Surgical Model for Analysis of Signal Transfer Through Biologic and Synthetic Materials.

BACKGROUND

High-fidelity volitional control of bioengineered prosthetic limbs with multiple degrees of freedom requires the implantation of multiple recording interfaces to detect independent control signals. However, interface utilization is complicated by interfering electrophysiological signals originating from surrounding muscles and nerves, leading to equivocal signal detection. We developed and validated a surgical model to characterize signal propagation through various biomaterials to identify insulating substrates for use in implantable interfaces. The identification of these insulating materials will facilitate the acquisition of noncontaminated prosthetic control signals, thus improving manipulation of advanced prosthetic limbs.

METHODS

Using a rat hindlimb model, 4 groups (n = 8/group) were tested. A medial gastrocnemius muscle flap was elevated, leaving the neurovascular pedicle intact. The flap was rotated into a chamber and secured to a silicone base. A stainless steel electrode was affixed to the surface of a muscle and encircled by 1-layer small intestinal submucosa (SIS), 4-layer SIS, silicone elastomer, or nothing (uninsulated). A superimposing electrode was attached, and an external silicone layer was wrapped around the construct and sutured in place. Electromyographic studies were then performed.

RESULTS

This model was found to correspond with expected signal isolation characteristics of the nonconductive silicone group, electrically inert single and multilayer SIS group, and the uninsulated group. Signal isolation of compound muscle action potential amplitude at stimulation threshold was significantly greater using silicone (51.4%) compared with the 1-layer SIS (-6.8%), 4-layer SIS (-3.3% ), or uninsulated groups (1.2%) (P = <0.001). Isolation of the maximum compound muscle action potential peak-to-peak amplitude was also greater with silicone (56.7%) versus the 1-layer SIS (1.5%), 4-layer SIS (1.1%), or uninsulated groups (-0.7%) (P = <0.001).

CONCLUSIONS

This study demonstrates and validates a novel surgical model to characterize in vivo signal propagation and subsequently identify insulating materials for use in implantable interface systems currently in development. Improved signal isolation through the utilization of these materials stands to greatly improve control fidelity of neuroprosthetic limbs.