BACKGROUND:Spinal cord stimulation (SCS) has demonstrated multiple benefits in treating chronic pain and other clinical disorders related to sensorimotor dysfunctions. However, the underlying mechanisms are still not fully understood, including how electrode placement in relation to the spinal cord neuroanatomy influences epidural spinal recordings (ESRs). To characterize this relationship, this study utilized stimulation applied at various anatomical sections of the spinal column, including at levels of the intervertebral disc and regions correlating to the dorsal root entry zone. METHOD:Two electrode arrays were surgically implanted into the dorsal epidural space of the swine. The stimulation leads were positioned such that the caudal-most electrode contact was at the level of a thoracic intervertebral segment. Intraoperative cone beam computed tomography (CBCT) images were utilized to precisely determine the location of the epidural leads relative to the spinal column. High-resolution microCT imaging and 3D-model reconstructions of the explanted spinal cord illustrated precise positioning and dimensions of the epidural leads in relation to the surrounding neuroanatomy, including the spinal rootlets of the dorsal and ventral columns of the spinal cord. In a separate swine cohort, implanted epidural leads were used for SCS and recording evoked ESRs. RESULTS:Reconstructed 3D-models of the swine spinal cord with epidural lead implants demonstrated considerable distinctions in the dimensions of a single electrode contact on a standard industry epidural stimulation lead compared to dorsal rootlets at the dorsal root entry zone (DREZ). At the intervertebral segment, it was observed that a single electrode contact may cover 20-25% of the DREZ if positioned laterally. Electrode contacts were estimated to be ~0.75 mm from the margins of the DREZ when placed at the midline. Furthermore, ventral rootlets were observed to travel in proximity and parallel to dorsal rootlets at this level prior to separation into their respective sides of the spinal cord. Cathodic stimulation at the level of the intervertebral disc, compared to an 'off-disc' stimulation (7 mm rostral), demonstrated considerable variations in the features of recorded ESRs, such as amplitude and shape, and evoked unintended motor activation at lower stimulation thresholds. This substantial change may be due to the influence of nearby ventral roots. To further illustrate the influence of rootlet activation vs. dorsal column activation, the stimulation lead was displaced laterally at ~2.88 mm from the midline, resulting in variances in both evoked compound action potential (ECAP) components and electromyography (EMG) components in ESRs at lower stimulation thresholds. CONCLUSION:The results of this study suggest that the ECAP and EMG components of recorded ESRs can vary depending on small differences in the location of the stimulating electrodes within the spinal anatomy, such as at the level of the intervertebral segment. Furthermore, the effects of sub-centimeter lateral displacement of the stimulation lead from the midline, leading to significant changes in electrophysiological metrics. The results of this pilot study reveal the importance of the small displacement of the electrodes that can cause significant changes to evoked responses SCS. These results may provide further valuable insights into the underlying mechanisms and assist in optimizing future SCS-related applications.

The BurstDRTM waveform (burst) for spinal cord stimulation to relieve chronic pain demonstrated superior to classical tonic stimulation efficacy. Burst stimulation allows effective stimulation at sub-paresthesia level, improving patient’s experience. Regarding sensing and closed-loop approach, burst waveform raises a question if a spinal cord stimulation evoked compound action potential (ECAP) can be recorded and analyzed using this complicated waveform.

Electrical epidural spinal cord stimulation (EES) has been successfully implemented to relieve chronic pain. Variations in the effect of EES to suppress chronic pain can be attributed to multiple factors such as exact location of the electrode contacts, patient position, and others. Evoked responses triggered during EES, provides a way that may measure and quantify these variations. These evoked responses, including evoked compound action potential (ECAP) from neuronal fiber activation as a major component, still have an enormous number of questions related to the response registration needed to be addressed. This study is focused on evoked response registration related to the latency, amplitude in Area Under the Curve (AUC), and the effect of displacement of the stimulation and recording leads along rostral caudal axis.

Lead migration is a frequent problem with Spinal Cord Stimulation (SCS) for chronic pain. The role of lead migration in efficacy of pain control is unclear. Using a large animal model, we investigated if a movement of stimulation lead could cause change in the evoked response, which could place impact on the evoked compound action potential (ECAP) component, seeking a way to notify and quantify lead migration during SCS.

The efficacy of BurstDRTM (burst) waveform has been shown to be superior to tonic stimulation and it is widely used to relieve chronic pain. The characteristics of its complex waveform and the sub-paresthesia therapy makes burst waveform itself difficult to sense an evoked compound action potential (ECAP). Here, we propose a robust sensing strategy by interleaving tonic stimulation pulses in the therapeutic burst waveform to enable continuous sensing capabilities while delivering therapy with burst waveform.

Epidural electrical stimulation (EES) of the spinal cord has been FDA approved and used therapeutically for decades. However, there is still not a clear understanding of the local neural substrates and consequently the mechanism of action responsible for the therapeutic effects. Epidural spinal recordings (ESR) are collected from the electrodes placed in the epidural space. ESR contains multi-modality signal components such as the evoked neural response (due to tonic or BurstDR™ waveforms), evoked muscle response, stimulation artifact, and cardiac response. The tonic stimulation evoked compound action potential (ECAP) is one of the components in ESR and has been proposed recently to measure the accumulative local potentials from large populations of neuronal fibers during EES. Here, we first review and investigate the referencing strategies, as they apply to ECAP component in ESR in the domestic swine animal model. We then examine how ECAP component can be used to sense lead migration, an adverse outcome following lead placement that can reduce therapeutic efficacy. Lastly, we show and isolate concurrent activation of local back and leg muscles during EES, demonstrating that the ESR obtained from the recording contacts contain both ECAP and EMG components. These findings may further guide the implementation of recording and reference contacts in an implantable EES system and provide preliminary evidence for the utility of ECAP component in ESR to detect lead migration. We expect these results to facilitate future development of EES methodology and implementation of use of different components in ESR to improve EES therapy.