Abstract: Partial Skeletal Muscle Grafts for Prosthetic Control (Plastic Surgery 2014)

Saturday, October 11, 2014: 10:45 AM

Shoshana L. Woo, MD , Plastic Surgery, University of Michigan, Ann Arbor, MI

Melanie G Urbanchek, PhD , Plastic Surgery, University of Michigan, Ann Arbor, MI

Xin Zheng, BS , Biomechanical Engineering, Plastic Surgery, University of Michigan, Ann Arbor, MI

Michelle K Leach, PhD , Plastic Surgery, University of Michigan, Ann Arbor, MI

Jana D Moon, BS , Plastic Surgery, University of Michigan, Ann Arbor, MI

Paul S Cederna, MD , Plastic Surgery, University of Michigan, Ann Arbor, MI

Nicholas B Langhals, PhD , Plastic Surgery, University of Michigan, Ann Arbor, MI

Presentation PDF file

Objective: Nonvascularized partial skeletal muscle grafts are notorious for their limited force-generating capacity and tendency to degenerate in the absence of reinnervation. Accompanied by peripheral nerve implantation, however, partial muscle grafts can survive and transmit detectable electromyographic (EMG) signals capable of prosthetic control. Our study investigated partial muscle graft survival in the construction of regenerative peripheral nerve interfaces (RPNIs) and further characterized their electrophysiological properties across various muscle donor sites.

Methods: Twenty F344 rats were assigned to 1 of 5 groups based on muscle graft type used for RPNI construction: 1) control-whole extensor digitorum longus; 2) partial biceps femoris; 3) partial rectus femoris; 4) partial lateral gastrocnemius; and 5) partial vastus medialis. Each graft (approximately 140-mg at initial harvest) was fixed to the femur, wrapped in small intestinal submucosa for tissue isolation, and implanted with the transected common peroneal nerve. After 4 months of recovery, in situ EMG and force testing were performed [Figure 1].

Results: All control RPNIs (n=4) transmitted detectable EMG signals, compared to 75% of partial muscle RPNIs (n=12 of 16). Significant differences between control and partial muscle RPNIs included average mass [118-mg (SD 42) vs. 66-mg (SD 25)], EMG peak-to-peak amplitude [6.7-mV (SD 2.3) vs. 1.16-mV (SD 1.5)], and maximum tetanic force [500-mN (SD 615) vs. 137-mN (SD 152)] [Figure 2]. Amongst partial muscle groups, donor muscle was not a significant predictor of EMG amplitude after adjusting for final RPNI mass and length at 4 months.

Conclusions: We report that partial muscle graft RPNIs transmit detectable EMG signals with a 75% success rate at 4 months. This proof of concept underscores the potential to develop and refine partial muscle graft-based interfaces to harness peripheral motor nerve signals for prosthetic control. While signal size remains favorable (i.e. 10-100 times larger than signals recorded directly from peripheral nerves), further studies are warranted for optimization of partial muscle graft regeneration and methods of signal acquisition.

Acknowledgements: This work was supported by DARPA (N66001-11-C-4190), the Plastic Surgery Foundation, and the Frederick A. Coller Surgical Society.

24743 Partial Skeletal Muscle Grafts for Prosthetic Control