Our purpose is to develop a flexible, multi-contact implant to promote and enhance long-term stable peripheral nerve stimulation. Micro scale implantable probe systems known as Micro-Electro-Mechanical Systems (MEMS) contain multi-channel actuators, sensors, and electronics on a silicon probe. We tested the null hypothesis that implantation of MEMS probes do not have a detrimental effect on peripheral nerve function.

A rat hindlimb, peroneal nerve model was utilized in all experimental groups: 1) No surgery (Control); 2) Nerve division and repair (Rep); and 3) Nerve division, insertion of MEMS probe in the distal end of the divided rat peroneal nerve, and repair (Rep+Probe).  Extensor digitorum longus (EDL) muscle contractile function, peroneal nerve electrophysiologic parameters, and nerve morphology were measured following an 11 week recovery.

When compared with Control, Rep and Rep+Probe groups had lower EDL muscle mass, smaller axon diameters, lower percent neural tissue in the peroneal nerve, 12% reduction in nerve conduction velocity, 25% reduction in relative action potential amplitude, and lower maximal muscle force. There were no differences between any of the groups in muscle specific force, nerve conduction latency, or number of axons (Table 1).  Most importantly, Rep+Probe did not vary from Rep group for any of the variables.  Though there was some local fibrosis around the MEMS probe, this did not lead to measurable detrimental effects.

The lack of a significant difference between repair group muscle force, nerve conduction, and nerve morphology indicate that MEMS electrodes are compatible with growing axons and show promise for establishing chemical and electrical interfaces with peripheral nerves.

Table 1.  Data summary for muscle contractile function, nerve conduction, and nerve morphology following recovery from micro scale electrode implantation during nerve repair.

Test	Dependent Variable	Peroneal Nerve Groups
		Control	Rep	Rep+Probe
EDL Muscle Force	Force (mN)	3868 ± 452 (n=9)	2025 ± 616* (n=6)	2348 ± 1024* (n=9)
	Muscle mass (mg)	159 ± 13	104 ± 13*	111 ± 28*
	Specific force (kN/m2)	324 ± 29	262 ± 62	277 ± 72
	Body mass (grams)	403 ± 42	433 ± 20.5	428 ± 14
	Post operative day (days)	.	76 ± 12	78 ± 9
Peroneal Nerve Conduction	Conduction speed† (%)	-22% (n=12)	-34% (n=2)	-34% (n=3)
	Relative Compound Motor Action Potential (mV/mg) †	1.9%	-1.2%	-1.7%
	Nerve Conduction Latency (ms, with Fibular head stimulation)	0.92 ± 0.21	0.78 ± 0.18	1.11 ± 0.55
Peroneal Nerve Morphology	Estimated Axon Count	1958 ± 1027 (n=7)	2383 ± 1120 (n=5)	1710 ± 540 (n=4)
	Average Axon Size (μm2)	48 ± 19	9.7 ± 4.1*	12 ± 4.4*
	Average Non- Neural Area (%)	57.8 ± 13.4	85.8 ±  4.1*	81.7 ± 4.6*

Values are means ± SD.  *Indicates different from the Control Nerve group, p < 0.05.  †Nerve conduction values are the difference between measurements made when the nerve was stimulated at the fibular head and at the sciatic notch divided by the value when stimulated at the fibular head.

The views expressed in this work are those of the authors and do not necessarily reflect official Army policy. This work was supported by the Department of Defense Multidisciplinary University Research Initiative (MURI) program administered by the Army Research Office under grant W911NF0610218.