Sensory Percepts Elicited by Macrosieve Stimulation of the Rat Sciatic Nerve
Nikhil S Chandra, MS; Weston M McCarron, BS; Harold Burton, PhD; Leonard S Green, PhD; Daniel W Moran, PhD; Wilson Z Ray, MD; Matthew R MacEwan, PhD
Washington University in St. Louis, St. Louis, MO
Introduction: Peripheral nerve stimulation by implanted electrode interfaces could provide sensory feedback originating in a prosthetic limb, enabling intuitive limb control. The newly developed macrosieve electrode (MSE) may be optimal for this role, due to its chronic stability and ability to selectively recruit subfascicular axon clusters through current steering. The MSE's success as a sensory interface depends on its ability to elicit sensory percepts at low current amplitudes without damaging interfaced axons over multiple activation cycles. We developed a rat behavioral model to assess whether detection of MSE stimuli can shape behavior and thus demonstrate the MSE's clinical viability as a sensory interface.
Methods: Rats learned a go/no-go auditory detection task in a Skinner box equipped with a house light, tone generator, snout detector, and food-dispenser. Each rat learned to initiate trials by inserting its snout into the detector. An uninterrupted snout insertion lasting a full, randomized interval (4 ± 2 s) triggered presentation of an auditory stimulus. Timely withdrawal within 500 ms of stimulus onset was reinforced. Premature withdrawal (prior to stimulus onset) extinguished the house light for a 10-s timeout. Late withdrawal (> 500 ms after stimulus onset) resulted in a 5-s inter-trial interval. Rats that learned this task successfully subsequently underwent a sterile surgery for MSE implantation. Under anesthesia, we exposed the right sciatic nerve, transected it, and sutured the proximal and distal stumps into a pair of silicone guidance conduits affixed to a prepared MSE/Omnetics assembly. The Omnetics assembly, with trailing microwires, was routed subcutaneously to the skull, passed through a scalp incision, and embedded in an acrylic head-cap. The connector provides an external interface to the MSE during behavioral experiments. Following implantation, rats will relearn a go/no-go analog of the auditory detection task by attending stimulation of the sciatic nerve now activated through the implanted MSE.
Results: Six adult male Lewis rats learned to withdraw from the detector within 500 ms of auditory stimulus onset. Each rat detected auditory stimuli in over 95% of trials, aggregated across the five most recent training sessions. Five rats were then implanted with MSE/Omnetics assemblies and head-caps. Implants remained stable for 4-12 months post-surgery, which is a sufficient interval for learning to attend MSE stimulation of the sciatic nerve.
Conclusion: Successful implementation of our behavioral model will advance development of sensory feedback originating from embedded prosthetic sensors and represent the MSE's potential as a clinically viable sensory interface.
Back to 2019 ePosters