| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Bioengineering, University of Utah, Salt Lake City, Utah; and 2Guru Nanak Dev University, Amritsar, Punjab, India
Submitted 8 September 2004; accepted in final form 9 December 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Cohen et al. 1991; Elbert et al. 1994; Hall et al. 1990; Jones et al. 2002; Merzenich and Jenkins 1993; Merzenich et al. 1984; Sanes et al. 1990). This central plasticity is thought to involve both immediate mechanisms, such as synaptic unmasking, and long-term effects, such as central neuronal sprouting (Calford 2002; Théoret et al. 2004; Wall et al. 2002). Moreover, chronic peripheral nerve transaction also produces atrophic changes in both sensory and motor neurons, including reduction of motor neuron dendritic arborizations (Carlson et al. 1979; Cragg and Thomas 1961; Hoffer et al. 1979; Horch 1978; Horch and Lisney 1981a,b; Kawamura and Dyck 1981; Kiraly and Krnjevic 1959; Mendell et al. 1974; Milner and Stein 1981; Sumner and Watson 1971; Sunderland 1978; Törnqvist and Aldskogius 1994). Although there is some evidence in animals that chronically amputated nerves retain at least some function (DeLuca and Gilmore 1976; DeLuca et al. 1982; Edell 1986), little is known about the extent to which nerve stumps in human amputees retain useful sensory or motor capabilities (Clippinger et al. 1974).
Work by different groups have shown that it is possible to interface microelectrodes to small clusters of motor and sensory neurons at a subfascicular level (Branner and Normann 2000; Branner et al. 2001; González and Rodríguez 1997; Goodall et al. 1991; Kovacs et al. 1992; 1994; Yoshida and Horch 1993, 1996). A recent study with long-term human amputees, involving implantation of intraneural electrodes in severed nerve stumps, demonstrated that it is possible through discrete stimulation of small micro-clusters of sensory neurons to provide feedback related to touch/pressure and joint position sense and to use these same electrodes to record motor neuron activity related to volitional attempts to move joints in a phantom limb (Dhillon et al. 2004). This indicates that central plastic changes notwithstanding, chronic section of peripheral nerves in humans does not eliminate all of their central sensory and motor pathway connections. What was not determined, however, is if repeated use of these pathways would result in a rapid and significant improvement in their functionality.
The present study was an attempt to explore the question of the effects of experience on function in residual neural pathways associated with peripheral nerve stumps. In addition to its importance in providing information about the limits of CNS plasticity in adult humans, this work has implications about the feasibility of interfacing with amputee nerve stumps to provide natural, closed-loop control of artificial limbs (Dhillon et al. 2004).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
7.3 yr) amputees voluntarily participated in the study. Institutional Review Board approval was obtained from both participating institutions, and the subjects were given adequate time to consider and provide informed consent. Although not selected on the basis of gender or handedness, all the subjects were right-hand dominant, adult males with an average age of 25.5 yr. All the amputations were of the right arm, above the elbow, and traumatic in nature. At the time of the study, the subjects were healthy and free from peripheral vascular disease or diabetes.
Fabrication procedures and performance characteristics of longitudinal intrafascicular electrodes (LIFEs) are described elsewhere (Lawrence et al. 2003, 2004; Lefurge et al. 1991; Malagodi et al. 1989; Malmstrom et al. 1998; McNaughton and Horch 1996; Nannini and Horch 1991). The distal ends of the LIFEs were attached to the pins of a miniature cable connector using conductive silver epoxy, which was thermally cured. A silicon rubber tube was used to provide strain relief for the fine LIFE wires. The connector assembly was embedded in glue and encased in a layer of silicone. A further layer of silicone was applied to the assembly to bond it to a circular silicon patch that was placed on the residual arm.
LIFEs were implanted within the healthy portion of the nerve proximal to any terminal neuroma to maximize recording signal-to-noise ratios and to ensure that recording and stimulation were performed in a part of the nerve that still maintained some degree of somatotopic organization. Partial epineurial dissection was performed at the site of implantation to allow visualization of the fascicle(s). The proximal ends of the electrodes were individually threaded through the skin using an 18-gauge needle as a trochar. Each electrode was then threaded through a 0.5- to 1.0-cm length of a fascicle with the aid of a 50-µm-diam tungsten needle attached to its leading end. Once the 1-mm recording/stimulating zone of the electrode was centered within the fascicle, the tungsten needle was removed. A reference LIFE electrode was sutured to the epineurial surface of the nerve, level with the intraneural electrodes. More details on the implantation procedure are given in Dhillon et al. (2004).
After the amputees had fully recovered from the anesthetic, they underwent 2 wk of computer-aided motor and sensory studies to map the functionality of the implanted electrodes and the nerve fibers they interfaced with (Dhillon et al. 2004). The primary statistical analysis techniques used were regression analysis and ANOVA. After completion of the study, the electrodes were removed percutaneously by applying gentle longitudinal traction. This did not require application of an anesthetic.
Sensory input
An initial mapping of the electrodes was made to determine which of them could be used to elicit tactile or proprioceptive sensations. To do so, each electrode was stimulated, through a current-controlled stimulus isolation unit driven by a D/A converter interface to a laptop computer, with charge balanced pulses of 300 µs duration, in a current range between 1 and 200 µA. Starting at a low current level, stimuli were increased until either the upper limit was reached or the subject reported a distinct sensation. If the subject reported a discrete, distally referred sensation, a staircase method of limits was used to identify threshold and upper limit pulse amplitudes for the sensation (Gelfand 1998). Threshold was defined as the lowest average stimulus pulse amplitude at which the subject could reliably feel a sensation. The upper limit was defined as the current at which the nature or the location of the sensation changed or when the sensation became uncomfortable.
A psychometric scaling task was employed to evaluate the relationship between stimulus amplitude and sensation magnitude (Stevens 1986). A stimulus amplitude midway between threshold and upper limit was selected and 500-ms-duration pulse trains, with various pulse frequencies, were used to stimulate the nerve. Initially, the subject was presented with sample trains at each of the pulse rates so he had a feel for the range of sensations that would be experienced. For testing, the pulse train frequencies were logarithmically distributed, presented a fixed number of times (typically five each) in pseudorandom order, with a time period of 
5 s between successive trains. Subjects were asked to verbally assign an (open-ended) number to the magnitude of the elicited sensation for each stimulus presentation. Other than being instructed that a stronger sensation should result in a higher number, the subjects were not constrained in how they were to assign the numerical values. This procedure was used to avoid biasing them in the task (Stevens 1986). The subjects were not blindfolded or asked to close their eyes, but because the computer generated the stimuli automatically after a variable period following entry of the subject's report, they had no external cue as to when a stimulus was to be delivered.
In addition to sensory psychophysical magnitude estimation as a function of stimulus frequency, the effects of changes in stimulus amplitude on perceived sensation magnitude and on the referred distributions of touch/pressure sensations were studied in some of the subjects. These changes were mapped by having the subject verbally report where on their phantom hand the sensations were perceived as coming from and having them assign perceived sensory magnitude values for the different areas of sensation that developed as the stimulus strength was increased. For some of the instances in which proprioceptive (finger flexion) sensations were reported, the subject was asked to indicate the perceived finger position with the contralateral finger. The angle of the matching joint was then measured with a goniometer.
Motor control
Candidate electrodes for evaluating motor control were identified by asking the amputee to attempt different movements in the amputated (phantom) part of the limb (e.g., the wrist or fingers) while recordings were made of neural activity from different implanted electrodes. Motor signals were recorded between the LIFE and the reference electrode with a differential amplifier (gain of 20,000), band-pass filtered (0.3–4 kHz), sent to a loudspeaker with a noise clipper, and fed through a 16-bit A/D converter to a battery-powered laptop computer (Dhillon et al. 2004). The subject was directed to select a phantom movement that resulted in maximum audible activity. Once the subject had learned to generate motor neuron activity, a simple computer game was used to evaluate his control over the rate of motor neuron action potential production.
Basically, the subject was asked to modulate recorded neural activity to control a cursor on the computer screen so that it would overlap and stay within a displayed target (Dhillon et al. 2004). At the start of a trial, the target would appear randomly in a screen area 480 pixels wide, and the cursor would appear at the left end of the screen. The subject's task was to move the cursor and place it in the target for 
500 ms (a "hit"). Simply placing the cursor in the target, no matter on how many occasions, counted as a failure unless the subject managed to hold the cursor in the target for the specified time.
As preparation for the task, background noise was recorded and displayed on the computer with the subject relaxed (in the absence of volitional motor output, recordings from the nerve stumps gave no indication of neural activity). This allowed the experimenter to set a minimum threshold level for detecting neural activity. The subject was then asked to generate neural activity, and the recorded signals were used to set an upper threshold for detecting action potentials. To simulate the properties of physiological motor control, the efferent activity (in the form of a pulse train from a Schmidt trigger based on these 2 threshold levels) was passed through a leaky integrator to provide the control signal. Time constants between 400 and 600 ms for the integrator were tested during the first day of training. With shorter time constants, amputees were able to control the cursor more precisely but found it difficult to hit targets on the far side of the practice screen area. With longer time constants amputees were able to freely move the cursor to all regions of the screen but precision was much reduced. As a compromise between these conditions, 500 ms was chosen as the standard time constant.
Given the settings for the spike detector and firing rate integrator, minimum output corresponded to the subject making no attempt at the phantom movement and maximum output was determined by having the subject make a maximal effort at the phantom movement. Minimum activity placed the cursor at the left edge of the screen, maximum activity placed it at the right edge.
Each testing set consisted of 20 10-s-long trials, and a subject was limited to completing two sets per day. All subjects started off with a target width of 96 pixels. After the subject managed to successfully perform this task with a success rate >75%, the task was made more difficult by reducing the size of the target to 68 and then 48 pixels. For one subject, who succeeded in hitting the smallest target >75% of the time, the dwell time requirement was increased to 750 ms for that target size.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
All of the subjects reported either tactile or proprioceptive sensations from one or more electrodes. In half of the subjects, both types of sensations could be elicited (through different electrodes). On the first day of testing or when electrical stimuli were first delivered on subsequent days, subjects sometimes reported sensations of spiders crawling on the region of the referred sensation. After stimuli were delivered for 30–60 s or when the stimulus amplitude was increased, this would stabilize to a sensation of pressure/touch, joint position or movement. Usually electrical stimulation resulted in unimodal (i.e., touch, movement, or static joint position) sensations that showed stable topography. In 10% of the cases, the location of the referred sensation tended to wander during the first 2 days. For tactile sensations, this movement was between the tips of different digits (middle and the index finger or thumb and the index finger). Over the duration of the study, these sensations eventually stabilized to one or more finger tips. In all cases, the sensations were in the fascicular projection territories of the implanted nerve, suggesting a stable electrode position, stimulating a small cluster of neurons. In 20% of the cases, sensations were confused as to being either movement or pressure, vibration in a digit, and a mixture of tugging and movement localized to the palm or the fingers. The majority (70%) of the cases resulted in stable sensations of touch/pressure or joint position/movement sense that were localized to the digits. In general, sensations were discrete, unimodal, repeatable, and could be painlessly elicited over the duration of the study. With increasing stimulus current, sensations of touch/ pressure usually spread from distal (digit tip) to proximal locations (Fig. 1).
|
During the course of the study, there was a small, but statistically significant (regression analysis, P < 0.01), increase in both the threshold and the upper limit for eliciting painless, unitary sensations of touch/pressure or joint movement, shown in Fig. 2 as values normalized to threshold for each subject on day 1. The mean value for threshold on day 1 was 7.0 ± 2.5 (SD) nC (n = 12).
|
|
250 Hz at 2 wk. Frequency of stimulation was also correlated with the perceived rate of digit flexion, but the dominant effect was the change in the sense of static joint position. Amputees could judge the rate of movement as fast or slow but were unable to quantify it on a numerical scale. In one amputee, 5 days after initial testing, perceived finger flexion could no longer be systematically controlled through modulation of stimulation parameters. Instead the subject reported only two positions, full extension (no pulses applied) or full flexion (for any value of pulse amplitude or stimulation frequency between the threshold and upper limit). This behavior was not seen with any other electrode in this or other amputees. In all other cases, over the duration of the study perceptions of digit flexion became smoother and less jerky.
Two amputees also reported a sensation of closing and opening of a pincer grip between the thumb, index, and middle fingers. For a stimulation frequency <10 Hz, the grip was felt as open. At 30 Hz, the index and the middle fingers were perceived as coming into contact with the thumb. With further increments in frequency, pinch force became stronger (Fig. 3A).
When defining threshold and upper limit stimulation parameters, subjects reported increasing flexion of fingers with increments in injected charge. This was formally investigated by keeping the frequency of nerve stimulation constant and varying the stimulus amplitude, which demonstrated that the perception of joint flexion/extension could be systematically modulated by varying the stimulus amplitude (Fig. 3C). In addition, the stimulus frequency required to provide a sense of full flexion depended on stimulus magnitude. In one subject, for example, with pulse amplitudes of 24, 28, and 32 µA (pulse width = 300 µs), the maximum frequency required for perception of full flexion of a digit was 510, 100, and 60 Hz, respectively. This dependence of joint-position sense on the interaction of frequency and charge was quantitatively investigated in one amputee by varying the amplitude and examining the frequency dependence of perceived joint excursion (Fig. 3D). In general, the higher the stimulus amplitude the lower the frequency required to produce the sensation of flexion to a given position. Similar observations were made in other subjects, although psychophysical evaluation was not conducted.
Motor output
One or more electrodes recording controllable efferent activity were identified in six of the eight subjects. The success rates with which the subjects could strike and stay within ("hit") the target increased with experience (Fig. 4). Once an amputee managed to score hits in >75% of the trials for a given target size, the target size was reduced. After a reduction in the target size, success rate initially declined and then eventually increased in subsequent sets of trials. Over a period of <70 min experience with the task, subjects demonstrated improved cursor control, progressing from successfully positioning it within a 96 pixel wide target to hitting a 48 pixel wide target (Fig. 4). One subject (not shown) even succeeded in hitting the smallest target 70% of the time with a 750-ms dwell time requirement.
|
Some, but not all, subjects showed an increase with time in the maximal neural output (as defined in METHODS) they could generate to drive the cursor (Fig. 5).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
We were unable to record volitionally induced neural activity in only two of the subjects. This does not mean that they did not have functional motor pathways, only that we were unable to identify them, perhaps due to the limited sampling provided by implanting only a few LIFEs in a given subject. Although motor performance did improve with time, given the generally limited increase in motor output seen with time, this improvement appears to be due more to practice with a novel task than to any significant change in central connectivity of the residual motor pathways. Given that the two longest (10 and 30 yr) amputees were among the subjects with good motor control, it appears that basic motor pathways are permanently established by early adulthood, even in the subsequent absence of effectors.
Sensory input
There was a small increase with time in the amount of charge per stimulus pulse needed to elicit a threshold sensation and somewhat larger increase in stimulus amplitude that could be delivered before the nature of the sensation changed. One might be tempted to interpret the former as being due to reactive changes around the active electrode stimulating sites, and the latter as being due to this plus, perhaps, a greater central tolerance for stimulation. However, note that all these measurements were made during the time when acute tissue responses to the surgery and implanted electrodes, such as edema, were active. One really needs to wait until these have resolved before making definitive statements about long-term effects and possible causes (Lefurge et al. 1991). One thing that did remain constant, though, was the relatively wide "safety zone" (the ratio of the upper limit to the threshold) for stimulation. Thus it was always easy to find a stimulus amplitude that reliably elicited a discrete, distally referred sensation without dropouts or spread, throughout the 2-wk period of the experiments.
In most cases, the elicited sensations could be systematically controlled through modulation of stimulus frequency and amplitude. Touch/pressure sensations were usually localized to the distal phalanx. With increasing stimulus amplitude, the sensation typically spread proximally. In the digits, the intensity gradient of referred phantom sensations was in the distal to proximal direction. This is consistent with the properties of the LIFEs as point stimulation electrodes, which stimulate progressively larger clusters of neurons with increasing charge injection (Meier et al. 1992; Nannini and Horch 1991), and the greater density of innervation as one moves distally along a digit (Vallbo and Johansson 1984).
Subjects showed a consistent ability to grade the intensity of elicited sensations over the duration of the study. However, there was no consistent pattern of improvement in this ability as evidenced by lack of definite trends in slopes of the regression lines or the variance of residuals around the regression lines. For the sensory studies, subjects were given random stimuli which elicited unitary, punctuate sensations for <10 min a day, and they graded the intensity of referred sensations without feedback as to what the stimulus level actually was. More experience, coupled with better feedback, may have improved their performance. Sensory reeducation has been shown to enhance sensory recovery after repair of nerve injuries (Dellon 1981; Mackin et al. 2002), but this recovery is not immediate: it occurs over many weeks. In the present study, sensory input was presented for only a short period of time (<75 min over the course of the study) and did not involve any formal training similar to that of sensory reeducation after nerve repair.
The location of elicited touch/pressure sensations either did not change or became better defined with time (e.g., row 2 of Fig. 1), the latter presumably because amputees could distinguish between intensity gradients, suggesting some beneficial effects of sensory stimulation in activating "silent" regions of the somatosensory cortex. The sensations elicited by the LIFEs tended to cover a larger area than those reported with microneurographic stimulation of intact sensory nerves (Schady and Torebjork 1983; Schady et al. 1983). This may be related to the fact that our study was not designed to precisely map the sizes and the shapes of projected fields. Rather we were more interested in the stability of the elicited sensations and their spread with increasing charge injection. Therefore when the subjects indicated, for example, a sensation referred to the thumb tip, no extra time was spent on elucidating its precise topography, unless the subject volunteered the information. A more-detailed study needs to be undertaken to more precisely define the locations, sizes and shapes of evoked receptive fields. Still, with increasing stimulus strength there was a clear increase in the spread of tactile sensations.
For proprioceptive sensations, amputees reported either movement of a given finger joint or movement of the entire digit. Subjects could reliably distinguish different degrees of joint flexion, through either stimulus frequency or stimulus amplitude modulation. Two subjects consistently reported a referred sensation of phantom grip opening and closing. The perceived magnitude of this pincer grip between thumb, index, and the middle fingers could be reliably controlled through stimulus frequency or amplitude modulation. This finding suggests that the sensory fibers which mediate complex movement and touch sensations of pincer grip opening and closing are topographically grouped and segregated in peripheral nerves. The frequency at which the joint was perceived as maximally flexed (250 Hz on average at 2 wk) is comparable to the maximal frequency of firing of muscle spindle fibers (Clark and Horch 1986). In contrast, frequencies 510 Hz were correlated with stronger touch/pressure sensations (Fig. 3A). With time sensations related to finger flexion became smoother and less "jerky," suggesting positive benefits of providing input to sensory cortex.
Although further studies are needed to more carefully investigate the relationship of stimulus pulse charge and frequency of stimulation to joint position sense, we did find that, in general, the lower the stimulus amplitude, the higher the frequency range needed to provide a sense of full joint excursion (Fig. 3D). This is consistent with encoding of joint position and movement information by the total afferent inflow from the pertinent sensory receptors.
Motor output
For motor control, in contrast to the sensory studies, the amputees were actively concentrating on improving their performance and were not simply passive subjects. The subjects were able to generate motor neuron activity related to phantom limb movements and demonstrated improvement in cursor control with practice. For example, the subjects managed to increase the precision of cursor control and score hits in 96 pixel targets on day 1, and by the end of the study, the majority of the subjects could proficiently control the position of the cursor in a 48 pixel wide target. For some, but not all, of the subjects there was gradual increase in the neural activity recorded by LIFEs associated with increased precision of cursor control. That is, through learning and practice, the subjects were able to improve motor control outflow to nerve fibers that had not been connected to muscle for periods of months or years. This is evidence of dynamic plasticity of CNS areas concerned with motor control even in the absence of proprioceptive feedback.
Implications for a neuroprosthetic arm
Upper limb amputees strongly desire a prosthetic control system that provides prehension feedback from the terminal device (Atkins et al. 1996). Referred sensations of pressure/touch resulting from intraneural stimulation were projected within the distribution of digital nerves and not scattered randomly throughout the hand. This is consistent with animal studies that have demonstrated that such stimulation is sufficiently localized that it is possible to elicit independent stimulation of microclusters within and between fascicles (Branner and Normann 2000; Branner et al. 2001; Yoshida and Horch 1993). Because the spread and topography of a pressure/touch sensation could be controlled through modulation of stimulation parameters, a given electrode could be used as a sensory channel for providing sensory feedback from a portion of the opposing ends of the gripper. With implants in separate fascicles, it is possible to get sensations in different digits. With implants of two or more electrodes in a given fascicle, different sensations could be distinguished within a given digit. Technologies under development may allow for stimulation of individual sensory neurons, increasing the number of sensory channels and resolution of sensory feedback (Dario et al. 1998; Donoghue 2002; González and Rodríguez 1997; Kovacs et al. 1992, 1994; Wallman et al. 2001).
For prehension feedback, sensors in the tip of the artificial hand/gripper would be adequate. Indeed, from a practical point of view the precise topography of elicited sensation does not need to be mapped for providing touch/pressure feedback from the gripper of the artificial arm. Rather it is only necessary that the modality of the elicited sensation be appropriate and that its referred location corresponds to the gripper. Because the intensity of the referred sensations can be reliably modulated, amputees would have appropriate information about contact and grip force.
Normal proprioception may not reflect a sense of joint angle (Scott and Loeb 1994; Soechting 1982), and recent work has demonstrated that extrinsic muscles of the hand signal finger-tip position sense more precisely than individual joint angles (Biggs et al. 1999). Even though our subjects could distinguish flexion of different IP joints, information about the location of the terminal aspect of the artificial hand/gripper may be all that is required for prosthetic control. Furthermore, joint position sense from the digits is relatively poor as compared with that from more proximal joints (Clark et al. 1995). When appropriate nerves are implanted for control of artificial limbs, we would expect better results for more proximal joint position sense.
In short, our study implies that if a neuroprosthetic arm were to be interfaced to the residual nerve stumps, amputees might be able to improve control over its movements and incorporate it into their body image through the effects of training, learning and central plasticity.
Limitations of this study
The subjects received <10 min of sensory experience on a daily basis. Because they were not informed of what the actual stimulus amplitude or frequency was, they received essentially no graded feedback to use in "improving" their performance. This was necessary to guard against them picking up on some cue other than the strength of a sensation for the tactile ratings or perceived finger position for the proprioception ratings to use in producing their responses. Subjects were provided with a similarly limited amount of motor experience, although in this case feedback was provided on their performance. Clearly amputees receiving a wealth of sensory re-education and motor control training may show greater improvements in sensory perception and motor control than reported here. On the other hand, the fact that even a short period of training can lead to measurable improvement in phantom limb motor control implies that human amputees show dynamic plasticity that may be comparable to that demonstrated in normal subjects after practice of simple movements.
The second limitation to the study is that we really can't say much about how the residual functions in the nerve stump central pathways compare with those in normal subjects. Ethical and regulatory constraints on human subject experimentation preclude implanting LIFEs in intact peripheral nerves, and similar data cannot be obtained via microneurographic studies that are not suitable for recording from and stimulating the same set of nerve fibers over periods of weeks.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Present addresses: G. S. Dhillon, Michael E. DeBakey Dept. of Surgery, Baylor College of Medicine, Houston, TX 77030; T. B. Krüger, Fraunhofer—Institute for Biomedical Engineering, Ensheimer Strasse 48, 66386 Sankt Ingbert, Germany; and J. S. Sandhu, Guru Nanak Dev University, Amritsar, Punjab, India
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: K. Horch, Dept. of Bioengineering, 50 S. Central Campus Dr., Rm. 2480, University of Utah, Salt Lake City, UT 84112 (E-mail: k.horch{at}m.cc.utah.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Biggs J, Horch K, and Clark FJ. Extrinsic muscles of the hand signal finger tip location more precisely than they signal the angles of individual finger joints. Exp Brain Res 125: 221–230, 1999.[CrossRef][ISI][Medline]
Branner A and Normann RA. A multielectrode array for intrafascicular recording and stimulation in sciatic nerve of cats. Brain Res Bull 51: 293–306, 2000.[CrossRef][ISI][Medline]
Branner A, Stein RB, and Normann RA. Selective stimulation of cat sciatic nerve using an array of varying-length microelectrodes. J Neurophysiol 85: 1585–1594, 2001.
Calford MB. Dynamic representational plasticity in sensory cortex. Neuroscience 111: 709–738, 2002.[CrossRef][ISI][Medline]
Carlson J, Lais AC, and Dyck PJ. Axonal atrophy from permanent peripheral axotomy in adult cat. J Neuropathol Exp Neurol 38: 579–585, 1979.[ISI][Medline]
Chen R, Cohen LG, and Hallett M. Nervous system reorganization following injury. Neuroscience 111: 761–773, 2002.[CrossRef][ISI][Medline]
Clark FJ and Horch KW. Kinesthesia. In: Handbook of Perception and Human Performance, edited by Boff KR, Kaufman L, and Thomas JP. New York: Wiley, 1986, p. 13-1–13-62.
Clark FJ, Larwood KJ, Davis ME, and Deffenbacher KA. A metric for assessing acuity in positioning joints and limbs. Exp Brain Res 107: 73–79, 1995.[ISI][Medline]
Clippinger FW, Avery R, and Titus BR. A sensory feedback system for an upper-limb amputation prosthesis. Bull Prosthet Res 10- 22: 247–258, 1974.
Cohen LG, Bandinelli S, Findley TW, and Hallet M. Motor reorganization after upper limb amputation in man. A study with focal magnetic stimulation. Brain 114: 615–627, 1991.
Cragg B and Thomas PK. Changes in conduction velocity and fibre size proximal to peripheral nerve lesions. J Physiol 157: 315–327, 1961.
Dario P, Garzella P, Toro M, Micera S, Alavi M, Meyer U, Valderrama E, Sebastiani L, Ghelarducci B, Mazzoni C, and Pastacaldi P. Neural interfaces for regenerated nerve stimulation and recording. IEEE Trans Rehab Eng 6: 353–363, 1998.[CrossRef][Medline]
Dellon AL. Evaluation of Sensibility and Reeducation of Sensation in the Hand. Baltimore, MD: Williams and Wilkins, 1981.
DeLuca C and Gilmore LD. Voluntary nerve signals from severed mammalian nerves: long-term recordings. Sci 191: 193–195, 1976.
DeLuca CJ, Gilmore LD, Bloom LJ, Thomson SJ, Cudworth AL, and Glimcher ML. Long-term neuroelectric signal recording from severed nerves. IEEE Trans Biomed Eng 29: 393–402, 1982.[ISI][Medline]
Dhillon GS, Lawrence SM, Hutchinson DT, and Horch KW. Residual function in peripheral nerve stumps of amputees: implications for neural control of artificial limbs. J Hand Surg 29A: 605–615, 2004.
Donoghue JP. Connecting cortex to machines: recent advances in brain interfaces. Nat Neurosci 5, Supplement: 1085–1088, 2002.
Edell DJ. A peripheral nerve information transducer for amputees: Long-term multichannel recordings from rabbit peripheral nerves. IEEE Trans Biomed Eng 33: 203–214, 1986.[ISI][Medline]
Elbert T, Flor H, Birbaumer N, Knecht S, Hampson S, Larbig W, and Taub E. Extensive reorganization of the somatosensory cortex in adult humans after nervous system injury. Neuroreport 5: 2593–2597, 1994.[ISI][Medline]
Gelfand SA. Hearing: An Introduction to Psychological and Physiological Acoustics. New York: Dekker, 1998.
González C and Rodríguez M. A flexible perforated microelectrode array probe for action potential recording in nerve and muscle tissues. J Neurosci Methods 72: 189–195, 1997.[CrossRef][ISI][Medline]
Goodall EV, Lefurge TM, and Horch KW. Information contained in sensory nerve recordings made with intrafascicular electrodes. IEEE Trans Biomed Eng 38: 846–850, 1991.[CrossRef][ISI][Medline]
Hall EJ, Flament D, Fraser C, and Lemon RN. Non-invasive brain stimulation reveals reorganised cortical outputs in amputees. Neurosci Lett 116: 379–386, 1990.[CrossRef][ISI][Medline]
Hoffer JA, Stein RB, and Gordon T. Differential atrophy of sensory and motor fibers following section of cat peripheral nerves. Brain Res 178: 347–361, 1979.[CrossRef][ISI][Medline]
Horch K. Central responses of cutaneous sensory neurons to peripheral nerve crush in the cat. Brain Res 151: 581–586, 1978.[CrossRef][ISI][Medline]
Horch KW and Lisney SJW. Changes in primary afferent depolarization of sensory neurons during peripheral nerve regeneration in the cat. J Physiol 313: 287–299, 1981a.
Horch KW and Lisney SJW. On the number and nature of regenerating myelinated axons after lesions of cutaneous nerves in the cat. J Physiol 313: 275–286, 1981b.
Jones EG, Woods TM, and Manger PR. Adaptive responses of monkey somatosensory cortex to peripheral and central deafferentation. Neuroscience 111: 775–797, 2002.[CrossRef][ISI][Medline]
Kawamura Y and Dyck PJ. Permanent axotomy by amputation results in loss of motor neurons in man. J Neuropathol Exp Neurol 40: 658–666, 1981.[ISI][Medline]
Kiraly JK and Krnjevic K. Some retrograde changes in function of nerves after peripheral section. Q J Exp Physiol Cogn Med Sci 44: 244–257, 1959.
Kovacs GT, Storment CW, Halks-Miller M, Belczynski CR, Jr, Della Santina CC, Lewis ER, and Maluf NI. Silicon-substrate microelectrode arrays for parallel recording of neural activity in peripheral and cranial nerves. IEEE Trans Biomed Eng 41: 567–577, 1994.[CrossRef][ISI][Medline]
Kovacs GTA, Storment CW, and Rosen JM. Regeneration microelectrode array for peripheral nerve recording and stimulation. IEEE Trans Biomed Eng 39: 893–902, 1992.[CrossRef][ISI][Medline]
Lawrence SM, Dhillon GS, and Horch KW. Fabrication and characteristics of an implantable, polymer-based, intrafascicular electrode. J Neurosci Methods 131: 9–26, 2003.[CrossRef][ISI][Medline]
Lawrence SM, Dhillon GS, Jensen W, Yoshida K, and Horch KW. Acute peripheral nerve recording characteristics of polymer-based longitudinal intrafascicular electrodes. IEEE Trans Neur Sys Rehab Eng 12: 345–348, 2004.[CrossRef]
Lefurge T, Goodall E, Horch K, Stensaas L, and Schoenberg A. Chronically implanted intrafascicular recording electrodes. Ann Biomed Eng 19: 197–207, 1991.[CrossRef][ISI][Medline]
Mackin EJ, Callahan AD, Osterman AL, Skirven TM, Schneider LH, and Hunter JM. Hunter, Mackin and Callahan's Rehabilitation of the Hand and Upper Extremity. St. Louis: Mosby, 2002.
Malagodi M, Horch KW, and Schoenberg AA. An intrafascicular electrode for recording of action potentials in peripheral nerves. Ann Biomed Eng 17: 397–410, 1989.[CrossRef][ISI][Medline]
Malmstrom JA, McNaughton TG, and Horch KW. Recording properties and biocompatibility of chronically implanted polymer-based intrafascicular electrodes. Ann Biomed Eng 26: 1055–1064, 1998.[CrossRef][ISI][Medline]
McNaughton TG and Horch KW. Metallized polymer fibers as leadwires and intrafascicular microelectrodes. J Neurosci Methods 70: 103–110, 1996.[CrossRef][ISI][Medline]
Meier JH, Rutten WLC, Zoutman AE, Boom HBK, and Bergveld P. Simulation of multipolar fiber selective neural stimulation using intrafascicular electrodes. IEEE Trans Biomed Eng 39: 122–134, 1992.[CrossRef][ISI][Medline]
Mendell LM, Munson JB, and Scott JG. Connectivity changes of Ia afferents on axotomized motoneurons. Brain Res 73: 338–342, 1974.[CrossRef][ISI][Medline]
Merzenich MM and Jenkins WM. Reorganization of cortical representations of the hand following alterations of skin inputs induced by nerve injury, skin island transfers, and experience. J Hand Ther 6: 89–104, 1993.[Medline]
Merzenich MM, Nelson RJ, Stryker MP, Cynader MS, Schoppmann A, and Zook JM. Somatosensory cortical map changes following digit amputation in adult monkeys. J Comp Neurol 224: 591–605, 1984.[CrossRef][ISI][Medline]
Milner TE and Stein RB. The effects of axotomy on the conduction of action potentials in peripheral sensory and motor nerve fibres. J Neurol Neurosurg Psychiat 44: 485–496, 1981.[Abstract]
Nannini N and Horch K. Muscle recruitment with intrafascicular electrodes. IEEE Trans Biomed Eng 38: 769–776, 1991.[CrossRef][ISI][Medline]
Sanes JN, Suner S, and Donoghue JP. Dynamic organization of primary motor cortex output to target muscles in adult rats. I. Long-term patterns of reorganization following motor or mixed peripheral nerve lesions. Exp Brain Res 79: 479–491, 1990.[ISI][Medline]
Schady WJL and Torebjork HE. Projected and receptive fields: a comparison of projected areas of sensations evoked by intraneural stimulation of mechanoreceptive units, and their innervation territories. Acta Physiol Scand 119: 267–275, 1983.[ISI][Medline]
Schady WJL, Torebjork HE, and Ochoa JL. Cerebral localisation function from the input of single mechanoreceptive units in man. Acta Physiol Scand 119: 277–285, 1983.[ISI][Medline]
Scott SH and Loeb GE. The computation of position sense from spindles in mono- and multiarticular muscles. J Neurosci 14: 7529–7540, 1994.[Abstract]
Soechting JF. Does position sense at the elbow reflect a sense of elbow joint angle or one of limb orientation? Brain Res 248: 392–395, 1982.[CrossRef][ISI][Medline]
Stevens SS. Psychophysics. New York: Wiley, 1986.
Sumner BEH and Watson WE. Retraction and expansion of the dendritic tree of motor neurones of adult rats induced in vivo. Nature 233: 273–275, 1971.[CrossRef][Medline]
Sunderland S. Nerves and Nerve Injuries. New York: Churchill Livingstone, 1978.
Théoret H, Merabet L, and Pascual-Leone A. Behavioral and neuroplastic changes in the blind: evidence for functionally relevant cross-modal interactions. J Physiol 98: 221–233, 2004.
Törnqvist E and Aldskogius H. Motoneuron survival is not affected by the proximo-distal level of axotomy but by the possibility of regenerating axons to gain access to the distal nerve stump. J Neurosci Res 39: 159–165, 1994.[CrossRef][ISI][Medline]
Vallbo AB and Johansson RS. Properties of cutaneous mechanoreceptors in the human hand related to touch sensation. Hum Neurobiol 3: 3–14, 1984.[ISI][Medline]
Wall JT, Xu J, and Wang X. Human brain plasticity: an emerging view of the multiple substrates and mechanisms that cause cortical changes and related sensory dysfunctions after injuries of sensory inputs from the body. Brain Research Reviews 39: 181–215, 2002.[CrossRef][Medline]
Wallman L, Zhang Y, Laurell T, and Danielsen N. The geometric design of micromachined silicon sieve electrodes influences functional nerve regeneration. Biomat 22: 1187–1193, 2001.[CrossRef]
Yoshida K and Horch K. Closed-loop control of ankle position using muscle afferent feedback with functional neuromuscular stimulation. IEEE Trans Biomed Eng 43: 167–176, 1996.[CrossRef][ISI][Medline]
Yoshida K and Horch K. Selective stimulation of peripheral nerve fibers using dual intrafascicular electrodes. IEEE Trans Biomed Eng 40: 492–494, 1993.[CrossRef][ISI][Medline]
This article has been cited by other articles:
|
|
 |
|
  K. T. Reilly and A. Sirigu The Motor Cortex and Its Role in Phantom Limb Phenomena Neuroscientist, April 1, 2008; 14(2): 195 - 202. [Abstract] [PDF]
|
|
|
|
 |
|
 K. T. Reilly, C. Mercier, M. H. Schieber, and A. Sirigu Persistent hand motor commands in the amputees' brain Brain, August 1, 2006; 129(8): 2211 - 2223. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Mercier, K. T. Reilly, C. D. Vargas, A. Aballea, and A. Sirigu Mapping phantom movement representations in the motor cortex of amputees Brain, August 1, 2006; 129(8): 2202 - 2210. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||
) and upper limit (
) stimulus amplitudes as a function of time. For each of the 12 electrodes included in the figure, amplitudes are normalized to the threshold stimulus amplitude on the 1st day of testing for that particular electrode. Not all electrodes were tested for the full 15 days due to limitations in subject availability.
, 
, 
) and 38 µA (
, 
, 
), 300-µs-duration pulses in subject 4532. In this case, the stimuli were presented in incremental, rather than pseudorandom, order.
