Revisiting nonvascularized partial muscle grafts: a novel use for prosthetic control.

Abstract

Sir: Although the popularity of fat grafting has skyrocketed over the past decade, nonvascularized partial skeletal muscle grafts have long been forgotten in light of their diminishing clinical applications. Historically, surgeons have used partial muscle grafts for palatopharyngeal reconstruction1 and facial reanimation,2 using the split extensor digitorum brevis muscle for eye closure and the palmaris longus for oral competence. To generate force, each muscle would rely on muscle-to-muscle neurotization, which is an inefficient way to reestablish innervation and force-generating capacity. Despite several reported successes, nonvascularized partial muscle grafts lost their popularity because they lacked the necessary force to optimally reanimate the face. Today, nonvascularized partial muscle grafts are used only as soft-tissue fillers or for obliterating the nasofrontal ducts during frontal sinus fracture repair.3 In such cases, the grafts are expected to die and be replaced by scar tissue, filling defects rather than reestablishing function. Partial muscle grafts have yet to serve a functional purpose in which muscle fiber survival is clinically necessary. Our laboratory has proposed a novel use for partial muscle grafts, namely, to harness signals from peripheral motor nerves for prosthetic control. We have developed a regenerative peripheral nerve interface model in the rat hind limb, in which a nonvascularized whole muscle graft is attached to a severed motor nerve. After undergoing a process of degeneration, regeneration, and reinnervation, this muscle-nerve construct serves as an interface for the transmission of electromyographic signals, which are detected by an epimysial electrode.4 Major benefits of this approach include decreased neuroma formation, reduced nerve trauma during recordings, and improved signal-to-noise ratio by means of neuromuscular signal amplification. To translate this regenerative peripheral nerve interface technology into humans, we have turned toward partial muscle grafts for their practical features: abundant donor tissue, decreased donor-site morbidity, ease of harvest, appropriate size selection, and overall “cookie-cutter” reproducibility. We hypothesize that partial muscle grafts can be used successfully in regenerative peripheral nerve interface construction. With our experiments underway, we report the first set of functional regenerative peripheral nerve interfaces constructed with partial muscle grafts (Fig. 1). At 4 months, maximum electromyographic peak-to-peak amplitudes ranged from 0.64 to 2.2 mV (Fig. 2).Fig. 1: In situ appearance of a regenerative peripheral nerve interface (RPNI) at 4 months. This particular regenerative peripheral nerve interface was constructed with a 130-mg graft from the lateral gastrocnemius muscle, placed along the femur, and implanted with the common peroneal nerve. Electrophysiologic testing was performed at 4 months by means of direct stimulation of the nerve and bipolar needle electromyographic recordings from the regenerative peripheral nerve interface.Fig. 2: Regenerative peripheral nerve interface signal waveforms and peak-to-peak amplitudes. After 4 months of maturation, each partial muscle graft regenerative peripheral nerve interface displayed discrete electromyographic signals, with maximum amplitudes ranging from 0.64 to 2.23 mV at threshold stimulation currents from 145 to 820 μA. Each bar represents a single regenerative peripheral nerve interface.This proof of concept underscores the potential for partial muscle grafts to be used in the clinical workplace for their signaling capabilities. There is a high demand for reliable peripheral nerve interfaces that provide multiple input control signals to take advantage of today’s most advanced hand prostheses, which are capable of many simultaneous degrees of freedom. Should partial muscle grafts prove reliable in regenerative peripheral nerve interface construction, they will support more intuitive voluntary prosthetic motion, bypass the limitations of skin-surface electromyography, and possibly serve as a platform for implantable, wireless prosthetic control. Signal detection will not be restricted by the remaining muscle volume within the residual limb or on the chest wall, which is a perceived disadvantage of targeted muscle reinnervation.5 In summary, partial muscle regenerative peripheral nerve interfaces are promising devices that may one day augment amputee functional recovery and quality of life. ACKNOWLEDGMENTS This work was supported by the Defense Advanced Research Projects Agency (N66001-11-C-4190), the Plastic Surgery Foundation, and the Frederick A. Coller Surgical Society. DISCLOSURE The authors have no financial interest to declare in relation to the content of this article. Shoshana L. Woo, M.D. Melanie G. Urbanchek, Ph.D. Paul S. Cederna, M.D. Nicholas B. Langhals, Ph.D. Section of Plastic Surgery University of Michigan Ann Arbor, Mich.