lower urinary tract function after paralysis. Front Neurosci. 2018;12: 432. pmid:30008661
  22. 22. Phillips AA, Squair JW, Sayenko DG, Edgerton VR, Gerasimenko Y, Krassioukov AV. An autonomic neuroprosthesis: noninvasive electrical spinal cord stimulation restores autonomic cardiovascular function in individuals with spinal cord injury. J Neurotrauma. 2018;35(3): 446–451. pmid:28967294
  23. 23. Rath M, Vette AH, Ramasubramaniam S, Li K, Burdick J, Edgerton VR, et al. Trunk stability enabled by noninvasive spinal electrical stimulation after spinal cord injury. J Neurotrauma. 2018;35(21): 2540–2553. pmid:29786465
  24. 24. Gorodnichev RM, Pivovarova EA, Pukhov A, Moiseev SA, Savokhin AA, Moshonkina TR, et al. Transcutaneous electrical stimulation of the spinal cord: a noninvasive tool for the activation of stepping pattern generators in humans. Hum Physiol. 2012;38(2): 158–167.
  25. 25. Hounsgaard J, Kiehn O. Calcium spikes and calcium plateaux evoked by differential polarization in dendrites of turtle motoneurones in vitro. J Physiol. 1993;468: 245–259. pmid:8254508
  26. 26. Elbasiouny SM, Mushahwar VK. Suppressing the excitability of spinal motoneurons by extracellularly applied electrical fields: insights from computer simulations. J Appl Physiol. 2007;103(5): 1824–1836. pmid:17702836
  27. 27. Li Y, Bennett DJ. Persistent sodium and calcium currents cause plateau potentials in motoneurons of chronic spinal rats. J Neurophysiol. 2003;90(2): 857–869. pmid:12724367
  28. 28. Hultborn H, Brownstone RB, Toth TI, Gossard J-P. Key mechanisms for setting the input–output gain across the motoneuron pool. Prog Brain Res. 2004;143: 75–95.
  29. 29. Hunanyan AS, Petrosyan HA, Alessi V, Arvanian VL. Repetitive spinal electromagnetic stimulation opens a window of synaptic plasticity in damaged spinal cord: role of NMDA receptors. J Neurophysiol. 2012;107(11): 3027–3039. pmid:22402659
  30. 30. Mendell LM, Munson JB, Arvanian VL. Neurotrophins and synaptic plasticity in the mammalian spinal cord. J Physiol. 2001;533(1): 91–97.
  31. 31. Knikou M, Dixon L, Santora D, Ibrahim MM. Transspinal constant-current long-lasting stimulation: a new method to induce cortical and corticospinal plasticity. J Neurophysiol. 2015;114(3): 1486–1499. pmid:26108955
  32. 32. Sayenko DG, Atkinson DA, Dy CJ, Gurley KM, Smith VL, Angeli C, et al. Spinal segment-specific transcutaneous stimulation differentially shapes activation pattern among motor pools in humans. J Appl Physiol. 2015;118(11): 1364–1374. pmid:25814642
  33. 33. Einhorn J, Li A, Hazan R, Knikou M. Cervicothoracic multisegmental transpinal evoked potentials in humans. PLoS One. 2013;8(10): e76940. pmid:24282479
  34. 34. Knikou M. Neurophysiological characterization of transpinal evoked potentials in human leg muscles. Bioelectromagnetics. 2013;34(8): 630–640. pmid:24115026
  35. 35. Galea JM, Celnik P. Brain polarization enhances the formation and retention of motor memories. J Neurophysiol. 2009;102(1): 294–301. pmid:19386757
  36. 36. Danner SM, Krenn M, Hofstoetter US, Toth A, Mayr W, Minassian K. Body position influences which neural structures are recruited by lumbar transcutaneous spinal cord stimulation. PLoS One. 2016;11(1): e0147479. pmid:26797502
  37. 37. Ellaway PH. Cumulative sum technique and its application to the analysis of peristimulus time histograms. Electroencephalogr Clin Neurophysiol. 1978;45(2): 302–304. pmid:78843
  38. 38. Klimstra M, Zehr EP. A sigmoid function is the best fit for the ascending limb of the Hoffmann reflex recruitment curve. Exp Brain Res. 2008;186(1): 93–105. pmid:18046545
  39. 39. Brinkworth RS, Tuncer M, Tucker KJ, Jaberzadeh S, Türker KS. Standardization of H-reflex analyses. J Neurosci Methods. 2007;162(1–2): 1–7. pmid:17257686
  40. 40. Knikou M. Transpinal and transcortical stimulation alter corticospinal excitability and increase spinal output. PLoS One. 2014;9(7): 1–14. pmid:25007330
  41. 41. Mallik A, Weir AI. Nerve conduction studies: essentials and pitfalls in practice. J Neurol Neurosurg Psychiatry. 2005;76(suppl 2): 23–31. pmid:15961865
  42. 42. Hultborn H, Pierrot-Deseilligny E. Input-output relations in the pathway of recurrent inhibition to motoneurones in the cat. J Physiol. 1979;297: 267–287. pmid:231651
  43. 43. Molnár E. Long-term potentiation in cultured hippocampal neurons. Semin Cell Dev Biol. 2011;22(5): 506–513. pmid:21807105
  44. 44. Yoon H, Walters G, Paulsen AR, Scarisbrick IA. Astrocyte heterogeneity across the brain and spinal cord occurs developmentally, in adulthood and in response to demyelination. PLoS One. 2017;12(7): e0180697. pmid:28700615
  45. 45. Bliss TVP, Collingridge GL. A synaptic model of memory: long-term potentiation in the hippocampus. Nature. 1993;361: 31–39. pmid:8421494
  46. 46. Bear MF, Malenka RC. Synaptic plasticity: LTP and LTD. Curr Opin Neurobiol. 1994;4(3): 389–399. pmid:7919934
  47. 47. Andrade-Talavera Y, Duque-Feria P, Paulsen O, Rodríguez-Moreno A. Presynaptic spike timing-dependent long-term depression in the mouse hippocampus. Cereb Cortex. 2016;26(8): 3637–3654. pmid:27282393
  48. 48. Kerchner GA, Nicoll RA. Silent synapses and the emergence of a postsynaptic mechanism for LTP. Nat Rev Neurosci. 2008;9: 813–825. pmid:18854855
  49. 49. Malenka RC. Synaptic plasticity in the hippocampus: LTP and LTD. Cell. 1994;78(4): 535–538. pmid:8069904
  50. 50. Volianskis A, France G, Jensen MS, Bortolotto ZA, Jane DE, Collingridge GL. Long-term potentiation and the role of N-methyl-d-aspartate receptors. Brain Res. 2015;1621: 5–16. pmid:25619552
  51. 51. Knikou M. The H-reflex as a probe: pathways and pitfalls. J Neurosci Methods. 2008;171(1): 1–12. pmid:18394711
  52. 52. Pierrot-Deseilligny E, Mazevet D. The monosynaptic reflex: a tool to investigate motor control in humans. Interest and limits. Clin Neurophysiol. 2000;30(2): 67–80.
  53. 53. Hounsgaard J, Hultborn H, Jespersen B, Kiehn O. Bistability of alpha-motoneurones in the decerebrate cat and in the acute spinal cat after intravenous 5-hydroxytryptophan. J. Physiol. 1988;405: 345–67. pmid:3267153
  54. 54. Minassian K, Hofstoetter US. Spinal cord stimulation and augmentative control strategies for leg movement after spinal paralysis in humans. CNS Neurosci. Ther. 2016;22: 262–70. pmid:26890324
  55. 55. Raithatha R, Carrico C, Powell ES, Westgate PM, CheletteIi KC, Lee K, et al. Non-invasive brain stimulation and robot-assisted gait training after incomplete spinal cord injury: A randomized pilot study. NeuroRehabilitation 2016;38: 15–25. pmid:26889794
  56. 56. Estes SP, Iddings JA, Field-Fote EC. Priming neural circuits to modulate spinal reflex excitability. Front Neurol 2017;8:17. pmid:28217104
  57. 57. Côté M-P, Murray LM, Knikou M. Spinal control of locomotion: individual neurons, their circuits and functions. Front Physiol. 2018;9:784. pmid:29988534
  58. 58. Hultborn H, Illert M, Nielsen JB, Paul A, Ballegaard M, Wiese H. On the mechanism of the post-activation depression of the H-reflex in human subjects. Exp Brain Res. 1996;108(3): 450–462. pmid:8801125
  59. 59. Maertens De Noordhout A, Rothwell JC, Thompson PD, Day BL, Marsden CD. Percutaneous electrical stimulation of lumbosacral roots in man. J Neurol Neurosurg Psychiatry. 1988;51(2): 174–181. pmid:2831302
  60. 60. Hammar I, Slawinska U, Jankowska E. A comparison of postactivation depression of synaptic actions evoked by different afferents and at different locations in the feline spinal cord. Exp Brain Res. 2002;145(1): 126–129. pmid:12070752
  61. 61. Knikou M. Spinal excitability changes after transspinal and transcortical paired associative stimulation in humans. Neural Plast. 2017;2017: 6751810. pmid:29123926
  62. 62. Knikou M, Murray LM. Neural interactions between transspinal evoked potentials and muscle spindle afferents in humans. J Electromyogr Kinesiol. 2018;43: 174–183. pmid:30352388
  63. 63. Dixon L, Ibrahim MM, Santora D, Knikou M. Paired associative transspinal and transcortical stimulation produces plasticity in human cortical and spinal neuronal circuits. J Neurophysiol. 2016;116(2): 904–916. pmid:27281748
  64. 64. Karunakaran KD, He J, Zhao J, Cui J, Zang Y-F, Zhang Z, et al. Differences in cortical gray matter atrophy of paraplegia and tetraplegia after complete spinal cord injury. J Neurotrauma. 2018; Nov 15. pmid:30430910