- Email: [email protected]
Mechanical and fatigue properties of wrist flexor muscles during repetitive contractions after cervical spinal cord injury
929
Mechanical and Fatigue Properties of Wrist Flexor Muscles During Repetitive Contractions After Cervical Spinal Cord Injury Tracy Cameron, MSBME, Blair Calancie, PhD ABSTRACT. Cameron T, Calancie B. Mechanical and fatigue properties of wrist flexor muscles during repetitive contractions after cervical spinal cord injury. Arch Phys Med Rehabil 1995;76:929-33.
Objectives: Force generation and fatigue properties of wrist flexor muscles were examined in subjects with chronic (>1 year) cervical spinal cord injury (SCI, n = 16), and also in a control group of able-bodied (AB, n = 9) subjects. Design: Using surface electrodes, wrist flexor muscles were stimulated with 126 trains of 26 stimuli at a frequency of 40Hz. The offset of each train was followed by a 1.5-second pause, for a total fatigue-test time of approximately 4.2 minutes. Isometric wrist flexion force was measured with a strain gauge. Setting: This study was conducted at a research and rehabilitation center for spinal cord injury. Main Outcome Measures: Force profiles were analyzed for the maximum (peak) amplitude, the rise time, and the time constant of relaxation. Results: At the outset, the average peak isometric measured in the SCI group was approximately one half that of the AB subjects. Although the relative decline in force with repeated stimulation was comparable between groups, the slowing of relaxation rate was much more pronounced in the SCI group. Conclusions: These findings are consistent with alterations in the metabolic profiles of wrist flexor muscles in the SCI group, probably reflected their altered activation pattern. When designing stimulation protocols for optimizing force and fatigue resistance in muscle left partially-paralyzed after spinal cord injury, particular care must be taken to allow adequate time for complete muscle relaxation, to avoid overdriving of the muscle and a loss of functional capacity. © 1995 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation EUROMUSCULAR electrical stimulation (NMS) has been used to strengthen muscles in paralyzed individuals ~ and to provide individuals with the ability to initiate movements that are of functional value, such as the restoration of limited grasping functions after cervical spinal cord injury. 2-4 For restoration of ambulation in the lower extremities, groups have investigated NMS alone, 5'6 or in combination with orthotics. 7-9 Although improvement has been seen in various engineering
N
From The Miami Project to Cure Paralysis and Department of Neurological Surgery (Ms. Cameron, Dr. Calancie), University of Miami School of Medicine, Miami, FL. Submitted for publication November 17, 1994. Accepted in revised form April 20., 1995. This work was supported by the Miami Project to Cure Paralysis. The authors have chosen not to select a disclosure statement. Reprint requests to Blair Calancie, The Miami Project to Cure Paralysis and Department of Neurological Surgery, University of Miami School of Medicine, Miami, FL 33136. © 1995 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003-9993/95/7610-331853.00/0
aspects of these systems, including electrode design and implantation ~° and reliable stimulator power supplies and controllers,Et a major limitation in the application of these technologies to human subjects continues to be neuromuscular fatigue. In this case, fatigue refers to decline in sustained muscle force with repetitive supramaximal stimulation of the muscle nerve. Neural activity is probably the major factor influencing the characteristic properties (contractility and fatigue resistance) of skeletal muscle. ~2Muscle fibers of the same motor unit contain identical isomyosins, suggesting that the neural input determines the phenotypic expression? 3 Supporting these observations are studies in which a muscle nerve from a slow-twitch fatigue-resistant muscle has been transposed to a fast-twitch fatigable muscle, resulting in modification of that muscle's properties to resemble those of a slow-twitch, fatigue-resistant muscle. ~4-~6In addition, recent evidence indicates that conversion in the opposite direction--from slow twitch to fast twitch--can also occur under some circumstances, t7 Thus, the characteristics of skeletal muscle fibers can be altered, depending on the total amount and pattern of neuronal input, t7"~8 Severe injury to the cervical spinal cord typically results in a mixed picture of both upper and lower motoneuron involvement for muscles of the distal upper extremity. ~9-2zMoreover, surviving motor units (whose somata reside caudal to the level of injury) will experience near total quiescence during the first few months postinjury. The onset of spasticity in subjects whose injury is neurologically complete will lead to a motor unit activity pattern very different from that experienced by the motor units before the injury. This new pattern is characterized by infrequent and relatively brief periods of intense motoneuron discharge (ie, spasms). Based on animal studies, such an activity pattern would seem to be ideal for maximizing the proportion of type IIb muscle fibers in the affected muscles. 23 Application of NMS for restoration of functional movements would then be hampered considerably by limited fatigue resistance of muscle(s) whose properties have been altered in this manner. Evidence for this phenomenon of altered mechanical and fatigue properties was recently obtained in subjects with chronic spinal cord injury (SCI). Stein and coworkers 24 showed profound changes in muscle properties of the tibialis anterior muscle, and in results similar to numerous animal studies, showed that chronic repetitive stimulation, gradually increasing to a total of 8 hours per day, dramatically increased the fatigue resistance of muscle. The purpose of this study was to examine force generation and fatigue properties of wrist flexor muscles in subjects with chronic (>1 year) injuries to the cervical spinal cord. This was accomplished by evoking a series of repeated maximal contractions. Although simple and rapid, this evaluation method was expected to provide useful information to guide development of optimal stimulation and training programs for this particular muscle group, considered essential for grasping with NMS. Because this muscle group presents with a more mixed pattern of upper and lower motoneuron involvement than does the tibialis anterior, 24 results from this latter study should not
Arch Phys Med Rehabil Vol 76, October 1995
930
FATIGUE PROPERTIES OF CSCI, Cameron
necessarily be extrapolated to the wrist flexors. Some of these results have been presented in abstract form. 25
electrodes
METHODS Subjects Two groups were examined in this study: able-bodied (AB) men (n = 6) and women (n = 3) between the ages of 18 and 35, and men (n = 14) and women (n = 2) subjects who had sustained injuries to their cervical spinal cord (ie, they were quadriplegic). Background information for the SCI subjects is given in table 1. The age of the SCI subjects was 27.6 _ 4.7 years (mean _+ standard deviation) with a time postinjury of 7.1 + 5.1 years. The level of injury describes the most caudal neurological level that retains some voluntary motor function. The mean age of the control subjects was 27.8 _+ 3.7 years. All procedures were performed identically on both groups. All subjects gave their informed consent to participate in this study.
Procedure Arm Supporting Device. A frame (fig 1) held the forearm and palm in a neutral position for the measurement of isometric force. Attached tO the frame by a hinge was a cuff that was strapped to the hand. A force transducer~ (10kg/mm) attached to and perpendicular to the cuff measured the force produced by the wrist flexor muscles at the palm (ie, approximately 10cm from the axis of rotation of the wrist), while the immobilization of the hand by the cuff reduced the effects of pronation and supination. Instrumentation and Fatigue Test. Subjects sat in a dental chair to which the arm support was affixed. The skin over the left wrist flexors was wiped with an alcohol swab and two stimulating electrodes (Unipatch 615 reusable, 5cm × 5cm) b were applied to the skin. One electrode was placed over the belly of the flexor carpi radialis muscle, approximately 5cm distal to the olecranon process. The second electrode was placed on the dorsal aspect of the wrist. Electrode placement was tested by briefly stimulating (Stayodynamic EMS/plus neuromuscular stimulator; biphasic stimulation at 40Hz total phase; 100#s pulse width) c and examining the movement of the wrist. If the correct movement (flexion) did not occur with a 40mA stimulus intensity, the position of the electrodes was changed slightly Table 1: Background Formation for Spinal Cord Injury Study Subjects Number
Age
Sex
Level of Injury
MMT
Years Postinjury
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
33 23 29 21 26 31 26 28 34 28 23 37 27 19 25 29
M F M M M M M M M M M M M M F M
C4, complete C6, complete C6, complete C5-6, incomplete C5-6, complete C5, incomplete C5, complete C5, complete C6, incomplete C5-6, incomplete C5, incomplete C5-6, incomplete C5-6, complete C5, complete C3-4, complete C5-6, complete
0 1 3 5 1 4 3 1 1 1 5 5 1 0 0 1
19 2 11 4 9 9 4.5 3 18 6.5 3 5 3 2 6 6
Level of injury based on medical evaluation at time of discharge. Subjects are denoted by incomplete with indication of most caudal segmental level within which they have at least some motor function. M M T performed on left wrist flexor muscles only. By chance, SCI population did not include anyone with a M M T score of 2. Abbreviation: MMT, manual muscle test.
Arch Phys Med Rehabil Vol 76, October 1995
arm
restrainer
force t r a n s d u c e r
Fig 1. Experimental arrangement, showing hand positioned within splint and force-recording arrangement. All studies were conducted on the left arm of subjects.
until the desired movement was obtained, after which the subject's arm was fastened into the device. The force signal was bridge-amplified,a displayed on a digital oscilloscope,e and recorded for off-line analysis (Vetter 4000 f and Mitsubishi U32 g videocassette recorder). The fatigue test consisted of 126 trains of 26 stimuli at a frequency of 40Hz. A frequency of 40Hz was choosen to supramaximally activate the wrist flexor group found to be composed of predominantly fast fatigable fibers. Thus, the same total number of stimuli were delivered in this fatigue test as for that first described by Burke and colleagues,26 but the present test required 4 minutes to complete rather than 2. This stimulus protocol resulted in a 1.5-second pause between successive stimulus trains, and was modified because more rapidly applied trains described by Burke resulted in partial fusion of the force profile in subjects with spinal cord injury (see Discussion). Manual Muscle Test. A manual muscle test (MMT) was used to determine the amount of function each spinal cord injured individual had in their left wrist flexor muscles. A registered occupational therapist assigned a grade on a scale of 0 to 5 as follows: 0, no visible or palpable flicker of contraction; 1, visible or palpable flicker of contraction but no movement of the joint; 2, wrist flexion in a gravity-neutral plane; 3, wrist flexion against gravity but not against mild resistance; 4, wrist flexion against mild resistance; and 5, normal muscle strength. Data Analysis. The individual force profiles for each subject were digitized by a personal computer containing an analog to digital board and averaging software, h For the fatigue test, the force profiles of 122 consecutive contractions were acquired for each subject. Averages were made of sweeps 1 to 5 (0), 18 to 22 (20), 38 to 42 (40), 58 to 62 (60), 78 to 82 (80), 98 to 102 (100), and 118 to 122 (120). The following values were determined for each the averaged force records: the maximum amplitude, the rise time, and the time constant of relaxation. The rise time was determined by measuring the time required for the force to increase to 50% of its final (maximum) value. The time constant of relaxation was determined as the time taken for the exponential portion of the relaxation curve to fall to one-third of its value at the end of stimulation. These values were measured manually with digital cursors within the software package, h The same individual made all measurements for all research subjects, and employed the same criteria for all measurements. Between and within group differences were examined using analysis of variance (ANOVA) and post hoc paired t test. RESULTS Figure 2 shows typical averaged force profiles taken during a fatigue test from an AB (panel A) and an SCI (panel B)
FATIGUE PROPERTIES OF CSCI, Cameron
931
100
A
.,_
X
•
Able-bodied
E 0
80
~.
6O
B
1-5
18-22
38-42
58-62
78-82
E
J
98-102
11B-122
Fig 2. Representative force profiles for an able-bodied (A) and a quadriplegic (B) subject during the fatigue test. Each trace is the average of 5 traces taken at equal intervals throughout the test, The calibration marks represent 1 second (x axis) and lkg (y axis),
0 U-
40
I
subject. Averages were made of five repetitions to allow us to study the time course of fatigue. In this example, the maximum force production in the AB subject (panel A) was approximately 2 times that measured from the SCI subject (panel B). Furthermore, the records show that the rate of relaxation remained rapid throughout the fatigue test for the AB subject, whereas relaxation-rate slowed considerably with successive contractions in the SCI subject. Figure 3 illustrates the averaged values of maximum force measured for the AB and SCI groups during the fatigue test. At the outset, the mean force in the AB group was approximately 2.2 times that of the SCI group. To our surprise, this ratio maintained constant throughout much of the fatigue test, as illustrated in figure 4. For this figure, the maximum force measured during the first series of five contractions was assigned a value of 100% for each subject, and all subsequent measures were expressed as a percentage of this starting point. The mean of these normalized values for the AB and SCI groups at each of the periods analyzed is shown in figure 4. The largest discrepancy between the two groups occurred at trial 80, but
6
7 v (b
6 5
E (l) O O LL
~
Able-bodied Quadriplegic
100 '
120
o)
E 500 400
•
Able-bodied Quadriplegic MM0 ~ MM1 ~J MM3/4 /
[]
/ -
o ~T
~,
300 %,-,
O
2
r--
200
t.-. o o (1)
100
1 • - - -
t--
i
0
80
neither these nor any other trials were significantly different between the two groups, as determined by ANOVA. In both groups the maximum averaged force at the end of the fatigue test was seen to decrease to approximately 40% of the initial value. Figure 5 shows the time constant of relaxation for different groups of subjects at different contraction times. Filled symbols represent the overall means and standard deviations of the AB
t1:I X
3
0t
6'0
Fig 4. Normalized force amplitude (percent of maximum force) versus sweep number, Both groups were found to decrease to approximately one half the maximum force by the end of the fatigue test. Able-bodied (n = 7) Quadriplegic (n = 14). Error bars indicate standard error measurements.
E O e •
I
40
Sweep Number
v
8 7
20
i
~
i
i
L
i
20
40
60
80
100
120
Sweep Number Fig 3. Peak force amplitude (mean) versus the sweep number for the two groups. The able-bodied group is shown with the solid circle (n = 7) whereas the quadriplegic group is shown with the solid square (n = 14). Error bars indicate standard error measurements.
0
I
0
I
"20
I
~
I
I
I
40
60
80
100
120
Sweep Number Fig 5. Time constant of force relaxation versus sweep number. The quadriplegic group (n = 14) exhibited a significantly greater relaxation time compared with the able-bodied group (n = 7) by the end of the test (p < .005}, Quadriplegic subjects are grouped by their score on the manual muscle test; MMO (n = 1), MM1 (n = 7), M M 3 / 4 (n = 3), and M M 5 (n = 3). Error bars indicate standard error measurements,
Arch Phys Med Rehabil Vol 76, October 1995
932
FATIGUE PROPERTIES OF CSCI, Cameron
and SCI groups, whereas the open symbols show the means and standard deviations of subsets of the SCI population. At the beginning of the fatigue test, the rate of relaxation in the AB group was faster (ie, smaller time constant) than that of the SCI groups as a whole or any of its subsets. Based on ANOVA, this difference between the averaged AB and SCI groups was found to be significant (F = 26.73, p < .0001). The rate of relaxation decreased with continuing contraction for all groups examined. Between the initial and second series of contractions studied (ie, averages of contractions 1 to 5 and 18 to 22), the slowing of relaxation was approximately constant for all groups. However, clear differences in the slopes became evident in the force profiles for the subsets of SCI subjects (open symbols), where slowing of relaxation was greater as the degree of paralysis increased. For example, by the end of the fatigue test the difference between the initial (81) and ending (439) relaxation rates in the MM 0 group was more than 5 times. By comparison, this difference was less than 3-fold (49.2 vs 133.7) in the AB group. For this latter group, the difference between starting and ending values was significant (F = 5.803; p < .005). For comparison between individual groups as plotted in figure 5, a significant difference was seen in initial and final relaxation rate values based on ANOVA (F = 5.803, p < .005). A post hoc t test comparing the mean values for the AB group with those of the SCI average group showed a significant difference (T = 3.44, p < .005). The time to 50% of maximum force (fig 6) is plotted for individual groups of subjects as a function of the sweep number. Filled symbols represent the overall means of the AB and SCI groups, whereas the open symbols show the means and standard deviations of subsets of the SCI population. AB subjects were found to take a longer time to rise to a 50% of maximum than the SCI group. However, as all the subjects fatigued, their times to 50% of maximum increased. Both the AB and SCI group showed a significant difference in rise-time between the initial
E
360
O 320
Able-bodied Quadriplegic
0 [] V
MM3/4
z ~
|
l / / ~ - O
~4~
o
E m
E
280
X o~
E 240 0
tn 0
200
E I--
160 I
I
F
I
I
[
I
0
20
40
60
80
100
120
Sweep Number Fig 6. Time to 50% maximum force versus the sweep number. The quadriplegic subjects are grouped as a whole (n = 16) and also are subdivided based on their manual muscle test scores. All groups were found to increase the time required to reach 50% maximum by the end of the test. The able-bodied group (n = g) was found to take longer at all times (p < ,005) throughout the test as compared with the quadriplegic group. Error bars indicate standard error measurements.
Arch Phys Med Rehabil Vol 76, October 1995
and final measures of the test (T = -5.47 and T = -5.72, respectively, p < .001), but did not differ from one another by a significant amount. DISCUSSION After neurologically complete injury to the cervical spinal cord, a variety of alterations in muscle fibers of distal upper extremity muscles will occur over different time courses. In the short term, neurologically complete injury typically results in atrophy of these muscles. This atrophy is caused by a combination of: (1) inactivity (or disuse atrophy); and (2) denervation, caused by loss of varying amounts of grey matter, including motoneurons of the ventral h o m J 9 Disuse atrophy may be reversed by activity, either through recovery of supraspinal conduction such that volitional movements are restored, or through the development of spasticity (although the limited duration of these latter contractions will be less effective in reversing atrophy). Denervation atrophy may be reversible to the extent that affected muscle fibers can remain viable until innervation by sprouts from surviving, adjacent motoneuron axons takes place. Following innervation, the same factors related to the reversal of disuse atrophy are in effect. As a group, the maximum force produced by supramaximal stimulation of wrist flexor muscles in SCI subjects at any given point during the fatigue test was approximately half that produced by the able-bodied subjects. Nevertheless, these force levels are high enough for the completion of a variety of functional tasks. The decreased but significant force levels seen in SCI subjects--particularly those with MMT scores of 0 or 1 indicate that: (1) muscles were innervated but typically lacked sufficient supraspinal input to cause useful voluntary contraction; and (2) atrophy of those fibers paralyzed to voluntary input contributed to the decreased forces seen. From this study it is not possible to determine the extent of partial denervation and sprouting that occurred. However, the level of injury of the majority of subjects (see table) coincides closely with recent studies of upper extremity muscles,2° and suggests that denervation and sprouting were probably widespread in the majority (if not all) of the SCI subjects examined. The rate of relaxation at the end of a stimulus train was decreased in SCI subjects compared with controls, and this difference was magnified by the fatigue test. Similar findings have been reported for the tibialis anterior of SCI subjects,24 and may reflect a decreased rate of Ca 2+ reuptake. A practical consequence of this property of the SCI wrist flexors is that repetitive contractions elicited by NMS, either for specific tasks or for exercise, 2 may not allow for complete relaxation before the onset of the next stimulus train if muscles in able-bodied subjects are used to establish the stimulus parameters (ie, time between contractions). In this study we experienced such a problem when applying the fatigue test protocol as described by Burke and colleagues.26 It became immediately apparent with the first SCI subject examined that we would have to modify the protocol to allow a longer period between successive trains. The risk of not allowing the muscle to fully relax between contractions is of overdriving the muscle and causing selective damage of the fatigable fibers, t2 Although this complication from NMS is most likely to occur during training of the relatively untrained muscle, it may also occur during functional procedures involving NMS ambulation and bicycle ergometry. In the present study, the strong positive relationship between increasing relaxation time and degree of paralysis of the wrist flexor muscles suggests that the pattern of use (and disuse) of these muscles was directly responsible for their performance during the fatigue test. It would be of interest to determine whether a protocol of NMS stimulation for forearm flexors in
FATIGUE PROPERTIES OF CSCI, Cameron
SCI subjects, incrementing in duration over a period of weeks, 24 would lead to relaxation trajectories more like those seen in AB subjects. In addition, demonstration of greater M M T scores at the end of such a stimulation program would be evidence supporting the utility of N M S stimulation in muscles under partial supraspinal control (ie, M M T score 1 to 3).
11. 12. 13.
CONCLUSIONS It is r e c o m m e n d e d by the authors that w h e n N M S is applied to relatively untrained muscle, care be taken to allow for the complete relaxation of the target muscle between contractions. The repetitive contractions elicited by many N M S systems, either for functional m o v e m e n t s or for exercise, may not allow for complete relaxation before the onset of the next stimulus train if muscles in able-bodied subjects are used to establish the stimulus parameters (ie, time between contractions). The risk of not allowing the muscle to fully relax between contractions m a y cause overdriving of the muscle and subsequent damage of the more susceptible fatigable fibers. Although this complication from N M S is most likely to occur during training of the relatively untrained muscle, it m a y also occur during functional procedures involving N M S ambulation and bicycle ergometry. The simple but important procedure of increasing the rest time between contractions m a y avoid unnecessary damage to fatigable fibers.
Acknowledgments: The authors thank Belinda Needham-Shropshire for performing the manual muscle tests, Diane Davies for advice regarding statistical analyses, and Jim Broton for providing a critical review of the manuscript.
14. 15. 16. 17. 18. 19. 20. 21. 22.
References 1. Fields RW. Electromyographically triggered electric muscle stimulation for chronic hemiplegia. Arch Phys Med Rehabil 1983; 68:40713. 2. Peckham PH, Mortimer JT. Restoration of hand function in the quadriplegic through electrical stimulation. In: Hambrecht FF, Reswick JB, editors. Functional electrical stimulation, applications in neural protheses. New York: Dekker, 1977:83-95. 3. Handa Y, Handa T, Nakatsuchi Y, Yagi R, Hoshimiya M. A voice controlled functional electrical stimulation system for the paralyzed hand. Jap J Med Electron Biol Eng 1985;23:292-8. 4. Wieler M, Prochazka A, Kenwell Z, Gauthier M, Isaacson G. "Electronic Glove" augments tenodesis-grip and hand-opening in people with quadriplegia. J Am Paraplegia Soc 1994; 17:132. 5. Klose KJ, Needham-Shropshire B, Lebwohl N, Green BA. The results of a 35 SCI subject experience with the Parastep-I walking system. J Am Paraplegia Soc 1994; 17:115. 6. Marsolais EB, Kobetic R. Functional walking in paralyzed patients by means of electrical stimulation. Clin Ortho 1983; 175:30-6. 7. Andrews BJ, Baxendale RM, Barnett RW, Phillips GF, Paul JP, Freeman PA. A hybrid orthosis for paraplegics incorporating feedback control. In: Popovic D, editor. Advances in external control of human extremities X. Belgrade: ETAN, 1987:297-3l 1. 8. Popovic D, Tomovic R, Schwirtlick L. Hybrid assistive system-the motor neuroprosthesis. IEEE Trans Biomed Eng 1989;36:39. 9. Solomonow M, Eldred E, Lyman J, Foster J. Control of muscle contractile force through indirect high-frequency stimulation. Am J Physiol 1983;62:71-82. 10. Naples GG, Mortimer JT, Yuen TGH. Overview of peripheral nerve electrode design and implantation. In: Agnew WF, McCreery DB,
23.
24. 25. 26.
933
editors. Neural prostheses. Fundamental studies. Newark: Prentice Hall, 1990:108-45. Prochazka A. Comparison of natural and artificial control of movement. IEEE Trans Rehab Engin 1993; 1:7-17. Pette D, Vrbova G. Invited Review: Neural control of phenotypic expression in mammalian muscle fibers. Muscle Nerve 1985;8:67689. Gauthier GF, Burke RE, Lowey S, Hobbs AW. Myosin isozymes in normal and cross-reinnervated cat skeletal muscle fibers. J Cell Biol 1983;97:756-71. Buller AJ, Eccles JC, Eccles RM. Differentiation of fast and slow muscles in the cat hind limb. J Physiol (Lond) 1960; 150:399-416. Salmons S, Vrbova G. The influence of activity on some contractile characteristics of mammalian fast and slow muscles. J Physiol (Lond) 1969;201:535-49. Luff AR, Proske V, Webb SN. The transformation of cross-reinnervated slow-twitch muscle after deafferentiation in the cat. Exp Brain Res 1984;55:152-7. Lehman AJR, Tulley FM, Vrbova G, Dimitrijevic MR, Towle JA. Muscle fatigue in some neurological disorders. Muscle Nerve 1989; 12:938-42. Vrbova G, Dimitrijevic MR, Partridge M, Halter J, Verhagen-Metman L. Reversal of increased muscle fatigue in paraplegic patients by electrical stimulation [abstract]. Soc Neurosci 1987; 13:253. Peckham PH, Mortimer JT, Marsolais EB. Upper and lower motor neuron lesions in the upper extremity muscles of tetraplegics. Paraplegia 1976; 14:115-21. Marino RJ, Herbison GJ, Ditunno JFJ. Peripheral sprouting in the zone of partial preservation in motor complete spinal cord injury. J Am Paraplegia Soc 1992; 15:122. Yang JF, Stein RB, Jhamandas J, Gordon T. Motor unit numbers and contractile properties after spinal cord injury. Ann Neurol 1990; 28:496-502. Calancie B, Gordon T, Stein RB, Davis LA, Dolphin B. Contractile properties of thenar motor units in normal and quadriplegic subjects [abstract]. Soc Neurosci 1987; 13:875. Burke RE. Motor units: anatomy, physiology, and functional organization~ In: Brooks VB, editor. Handbook of physiology. The nervous system. Bethesda, MD: American Physiological Society, 1981:345442. Stein RB, Gordon T, Jefferson J, Sharfenberger A, Yang JF, Totosy de Zepetnek J, et al. Optimal stimulation of paralyzed muscle after human spinal cord injury. J Appl Physiol 1992;72:1393-1400. Cameron T, Calancie B. Fatigue properties of wrist flexor muscles in able-bodied and quadriplegic subjects [abstract]. J Am Paraplegia Soc 1990; 13:120. Burke RE, Levine DN, Tsairis P, Zajac FE III. Physiological types and histochemical profiles in motor units of the cat gastrocnemius. J Physiol (Lond) 1973;234:723-48.
Suppliers a. Grass FTI0; Grass, 101 Old Colony Ave., PO Box 51@ Quincy, MA 02269. b. Unipatch, PO Box 271, Wabasha, MN 55981. c. Stayodynamics, 1225 Florida Ave., PO Box 1379, Longmont, CO 80501. d. Measurements Group 2120A; Measurements Group, PO Box 27777, Raleigh, NC 27611. e. Gould 1425; Gould, 774 North Lake Boulevard, no. 1000, Altamonte Springs, FL 32701. f. Vetter, Box 143, Rebersburg, PA 16872. g. Mitsubishi, Denki Building, Marunouchi, Tokyo. h. RC Electronics Computerscop, $386-D Hollister Ave., Santa Barbara, CA 93111.
Arch Phys Med Rehabil Vol 76, October 1995
Mechanical and Fatigue Properties of Wrist Flexor Muscles During Repetitive Contractions After Cervical Spinal Cord Injury Tracy Cameron, MSBME, Blair Calancie, PhD ABSTRACT. Cameron T, Calancie B. Mechanical and fatigue properties of wrist flexor muscles during repetitive contractions after cervical spinal cord injury. Arch Phys Med Rehabil 1995;76:929-33.
Objectives: Force generation and fatigue properties of wrist flexor muscles were examined in subjects with chronic (>1 year) cervical spinal cord injury (SCI, n = 16), and also in a control group of able-bodied (AB, n = 9) subjects. Design: Using surface electrodes, wrist flexor muscles were stimulated with 126 trains of 26 stimuli at a frequency of 40Hz. The offset of each train was followed by a 1.5-second pause, for a total fatigue-test time of approximately 4.2 minutes. Isometric wrist flexion force was measured with a strain gauge. Setting: This study was conducted at a research and rehabilitation center for spinal cord injury. Main Outcome Measures: Force profiles were analyzed for the maximum (peak) amplitude, the rise time, and the time constant of relaxation. Results: At the outset, the average peak isometric measured in the SCI group was approximately one half that of the AB subjects. Although the relative decline in force with repeated stimulation was comparable between groups, the slowing of relaxation rate was much more pronounced in the SCI group. Conclusions: These findings are consistent with alterations in the metabolic profiles of wrist flexor muscles in the SCI group, probably reflected their altered activation pattern. When designing stimulation protocols for optimizing force and fatigue resistance in muscle left partially-paralyzed after spinal cord injury, particular care must be taken to allow adequate time for complete muscle relaxation, to avoid overdriving of the muscle and a loss of functional capacity. © 1995 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation EUROMUSCULAR electrical stimulation (NMS) has been used to strengthen muscles in paralyzed individuals ~ and to provide individuals with the ability to initiate movements that are of functional value, such as the restoration of limited grasping functions after cervical spinal cord injury. 2-4 For restoration of ambulation in the lower extremities, groups have investigated NMS alone, 5'6 or in combination with orthotics. 7-9 Although improvement has been seen in various engineering
N
From The Miami Project to Cure Paralysis and Department of Neurological Surgery (Ms. Cameron, Dr. Calancie), University of Miami School of Medicine, Miami, FL. Submitted for publication November 17, 1994. Accepted in revised form April 20., 1995. This work was supported by the Miami Project to Cure Paralysis. The authors have chosen not to select a disclosure statement. Reprint requests to Blair Calancie, The Miami Project to Cure Paralysis and Department of Neurological Surgery, University of Miami School of Medicine, Miami, FL 33136. © 1995 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003-9993/95/7610-331853.00/0
aspects of these systems, including electrode design and implantation ~° and reliable stimulator power supplies and controllers,Et a major limitation in the application of these technologies to human subjects continues to be neuromuscular fatigue. In this case, fatigue refers to decline in sustained muscle force with repetitive supramaximal stimulation of the muscle nerve. Neural activity is probably the major factor influencing the characteristic properties (contractility and fatigue resistance) of skeletal muscle. ~2Muscle fibers of the same motor unit contain identical isomyosins, suggesting that the neural input determines the phenotypic expression? 3 Supporting these observations are studies in which a muscle nerve from a slow-twitch fatigue-resistant muscle has been transposed to a fast-twitch fatigable muscle, resulting in modification of that muscle's properties to resemble those of a slow-twitch, fatigue-resistant muscle. ~4-~6In addition, recent evidence indicates that conversion in the opposite direction--from slow twitch to fast twitch--can also occur under some circumstances, t7 Thus, the characteristics of skeletal muscle fibers can be altered, depending on the total amount and pattern of neuronal input, t7"~8 Severe injury to the cervical spinal cord typically results in a mixed picture of both upper and lower motoneuron involvement for muscles of the distal upper extremity. ~9-2zMoreover, surviving motor units (whose somata reside caudal to the level of injury) will experience near total quiescence during the first few months postinjury. The onset of spasticity in subjects whose injury is neurologically complete will lead to a motor unit activity pattern very different from that experienced by the motor units before the injury. This new pattern is characterized by infrequent and relatively brief periods of intense motoneuron discharge (ie, spasms). Based on animal studies, such an activity pattern would seem to be ideal for maximizing the proportion of type IIb muscle fibers in the affected muscles. 23 Application of NMS for restoration of functional movements would then be hampered considerably by limited fatigue resistance of muscle(s) whose properties have been altered in this manner. Evidence for this phenomenon of altered mechanical and fatigue properties was recently obtained in subjects with chronic spinal cord injury (SCI). Stein and coworkers 24 showed profound changes in muscle properties of the tibialis anterior muscle, and in results similar to numerous animal studies, showed that chronic repetitive stimulation, gradually increasing to a total of 8 hours per day, dramatically increased the fatigue resistance of muscle. The purpose of this study was to examine force generation and fatigue properties of wrist flexor muscles in subjects with chronic (>1 year) injuries to the cervical spinal cord. This was accomplished by evoking a series of repeated maximal contractions. Although simple and rapid, this evaluation method was expected to provide useful information to guide development of optimal stimulation and training programs for this particular muscle group, considered essential for grasping with NMS. Because this muscle group presents with a more mixed pattern of upper and lower motoneuron involvement than does the tibialis anterior, 24 results from this latter study should not
Arch Phys Med Rehabil Vol 76, October 1995
930
FATIGUE PROPERTIES OF CSCI, Cameron
necessarily be extrapolated to the wrist flexors. Some of these results have been presented in abstract form. 25
electrodes
METHODS Subjects Two groups were examined in this study: able-bodied (AB) men (n = 6) and women (n = 3) between the ages of 18 and 35, and men (n = 14) and women (n = 2) subjects who had sustained injuries to their cervical spinal cord (ie, they were quadriplegic). Background information for the SCI subjects is given in table 1. The age of the SCI subjects was 27.6 _ 4.7 years (mean _+ standard deviation) with a time postinjury of 7.1 + 5.1 years. The level of injury describes the most caudal neurological level that retains some voluntary motor function. The mean age of the control subjects was 27.8 _+ 3.7 years. All procedures were performed identically on both groups. All subjects gave their informed consent to participate in this study.
Procedure Arm Supporting Device. A frame (fig 1) held the forearm and palm in a neutral position for the measurement of isometric force. Attached tO the frame by a hinge was a cuff that was strapped to the hand. A force transducer~ (10kg/mm) attached to and perpendicular to the cuff measured the force produced by the wrist flexor muscles at the palm (ie, approximately 10cm from the axis of rotation of the wrist), while the immobilization of the hand by the cuff reduced the effects of pronation and supination. Instrumentation and Fatigue Test. Subjects sat in a dental chair to which the arm support was affixed. The skin over the left wrist flexors was wiped with an alcohol swab and two stimulating electrodes (Unipatch 615 reusable, 5cm × 5cm) b were applied to the skin. One electrode was placed over the belly of the flexor carpi radialis muscle, approximately 5cm distal to the olecranon process. The second electrode was placed on the dorsal aspect of the wrist. Electrode placement was tested by briefly stimulating (Stayodynamic EMS/plus neuromuscular stimulator; biphasic stimulation at 40Hz total phase; 100#s pulse width) c and examining the movement of the wrist. If the correct movement (flexion) did not occur with a 40mA stimulus intensity, the position of the electrodes was changed slightly Table 1: Background Formation for Spinal Cord Injury Study Subjects Number
Age
Sex
Level of Injury
MMT
Years Postinjury
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
33 23 29 21 26 31 26 28 34 28 23 37 27 19 25 29
M F M M M M M M M M M M M M F M
C4, complete C6, complete C6, complete C5-6, incomplete C5-6, complete C5, incomplete C5, complete C5, complete C6, incomplete C5-6, incomplete C5, incomplete C5-6, incomplete C5-6, complete C5, complete C3-4, complete C5-6, complete
0 1 3 5 1 4 3 1 1 1 5 5 1 0 0 1
19 2 11 4 9 9 4.5 3 18 6.5 3 5 3 2 6 6
Level of injury based on medical evaluation at time of discharge. Subjects are denoted by incomplete with indication of most caudal segmental level within which they have at least some motor function. M M T performed on left wrist flexor muscles only. By chance, SCI population did not include anyone with a M M T score of 2. Abbreviation: MMT, manual muscle test.
Arch Phys Med Rehabil Vol 76, October 1995
arm
restrainer
force t r a n s d u c e r
Fig 1. Experimental arrangement, showing hand positioned within splint and force-recording arrangement. All studies were conducted on the left arm of subjects.
until the desired movement was obtained, after which the subject's arm was fastened into the device. The force signal was bridge-amplified,a displayed on a digital oscilloscope,e and recorded for off-line analysis (Vetter 4000 f and Mitsubishi U32 g videocassette recorder). The fatigue test consisted of 126 trains of 26 stimuli at a frequency of 40Hz. A frequency of 40Hz was choosen to supramaximally activate the wrist flexor group found to be composed of predominantly fast fatigable fibers. Thus, the same total number of stimuli were delivered in this fatigue test as for that first described by Burke and colleagues,26 but the present test required 4 minutes to complete rather than 2. This stimulus protocol resulted in a 1.5-second pause between successive stimulus trains, and was modified because more rapidly applied trains described by Burke resulted in partial fusion of the force profile in subjects with spinal cord injury (see Discussion). Manual Muscle Test. A manual muscle test (MMT) was used to determine the amount of function each spinal cord injured individual had in their left wrist flexor muscles. A registered occupational therapist assigned a grade on a scale of 0 to 5 as follows: 0, no visible or palpable flicker of contraction; 1, visible or palpable flicker of contraction but no movement of the joint; 2, wrist flexion in a gravity-neutral plane; 3, wrist flexion against gravity but not against mild resistance; 4, wrist flexion against mild resistance; and 5, normal muscle strength. Data Analysis. The individual force profiles for each subject were digitized by a personal computer containing an analog to digital board and averaging software, h For the fatigue test, the force profiles of 122 consecutive contractions were acquired for each subject. Averages were made of sweeps 1 to 5 (0), 18 to 22 (20), 38 to 42 (40), 58 to 62 (60), 78 to 82 (80), 98 to 102 (100), and 118 to 122 (120). The following values were determined for each the averaged force records: the maximum amplitude, the rise time, and the time constant of relaxation. The rise time was determined by measuring the time required for the force to increase to 50% of its final (maximum) value. The time constant of relaxation was determined as the time taken for the exponential portion of the relaxation curve to fall to one-third of its value at the end of stimulation. These values were measured manually with digital cursors within the software package, h The same individual made all measurements for all research subjects, and employed the same criteria for all measurements. Between and within group differences were examined using analysis of variance (ANOVA) and post hoc paired t test. RESULTS Figure 2 shows typical averaged force profiles taken during a fatigue test from an AB (panel A) and an SCI (panel B)
FATIGUE PROPERTIES OF CSCI, Cameron
931
100
A
.,_
X
•
Able-bodied
E 0
80
~.
6O
B
1-5
18-22
38-42
58-62
78-82
E
J
98-102
11B-122
Fig 2. Representative force profiles for an able-bodied (A) and a quadriplegic (B) subject during the fatigue test. Each trace is the average of 5 traces taken at equal intervals throughout the test, The calibration marks represent 1 second (x axis) and lkg (y axis),
0 U-
40
I
subject. Averages were made of five repetitions to allow us to study the time course of fatigue. In this example, the maximum force production in the AB subject (panel A) was approximately 2 times that measured from the SCI subject (panel B). Furthermore, the records show that the rate of relaxation remained rapid throughout the fatigue test for the AB subject, whereas relaxation-rate slowed considerably with successive contractions in the SCI subject. Figure 3 illustrates the averaged values of maximum force measured for the AB and SCI groups during the fatigue test. At the outset, the mean force in the AB group was approximately 2.2 times that of the SCI group. To our surprise, this ratio maintained constant throughout much of the fatigue test, as illustrated in figure 4. For this figure, the maximum force measured during the first series of five contractions was assigned a value of 100% for each subject, and all subsequent measures were expressed as a percentage of this starting point. The mean of these normalized values for the AB and SCI groups at each of the periods analyzed is shown in figure 4. The largest discrepancy between the two groups occurred at trial 80, but
6
7 v (b
6 5
E (l) O O LL
~
Able-bodied Quadriplegic
100 '
120
o)
E 500 400
•
Able-bodied Quadriplegic MM0 ~ MM1 ~J MM3/4 /
[]
/ -
o ~T
~,
300 %,-,
O
2
r--
200
t.-. o o (1)
100
1 • - - -
t--
i
0
80
neither these nor any other trials were significantly different between the two groups, as determined by ANOVA. In both groups the maximum averaged force at the end of the fatigue test was seen to decrease to approximately 40% of the initial value. Figure 5 shows the time constant of relaxation for different groups of subjects at different contraction times. Filled symbols represent the overall means and standard deviations of the AB
t1:I X
3
0t
6'0
Fig 4. Normalized force amplitude (percent of maximum force) versus sweep number, Both groups were found to decrease to approximately one half the maximum force by the end of the fatigue test. Able-bodied (n = 7) Quadriplegic (n = 14). Error bars indicate standard error measurements.
E O e •
I
40
Sweep Number
v
8 7
20
i
~
i
i
L
i
20
40
60
80
100
120
Sweep Number Fig 3. Peak force amplitude (mean) versus the sweep number for the two groups. The able-bodied group is shown with the solid circle (n = 7) whereas the quadriplegic group is shown with the solid square (n = 14). Error bars indicate standard error measurements.
0
I
0
I
"20
I
~
I
I
I
40
60
80
100
120
Sweep Number Fig 5. Time constant of force relaxation versus sweep number. The quadriplegic group (n = 14) exhibited a significantly greater relaxation time compared with the able-bodied group (n = 7) by the end of the test (p < .005}, Quadriplegic subjects are grouped by their score on the manual muscle test; MMO (n = 1), MM1 (n = 7), M M 3 / 4 (n = 3), and M M 5 (n = 3). Error bars indicate standard error measurements,
Arch Phys Med Rehabil Vol 76, October 1995
932
FATIGUE PROPERTIES OF CSCI, Cameron
and SCI groups, whereas the open symbols show the means and standard deviations of subsets of the SCI population. At the beginning of the fatigue test, the rate of relaxation in the AB group was faster (ie, smaller time constant) than that of the SCI groups as a whole or any of its subsets. Based on ANOVA, this difference between the averaged AB and SCI groups was found to be significant (F = 26.73, p < .0001). The rate of relaxation decreased with continuing contraction for all groups examined. Between the initial and second series of contractions studied (ie, averages of contractions 1 to 5 and 18 to 22), the slowing of relaxation was approximately constant for all groups. However, clear differences in the slopes became evident in the force profiles for the subsets of SCI subjects (open symbols), where slowing of relaxation was greater as the degree of paralysis increased. For example, by the end of the fatigue test the difference between the initial (81) and ending (439) relaxation rates in the MM 0 group was more than 5 times. By comparison, this difference was less than 3-fold (49.2 vs 133.7) in the AB group. For this latter group, the difference between starting and ending values was significant (F = 5.803; p < .005). For comparison between individual groups as plotted in figure 5, a significant difference was seen in initial and final relaxation rate values based on ANOVA (F = 5.803, p < .005). A post hoc t test comparing the mean values for the AB group with those of the SCI average group showed a significant difference (T = 3.44, p < .005). The time to 50% of maximum force (fig 6) is plotted for individual groups of subjects as a function of the sweep number. Filled symbols represent the overall means of the AB and SCI groups, whereas the open symbols show the means and standard deviations of subsets of the SCI population. AB subjects were found to take a longer time to rise to a 50% of maximum than the SCI group. However, as all the subjects fatigued, their times to 50% of maximum increased. Both the AB and SCI group showed a significant difference in rise-time between the initial
E
360
O 320
Able-bodied Quadriplegic
0 [] V
MM3/4
z ~
|
l / / ~ - O
~4~
o
E m
E
280
X o~
E 240 0
tn 0
200
E I--
160 I
I
F
I
I
[
I
0
20
40
60
80
100
120
Sweep Number Fig 6. Time to 50% maximum force versus the sweep number. The quadriplegic subjects are grouped as a whole (n = 16) and also are subdivided based on their manual muscle test scores. All groups were found to increase the time required to reach 50% maximum by the end of the test. The able-bodied group (n = g) was found to take longer at all times (p < ,005) throughout the test as compared with the quadriplegic group. Error bars indicate standard error measurements.
Arch Phys Med Rehabil Vol 76, October 1995
and final measures of the test (T = -5.47 and T = -5.72, respectively, p < .001), but did not differ from one another by a significant amount. DISCUSSION After neurologically complete injury to the cervical spinal cord, a variety of alterations in muscle fibers of distal upper extremity muscles will occur over different time courses. In the short term, neurologically complete injury typically results in atrophy of these muscles. This atrophy is caused by a combination of: (1) inactivity (or disuse atrophy); and (2) denervation, caused by loss of varying amounts of grey matter, including motoneurons of the ventral h o m J 9 Disuse atrophy may be reversed by activity, either through recovery of supraspinal conduction such that volitional movements are restored, or through the development of spasticity (although the limited duration of these latter contractions will be less effective in reversing atrophy). Denervation atrophy may be reversible to the extent that affected muscle fibers can remain viable until innervation by sprouts from surviving, adjacent motoneuron axons takes place. Following innervation, the same factors related to the reversal of disuse atrophy are in effect. As a group, the maximum force produced by supramaximal stimulation of wrist flexor muscles in SCI subjects at any given point during the fatigue test was approximately half that produced by the able-bodied subjects. Nevertheless, these force levels are high enough for the completion of a variety of functional tasks. The decreased but significant force levels seen in SCI subjects--particularly those with MMT scores of 0 or 1 indicate that: (1) muscles were innervated but typically lacked sufficient supraspinal input to cause useful voluntary contraction; and (2) atrophy of those fibers paralyzed to voluntary input contributed to the decreased forces seen. From this study it is not possible to determine the extent of partial denervation and sprouting that occurred. However, the level of injury of the majority of subjects (see table) coincides closely with recent studies of upper extremity muscles,2° and suggests that denervation and sprouting were probably widespread in the majority (if not all) of the SCI subjects examined. The rate of relaxation at the end of a stimulus train was decreased in SCI subjects compared with controls, and this difference was magnified by the fatigue test. Similar findings have been reported for the tibialis anterior of SCI subjects,24 and may reflect a decreased rate of Ca 2+ reuptake. A practical consequence of this property of the SCI wrist flexors is that repetitive contractions elicited by NMS, either for specific tasks or for exercise, 2 may not allow for complete relaxation before the onset of the next stimulus train if muscles in able-bodied subjects are used to establish the stimulus parameters (ie, time between contractions). In this study we experienced such a problem when applying the fatigue test protocol as described by Burke and colleagues.26 It became immediately apparent with the first SCI subject examined that we would have to modify the protocol to allow a longer period between successive trains. The risk of not allowing the muscle to fully relax between contractions is of overdriving the muscle and causing selective damage of the fatigable fibers, t2 Although this complication from NMS is most likely to occur during training of the relatively untrained muscle, it may also occur during functional procedures involving NMS ambulation and bicycle ergometry. In the present study, the strong positive relationship between increasing relaxation time and degree of paralysis of the wrist flexor muscles suggests that the pattern of use (and disuse) of these muscles was directly responsible for their performance during the fatigue test. It would be of interest to determine whether a protocol of NMS stimulation for forearm flexors in
FATIGUE PROPERTIES OF CSCI, Cameron
SCI subjects, incrementing in duration over a period of weeks, 24 would lead to relaxation trajectories more like those seen in AB subjects. In addition, demonstration of greater M M T scores at the end of such a stimulation program would be evidence supporting the utility of N M S stimulation in muscles under partial supraspinal control (ie, M M T score 1 to 3).
11. 12. 13.
CONCLUSIONS It is r e c o m m e n d e d by the authors that w h e n N M S is applied to relatively untrained muscle, care be taken to allow for the complete relaxation of the target muscle between contractions. The repetitive contractions elicited by many N M S systems, either for functional m o v e m e n t s or for exercise, may not allow for complete relaxation before the onset of the next stimulus train if muscles in able-bodied subjects are used to establish the stimulus parameters (ie, time between contractions). The risk of not allowing the muscle to fully relax between contractions m a y cause overdriving of the muscle and subsequent damage of the more susceptible fatigable fibers. Although this complication from N M S is most likely to occur during training of the relatively untrained muscle, it m a y also occur during functional procedures involving N M S ambulation and bicycle ergometry. The simple but important procedure of increasing the rest time between contractions m a y avoid unnecessary damage to fatigable fibers.
Acknowledgments: The authors thank Belinda Needham-Shropshire for performing the manual muscle tests, Diane Davies for advice regarding statistical analyses, and Jim Broton for providing a critical review of the manuscript.
14. 15. 16. 17. 18. 19. 20. 21. 22.
References 1. Fields RW. Electromyographically triggered electric muscle stimulation for chronic hemiplegia. Arch Phys Med Rehabil 1983; 68:40713. 2. Peckham PH, Mortimer JT. Restoration of hand function in the quadriplegic through electrical stimulation. In: Hambrecht FF, Reswick JB, editors. Functional electrical stimulation, applications in neural protheses. New York: Dekker, 1977:83-95. 3. Handa Y, Handa T, Nakatsuchi Y, Yagi R, Hoshimiya M. A voice controlled functional electrical stimulation system for the paralyzed hand. Jap J Med Electron Biol Eng 1985;23:292-8. 4. Wieler M, Prochazka A, Kenwell Z, Gauthier M, Isaacson G. "Electronic Glove" augments tenodesis-grip and hand-opening in people with quadriplegia. J Am Paraplegia Soc 1994; 17:132. 5. Klose KJ, Needham-Shropshire B, Lebwohl N, Green BA. The results of a 35 SCI subject experience with the Parastep-I walking system. J Am Paraplegia Soc 1994; 17:115. 6. Marsolais EB, Kobetic R. Functional walking in paralyzed patients by means of electrical stimulation. Clin Ortho 1983; 175:30-6. 7. Andrews BJ, Baxendale RM, Barnett RW, Phillips GF, Paul JP, Freeman PA. A hybrid orthosis for paraplegics incorporating feedback control. In: Popovic D, editor. Advances in external control of human extremities X. Belgrade: ETAN, 1987:297-3l 1. 8. Popovic D, Tomovic R, Schwirtlick L. Hybrid assistive system-the motor neuroprosthesis. IEEE Trans Biomed Eng 1989;36:39. 9. Solomonow M, Eldred E, Lyman J, Foster J. Control of muscle contractile force through indirect high-frequency stimulation. Am J Physiol 1983;62:71-82. 10. Naples GG, Mortimer JT, Yuen TGH. Overview of peripheral nerve electrode design and implantation. In: Agnew WF, McCreery DB,
23.
24. 25. 26.
933
editors. Neural prostheses. Fundamental studies. Newark: Prentice Hall, 1990:108-45. Prochazka A. Comparison of natural and artificial control of movement. IEEE Trans Rehab Engin 1993; 1:7-17. Pette D, Vrbova G. Invited Review: Neural control of phenotypic expression in mammalian muscle fibers. Muscle Nerve 1985;8:67689. Gauthier GF, Burke RE, Lowey S, Hobbs AW. Myosin isozymes in normal and cross-reinnervated cat skeletal muscle fibers. J Cell Biol 1983;97:756-71. Buller AJ, Eccles JC, Eccles RM. Differentiation of fast and slow muscles in the cat hind limb. J Physiol (Lond) 1960; 150:399-416. Salmons S, Vrbova G. The influence of activity on some contractile characteristics of mammalian fast and slow muscles. J Physiol (Lond) 1969;201:535-49. Luff AR, Proske V, Webb SN. The transformation of cross-reinnervated slow-twitch muscle after deafferentiation in the cat. Exp Brain Res 1984;55:152-7. Lehman AJR, Tulley FM, Vrbova G, Dimitrijevic MR, Towle JA. Muscle fatigue in some neurological disorders. Muscle Nerve 1989; 12:938-42. Vrbova G, Dimitrijevic MR, Partridge M, Halter J, Verhagen-Metman L. Reversal of increased muscle fatigue in paraplegic patients by electrical stimulation [abstract]. Soc Neurosci 1987; 13:253. Peckham PH, Mortimer JT, Marsolais EB. Upper and lower motor neuron lesions in the upper extremity muscles of tetraplegics. Paraplegia 1976; 14:115-21. Marino RJ, Herbison GJ, Ditunno JFJ. Peripheral sprouting in the zone of partial preservation in motor complete spinal cord injury. J Am Paraplegia Soc 1992; 15:122. Yang JF, Stein RB, Jhamandas J, Gordon T. Motor unit numbers and contractile properties after spinal cord injury. Ann Neurol 1990; 28:496-502. Calancie B, Gordon T, Stein RB, Davis LA, Dolphin B. Contractile properties of thenar motor units in normal and quadriplegic subjects [abstract]. Soc Neurosci 1987; 13:875. Burke RE. Motor units: anatomy, physiology, and functional organization~ In: Brooks VB, editor. Handbook of physiology. The nervous system. Bethesda, MD: American Physiological Society, 1981:345442. Stein RB, Gordon T, Jefferson J, Sharfenberger A, Yang JF, Totosy de Zepetnek J, et al. Optimal stimulation of paralyzed muscle after human spinal cord injury. J Appl Physiol 1992;72:1393-1400. Cameron T, Calancie B. Fatigue properties of wrist flexor muscles in able-bodied and quadriplegic subjects [abstract]. J Am Paraplegia Soc 1990; 13:120. Burke RE, Levine DN, Tsairis P, Zajac FE III. Physiological types and histochemical profiles in motor units of the cat gastrocnemius. J Physiol (Lond) 1973;234:723-48.
Suppliers a. Grass FTI0; Grass, 101 Old Colony Ave., PO Box 51@ Quincy, MA 02269. b. Unipatch, PO Box 271, Wabasha, MN 55981. c. Stayodynamics, 1225 Florida Ave., PO Box 1379, Longmont, CO 80501. d. Measurements Group 2120A; Measurements Group, PO Box 27777, Raleigh, NC 27611. e. Gould 1425; Gould, 774 North Lake Boulevard, no. 1000, Altamonte Springs, FL 32701. f. Vetter, Box 143, Rebersburg, PA 16872. g. Mitsubishi, Denki Building, Marunouchi, Tokyo. h. RC Electronics Computerscop, $386-D Hollister Ave., Santa Barbara, CA 93111.
Arch Phys Med Rehabil Vol 76, October 1995