Optimizing Transcutaneous Spinal Stimulation: Excitability of Evoked Spinal Reflexes is Dependent on Electrode Montage

doi:10.21203/rs.3.rs-4719031/v1

Download PDF

Research Article

Optimizing Transcutaneous Spinal Stimulation: Excitability of Evoked Spinal Reflexes is Dependent on Electrode Montage

https://doi.org/10.21203/rs.3.rs-4719031/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

There is growing interest in use of transcutaneous spinal stimulation (TSS) for people with neurologic conditions both to augment volitional control (by facilitating motoneuron excitability), and to decrease spasticity (by activating inhibitory networks). Various electrode montages are used during TSS, with little understanding of how electrode position influences spinal circuit activation. We sought to identify the thoracolumbar electrode montage associated with the most robust activation of spinal circuits by comparing posterior root-muscle reflexes (PRM reflexes) elicited by 6 montages. Additionally, we assessed tolerability of the stimulation during PRM reflex testing.

Methods

Fifteen adults with intact neurological systems participated in this randomized crossover study. PRM reflexes were evoked transcutaneously using electrode montages with dorsal-ventral (DV) or dorsal-midline (DM) current flow. DV montages included: [1] cathode over T11/T12, anodes over iliac crests (DV-I), [2] cathode over T11/T12, anodes over umbilicus (DV-U), [3] dual paraspinal cathodes at T11/12, anodes over iliac crests (DV-PI), and [4] dual paraspinal cathodes at T11/12, anodes over umbilicus (DV-PU). DM montages included: [5] cathode over T11/12, anode 5cm caudal (DM-C), and [6] cathode over T11/12, anode 5cm rostral (DM-R). PRM reflex recruitment curves were obtained in the soleus muscle of both lower extremities.

Results

DV-U and DV-I montages elicited bilateral reflexes with lower reflex thresholds and larger recruitment curve area than other montages. There were no differences in response amplitude at 120% of RT(1.2xRT) or tolerability among montages.

Conclusions

Differences in spinal circuit recruitment are reflected in the response amplitude of the PRM reflexes. DV-I and DV-U montages were associated with lower reflex thresholds, indicating that motor responses can be evoked with lower stimulation intensity. DV-I and DV-U montages therefore have the potential for lower and more tolerable interventional stimulation intensities. Our findings optimize electrode placement for interventional TSS and PRM reflex assessments.

Clinical Trial Number:

NCT04243044

Neuromodulation

Transcutaneous Spinal Stimulation

Spinal Excitability

Electrodes

Transcutaneous spinal stimulation (TSS) is increasingly being utilized as a neurorehabilitation tool for improving walking function,[1–4] spasticity,[5, 6] and postural control[7] in people with neurologic conditions. TSS can be applied using clinically accessible devices, making it optimal for pairing with physical therapy interventions. While early studies have focused on people with spinal cord injury (SCI), emerging reports are investigating the effects of TSS in people with multiple sclerosis,[8] stroke,[9] traumatic brain injury,[10] and cerebral palsy.[11]

While frequently referred to as “spinal cord stimulation”, modeling studies indicate TSS activates large-diameter Ia afferents at the dorsal nerve roots, similar to epidural spinal stimulation.[12, 13] Hence, our use of the term transcutaneous spinal stimulation indicates electrode location and does not imply direct activation of spinal cord structures. Ia afferents synapse directly with spinal motoneurons, with subthreshold stimuli bringing motoneurons closer to activation threshold, and suprathreshold stimuli eliciting a monosynaptic reflex.[12, 13] TSS thereby improves motor output in persons with neurologic conditions by augmenting the motoneuronal excitability produced by volitional effort.[7, 14, 15]

The same electrode positions used during interventional TSS can be used to measure spinal reflex excitability. Single pulses of transcutaneous stimulation over the lumbosacral enlargement activate Ia afferents at the dorsal nerve roots and elicit short-latency evoked potentials, termed posterior root-muscle reflexes (PRM reflexes).[16] PRM reflexes bear similarities to H-reflexes, the electrical analogue of the stretch reflex.[16–18] PRM reflexes demonstrate potential utility in measuring responses to neuromodulatory interventions that target Ia afferents, such as TSS, whole body vibration, and peripheral nerve stimulation.

A variety of electrode placements have been used for lower extremity PRM reflex testing and interventional TSS, with little understanding of which montage more efficaciously activates spinal circuitry. Reported cathode positions range from T9 to L2 including a single midline electrode over the interspinous space,[1, 19] paired electrodes perpendicular to the spinal column,[12, 16] or paired electrodes vertical to the spinal column.[2, 4] Reported anode positions include paired electrodes over the umbilicus,[1, 20, 21] paired electrodes over the iliac crests,[7, 22, 23] or a single electrode over the interspinous space rostral or caudal to the cathode.[24] Although some studies utilize multiple cathodes for interventional TSS, evidence supporting use of multiple cathodes is lacking. In fact, a recent study found force production was lower for plantarflexion and knee extension when using dual cathodes for TSS in comparison to a single cathode.[25]

When considering electrode placement, it is important to appreciate electrode placement determines current flow, directly influencing activation of neural circuitry. A posterior cathode with anterior anode creates dorsal-ventral (DV) current flow, while a posterior cathode with anode creates dorsal-midline (DM) current flow. Understanding differences in PRM reflex responses among different electrode montages is important for standardizing spinal reflex excitability measurement and optimizing interventional TSS. A limited number of studies have compared PRM reflex outcomes using different electrode positions.[24, 26–28] However, electrode montages commonly used in the SCI literature have not been systematically compared.

The goal of this study was to compare PRM reflexes of 6 commonly reported electrode montages in individuals with intact neurological systems. We aimed to identify montage(s) associated with the largest PRM reflex response at the lowest stimulation intensity and to assess stimulation tolerability among montages. Based on what is known about electrical current penetration, we hypothesized montages generating DV current flow would activate Ia afferents more efficaciously than montages generating DM current flow, as indicated by lower soleus reflex thresholds (RTs) and larger area under the PRM reflex recruitment curve (AUC). We additionally hypothesized montages with DM current flow would be better tolerated than DV current flow due to the absence of abdominal contractions generated by anode placement.

This study was carried out with approval of the Shepherd Center Research Review Committee. All participants gave written informed consent prior to study enrollment in accordance with the Declaration of Helsinki. This study was funded by National Institutes of Health grant R01HD101812 (ECF-F) and the Hulse SCI Research Fund. The funders played no role in the design, conduct, or reporting of this study.

Participants

Individuals who met the following inclusion criteria were eligible for study participation: ≥18 years of age, no changes in prescription medication use over the prior 2 weeks, ability and willingness to authorize use of protected health information, ability to follow multiple directions, and ability to communicate pain/discomfort. Individuals were excluded from study participation if they had any of the following exclusion criteria: history of neurologic injury/disease, or cardiovascular irregularities, current pregnancy, implanted stimulators, or skin lesions, irregularities, or sensitivities.

Electrode Montages

Utilizing a randomized, crossover design, six electrode montages were tested in pairs over three sessions. Electrode montages evaluated in this study were selected based on use in recent SCI literature and named based on expected current flow. Order of montage testing was randomized both between and within sessions. Participant allocation to each montage order is depicted in the CONSORT diagram (Fig. 1A). Sessions were separated by ≥ 24 hours to prevent potential carryover effects of repeated electrical stimulation. Four DV and two DM montages were evaluated (Fig. 2). DV montages included [1] dorsal-ventral iliac crests (DV-I): cathode over T11/T12 and anodes over iliac crests, [2] dorsal-ventral umbilicus (DV-U): cathode over T11/T12 and anodes over umbilicus, [3] dorsal-ventral paraspinal iliac crests (DV-PI): paraspinal cathodes at T11/12 and anodes over iliac crests, and [4] dorsal-ventral paraspinal umbilicus (DV-PU): paraspinal cathodes at T11/12 and anodes over umbilicus. DM montages included [5] dorsal-midline caudal (DM-C): cathode over T11/12 and anode 5cm caudal, and [6] dorsal-midline rostral (DM-R): cathode over T11/12 and anode 5cm rostral. For DV montages, cathode(s) were 5cm round electrodes and anodes were 9x5cm rectangular interconnected electrodes. For DM montages, 5cm round electrodes were used for cathode and anode. For umbilical montages, anodes were placed 5cm apart on either side of the umbilicus. For iliac crest montages, anodes were placed with superior border oriented laterally and inferior border oriented medially. Manual palpation of spinous processes by a physical therapist (author KLT/EBS) was used to determine cathode placement. Skin under the cathode was swabbed with isopropyl alcohol and abraded (NuPrep, Weaver and Company, Aurora, CO) to decrease impedance. Conductive gel (Spectra 360, Parker Laboratories, Fairfield, NJ) was placed around the border of the cathode(s) to reduce risk of skin irritation at the interface between skin tissue and electrode edge. To ensure consistent stimulating electrode placement across sessions, cathode location and referencing anatomical landmarks were marked on transparent film and used for reference in subsequent sessions.

Figure 1. A) Consort diagram outlining participant enrollment and randomization order. In this cross-over design, participants were randomized into groups dictating the order they received six electrode montages. All participants received all six montages across three sessions. B) Participant demographics and reflex thresholds obtained per individual in each montage.

Figure 2. Cathode (black) and anode (red) positions for 6 electrode montages (left) with associated recruitment curves (center) and posterior root muscle reflex response traces elicited by a stimulation intensity of 1.2xRT (right) from a representative participant (P13). Dorsal-ventral umbilicus (DV-U): cathode over T11/T12 with anodes over the umbilicus; dorsal-ventral iliac crests (DV-I): cathode over T11/T12 with anodes over iliac crests; dorsal-ventral paraspinal umbilicus (DV-PU): paraspinal cathodes at T11/12 with anodes over the umbilicus; dorsal-ventral paraspinal iliac crests (DV-PI): paraspinal cathodes at T11/12 with anodes over iliac crests; dorsal-midline caudal (DM-C): cathode over T11/12 with an anode 5 cm caudal; and dorsal-midline rostral (DM-R): cathode over T11/12 with an anode 5 cm rostral. Blue = dominant lower extremity, red = nondominant lower extremity, inverted triangles = time of stimulus application. Note the different scales for response traces across montages.

PRM Reflexes

Electromyographic activity (EMG) was measured in the soleus muscle bilaterally using pre-amplified surface EMG electrodes (Motion Lab Systems, Baton Rouge, LA). Prior to EMG electrode placement, the skin over the soleus was swabbed with isopropyl alcohol and abraded to decrease impedance. To ensure consistent EMG electrode placement across sessions, EMG electrode location and referencing anatomical landmarks were marked on stockinette sleeves placed on each lower extremity. EMG signals were acquired (MA300, Motion Lab Systems, Baton Rouge, LA), digitized (Power 1401, Cambridge Electronic Design, Cambridge, UK), and recorded at a sampling rate of 2 kHz for offline analysis [17, 29, 30] using sweep-based data capture and analysis software (Signal, Cambridge Electronic Design, Cambridge, UK).

Data were acquired with the participant lying supine. Pillows supported the participant’s head and knees as needed for comfort. As position influences PRM reflex responses,[31] position was determined during the first session and remained consistent across subsequent sessions within each participant. PRM reflexes were elicited by monophasic, rectangular stimulation pulses with a 1ms pulse width[32] using a constant current stimulator (Digitimer DS7AH, Hertforshire, UK). Paired stimulation pulses (40ms inter-pulse interval)[33] were delivered (Grass S88X, Natus Neurology, Middleton, WI) with a minimum of 7 seconds between pulse pairs.[24] Paired pulses confirmed the reflex origin of the evoked responses, as depression of the second response confirms activation of primary afferent neurons (as opposed to direct activation of the anterior motor root).[13, 16] Depression was determined by calculating the difference between response amplitudes for the first and second stimuli and dividing by the response amplitude of the first stimulus. RT was defined as the stimulation intensity required to elicit a peak-to-peak PRM reflex response amplitude of ≥ 100 µV in at least 50% of trials[20] in each lower extremity independently. PRM reflex recruitment curves were collected beginning at a subthreshold stimulation intensity (≤ 30mA, dependent upon RT). Stimulation intensity was increased in increments of 5mA until the soleus response plateaued or a stimulation intensity of 100mA was achieved. A maximum stimulation intensity of 100mA was imposed to avoid acute discomfort accompanying higher stimulation intensities. In addition to the 5mA increments in stimulation intensity, PRM reflexes were collected at the stimulation intensity equivalent to 120% of RT (1.2xRT) if 1.2xRT fell below 100mA. Three or five stimuli were repeated at each stimulation intensity for subthreshold responses and responses ≥ RT, respectively. Following recruitment curve acquisition, participants reported stimulation tolerability for each montage on a 0–10 visual analog scale with 0 indicating “absolutely tolerable” and 10 indicating “not at all tolerable.” Sensation descriptors contributing to the participant’s tolerability rating were recorded.

Extremity Dominance

The dominant lower extremity was determined by the following question taken from the Waterloo Footedness Questionnaire-Revised, “If you were asked to shoot a ball on target, which leg would you use to shoot the ball?”.[34]

Data Processing

Peak-to-peak soleus PRM reflex response amplitudes were exported and processed using custom MATLAB codes (MathWorks, Inc., Natick, MA). For all participants, soleus recruitment curves were generated for the dominant and nondominant leg for each montage by averaging PRM reflex response amplitudes for each stimulation intensity tested. Within each recruitment curve, stimulation intensity, s, was normalized to the acquired RT (i.e. s/RT) to account for inter-individual variability. Using a modified Boltzmann equation, non-linear curve fitting was performed for all recruitment curves for which both RT and 1.2xRT were obtained:

$$\:PRMR\left(s\right)=\:\frac{{PRMR}_{max}}{1+\:{e}^{m({S}_{50}-s)}}\:,$$

where PRMR_max is the maximum response amplitude estimated by the function, S₅₀ is the stimulation intensity required to elicit a response amplitude 50% of PRMR_max, and m is the slope parameter of the Boltzmann function (Fig. 3).[29, 35] The Levenberg-Marquardt algorithm (lsqcurvefit, Optimization Toolbox, The MathWorks, Natick, MA) was used to solve non-linear least-squares curve fitting. Initial guess inputs for the Levenberg-Marquardt algorithm were calculated as follows: PRMR_max – maximum average response amplitude acquired during data collection; S₅₀ – mean normalized stimulation intensity averaged from the nearest data point above and below the value of 50% of PRMR_max; m – slope of a linear regression fitted to the data points from RT to 1.2xRT, inclusive. Default termination criteria for the optimization function were used (600 for the maximum number of function evaluations and 1e^− 6 for function tolerance). Area under the recruitment curve (AUC) was calculated by numerically integrating the optimized Boltzmann equation from RT to S₅₀.

Figure 3. Model posterior root muscle reflex recruitment curve with labeled outcomes. Values acquired during data collection include RT. Values derived from curve fitting include S₅₀, AUC, and PRMRmax. Reflex threshold (RT): stimulation intensity required to elicit a reflex response > 100µV in at least 50% of trials; S₅₀: stimulation intensity required to elicit a response that is 50% of maximum posterior root muscle reflex amplitude (PRMRmax); Area under the curve (AUC): integrated recruitment curve area from RT to S₅₀.

Statistical Analysis

Statistical analyses were completed in R (R Foundation for Statistical Computing, Vienna, Austria). Outcomes were visualized and descriptive statistics calculated to check model assumptions and identify potential outliers. For each outcome, multilevel models with random intercept for participant and fixed effects for montage, lower extremity, and interaction between montage and lower extremity were the starting point for analyses, with modifications described below for specific tests. Pairwise comparisons were calculated as differences in estimated marginal means using the Tukey method for p-value adjustment for multiple testing and Kenward-Roger degrees of freedom following estimation of the multilevel model.

To assess the proportion of variability in each outcome explained by the participant for each montage, we used ICC model ICC(1,1). We selected this model as it does not include lower extremity in the underlying multilevel model, thereby reflecting an assumption that the dominant and nondominant lower extremities are exchangeable.

Fifteen adults with intact neurological systems, aged 24–60 years, participated in this study (Fig. 1B). Data from 134 recruitment curves were included in final analyses. Of the 180 recruitment curves collected (15 participants x 2 lower extremities x 6 montages), RT and 1.2xRT were not acquired for 36 and 10 curves respectively, as acquiring these data would have required us to exceed our a priori stimulation tolerability limit of 100mA. For all study outcomes, estimated marginal means and confidence intervals are reported in Table 1.

Table 1

Marginal mean values followed by model-based 95% confidence intervals [lower limit, upper limit] for dominant (D) and nondominant (ND) lower extremities for reflex threshold (RT), response amplitude (RA) at 1.2xRT, area under the curve (AUC) from RT to S50, 1.2xRT, and S50.
		DV-U	DV-I	DV-PU	DV-PI	DM-C	DM-R
RT (mA)	D	46.7 [33.9, 59.4]	48.1 [35.3, 60.8]	64.3 [51.4, 77.1]	64.9 [52.1, 77.7]	60.0 [46.9, 73.1]	63.1 [50.3, 76.0]
RT (mA)	ND	45.4 [32.5, 58.2]	45.4 [32.5, 58.2]	61.7 [48.8, 74.6]	61.4 [48.5, 75.5]	63.6 [50.8, 76.5]	62.6 [49.8, 75.5]
RA @ 1.2xRT (µV)	D	2344 [1192, 3497]	2826 [1622, 4029]	2856 [1651, 4061]	2983 [1806, 4160]	2429 [1191, 3666]	2554 [1378, 3730]
RA @ 1.2xRT (µV)	ND	1632 [456, 2808]	1545 [368, 2722]	1791 [614, 2969]	1607 [430, 2784]	1893 [697, 3089]	2629 [1476, 3781]
AUC (µVxRT)	D	422 [275, 568]	382 [235, 528]	266 [116, 416]	264 [114, 414]	319 [160, 477]	276 [126, 426]
AUC (µVxRT)	ND	484 [334, 633]	518 [372, 665]	329 [179, 479]	278 [128, 428]	347 [188, 505]	250 [103, 396]
1.2xRT (mA)	D	49.1 [38.5, 59.8]	49.8 [39.1, 60.5]	68.7 [58.0, 79.4]	68.4 [57.7, 79.1]	67.0 [56.3, 77.7]	68.2 [57.6, 78.9]
1.2xRT (mA)	ND	49.6 [39.0, 60.2]	50.2 [39.5, 60.9]	69.2 [58.5, 79.9]	68.9 [58.2, 79.5]	67.4 [56.7, 78.1]	68.7 [58.1, 79.3]
S50 (mA)	D	52.8 [42.1, 63.4]	53.2 [42.5, 63.8]	71.5 [60.8, 82.1]	72.2 [61.5, 82.9]	70.1 [59.2, 80.9]	69.9 [59.2, 80.5]
S50 (mA)	ND	53.2 [42.6, 63.9]	53.6 [43.0, 64.2]	71.9 [61.3, 82.6]	72.6 [62.0, 83.3]	70.5 [59.7, 81.3]	70.3 [59.7, 81.0]

Confirmation of Reflex Responses:

Depression of the second stimulus response of the PRM reflex was used to confirm the evoked response was due to afferent fiber activation.[13, 16] The response amplitude corresponding to the second stimulus of the paired pulse was lower than the first response in 98.75% of RT trials and 99.21% of 1.2xRT trials.

Reflex Threshold (RT):

RT indicates the minimum stimulation intensity required to elicit a reflex response. Due to the a priori stimulation intensity maximum of 100mA, complete RT data (all 6 montages and both lower extremities) were acquired for 9 participants, partial RT data were acquired for 5 participants, and no RT data was obtained for 1 participant (P12). The DM-C montage had the lowest RT acquisition rate (Fig. 1B). Sensitivity analyses were undertaken to explore potential impacts of missing values on RT, including comparison of the lower extremity and montage of missing compared to non-missing RT, repeating analyses with values of 100mA and 120mA substituted for missing RT values, and repeating analyses excluding participants with partially missing RT. All analyses resulted in findings consistent with the following results.

Significant differences in RT were identified across montages (F(5, 117.92) = 23.95, p < 0.001). DV-U and DV-I demonstrated significantly lower soleus RT (mA) in dominant and nondominant lower extremities compared to other montages (Fig. 4A). Mean soleus RT for DV-U and DV-I were not statistically different (D:1.39, p = 1.00; ND:0, p = 1.00). No significant differences in RT were identified between dominant and nondominant lower extremities (F(1, 117.87) = 0.67, p = 0.42).

Figure 4. Recruitment curve outcomes in dominant (blue) and nondominant (red) lower extremities. * = p < 0.05; X = mean. For reflex threshold (A), DV-U is significantly lower than DV-PU, DV-PI, DM-C, and DM-R for both dominant and nondominant lower extremities. DV-I is significantly lower than DV-PU, DV-PI, DM-C, and DM-R for both dominant and nondominant lower extremities. For response amplitude at 1.2xRT (B), there are no significant differences across montages for either dominant or nondominant lower extremities.

Response Amplitude at 1.2x Reflex Threshold (1.2xRT):

Response amplitude at 1.2xRT reflects output of spinal circuitry within the ascending portion of the recruitment curve. Due to the stimulation intensity maximum of 100mA, complete response amplitude at 1.2xRT data were acquired for 6 participants, partial data were acquired for 7 participants, and no data was obtained for 1 participant (P12). There were no significant differences in peak-to-peak response amplitude at 1.2xRT among montages (F(5, 106.33) = 0.51, p = 0.77). Significant differences in response amplitude at 1.2xRT were identified between dominant and nondominant lower extremities (F(1, 106.09) = 11.73, p < 0.001), with dominant response amplitudes typically exceeding nondominant response amplitudes. On average, responses in the dominant lower extremity were 144% larger compared to the nondominant lower extremity (Fig. 4B).

Area Under the Curve (AUC):

The total output of activated spinal circuits can be assessed by calculating AUC between RT and S₅₀. Due to stimulation intensity maximum of 100 mA, complete AUC data were acquired for 8 participants, partial data were acquired for 4 participants, and no data was obtained for 1 participant (P12). There were significant differences in AUC among montages (F(5, 108.99) = 5.24, p < 0.001). Overall, AUC values were larger for DV-U and DV-I compared to other montages (Fig. 5C). Significant differences in AUC among montages were identified in the nondominant lower extremity only, with AUC for DV-U significantly larger than DM-R (233.69, p = 0.03, 95%CI: 15.7-451.7), and DV-I significantly larger than DM-R (268.46, p = 0.01, 95%CI: 55.7-481.3) and DV-PI (240.46, p = 0.02, 95%CI: 22.2-458.8). There was no significant difference in AUC between dominant and nondominant lower extremities (F(1, 108.60) = 2.14, p = 0.15).

Figure 5. Recruitment curve outcomes in dominant and nondominant lower extremities. * = p < 0.05; X = mean. There are no significant differences between 1.2xRT and S50 in the dominant (A) and nondominant (B) lower extremity. AUC (C) was generally higher in DV-U and DV-I montages with significance found only in the nondominant lower extremity where DV-U is significantly larger than DM-R and DV-I is significantly larger than DM-R and DV-PI.

Stimulation Intensity at S₅₀ compared to 1.2xRT:

S₅₀ is the stimulation intensity required to elicit a response that is 50% of PRMR_max. S₅₀ represents the point of the recruitment curve with the steepest slope and, therefore, has the greatest sensitivity to change when used as a measurement. Due to the stimulation intensity maximum of 100mA, complete data for S₅₀ were derived for 8 participants, partial data were derived for 4 participants, and no data was obtained for 1 participant (P12). As with 1.2xRT, on average, S₅₀ was lower in DV-U and DV-I compared to other montages (Table 1). When comparing the stimulation intensities of 1.2xRT and S₅₀ within montages, post-hoc tests did not identify any significant differences between these two measures in either lower extremity (Fig. 5A & 5B).

Tolerability:

Twelve participants (P4-P15) provided tolerability ratings. Differences in tolerability ratings among montages were not significant (F(5, 55) = 1.13, p = 0.35). The estimated marginal mean tolerability rating and corresponding 95% CI per montage were as follows, DV-U: 4.17 [2.78, 5.55], DV-I: 4.33 [3.10, 5.57], DV-PU: 3.92 [2.97, 4.87], DV-PI: 3.75 [2.54, 4.96], DM-C: 4.92 [3.38, 6.45], DM-R: 4.17 [2.72, 5.61]. “Muscle contraction” and “sharpness” were the most commonly reported sensory descriptors contributing to higher tolerability ratings.

Intra-individual Variability:

Separating observations by lower extremity (dominant or nondominant) and response amplitude (RT or 1.2xRT), we computed intra-individual means (iMeans) and intra-individual standard deviations (iSDs) for peak-to-peak response amplitudes for each montage. We calculated proportional iSDs (iSD/iMean) to produce a single value reflecting the intra-individual variability for each montage and limb. Proportional iSDs were generally lower for 1.2xRT as compared to RT. Separate analyses for RT and 1.2xRT found no differences in proportional iSD among montages (RT: F(5, 121.82) = 0.62, p = 0.68; 1.2xRT: F(5, 103.71) = 0.42, p = 0.83). Proportional iSD was lower for the nondominant lower extremity at RT (F(1, 121.45) = 6.76, p = 0.01, estimated difference = 0.08, 95% CI: 0.02–0.15), but generally lower for the dominant lower extremity at 1.2xRT (F(1, 102.8) = 4.16, p = 0.04, estimated difference = 0.03, 95% CI: 0.00-0.06).

Neuromodulatory interventions can be a valuable adjuvant to physical therapy. The activation of Ia afferents can both increase motoneuron excitability to augment volitional activation] and activate inhibitory circuits to decrease reflex excitability.[36] For this reason, it is valuable for physical therapists to know what electrode montages best activate these neural structures. We found DV-U and DV-I montages were associated with significantly lower soleus RT in both dominant and nondominant lower extremities compared to other montages, indicating activation of spinal circuitry was achieved at lower stimulation intensities using these montages. DV-U and DV-I montages also had larger AUCs, with significant differences among montages observed only in the nondominant lower extremity. This indicates greater overall activation of spinal circuits across stimulation intensities between RT and S₅₀ with these two montages compared to the other montages evaluated.

TSS is typically delivered at stimulation intensities near RT.[6, 19] Given DV-U and DV-I montages demonstrated lower RTs, these montages may require lower stimulation intensities to activate neural structures targeted during interventional TSS. Lower stimulation intensities decrease risk for skin burns and improve tolerability. Our results differ from a previous study that found no significant differences between montages analogous to DV-U and DM-R.[24] A more recent study found montages using two cathodes at midline and a montage using a rectangular cathode at midline had lower RT values than a montage analogous to DV-I.[28] We used conventional electrophysiologic analyses to compare the TSS montages most commonly used in the SCI literature, some differences identified with other montages could be attributable to differences in methodology. For example, while most studies construct recruitment curves based on response amplitude at a specified stimulation intensity, both of the aforementioned studies constructed recruitment curves based on area under the full-wave rectified waveform. Also while most studies base RT on values that have been acquired through direct measurement, both of these studies used RT values derived from curve fitting. Finally, while most investigators define RT based on the stimulation intensity required to elicit a response of a pre-specified amplitude, both of these studies used a nonstandard definition of RT using calculated values based on slope of the recruitment curve. [24, 28] Therefore, methodological variations may account for differences in results between previous studies and this study.

At 1.2xRT, there were no significant differences in peak-to-peak response amplitude among montages. While some investigators have analyzed differences in response amplitudes at PRMRmax between montages,[24, 28] there has been no prior comparison of response amplitude at a normalized stimulation intensity on the ascending portion of the curve. Interventional TSS commonly uses stimulation intensities below those required to elicit maximum responses. For this reason, it is important to understand responses elicited at intensities used in clinical application.

AUC provides insight into the overall output of neural circuits for a given individual. DV-U and DV-I montages have larger AUCs in comparison to other montages, indicating recruitment of more neural structures at stimulation intensities between RT and S₅₀. To our knowledge, this is the first study to analyze differences in AUC among montages. When stimulation is delivered concurrently with volitional effort, intensities at and above RT have been demonstrated to increase muscle activation in persons with neurological injury.[7] Greater recruitment of neural structures by TSS, as denoted by larger AUC, may lead to enhanced rehabilitation outcomes.

S₅₀ is another commonly used measure of neural excitability, including motor evoked potentials from transcranial magnetic stimulation and H-reflex testing. [36–38] Stimulating at an intensity corresponding to S₅₀, the point with greatest potential for modulation, allows for greater sensitivity to identify change due to an intervention. Within a specific montage, when we compared values for 1.2xRT versus S₅₀, there were no significant differences found between these measures. This finding suggests these two measures could be used interchangeably when using PRM reflexes to measure changes in spinal reflex excitability. Using 1.2xRT as a proxy for S₅₀ would make it possible to measure changes in PRM reflexes following an intervention without the need to obtain a full recruitment curve.

We hypothesized DM montages may be more tolerable due to absence of abdominal contractions. However, our results demonstrated no significant differences in tolerability rating among montages. Given DV-U and DV-I demonstrate similar electrophysiological outcomes, participant preference may be used to determine anode placement between these two options.[30, 39]

Based on our findings that DV-U and DV-I montages demonstrated lower RTs and larger AUCs, we recommend utilizing DV-U or DV-I montages to optimize Ia afferent recruitment during interventional TSS and PRM reflex assessments for the soleus. Our participants had no history of neurological injury and PRM reflexes were elicited under the same conditions; thus, we conclude the observed results reflect differences in recruitment of neural structures, particularly afferent fibers at the dorsal nerve roots. Future studies should assess the impact of different montages when paired with physical therapy interventions.

While the stimulation devices used in this study were not clinical devices, interventional TSS can be applied with meaningful effects using clinically accessible stimulators.[1, 5, 40] TSS has also been used to modulate spinal reflex excitability and reduce spasticity based on activation of spinal inhibitory circuits,[5, 6] with one study using stimulators commonly available in the clinic.[5] Generally, lower frequencies (e.g., 30 Hz) are used for motor activation while higher frequencies (e.g., 50Hz) are used to reduce spasticity. Some devices that are not yet clinically available use a high frequency carrier wave, which is believed to improve comfort of the stimulation. However, this claim has been questioned, with evidence that high frequency carrier waves are associated with lower levels of neural activation, leading to less motor activation along with perception of greater comfort.[41, 42]

Since the effects of TSS are based on activation of peripheral sensory fibers, it bears similarity to other forms of afferent input used for neuromodulation.[43–45] In addition to the activation of spinal circuits, stimulation of peripheral sensory afferents has been shown to influence the excitability of cortical circuits and to promote adaptive neuroplasticity when combined with training.[45–47] The advantage of TSS is that a single cathodal electrode can activate multiple spinal roots, while peripheral nerve stimulation is more selective by virtue of electrode location.

Limitations:

Our study acquired PRM reflexes up to 100 mA to avoid acute discomfort; therefore, our results cannot be applied to individuals whose RT falls above 100 mA. Although this study only included individuals with intact neurological systems, the anatomical location of the dorsal nerve roots, which influences PRM reflex acquisition, is the same in individuals with SCI. Therefore, we predict our findings of montage differences will apply to people with SCI. Since stimulation during PRM reflex assessments involved paired pulses separated by at least 7 seconds compared to the continuous stimulation typically applied during intervention, our results may not have a direct correlation with tolerability to interventional TSS. Additionally, the tolerability of a monophasic waveform may be different than a biphasic waveform, most commonly used for interventional TSS.

TSS: Transcutaneous spinal stimulation

SCI: Spinal cord injury

PRM reflexes: Posterior root-muscle reflexes

DV: Dorsal-Ventral

DM: Dorsal-Midline

RT: Reflex Threshold

DV-I: Dorsal-Ventral Iliac Crests

DV-U: Dorsal-Ventral Umbilicus

DV-PI: Dorsal-Ventral Paravertebral Iliac Crests

DV-PU: Dorsal-Ventral Paravertebral Umbilicus

DM-C: Dorsal-Ventral Caudal

DM-R: Dorsal-Ventral Rostral

EMG: Electromyographic

PRMRmax: Posterior root-muscle reflex maximum response amplitude of the recruitment curve

AUC: Area Under the Recruitment Curve

iMean: Intra-Individual Means

iSD: Intra-Individual Standard Deviations

Ethics Approval and Consent to Participate: This study was carried out with approval of the Shepherd Center Research Review Committee. All participants gave written informed consent prior to study enrollment in accordance with the Declaration of Helsinki.

Consent for Publication: All participants gave written informed consent for the publication of the study.

Availability of Data and Materials: The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing Interests: Author ECF-F serves as a consultant to the medical device company ONWARD Inc.

Funding: This study was funded by National Institutes of Health grant R01HD101812 (ECF-F) and the Hulse SCI Research Fund.

Authors’ Contributions: Conceptualization: KLT, KEN, EBS, JAI, ECF-F; Data Curation: KLT, JAI, EBS; Formal Analysis: EBS, JAI, KEN; Funding Acquisition: ECF-F; Investigation: KLT, EBS, OD, JAI; Methodology: KLT, KEN, EBS, JAI, ECF-F; Project Administration: KLT, EBS, JAI; Resources: KLT, ECF-F; Software: OD; Supervision: EBS, JAI, ECF-F; Validation: KLT, EBS, OD, JAI; Visualization: KLT, KEN, EBS, JAI; Writing (original draft): KLT, KEN, EBS, OD, JAI; Writing (review & editing): KLT, KEN, EBS, OD, JAI, ECF-F

Acknowledgements: The authors gratefully acknowledge the mentorship of Dr. Karen Minassian, PhD.

Estes S, Zarkou A, Hope JM, Suri C, Field-Fote EC. Combined Transcutaneous Spinal Stimulation and Locomotor Training to Improve Walking Function and Reduce Spasticity in Subacute Spinal Cord Injury: A Randomized Study of Clinical Feasibility and Efficacy. J Clin Med. 2021;10(6).
Samejima S, Caskey CD, Inanici F, Shrivastav SR, Brighton LN, Pradarelli J et al. Multisite Transcutaneous Spinal Stimulation for Walking and Autonomic Recovery in Motor-Incomplete Tetraplegia: A Single-Subject Design. Phys Ther. 2022;102(1).
McHugh LV, Miller AA, Leech KA, Salorio C, Martin RH. Feasibility and utility of transcutaneous spinal cord stimulation combined with walking-based therapy for people with motor incomplete spinal cord injury. Spinal Cord Ser Cases. 2020;6(1):104.
Tefertiller C, Rozwod M, VandeGriend E, Bartelt P, Sevigny M, Smith AC. Transcutaneous Electrical Spinal Cord Stimulation to Promote Recovery in Chronic Spinal Cord Injury. Front Rehabil Sci. 2021;2.
Estes SP, Iddings JA, Field-Fote EC. Priming Neural Circuits to Modulate Spinal Reflex Excitability. Front Neurol. 2017;8:17.
Hofstoetter US, Freundl B, Danner SM, Krenn MJ, Mayr W, Binder H, et al. Transcutaneous Spinal Cord Stimulation Induces Temporary Attenuation of Spasticity in Individuals with Spinal Cord Injury. J Neurotrauma. 2020;37(3):481–93.
Sayenko DG, Rath M, Ferguson AR, Burdick JW, Havton LA, Edgerton VR, et al. Self-Assisted Standing Enabled by Non-Invasive Spinal Stimulation after Spinal Cord Injury. J Neurotrauma. 2019;36(9):1435–50.
Hofstoetter US, Freundl B, Lackner P, Binder H. Transcutaneous Spinal Cord Stimulation Enhances Walking Performance and Reduces Spasticity in Individuals with Multiple Sclerosis. Brain Sci. 2021;11(4).
Kreydin EI, Abedi A, Montero VS, Morales L, Jen R, Perez L et al. A Pilot Study of the Effect of Transcutaneous Spinal Cord Stimulation on Micturition-Related Brain Activity and Lower Urinary Tract Symptoms After Stroke. J Urol. 2023:101097JU0000000000003776.
Qian Q, Ling YT, Zhong H, Zheng YP, Alam M. Restoration of arm and hand functions via noninvasive cervical cord neuromodulation after traumatic brain injury: a case study. Brain Inj. 2020;34(13–14):1771–80.
Gad P, Hastings S, Zhong H, Seth G, Kandhari S, Edgerton VR. Transcutaneous Spinal Neuromodulation Reorganizes Neural Networks in Patients with Cerebral Palsy. Neurotherapeutics. 2021;18(3):1953–62.
Hofstoetter US, Freundl B, Binder H, Minassian K. Common neural structures activated by epidural and transcutaneous lumbar spinal cord stimulation: Elicitation of posterior root-muscle reflexes. PLoS ONE. 2018;13(1):e0192013.
Danner SM, Hofstoetter US, Ladenbauer J, Rattay F, Minassian K. Can the human lumbar posterior columns be stimulated by transcutaneous spinal cord stimulation? A modeling study. Artif Organs. 2011;35(3):257–62.
Hofstoetter US, Hofer C, Kern H, Danner SM, Mayr W, Dimitrijevic MR et al. Effects of transcutaneous spinal cord stimulation on voluntary locomotor activity in an incomplete spinal cord injured individual. Biomed Tech (Berl). 2013;58 Suppl 1.
Minassian K, Hofstoetter US, Danner SM, Mayr W, Bruce JA, McKay WB, et al. Spinal Rhythm Generation by Step-Induced Feedback and Transcutaneous Posterior Root Stimulation in Complete Spinal Cord-Injured Individuals. Neurorehabil Neural Repair. 2016;30(3):233–43.
Minassian K, Persy I, Rattay F, Dimitrijevic MR, Hofer C, Kern H. Posterior root-muscle reflexes elicited by transcutaneous stimulation of the human lumbosacral cord. Muscle Nerve. 2007;35(3):327–36.
Courtine G, Harkema SJ, Dy CJ, Gerasimenko YP, Dyhre-Poulsen P. Modulation of multisegmental monosynaptic responses in a variety of leg muscles during walking and running in humans. J Physiol. 2007;582(Pt 3):1125–39.
Andrews JC, Stein RB, Roy FD. Post-activation depression in the human soleus muscle using peripheral nerve and transcutaneous spinal stimulation. Neurosci Lett. 2015;589:144–9.
Meyer C, Hofstoetter US, Hubli M, Hassani RH, Rinaldo C, Curt A et al. Immediate Effects of Transcutaneous Spinal Cord Stimulation on Motor Function in Chronic, Sensorimotor Incomplete Spinal Cord Injury. J Clin Med. 2020;9(11).
Megía-García Á, Serrano-Muñoz D, Comino-Suárez N, Del-Ama AJ, Moreno JC, Gil-Agudo A, et al. Effect of posture and body weight loading on spinal posterior root reflex responses. Eur J Neurosci. 2021;54(7):6575–86.
Hofstoetter US, Krenn M, Danner SM, Hofer C, Kern H, McKay WB, et al. Augmentation of Voluntary Locomotor Activity by Transcutaneous Spinal Cord Stimulation in Motor-Incomplete Spinal Cord-Injured Individuals. Artif Organs. 2015;39(10):E176–86.
Gerasimenko Y, Gorodnichev R, Moshonkina T, Sayenko D, Gad P, Reggie Edgerton V. Transcutaneous electrical spinal-cord stimulation in humans. Ann Phys Rehabil Med. 2015;58(4):225–31.
Dy CJ, Gerasimenko YP, Edgerton VR, Dyhre-Poulsen P, Courtine G, Harkema SJ. Phase-dependent modulation of percutaneously elicited multisegmental muscle responses after spinal cord injury. J Neurophysiol. 2010;103(5):2808–20.
Krenn MJ, Vargas Luna JL, Mayr W, Stokic DS. Bipolar transcutaneous spinal stimulation evokes short-latency reflex responses in human lower limbs alike standard unipolar electrode configuration. J Neurophysiol. 2020;124(4):1072–82.
Tran K, Steele A, Crossnoe R, Martin C, Sayenko DG. Multi-site lumbar transcutaneous spinal cord stimulation: When less is more. Neurosci Lett. 2024;820:137579.
Roy FD, Gibson G, Stein RB. Effect of percutaneous stimulation at different spinal levels on the activation of sensory and motor roots. Exp Brain Res. 2012;223(2):281–9.
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):174–81.
Skiadopoulos A, Pulverenti TS, Knikou M. Physiological effects of cathodal electrode configuration for transspinal stimulation in humans. J Neurophysiol. 2022;128(6):1663–82.
Knikou M. Spinal Excitability Changes after Transspinal and Transcortical Paired Associative Stimulation in Humans. Neural Plast. 2017;2017:6751810.
Murray LM, Knikou M. Transspinal stimulation increases motoneuron output of multiple segments in human spinal cord injury. PLoS ONE. 2019;14(3):e0213696.
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.
Militskova A, Mukhametova E, Fatykhova E, Sharifullin S, Cuellar CA, Calvert JS, et al. Supraspinal and Afferent Signaling Facilitate Spinal Sensorimotor Network Excitability After Discomplete Spinal Cord Injury: A Case Report. Front Neurosci. 2020;14:552.
Minassian K, Jilge B, Rattay F, Pinter MM, Binder H, Gerstenbrand F, et al. Stepping-like movements in humans with complete spinal cord injury induced by epidural stimulation of the lumbar cord: electromyographic study of compound muscle action potentials. Spinal Cord. 2004;42(7):401–16.
van Melick N, Meddeler BM, Hoogeboom TJ, Nijhuis-van der Sanden MWG, van Cingel REH. How to determine leg dominance: The agreement between self-reported and observed performance in healthy adults. PLoS ONE. 2017;12(12):e0189876.
Iyer PC, Madhavan S. Characterization of stimulus response curves obtained with transcranial magnetic stimulation from bilateral tibialis anterior muscles post stroke. Neurosci Lett. 2019;713:134530.
Knikou M, Murray LM. Repeated transspinal stimulation decreases soleus H-reflex excitability and restores spinal inhibition in human spinal cord injury. PLoS ONE. 2019;14(9):e0223135.
Temesi J, Gruet M, Rupp T, Verges S, Millet GY. Resting and active motor thresholds versus stimulus-response curves to determine transcranial magnetic stimulation intensity in quadriceps femoris. J Neuroeng Rehabil. 2014;11:40.
Kukke SN, Paine RW, Chao CC, de Campos AC, Hallett M. Efficient and reliable characterization of the corticospinal system using transcranial magnetic stimulation. J Clin Neurophysiol. 2014;31(3):246–52.
Islam MA, Pulverenti TS, Knikou M. Neuronal Actions of Transspinal Stimulation on Locomotor Networks and Reflex Excitability During Walking in Humans With and Without Spinal Cord Injury. Front Hum Neurosci. 2021;15:620414.
Sandler EB, Condon K, Field-Fote EC. Efficacy of Transcutaneous Spinal Stimulation versus Whole Body Vibration for Spasticity Reduction in Persons with Spinal Cord Injury. J Clin Med. 2021;10(15).
Dalrymple AN, Hooper CA, Kuriakose MG, Capogrosso M, Weber DJ. Using a high-frequency carrier does not improve comfort of transcutaneous spinal cord stimulation. J Neural Eng. 2023;20(1).
Manson GA, Calvert JS, Ling J, Tychhon B, Ali A, Sayenko DG. The relationship between maximum tolerance and motor activation during transcutaneous spinal stimulation is unaffected by the carrier frequency or vibration. Physiol Rep. 2020;8(5):e14397.
Beekhuizen KS, Field-Fote EC. Sensory stimulation augments the effects of massed practice training in persons with tetraplegia. Arch Phys Med Rehabil. 2008;89(4):602–8.
Hoffman L, Field-Fote E. Effects of practice combined with somatosensory or motor stimulation on hand function in persons with spinal cord injury. Top Spinal Cord Inj Rehabil. 2013;19(4):288–99.
Gomes-Osman J, Field-Fote EC. Cortical vs. afferent stimulation as an adjunct to functional task practice training: a randomized, comparative pilot study in people with cervical spinal cord injury. Clin Rehabil. 2015;29(8):771–82.
Iddings JA, Zarkou A, Field-Fote EC. Noninvasive neuromodulation and rehabilitation to promote functional restoration in persons with spinal cord injury. Curr Opin Neurol. 2021;34(6):812–8.
Ji Q, He H, Zhang C, Lu C, Zheng Y, Luo XT, et al. Effects of whole-body vibration on neuromuscular performance in individuals with spinal cord injury: a systematic review. Clin Rehabil. 2017;31(10):1279–91.

Competing interest reported. Author ECF-F serves as a consultant to the medical device company ONWARD Inc.

Download PDF

Reviews received at journal
12 Sep, 2024
Reviews received at journal
07 Sep, 2024
Reviewers agreed at journal
24 Aug, 2024
Reviewers agreed at journal
23 Aug, 2024
Reviewers agreed at journal
20 Aug, 2024
Reviewers agreed at journal
19 Aug, 2024
Reviewers invited by journal
18 Aug, 2024
Editor assigned by journal
15 Jul, 2024
Submission checks completed at journal
15 Jul, 2024
First submitted to journal
10 Jul, 2024

You are reading this latest preprint version

Optimizing Transcutaneous Spinal Stimulation: Excitability of Evoked Spinal Reflexes is Dependent on Electrode Montage

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Clinical Trial Number:

Figures

BACKGROUND

METHODS

Participants

Electrode Montages

PRM Reflexes

Extremity Dominance

Data Processing

Statistical Analysis

RESULTS

Confirmation of Reflex Responses:

Reflex Threshold (RT):

Response Amplitude at 1.2x Reflex Threshold (1.2xRT):

Area Under the Curve (AUC):

Stimulation Intensity at S₅₀ compared to 1.2xRT:

Tolerability:

Intra-individual Variability:

DISCUSSION

CONCLUSIONS

Limitations:

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1

Optimizing Transcutaneous Spinal Stimulation: Excitability of Evoked Spinal Reflexes is Dependent on Electrode Montage

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Clinical Trial Number:

Figures

BACKGROUND

METHODS

Participants

Electrode Montages

PRM Reflexes

Extremity Dominance

Data Processing

Statistical Analysis

RESULTS

Confirmation of Reflex Responses:

Reflex Threshold (RT):

Response Amplitude at 1.2x Reflex Threshold (1.2xRT):

Area Under the Curve (AUC):

Stimulation Intensity at S50 compared to 1.2xRT:

Tolerability:

Intra-individual Variability:

DISCUSSION

CONCLUSIONS

Limitations:

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1

Stimulation Intensity at S₅₀ compared to 1.2xRT: