Muscle atrophy and procedures for training after spinal cord injury.

Spinal cord injuries debilitate thousands of people each year. These injuries are most frequently sustained in automobile accidents; sports activities such as diving, skiing, football, and gymnastics; and, even in peace-time, gunshots. Many of the individuals are young and will depend on external assistance as a substitute for voluntary movement for the rest of their lives. As a result of the considerable advances made in care and rehabilitation in recent years, the life expectancy of the patients who have sustained spinal cord injuries is now only 10% lower than that of nondisabled individuals.[1-3]

Attempts to replace the central activation of muscle contraction by electrical stimulation of the paralyzed muscles have been made in several rehabilitation centers around the world. Functional electrical stimulation (FES), particularly when combined with external bracing, holds considerable promise as a strategy for assisting patients in executing functional movements.[4-8] For people who are quadriplegic, FES has been used to activate hand and ann muscles to assist in the performance of functional movements of the upper extremities.[4-6,9] In patients with a loss of lower-extremity function (paraplegia, incomplete quadriplegia, and hemiplegia), FES has been implemented for standing and walking.[4-6,10-12] In addition, some success has been reported for the use of FES in preventing pressure sores, joint contractures, muscle adhesions, and spinal curvature.[4-6]

Several factors must be considered in the application of FES as a means of effective rehabilitation. These factors include (1) the replacement of the central command by the external control, (2) the strength of the bones in the paralyzed limbs, (3) retention of the range of movements of the joints, and (4) the prevention and reversal of muscle atrophy. The external control is provided by the intelligent devices that activate both agonist and antagonist muscles. Movements such as extending an arm require the temporal and spatial activation of muscles normally coordinated by higher brain centers, which rely in part on the sensory feedback from the moving arm. In the absence of the central control and sensory feedback, the external control must be programmed to achieve the same goal by activating muscles at the appropriate time and sequence in order to execute the desired movements. Effective external control has been discussed in detail in recent reviews.[4,7,8]

After spinal cord injuries, bone breakage due to osteoporosis is a well-recognized problem, particularly if subjects stand and walk with FES.[4,5,12-17] In addition, joint contractures may prevent full extension of hip and knee joints and may limit ankle dorsiflexion, which are prerequisite for standing and walking with FES.[4,12] The effectiveness of exercise and FES regimens in increasing bone strength and range of joint movements remains controversial.[12-17] Muscles supplied by motoneurons in the spinal cord segments at or below the injury site undergo denervation atrophy as a result of damage to their spinal motoneurons or disuse atrophy as a result of damage to central pathways, with subsequent loss of synaptic input to the muscles' spinal motoneurons. Muscles are usually referred to as paralyzed in both cases. Muscle force, measured as joint torque in response to external stimulation, may drop to less than 20% to 30%, as compared with the same torque in nondisabled individuals, as a result of disuse atrophy.[5] No force will be generated when muscles are completely denervated.

This review focuses on four issues: (1) magnitude of muscle atrophy after spinal cord injuries and the distinction between denervation and disuse atrophies, (2) increased susceptibility of paralyzed muscles to fatigue, (3) capacity of surviving motor nerves to sprout and reinnervate denervated muscle fibers in partially denervated muscles, and (4) effective use of FES to increase the strength and endurance of paralyzed muscles.

Muscle Atrophy After Spinal

Cord Injury

Since the pioneering work of Tower,[18] it has generally been accepted that muscles that are paralyzed as a result of spinal cord injuries undergo atrophy and develop less force. Muscle atrophy, a reduction in the size and/or number of muscle fibers, may be present as denervation atrophy or disuse atrophy.[19,20] Denervation atrophy results from injury to motoneurons in the spinal cord or to the motor nerves in the ventral roots through which they exit.[5,21] Disuse atrophy occurs as a result of loss of muscle activation due to disruption to the central and segmental synaptic drive onto the surviving spinal motoneurons.[5,22-24]

Denervation Atrophy

With spinal cord injuries, a number of neurons including the motoneurons in the ventral horn may be fatally damaged, and the ventral and dorsal roots may be traumatized even when the cell bodies are not directly affected. Thus, the segmental trauma may lead to substantial denervation of muscles supplied by motoneurons in the spinal cord segment and by motor nerves that exit the spinal cord through the ventral roots at the level damaged.

The muscles that lose all of their innervation undergo drastic and rapid wasting.[18-21] Generally, the proportion of muscles that suffer complete denervation after spinal cord injuries is Small.[22] However, many muscle fibers that receive their innervation from the affected spinal cord segments will suffer partial denervation as a result of the irreversible damage to their motoneurons. For example, in patients with complete lesions at the C-5 to C-6 levels, the paralyzed thenar muscles lose as much as 50% to 90% of their normal complement of motor innervation.[25] Prevention or reversal of denervation atrophy in these cases will depend on the capacity of the nerves of surviving motoneurons to sprout and reinnervate as many denervated muscle fibers as possible. The greater the sprouting, the better the reinnervation of denervated muscle fibers. As a result, muscle fibers may survive and contract in response to FES to develop sufficient forces to perform functional movements. Because the remaining motor nerves may not always succeed in reinnervating all the denervated muscle fibers, denervation atrophy may still contribute to the weakness of paralyzed muscles that receive their innervation from spinal segments at or near the lesion site.

Some reports have suggested that there may be a loss of motoneurons several segments below a spinal cord lesion in humans; the loss has been attributed to transsynaptic degeneration of motoneurons.[26,27] The issue is not fully settled, however, because several studies[28-30] have shown that the number of surviving motoneurons is not significantly reduced. One study[26] demonstrated a 20% reduction in the number of motoneurons. Because the remaining nerves sprout and reinnervate the denervated muscle fibers, denervation atrophy is unlikely to contribute to wasting of muscles that receive their innervation from spinal segments below the lesion site.

Disuse Atrophy

Muscle wasting after spinal cord injury is generally attributed to the muscle inactivity that ensues after loss of the synaptic inputs from higher centers and from spinal cord segments to spinal motoneurons.[18,21] Studies to date, however, suggest that much of the disuse atrophy of the paralyzed muscles should be attributed to concurrent changes in muscle length or loading conditions, rather than decline in neuromuscular activity.[23,24] The magnitude of disuse atrophy varies widely from study to study in both human and animals after spinal cord lesions but does not necessarily correlate with the decline in neuromuscular activity.[23,24,31-42] Neural activity that results in neuromuscular activity is generally reduced after spinal cord lesions but varies considerably depending on the type of lesion and the level of spasticity.[23,24,28,37,38]

Disuse atrophy is more pronounced in paralyzed muscles that normally bear weight, especially those that cross single joints.[23,24,31-35] These muscles often contain a large proportion of slow fatigue-resistant muscle fibers, which are largely responsible for maintaining posture and bearing weight.[19,20] For example, the soleus muscle, a postural muscle that extends the ankle, undergoes significant atrophy. In contrast, the atrophy may be negligible in other muscles in the lower limb that do not bear weight or that cross more than one joint. For example, the tibialis anterior (TA) muscle, which flexes the ankle and does not normally contract against resistance, does not atrophy as much as the soleus muscle in a number of species, including humans.(23,24) The medial gastrocnemius (MG) muscle, which crosses both the knee and ankle joints, also undergoes less atrophy than the soleus muscle even though it is a synergist to the soleus muscle. The preferential atrophy of soleus muscles of the cat is illustrated in Figure 1. The ability of paralyzed MG muscles of the cat to develop as much tetanic force as normal muscles is illustrated in Figure 2.

The same principles of preferential atrophy of weight-bearing muscles apply to humans. Non-weight-bearing muscles demonstrate little atrophy when paralyzed.[33-35] For example, in patients with complete C-5 to C-6 lesions, the paralyzed thenar[25] and TA[28] muscles developed isometric forces very similar to those in nondisabled individuals. In contrast, the quadriceps femoris muscle, which normally lifts the lower limb by extending the knee, shows significant atrophy after spinal cord injuries.[5,36]

A similar pattern of atrophy of limb muscles is seen after space flight, hind-limb suspension, limb immobilization, or tenotomy, conditions in which muscles undergo shortening contractions that are not resisted by a normal load.[23,24,43-47] These findings also suggest that changes in loading or length of paralyzed muscles after spinal cord lesions are responsible, at least in part, for the atrophy that occurs.

The most severe disuse atrophy occurs in unloaded muscles that are immobilized at a shortened length[45,46] or tenotomized.[47] Muscle fiber degeneration is particularly widespread in tenotomized muscles that undergo unopposed shortening contractions.[48,49] The detrimental effects of unopposed shortening contractions must be considered when FES regimens are developed.

Fatigue in Paralyzed Muscles

The ability of muscles to sustain force over time depends on their fiber type composition, their metabolic profile, and the general nutritional and cardiovascular state of the organism. Slow-twitch muscles contain mainly slow oxidative fibers, which do not fatigue readily. Fast-twitch muscles contain a small proportion of the slow fibers and mostly fast fibers, which vary in their oxidative and glycolytic enzyme profiles and their corresponding susceptibility to fatigue.[23,24,50-52] Fast fatigue-resistant units contain fibers with high oxidative and low glycolytic enzyme activities; fast fatigable units have low oxidative capacity and high glycolytic enzyme activities.[53,54] In the cat MG muscle, for example, tetanic force declines to 33% of initial values during repetitive activity (Fig. 3); the remaining force generated by the fatigued muscle corresponds well with the proportion of the total tetanic force that is generated by the slow fatigue and fast fatigue-resistant motor units in the MG muscle (Fig. 4).

In patients or animal models of spinal cord injuries, the capacity of paralyzed muscles to sustain contractions is dramatically reduced.[5,23,24,28,36,55] This effect of spinal cord injury on muscle endurance[39,40] is illustrated in an animal model of spinal cord injury in Fig. 4. Within 4 minutes of repetitive activity, the tetanic force of paralyzed muscles declines to 3% of initial values as compared with 33% in the contralateral control muscles. The increased susceptibility to fatigue is accounted for by a reduced number of fatigue-resistant motor units in the paralyzed muscles (Fig. 5), which, in turn, reflects a reduction in oxidative capacity of the muscle fibers.[23,24,35,41,42]

Disuse atrophy and low endurance in paralyzed muscles in patients with spinal cord injuries makes effective and safe use of FES problematic unless the muscles are adequately prepared by some training protocols and used under the commands of the appropriate external control.[5] For example, the fatigable quadriceps femoris muscle must be adequately trained in order for the patient to stand up safely with a control strategy that will reduce the duration of stimulation.[8]

Sprouting in Partially

Denervated Muscles

Reversal of denervation atrophy in a partially denervated muscle depends on how many motoneurons survive the spinal cord injury and their ability to increase the number of muscle fibers that they supply by sprouting. Sprouting occurs from the terminal regions of the intramuscular nerve branches and serves to reinnervate denervated muscle fibers that lie nearby.[56,57] Normally, motoneurons innervate hundreds or even thousands of muscle fibers. The motoneuron and its muscle fibers form a motor unit.[50] As a result of sprouting, each motoneuron supplies an increased number of muscle fibers, and activation of the motor unit generates more force than normal.

In animal experiments in which the number of muscle fibers per motoneuron or the motor unit force, or both, were recorded, the results showed that motoneurons can supply up to 5 times as many muscle fibers as they normally do (Fig. 5).[58-66] The details of these studies are described elsewhere.[64-68] In humans, results of single motor unit recordings show that motoneurons that have survived cervical spinal cord injuries have the ability to reinnervate denervated muscle fibers by sprouting.[25,65] The methods used to estimate the number of surviving motoneurons in a muscle and increased motor unit size in human subjects have been described in detail.[69-71]

For injuries in which at least 15% of the motor supply to a partially denervated muscle is left intact, the number of muscle fibers per motoneuron and motor unit force increase in direct proportion to the extent of denervation.[63,65] In these cases, all denervated muscle fibers are reinnervated, and the partially denervated muscle recovers completely from denervation atrophy. However, for lesions in which more than 85% of the motoneurons to a muscle are fatally injured, sprouting does not fully compensate for the loss. Motoneurons appear to be limited to a fivefold increase in the number of muscle fibers they Supply.[64,65] The underlying mechanism of this phenomenon is not yet understood.

Evidence suggests that the limit in the number of muscle fibers per motoneuron is not set by the motoneuron itself but rather by physical constraints within the partially denervated muscle that limit the distance over which the sprouts can grow to reach denervated muscle fibers.[63-66] Normally, the muscle fibers of a single motor unit are distributed in a discrete area, and fibers belonging to different motor units are interspersed (Fig. 5C). In partially denervated muscles, the size of the unit territories do not increase significantly, but an increasing number of muscle fibers are incorporated in each territory (Fig. 5D). These observations indicate that motor nerve sprouts tend to reinnervate muscle fibers in close proximity. Under conditions of extensive denervation, nerve sprouts are therefore unlikely to grow far enough to reinnervate the many denervated fibers that are outside their territories.[64,65] This limitation may also account for the inability of nerve sprouts from one muscle to reinnervate adjacent denervated muscles. Muscles that lose all of their innervation are not reinnervated by nerve sprouts belonging to surviving motoneurons that supply other muscles.

The studies discussed show that nerve sprouts from intramuscular nerve branches of surviving motoneurons reinnervate denervated muscle fibers and thereby reduce denervation atrophy. When at least 15% of the normal complement of motor nerves remain, nerve sprouting is a very effective mechanism that compensates for the loss of innervation of many of the muscle fibers.[64,65] As a result, denervation atrophy may be fully reversed in partially denervated muscles,[64] and thus the muscles can be activated by FES to generate functional movements.

Muscle Training for

Functional Electrical

Stimulation

Low-frequency (eg, 20-Hz) electrical stimulation via implanted or skin surface electrodes at the motor point has been used in several rehabilitation centers to train muscles prior to application of FES to generate movement.[5,12,15,17,28,36,72] Stimulation has been used to counteract disuse atrophy and low endurance of paralyzed muscles in order for FES to be applied successfully and safely for performance of functional movements.

Training methods vary among centers and for different muscles, particularly with respect to the duration of stimulation, the loading of contracting muscles, and the muscle type. A short stimulation duration and muscle loading regimen is exemplified in a situation in which the subject rides an exercise bicycle that provides loading to the quadriceps femoris muscle while receiving 30 minutes of electrical stimulation daily in three weekly sessions.[72] In other regimens, electrical stimulation is applied for longer durations (eg, 1-2 hours per day) to the same muscles but with less loading.[5] The duration of stimulation may be as long as 8 hours on a daily basis for muscles in the upper extremities,[36] whereas 2-hour sessions are effective in increasing the endurance of lower-limb muscles without external loading.[28] Outcome measures are difficult to compare from study to study because the variables for stimulation are so different. As a result, comparisons of the efficacy of different training methods have been difficult.

Low-frequency stimulation paradigms that increase muscle endurance have had different effects on muscle strength, depending on conditions of stimulation. The results of many animal studies in which muscles were stimulated at low frequencies (10-20 Hz) for 2 or more hours per day show that an increase in muscle endurance and oxidative capacity is often accompanied by an undesirable decrease in muscle fiber size and muscle force.[66,73-79] For example, endurance of the MG muscle of the cat was dramatically increased by 20-Hz stimulation for 12 hours per day, but the muscle's peak tetanic force was significantly reduced (Fig. 6). These results are consistent with the normal reciprocal relationship between endurance and strength in normal motor units.[23,50,80-82] In human subjects, daily sessions or three sessions per week of low-frequency stimulation increased muscle endurance and strength when the stimulated muscles contracted against set resistances.[15,36,72,83] Muscle Strength did not increase significantly under conditions in which the stimulated muscles did not contract against a load (Fig. 7).[5,28]

Several factors may influence muscle strength and should be considered in designing training regimens, determining their efficacy, and interpreting data derived from many of the animal and human studies. These factors include (1) muscle length, (2) muscle activity, (3) muscle loading, (4) muscle type and function, (5) species, and (6) the interaction of these factors.[23,24]

Length

Goldspink[84,85] showed that both innervated and denervated muscles pinned at long length underwent substantially less atrophy initially than muscles fixed at short length. Others have confirmed and extended these original findings. Generally, there is an inverse relationship between the initial degree of atrophy and the amount of stretch imposed on an immobilized muscle.[23,24,40,86] Lengthening increases protein synthesis and decreases protein degradation; the reverse is true for muscles fixed in a shortened position. These underlying alterations of protein turnover are responsible for the hypertrophy and atrophy of lengthened and shortened muscles, respectively.[87] Similar patterns of alterations are found in bone metabolism.[88]

Activity

Salmons and Vrbova[89] first demonstrated that increased activity could convert fast-twitch muscles to slow-twitch muscles. Their work has been confirmed many times.[23,80,81,90-92] Using the cat model described in Figures 1 and 2, in which the spinal cord is hemisected and the ipsilateral hind limb is deafferented, Kernell and colleagues[74-80] were able to greatly reduce the spontaneous nerve activity and study in detail the effects of stimulation on muscle properties. Increased amount of stimulation to the cat peroneus longus muscle induced an increase in fatigue resistance, but a decrease in muscle strength.[74-78] Superimposing a brief period of high-frequency stimulation (100 Hz for 0.5% of the day) on continuous low-frequency stimulation (10 Hz for 5% of the day) prevented most of this loss of strength.[76] Kernell and colleagues[75] suggested that the "force stress" provided by the high-frequency tetanic burst favored the maintenance of factors of relevance to contractile force. Recent attempts to reproduce these findings in human subjects, however, have not been successful (DB Popovic, unpublished observations).

Loading

It is generally accepted[90-96] that to build strength a muscle must generate close to maximal forces for short periods of time, whereas to improve endurance a muscle must maintain small forces for long periods of time (eg, sprinting versus long-distance running). Thus, isometric or even eccentric forceful contractions may be necessary for building strength, whereas concentric contractions against light loads may increase endurance. These concepts, taken from sports medicine literature on athletes and nonathletes and from animal studies of exercise and muscle overload, have been applied to the training of paralyzed muscles using FES,[5,72,97] but their application still requires rigorous testing.

Muscle Type and Function

As previously described, a general finding in both patients with spinal cord injuries and animal models is the more severe atrophy of extensor muscles, especially slow-twitch muscles that cross a single joint.[23,24,31-34,98] These muscles are largely responsible for maintaining posture and bearing weight.[99] In patients, the quadriceps femoris muscle (a knee extensor) undergoes more significant atrophy than the TA muscle (a fast-twitch ankle flexor). The strength of the quadriceps femoris muscle does improve after retraining,[5] and the strength of the TA muscle changes very little.[28]

Species

The speed at which muscles contract varies widely among species. For the same muscle in the species commonly studied, contractile speed will generally follow the following descending order: mouse [right arrow] rat [right arrow] cat [right arrow] human.[100] Differences in muscle contractile speed are associated with differences in the proportions of fast and slow muscle fibers in the same muscles across species. In addition, differences in muscle strength and endurance are associated with the differences in the fiber type composition. Thus, one cannot assume that data from experiments on rodents, for example, will automatically apply to human subjects. Thus, parallel experiments in both human and nonhuman muscles may provide insight into preparation of muscles for FES regimens. Differences in the resting length of paralyzed muscles in patients and animals must also be taken into account in assessing efficacy of training with FES. For example, patients with spinal cord injuries sitting in wheelchairs much of the day usually keep their ankles at approximately a right angle, whereas a spinalized cat will fully extend the paralyzed hind limb by dragging it. Thus, species differences must be taken into account in comparisons of the outcome of stimulation protocols.

Interactions

Although some of the factors discussed earlier have been individually studied, little is known about the interaction among them. For example, a muscle contracting at a long length would produce a greater load because of the passive length-tension properties of muscle. Unloaded muscles lose weight and are less forceful even if they are exercised[101] or stimulated for 20 minutes per day[102]; the muscles are only able to maintain weight and force if they are passively stretched by changing joint angle.[103,104] Also, physiological extensors that maintain body posture against gravity are activated more frequently than flexors.[102] Thus, it is not surprising that spinalization has a much more drastic impact on the activity of extensors with correspondingly greater changes in their fiber type proportions than that of the flexors. Shortening contractions in unloaded muscles will induce fatigue more rapidly than isometric contractions because more oxygen is consumed (Fenn effect[105]). Thus, changes in muscle properties in any case could be due to the combined effects of load, length and phosphate metabolism.

Conclusions

Muscles undergo atrophy after spinal cord injury. In muscles that receive their innervation from spinal cord segments at or close to the lesion site, denervation atrophy may be extensive. Nerve sprouting from the intramuscular branches of the surviving motoneurons is very effective in reducing the atrophy in partially (but not completely) denervated muscles. Muscles that are paralyzed by loss of central and segmental input to their motoneurons and not by denervation undergo disuse atrophy. Weight-bearing muscles, particularly those that cross a single joint, are most susceptible. Several factors are likely to contribute to disuse atrophy including changes in muscle length, loading, and activity, all of which vary with muscle type and function. These factors should be systematically controlled in the development and use of electrical stimulation protocols for preparing muscles prior to the application of FES to elicit functional movements.

References

[1] Green EA, Eismont FJ, O'Heir JT. Prehospital management of spinal cord injuries. Paraplegia. 1987;5:229-238. [2] Bedbrook GM. The development and care of spinal cord paralysis. Paraplegia. 1987;25: 172-184. [3] Stauffer ES. Long-term management of traumatic quadriplegia. In: Pierce DS, Nickel VH, eds. The Total Care of Spinal Cord Injuries Boston, Mass: Little, Brown & Co Inc; 1978:81-102. [4] Stein RB, Peckham PH, Popovic' DB. Neural Prostheses: Replacing Motor Function After Disease or Disability. New York, NY: Oxford University Press Inc: 1992. [5] Kralj A, Bajd T. Functional Electrical Stimulation: Standing and Walking After Spinal Cord Injury. Boca Raton, Fla: CRC Press Inc; 1989. [6] Peckham PH. Functional electrical stimulation: current status and future prospects of applications to neuromuscular system in spinal cord injury. Paraplegia. 1987;25:279-288. [7] Prochazka A. Comparison of natural and artificial control of movement. IEEE Trans Rehabil Eng. In press. [8] Stein RB, Belanger M, Wheeler G, et al. Assessment of electrical systems for improving locomotion after incomplete spinal cord injury. Arch Phys Med Rehabil. In press. [9] Hambrecht FT. A brief history of neural prostheses for motor control of paralysed extremities. In: Stein RB, Peckham PH, Popovic DB, eds. Neural Prostheses Replacing Motor Function After Disease or Disability. New York, NY: Oxford University Press Inc; 1992:3-14. [10] Yarkony GM, Roth EJ, Cybulski GR, Jaeger RJ. Neuromuscular stimulation in spinal cord injury, II: prevention of secondary complications. Arch Phys Med Rehabil. 1992;73:195-200. [11] Phillips CA, Hendershot DM. A systems approach to medically prescribed functional electrical stimulation: ambulation after spinal cord injury. Paraplegia. 1991;29:505-513. [12] Popovic DB, Watt J, Stein RB, et al. Functional electrical stimulation in the rehabilitation of the spinal cord injured child. In: Proceedings of the International Conference of the IEEE Engineering in Biology and Medicine Society, Orlando, Fla, 1991. 1991;13:938-939. [13] Comarr AE, Hutchinson RH, Bors E. Extremity fractures of patients with spinal cord injury. Am J Surg. 1962;103:732-739. [14] Ragnarsson KT, Sell GH. Lower extremity fractures after spinal cord injury: a retrospective study. Arch Phys Med Rehabil. 1981;62:418-423. [15] Belanger M, Wheeler GD, Wieler M, et al. Resistance training and functional electrical stimulation in preparation for standing and walking in complete spinal cord injured patients. J Biomech. In press. [16] Leeds EM, Klose KJ, Ganz W, et al. Bone mineral density after bicycle ergometry training. Arch Phys Med Rehabil. 1990;71:207-209. [17] Rodgers MM, Glaser RM, Figoni SF, et al. Musculoskeletal responses of spinal cord injured individuals to functional neuromuscular stimulation-induced knee extension exercise training. J Rehabil Res Dev. 1991;28:19-26. [18] Tower SS. Function and structure in the chronically isolated lumbosacral spinal cord of the dog. J Comp Neurol, 1937;67:109-131. [19] McComas AJ. Neuromuscular Functional and Disorders. London, England: Butterworth & Co (Publishers) Ltd; 1977. [20] Vrbova G, Gordon T, Jones R. Nerve-Muscle Interaction. London, England: Chapman & Hall Ltd. In press. [21] Solandt DY, Magladery JW. A comparison of effects of upper and lower motor neurone lesions on skeletal muscle. J Neurophysiol. 1942;5:373-380. [22] Peckham PH, Mortimer JT, Marsolais EB. Upper and lower motor neuron lesions in the upper extremity muscles of tetraplegics. Paraplegia. 1976;14:115-121. [23] Gordon T, Pattullo MC. Plasticity of muscle fiber and motor unit types. Exerc Sport Sci Rev. 1993;21:1331-362. [24] Roy RR, Baldwin KM, Edgerton VR. The plasticity of skeletal muscle: effects of neuromuscular activity. Exerc Sports Sci Rev. 1991;19: 269-312. [25] Yang JF, Stein RB, Jhamandas J, Gordon T. Motor unit numbers and contractile properties after spinal cord injury. Ann Neurol. 1990;28: 496-502. [26] Hunter J, Ashby P, Secondary changes in segmental neurons below a spinal cord lesion in man. Arch Phys Med Rehabil. 1984;65:702-705. [27] Eidelberg E, Nguyen H, Polich R, Walden JG. Transsynaptic degeneration of motoneurones caudal to spinal cord lesions. Brain Res Bull. 1989;22:39-45. [28] Stein RB, Gordon T, Jefferson J, et al. Optimal stimulation of paralysed muscle in spinal cord injured patients, J Appl Physiol. 1992;72: 1393-1400. [29] Brandstater ME, Dinsdale SM. Electrophysiological studies in the assessment of spinal cord lesions Arch Phys Med Rehabil. 1984;57: 70-74. [30] McComas AJ, Sica REP, Upton AR, Aguilera N. Functional changes in motoneurones in hemiparetic patients. J Neurol Neurosurg Psychiatry, 1973;35:183-193. [31] Roy RR, Acosta R Jr. Fiber type fiber size changes in selected thigh muscles six months after low thoracic spinal cord transection in adult cats: exercise effects. Exp Neurol, 1986; 92:675-685. [32] West SP, Roy RR, Edgerton VR. Fiber type and fiber size of cat ankle, knee, and hip extensors and flexors following low thoracic spinal cord transection at an early age. Exp Neurol 1986;91:174-182 [33] Grimby G, Broberg C, Krotkiewska I, Krotkiewski M. Muscle fiber composition in patients with traumatic cord lesion. Scand J Rebabil Med. 1976;8:37-42, [34] Scelsi R, Marchetti C, Poggi P, et al. Muscle fiber type morphology and distribution in paraplegic patients with traumatic cord lesion: histochemical and ultrastructural aspects of rectus remoris muscle. Arch Neuropathol (Berl). 1982;57:243-248. [35] Martin TP, Stein RB, Hoepner PT Reid DR. The influence of functional electrical stimulation on the morphological metabolic properties of paralyzed muscle. J Appl Physiot. 1992; 72:1401-1406. [36] Peckham PH, Mortimer JT, Marsolais EB. Alteration in the force and fatigability of skeletal murcle in quadriplegic humans following exercise induced by chronic elcctrical stimulation. Clin Orthop. 1976;114:326-333. [37] Alaimo AM, Smith JL, Roy R, Edgerton VR. Electromyographic activity of slow and fast ankle extensors following spinal cord transection. J Appl physiol. 1984;56:1608-1613. [38] Lovely RG, Gregor RJ, Roy RR, Edgerton VR. Effects of training on the recovery of fullweight-bearing stepping in the adult spinal cat. Exp Neurol. 1986;92:421-435. [39] Cope TC, Bodine SC, Fournier M, Edgerton VR. Soleus motor units in chronic spinal transected cats: physiological and morphological alterations. J Neurophysiol. 1986;55:1202-1220. [40] Lieber RL, Johansson CB, Vahlsing HL, et al. Long-term effects of spinal cord transection on fast and slow rat skeletal muscle, I: contractile properties. Exp Neurol. 1986;91:423-434. [41] Lieber RL, Friden JO, Hargens R, Feringa EP, Long-term effects of spinal cord transection on fast and slow rat skeletal muscle, II: morphometric properties. Exp Neurol. 1986;91: 435-448. [42] Pierotti DJ, Roy RR, Bodine-Fowler SC, et al. Mechanical and morphological properties of chronically inactive cat tibialis anterior motor units. J Physiol (Lond). 1991;444:175-192. [43] Alford EK, Roy RR, Hodgson JA, Edgerton VR. Mectromyography of rat soleus, medial gastrocnemius, and tibialis anterior during hind-limb suspension. Exp Neurol. 1989;96: 635-649. [44] Winiarski AM, Roy RR, Alford EK, et al. Mechanical properties of rat skeletal muscle after hind limb suspension. Exp Neurol, 1987;96: 650-M. [45] Baker JH, Matsumoto DE. Adaptation of skeletal muscle to immobilization in a shortened position. Muscle Nerve. 1988;11:231-344. [46] Pachter BR, Eberstein A. Neuromuscular plasticity following limb immobilization. J Neurocytol. 1984;13:1013-1025. [47] Baker JH. Segmental necrosis in tenotomized muscle fibres. Muscle Nerve. 1983;6:29-39. [48] McMinn RMH, Vrbova G. Motoneurone activity as a cause of degeneration in the soleus of the rabbit. Q J Exp Physiol. 1967;52: 411-415. [49] Pattullo MC. The Role of Activity in the Determination of Skeletal Muscle Properties Aberdeen, Scotland: University of Aberdeen; 1989. PhD thesis. [50] Burke RE. Motor units: anatomy, physiology, and functional organization. In: Brookhart JM, Mountcastle VB, eds. Handbook of Physiology: The Nervous System. Bethesda, Md: American Physiological Society, 1981;2:345-422. [51] Sargeant AJ, Kernell D, eds. Neuromuscular Fatigue Amsterdarn, the Netherlands: Elsevier Science Publishers BV; 1993. [52] Enoka RM. Muscle strength and its development: new perspectives. Sports Med. 1988;6: 146-168. [53] Burke RE, Levine DN, Tsairis P, Zajac FE. Physiological types and histochemical profiles in motor units of the cat gastrocnemius. J Physiol (Lond). 1973;234:723-748. [54] Totosy de Zepetnek JE, Zung HV, Erdebil S, Gordon T. Motor unit categorization on the basis of contractile histochemical properties: a glycogen depletion analysis of normal and reinnervated rat tibialis anterior muscle. J neurophysiol. 1992;67:1404-1415. [55] Miller RG, Green AT, Moussavi RS, et al. Excessive muscular fatigue in patients with spastic paraparesis. Neurology. 1990;40:1271-1274. [56] Brown MC, Holland RL, Hopkins WG. Motor nerve sprouting. Annu Rev Neurosci. 1981; 4:17-42. [57] Wernig A, Herrarra AA. Sprouting and remodelling at the nerve-muscle junction. Prog Neurobiol 1986;27:251-291. [58] Kugelberg E, Edstrom L, Abbruzzese M. Mapping of motor units in experimentally reinnervated rat muscle. J Neurol Neurosurg Psychiatry. 1970;33:319-329. [59] Thomson W, Jansen JKS. The extent of sprouting of remaining motor units in partly denervated immature and adult rat soleus muscle. Neuroscience. 1977;2:523-535. [60] Brown MC, Ironton R. Sprouting and regression of neuromuscular synapses in partially denervated mammalian muscles. J Physiol (Lond). 1978;278:325-348. [61] Gorio A, Carmignoto G, Finesso M, et al. Muscle reinnervation, II: sprouting, synapse formation and repression. Neuroscience. 1983; 8:403-416. [62] Fisher TJ, Vrbova G, Wijetunge A. Partial denervation of the rat soleus muscle at two different developmental stages Neuroscience. 1989;28:755-763. [63] Rafuse VF, Gordon T, Orozco R. Proportional sprouting in partially denervated triceps surae muscles in the cat. J Neurophysiol. 1992; 68:1261-1276. [64] Gordon T, Totosy de Zepetnek JE, Rafuse V, Erdebil S. Motoneuronal branching and motor unit size after complete and partial nerve injuries. In: Wernig E, ed. Motoneuronal Plasticity. Berlin, Federal Republic of Germany: Springer-Verlag; 1991:207-216. [65] Gordon T, Yang JF, Ayer K, et al. Recovery potential of the neuromuscular system after nerve injuries. Brain Res Bull, 1993;30:477-482. [66] Rafuse VF. Regulation of Motor Unit Size and Contractile Properties After Nerve Injuries Edmonton, Alberta, Canada: University of Alberta; 1993. PhD thesis. [67] Totosy de Zepetnek JE, Zung HV, Erdebil S, Gordon T. Innervation ratio is an important determinant of force in normal and reinnervated rat tibialis anterior muscle, J Neurophysiol. 1992; 67:1385-1403. [68] Totosy de Zepetnek JE, Zung HV, Gordon T, Stein RB. Comparison of force EMG measures in normal and reinnervated tibialis anterior muscles of the rat. Can J Physiol Pharmacol. 1991;69:1774-1783. [69] Stein RB, Yang JF. Methods for estimating the number of motor units in human muscles. Ann Neurol. 1990;28:487495, [70] McComas AJ. Motor unit estimation: methods, results, and present status. Muscle Nerve. 1991;14:585-597. Invited review. [71] McComas AJ, Galea V, de Bruin H. Motor unit populations in healthy and diseased muscles. Phys Ther. 1993;73:868-.877. [72] Petrofsky JS, Phillips CA. Active physical therapy: a modern approach for rehabilitation of the disabled. J Neurol Orthop Surg. 1983;4: 165-173. [73] Pette D, Muller W, Leisner E, Vrbova G. Time-dependent effects on contractile properties, fibre population, myosin light chains and enzymes of energy metabolism in intermittendy and continuously stimulated fast twitch muscles of the rabbit. Pflugers Arch. 1976;364: 103-112, [74] Eerbeek O, Kernell D, Verhey BA. Effects of fast and slow patterns of tonic long-term stimulation on contractile properties of fast muscle in cat. J Physiol (Lond). 1984;352:73-90. [75] Donselaar Y, Eerbeek O, Kernell D, Verhey BA. Fibre sizes and histochemical staining characteristics in normal and chronically stimulated fast muscle of cat. J Physiot (Lond). 1987;382:237-254. [76] Kernell D, Eerbeek O, Verhey BA, Donselaar Y. Effects of physiological amounts of high- and low-rate chronic stimulation on fast-twitch muscle of the cat hindlimb, I: speed force-related properties. J Neurophysiol. 1987; 58:598-613. [77] Kernell D, Donselaar Y, Eerbeek O. Effects of physiological amounts of high- and low-rate chronic stimulation on fast-twitch muscle of the cat hindlimb, II: endurance-related properties. J neurophysiol. 1987;58:614-627. [78] Kernell D, Eerbeek O. Physiological effects of different patterns of chronic stimulation on muscle properties. In: Rose F, Jones P, eds. Neuromuscular Stimulation. New York, NY: Dem6os Publications; 1989:193-200. [79] Gordon T, Pattullo MC, Rafuse VF. Motor unit heterogeneity with respect to speed and fatiguability in cat muscles after chronic stimulation or paralysis. In: Sargeant AJ, Kernell D, eds. Neuromuscular Fatigue. Amsterdam, the Netherlands: Elsevier Science Publishers BV; 1993:63-66. [80] Pette D, Vrbova G. Neural control of phenotypic expression in mammalian muscle fiber. Muscle Nerve. 1985;8:676-689. Invited review. [81] Pette D, Vrbova G. Adaptation of mammalian skeletal muscle to chronic electrical stimulation. Rev Physiol Biochem Pharmacol. 1992; 120:115-202. [82] Kernell D. Organized variability in the neuromuscular system: a survey of task-related adaptations. Arch Ital Biol. 1992; 130:19-66. [83] Bald T, Kralj A, Turk R, et al. Use of functional electrical stimulation in the rehabilitation of patients with incomplete spinal cord injuries. J Biomed Eng. 1989;11:96-102. [84] Goldspink DF. The influence of immobilization and stretch on protein turnover of rat skeletal muscle. J Physiol (Lond). 1977;264: 267-282. [85] Goldspink DF. The influence of passive stretch on the growth and protein turnover of the denervated extensor digitorum longus. Biochem J 1978;174:595-602. [86] Spector SA, Simard CP, Fournier M, et al. Architectural alterations of rat hind-limb muscles immobilized at different lengths. Exp Neurol. 1982;76:94-110. [87] Williams PE, Goldspink G. The effect of immobilization on the longitudinal growth of striated muscle fibres. J Anat. 1973;116:45-55. 88 Leblanc A, Marsh CL, Evans H, et al. Bone and muscle atrophy with suspension of the rat. J Appl physiol. 1985;58:166-167. [89] Salmons S, Vrbova G. The influence of activity on some contractile characteristics of mammalian fast slow muscles. J Physiol (Lond). 1969;201:535-549. [90] Salmons S, Henriksson J. The adaptive response of skeletal muscle to increased use. Muscle Nene. 1981;4:94-105. [91] Edstrom L, Grimby L. Effect of exercise on the motor unit. Muscle Nerve. 1986;9:104-126. [92] Saltin B, Gollnick PD. Skeletal muscle adaptability: significance for metabolism and performance. In: Peachey ID, ed. Handbook of Physiology. Bethesda, Md: American Physiological Society; 1983:539-554. [93] Sale DG, MacDougall JD, Jacobs I, Garner S. Interaction between concurrent strength and endurance training. J Appl Physiol. 1990; 68:260-270. [94] Narici MV, Roi GS, Landoni L, et al. Changes in force, cross-sectional area and neural activation during strength training and detraining of the human quadriceps. Eur J Appl Physiol 1989;59:310-319. [95] Delitto A, Brown M, Strube MJ, et al. Electrical stimulation of quadriceps femoris in an elite weight lifter: a single-subject experiment. Int J Sports Med. 1989;10:187-191. [96] Lieber RL. Skeletal Muscle Structure and Function: Implications for Rehabilitation and Sports Medicine. 1992; Baltimore, Md: Williams & Wilkins; 1992. [97] Mao J, Tyreman N, Stein RB, Gordon T. Use of functional electrical stimulation to enhance paralyzed cat hindlimb muscles. Neuroscience Abstracts. In press. [98] Lieber RL, Friden JO, Hargens R, et al. Differential response of the dog quadriceps muscle to external skeletal fixation of the knee. Muscle Nerve. 1988;11:193-201. [99] Denny-Brown D. On the nature of postural reflexes. Proc R Soc Lond [Biol]. 1929;104:252-301. [100] Close RI. Dynamic properties of mammalian skeletal muscles. Physiol Rev 1972;52:129-197. [101] Herbison GJ, Talbot JM. Muscle atrophy during space flight: research needs and opportunities. Physiol". 1985;28:520-527. [102] Gardiner PF, Lapointe MA. Daily in vivo neuromuscular stimulation effects on immobilized rat hindlimb muscles. J Appl Physiol. 1988;53:960.-966. [103] Goldspink DF, Goldspink G. The role of passive stretch in retarding muscle atrophy. In: Nix WA, Vrbova G, eds. Electrical Stimulation and Neuromuscular Disorders. Berlin, Federal Republic of Germany: Springer-Verlag; 1975: 91-100. [104] Jaspers SR, Fagan JM, Sararug S, et al. Effects of immobilization on rat hind limb muscles under non-weight-beating conditions. Muscle Nerve. 1988;11:458-466. [105] Fenn WO. A quantitative comparison between the energy liberated and the work performed by the isolated sartorius muscle of the frog. J Physiol (Lond). 1923;58:175-203.