The Osseointegrated Neural Interface For Prosthetic Control: A Chronic, Percutaneous, Rabbit Model
Aaron M. Dingle, PhD, Jared P. Ness, MS, Joseph Novello, MS, Augusto X. Millevolte, BS, Rashea L. Minor, BS, Nicholas J. Albano, MD, Ruston Sanchez, MD, Brett Nemke, BS, Yan Lu, MD, Aaron J. Suminski, PhD, Mark D. Markel, DVM, Justin C. Williams, PhD, Samuel O. Poore, MD, PhD.
University of Wisconsin- Madison, Madison, WI, USA.
Purpose: Advanced prosthetic limbs can be controlled directly via the peripheral nervous system, however, seamless control of both motor and sensory components remain a futuristic goal. The Osseointegrated Neural Interface (ONI) builds on the clinical treatment for amputation neuromas, neural interfacing and osseointegration to create a chronic, stable neural interface with direct transcutaneous connection to a prostheses required for naturalistic prosthetic control. The objective of our current research is demonstrate the capacity of nerves transposed into bone for chronic bi-directional electrophysiology in rabbits, towards future prosthetic control.
Methods: Above knee amputation was performed in New Zealand white rabbits. Briefly, the sciatic nerve was isolated and severed above the trifurcation. The femur was amputated at the midpoint and the nerve passed through a corticotomy. The terminal end of the nerve was sutured into a bipolar cuff electrode, and pressed back into the medullary canal. A second bi-polar cuff electrode was secured proximal to the corticotomy in order to stimulate and record efferent and afferent signals between the proximal and distal electrodes respectively. Both electrodes were connected to independent printed circuit boards, which were intern secured to a stainless steel screw. The stainless steel screw served as both the osseointegrated and percutaneous portion of the ONI device. The muscle and skin were closed over the femur. Animals underwent electrophysiological recordings of compound nerve action potentials (CNAPs) at weeks 3, 5, 8 and 12 weeks under anesthesia, as well as terminal recordings of somatosensory evoked potentials (SSEPs) at week 12.
Results: Chronic bi-directional signaling was achieved via an ONI, as demonstrated by the ability to record both afferent and efferent CNAPs over 12 weeks. Furthermore, physiological function of amputated nerves transposed into bone improve over time, indicated by achieving higher peak amplitudes from lower stimulation over time. Finally, the ability to record SSEPs generated from stimulation of the nerve via the ONI demonstrates the ability to write sensory information to the central nervous system, required for true haptic feedback.
Conclusion: The ONI represents a viable model for directly interfacing peripheral nerves with advanced prosthesis to improved prosthetic control.
Back to 2020 Abstracts