Evaluation of optimal electrode configurations for epidural spinal cord stimulation in cervical spinal cord injured rats

Abstract

Background: Epidural spinal cord stimulation is a promising technique for modulating the level of excitability and reactivation of dormant spinal neuronal circuits after spinal cord injury (SCI). We examined the ability of chronically implanted epidural stimulation electrodes within the cervical spinal cord to (1) directly elicit spinal motor evoked potentials (sMEPs) in forelimb muscles and (2) determine whether these sMEPs can serve as a biomarker of forelimb motor function after SCI.

New method: We implanted EMG electrodes in forelimb muscles and epidural stimulation electrodes at C6 and C8 in adult rats. After recovering from a dorsal funiculi crush (C4), rats were tested with different stimulation configurations and current intensities to elicit sMEPs and determined forelimb grip strength.

Results: sMEPs were evoked in all muscles tested and their characteristics were dependent on electrode configurations and current intensities. C6(-) stimulation elicited more robust sMEPs than stimulation at C8(-). Stimulating C6 and C8 simultaneously produced better muscle recruitment and higher grip strengths than stimulation at one site.

Comparison with existing method(s): Classical method to select the most optimal stimulation configuration is to empirically test each combination individually for every subject and relate to functional improvements. This approach is impractical, requiring extensively long experimental time to determine the more effective stimulation parameters. Our proposed method is fast and physiologically sound.

Conclusions: Results suggest that sMEPs from forelimb muscles can be useful biomarkers for identifying optimal parameters for epidural stimulation of the cervical spinal cord after SCI.

Keywords: Cervical spinal cord injury; Dorsal funiculi crush; Epidural stimulation; Motor evoked potentials.

Fig. 1

Rat spinal cord injury model. A dorsal funiculi crush at the C4 spinal segment (shown by a cross) damages the dorsal corticospinal tract projections (shown in inset on right with a 0.2 mm white scale bar; * spinal cord lesion site). Epidural electrodes are implanted at the C6 and C8 spinal cord segments (shown by rectangles on the spinal cord).

Fig. 2

Average minimum threshold current for each stimulation electrode configuration to elicit motor evoked responses (sMEPs) in all five forelimb muscles implanted with EMG electrodes (deltoid, biceps brachii, pronator teres, flexor digitorum, and extensor digitorum).

Fig. 3

(A) The threshold stimulation currents for inducing sMEPs in all five forelimb muscles for the six electrode configurations are shown. Each panel shows multiple plots (superimposed) of sMEPs induced by a single stimulation current. The last row indicates the stimulation pulse at time 0 s. (B) Average (±SEM) peak amplitude of the early (0–9 ms) and late (9–25 ms) responses of sMEPs for all six stimulation electrode configurations for each of the five muscles. * Significantly different from C6– C8+. + Significantly different from C6– Ref+.

Fig. 4

Average rectified sMEPs for each of the five muscles induced by spinal cord stimulation at different current intensities using all six stimulation electrode configurations. Each 3-D panel indicates the rectified sMEPs with increasing stimulation current intensities for one stimulation electrode configuration. X-axis is latency, Y-axis is current intensity, and Z-axis is rectified voltage amplitudes.

Fig. 5

Muscle recruitment patterns for all six stimulation electrode configurations. Each panel indicates recruitment pattern for each muscle as a function of increasing stimulation current intensities. X-axis indicates an increase in the stimulation current from threshold to saturation expressed as a percentage of saturation and the Y-axis indicates the integrated area under the rectified sMEP curves.

Fig. 6

Mean (±SEM) normalized (to maximum force) grip strength with no stimulation and with 40-Hz stimulation at sub-threshold current levels for all six stimulation electrode configurations.