2: Sensorimotor Myoelectric Control Using Surface Based Regenerative Peripheral Nerve Interface (RPNI)

Abstract

Purpose: Although advanced prosthetic devices have the potential to allow fine-motor movements and extract somatosensory signals via sensitive pressure sensors, an ideal interface to integrate the human nervous system with the prosthetic doesn’t exist. Furthermore, the requirement for the implantation of indwelling electrodes and prohibitive costs limits the application of current technologies. The Regenerative Peripheral Nerve Interface (RPNI) was developed as a stable biologic interface on the notion of providing physiologic end-organ targets for regenerating axons by implantation of a residual nerve into an autogenous free muscle graft. Despite providing intuitive motor control, RPNI sensory feedback is limited and also relies on implantable electrodes for myoelectric signal transmission. To address these challenges, we investigated the placement of RPNIs underneath the defatted skin in rats to capture myoelectric signals using surface electrodes. This strategy simultaneously provides sensory feedback through the sensory reinnervation of the overlying skin. Methods: Utilizing six male F344 rats, the right tibial nerve was transected distally in the thigh before entering the leg’s posterior compartment. Subsequently, the left side extensor digitorum longus (EDL) muscle was harvested and co-apted with the proximal segment of the tibial nerve for RPNI fabrication. The RPNI was placed and secured in between the biceps femoris muscle near the skin while the side opposite to the nerve coaptation was facing the dermis. The overlying skin was defatted and fixed on top of the superficial RPNI (S-RPNI). At two months post-surgery, functional motor reinnervation was evaluated by electrical stimulation of the tibial nerve and compound muscle action potentials (CMAPs) were recorded using surface electrodes. Sensory feedback was assessed by electrical and mechanical stimulation of the skin to respectively record sensory nerve action potentials (SNAPs) and sensory afferent signals. The S-RPNI construct and its overlying skin were subsequently harvested and processed for immunohistochemistry and whole-mount immunostaining. Results: Recording muscle CMAPs from the skin was readily feasible in all animals, showing robust signals with minimal noise distortion (366 µV ± 86.3). Electrical stimulation of the skin on the lateral thigh, which is not innervated by the tibial nerve normally, resulted in classic tri-phasic SNAP generation in this nerve. Brushing the overlying skin using a cotton swab generated synchronized monomorphic afferent signals. Sensory reinnervation of the skin was shown using IHC. Whole-mount staining of the muscle component showed regenerated muscle with new neuromuscular junctions (NMJs) and spatial segregation of sensory fibers toward their end-target organ. Conclusion: Superficially placed RPNI is a viable and an alternate methodology that is simple to implicate and has the potential to transmit simultaneous, real-time, and independent sensory and motor signals between the residual nerve and the prostheses.