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

Abstract

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.

Keywords: Electrodes; Neuromodulation; Spinal Excitability; Transcutaneous Spinal Stimulation.

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.

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 ${\text{S}}_{50}$ AUC, and PRMRmax. Reflex threshold (RT): stimulation intensity required to elicit a reflex response >100μV in at least 50% of trials; ${\text{S}}_{50}$: 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 ${\text{S}}_{50}$.

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.

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.