SPARC: Modelling a Bidirectional Bioelectric Pelvic Nerve Interface

Abstract

BackgroundMicturition (urination) and continence are controlled by the autonomic nervous system; disruptions to urinary control due to disease or injury significantly impair patients’ quality‐of‐life. The pelvic nerves (in human, the pelvic splanchnic nerves) provide the primary source of autonomic sensory and motor innervation to the bladder and are prime targets for electrical stimulation to treat over‐ or under‐active bladder. The pelvic nerves in both human and rat are fasciculated, but differ from other well‐studied nerves such as the vagus or sciatic nerves in largely lacking a resistive epineurium. The pelvic nerves also comprise a much higher proportion of smaller‐caliber axons than are currently implemented in existing myelinated axon models.ObjectiveTo determine the significance of the macro‐ and micro‐anatomical properties of the pelvic nerve on the stimulation and recording performance of bioelectric neural interfaces.MethodsThe functional composition of the different fascicles of the pelvic nerve were determined using a combination of immunolabeling and tract tracing from the specific regions of the lower urinary tract or large intestine in male rats. Using this data, a model of the pelvic nerves and electrode array was implemented in SIM4LIFE (zmt.swiss) and EIDORS (eidors.org) to simulate electrical stimulation and recording, respectively. Established models for myelinated and unmyelinated axons were adapted to the small‐diameter axons found in the pelvic nerve and implemented in NEURON (neuron.yale.edu), and parameter sensitivity was established using Sobel’ sensitivity analysis for different fascicle arrangements and orientations.ResultsThe pelvic nerve consists of mixed sensory and motor fascicles, with sensory projections to lower urinary or intestinal tracts projecting in distinct fascicles. Sensitivity profiles to extracellular stimulation depend on the size and orientation of the fascicles, as well as their axon type and diameter. Within‐fascicle selectivity is enhanced in the largest fascicles of the pelvic nerve, and overall thresholds are reduced in the smallest fascicles. The stimulus thresholds and conduction velocities of the small myelinated axons depend on the ultrastructural geometry of the nodal and internodal regions, with less dependence on the juxtanode, in an axon‐type‐dependent manner. Extracellular recording sensitivity profiles for clinically realistic electrodes were computed in EIDORS and validated in SIM4LIFE for predicting spontaneous and electrically evoked activity recordings. Extracellular recording sensitivity and stimulus thresholds were both most closely correlated with axon‐electrode distance.ConclusionThe macroscopic and microscopic structure of the pelvic nerve alters its sensitivity to electrical stimulation, as well as the selectivity of extracellular electrical recording. EIDORS can be used to accurately simulate extracellular recording in complex, heterogenous neural geometries.Support or Funding InformationNIH SPARC 3OT2OD023872‐01S4