Experimental Study Of The Selective Stimulation And Close-up Feedback Controlling Function Of Multi-LIFEs Implantation

BackgroundNeuroprosthesis should develop a good interface between the peripheral nerve and device for the information exchange. The good geometry of Longitudinally implanted intrafascicular electrodes( LIFEs) makes implantion easy. The electrode contacts with axon directly, so it has better recording or stimulating ability. But single LIFE can only record or stimulate the adjacent axon .So the recorded information is not integrity.Only controlled synergically by different responsible signal source, can multifreedom electronic hand prosthesis play its multifunction. Only providing grade of stimulation selectively to the specified muscle involved in an action, can functional electrical stimulation restore the natural motion pattern. Now scholars are trying to implant more electrodes (multielectrode array) into nerve trunk to abtain integrity sensory and motion information .But minor progress had been made until now and there were few reports on the long-term implantation of multielectrode arrays. Lack of sensory feedback of available neuroprostheses is another fatal defect. These devices were manipulated by open-loop controlled patterns which may result in poor coordination and accuracy of prosthesis That was the reason why many patients refused to use such kind of devices. Based on the principle of sensory information included in the discharging rate and firing pattern of the action potential by means of coding form, recording sensory information and realizing close-up feedback regulation of neuroprostheses become research focus now.Part 1The observation of neural stimulation and signal recording ofimplanted multi-LIFESObjective1. To investigate the possibility of multi-LIFEs implanted according to peripheral fasciculi location; 2. To observe the selective stimulation of multi-LIFES; 3. To use LIFEs to collect sensory nerve action potentials (SNAP) and carry out preliminary quantitative analysis to explore the releasing rule of sensory signals . MethodsThrough observe the microdissection and histology of common peroneal nerve in feline, we identified the fasciculi location of the peripheral nerve in order to direct the implantation of multi-LIFEs. Electrodes were fabricated using 25 u m Teflon-insulated 95% Pt-5% Ir wires, and they were threaded into the branch of tibialis anterior, long extensor muscle digits, peroneal muscle and sensory nerve fasciculi of superficial peroneal nerve in six adult cats. We compared the action potential of common peroneal nerve before and after electrodes implantation to evalue the influence of implantation on the conduction. We drew the curve of muscle recruitment by measuring the tension of corresponding muscle to observe the selectivity of multi-LIFEs. We also used the opposite side as control and chose two parameter of recruitment curve, Xo (the current required to reach 50% the maximum force) and W (the width of recruitment curve), to compare the stimulating difference between LIFEs and traditional extraneural electrodes (hook electrodes). The sensory nerve action potential (SNAP) under rest status and induced with various stimulations applied on the dorsal skin of the foot were recorded.The action potential area, frequency, coefficient of variation (CV) of peak and functional spectrum were then analyzed by MFLab3.01 software package. ResultsThe common peroneal nerve of feline accords with the natural fasciculus segement principle. There is an optimum implantation region 0.5-lcm long after the branch of tibialis anterior long extensor muscle digits departing, because there is no more slender fasciculi in this segment. The nerve action potential before and after multi-LIFEs implantation showed that the electrodes implantation resulted in a temporary interference in the nerve conduction . The peak amplitude of action potential decreased about 13% after implantation instantly (p<0.01) . While after one hour the waveform recovered nearly normal and the peak amplitude were about 97 percent of that before implantation. The electrical stimulation to different fasciculi showed that LIFEs had good selectivity and low current (threshold current was about 10-20uA) can evoke corresponding muscle contraction. When we stimulated the nerve fasciculi dominatingtibialis anterior , the force of tibialis anterior had reached 66% the maximum at the time of the long extensor muscle digits beginning to excite ,while peroneal muscle still had no reaction. The selectivity of hook electrodes is poor . Only high current can induce muscle exciting. We compared the two parameters Xo and W of recruitment curves. The Xo of intrafascicular electrode group was only one tenth of that of hook electrode group, but the width was almost 4 times broader than that of hook electrode. Statistics analysis showed that there were significant differences (p<0.01) .All these results demonstrated that the stimulating current of intrafascicular electrode was smaller and safer .So the nerve injury caused by high currents was reduced to the minimum. The broader recruitment curve allowed us grade stimulation. The sensory signals under rest status and induced with various stimulation were recorded. The results showed that the sensory nerve action potentials (SNAPs) were 0-2 spikes per second at the rest state.The count increased when stimulation was administered. When scraping stimulation was applied, the SNAPs reach 16-24 spikes per second with consistent pulse intervals and waveform,and when stress stimulation was given, the SNAPs burst in cluster. We compared the action potential area, frequency, coefficient of variation (CV) of peak and functional spectrum using MFLab3.01 software package and showed that there were significant differences between different status. The functional spectrum analysis showed that the frequency of action potential increased when the stress stimulation was applied. ConclusionIt was feasible to implant multi-LIFEs according to peripheral fasciculi location and we can stimulate or record the corresponding nerve fasciculus point to point. The stimulating currents of intrafascicular electrodes were smaller and safer than extraneural electrode .Nerve injury caused by high current stimulation was reduced to minimum. The broader recruitment curve allowed us to stimulate by grade. The self-made .LIFEs can collected the sensory signals sensitively and provided a good interface.The action potential area, frequency, coefficient of variation (CV) of peak and functional spectrum were valid parameters to discriminate different types of stimulations.Part 2 Long-term biocompatibility of implanted multi-LIFEsObject1 .To investigate stimulative function and biocompatibility of long-term implanted multi-LIFEs. 2.To observe adverse reaction on the neurofibra of long-term implanted multi-LIFEs. 3.To probe the feasibility of multi-LIFEs long-term implantation according to peripheral fasciculi location. MethodsMulti-LIFEs were made of 25um diameter Teflon-insulated 95%Platinum 5%Iridium wires .They were implanted into the muscular branch of tibialis anterior and peroneal muscle according to peripheral fasciculi location and 10-0 nylon sutures were implanted into the opposited fasciculus as control. The threshold currents evoking muscle contraction were recorded at 0, 0.5, 1, 2, 3 ,4, 6 months after operation respectively to observe the change of stimulatory function of multi-LIFEs . We chose the specimens that both electrodes were disabled to study the microstructural changes with light microscope and transmission electron microscope. At the end of the experiment, the common peroneal nerves at the level of implant and the level 1.5cm distal to the implant zone were harvested. The myelinated fibers were strained by toluidine blue. The density of myelinated fibers at the level 1.5cm distal to the implant zone and the thickness of connective tissue sheath around implant were determinded by Image Analysis System. We also compared the area of nerve fiber ,axon and mylin sheath within 200um around implant and that of fascicular periphery. ResultThere were five cats whose both electrodes lost the stimulative function .The nerves were anatomized for pathological examination. The LIFEs became invalid mainly due to the fracture or withdrawal of electrodes except one case due to the deep infection. Half of 24 electrodes retained the stimulative function.The statistics analysis showed that threshold current rose greatly at 0.5 month and reached the peak at 2 months which was almost 4-6 times higher than the original value (pO.Ol). Then it showed a slowly descending trend and became stable gradually. But the threshold was still about 3.3-4.2 times as high as the original value .There was no statistic significance between the threshold of the last 3 month .The tissue section stained with hematoxylin and Eosin showed that neurofibra around electrodes arranged disorderly and the neural...