Acta Physiol Scand 1998, 162, 275±283

Neural control in human muscle fatigue: changes in muscle afferents, moto neurones and moto cortical drive S. C. GANDEVIA Prince of Wales Medical Research Institute and University of New South Wales, Sydney, Australia ABSTRACT To understand the neural factors which contribute to fatigue, it is not satisfactory to regard fatigue as occurring only when a task can no longer be performed. Changes in muscle afferent feedback, moto neuronal discharge, moto cortical output, and perceived effort develop well before an endurance limit in limb muscles. During sustained maximal contractions the discharge of moto neurones declines, commonly to below the level required to produce maximal force from the muscle whose contractile speed is usually slowed. Thus, some `central' fatigue develops. Recent findings using transcranial stimulation have revealed that the moto cortex is one site at which suboptimal output develops during human muscle fatigue. There is a need to study the reflex effects on moto neurones and the excitability of the moto cortex in experimental animals, as well as to apply rigorous methods to assess these processes in voluntary exercise in human subjects. Keywords central fatigue, cortical stimulation, fatigue, inhibition, moto neurones, moto cortex, muscle spindles, re¯exes. Received 2 April 1997, accepted 16 June 1997

Fatigue is a time-dependent exercise-induced reduction in the maximal force generating capacity of a muscle. During a maximal voluntary contraction fatigue begins rapidly. It arises more slowly for weaker contractions but occurs even if a submaximal voluntary force is sustained. Manifestations of fatigue re¯ecting largely peripheral factors include the reduction in maximal force and power and also the slowing of muscle relaxation which often accompanies sustained high-intensity effort. However, the central nervous system is involved not only in driving moto neurones but also in the increased tremor of the exercising limb, the recruitment of muscles initially uninvolved in the task and the subjective increase in effort. Some of these `central' features may disrupt performance more than the reduction in maximal muscle force. The increased perceived `effort', a signal requiring the moto cortex, re¯ects the need to recruit more moto neurones and muscles, and to drive them harder (for review see Jones 1995, Gandevia 1996). However, the interaction between re¯ex inputs, moto neuronal properties and supraspinal factors during fatigue is not well understood (for review see Enoka & Stuart 1992, Fuglevand 1996). There is a question as to what extent the available moto neurones are recruited in fatigue. Studies using

sensitive forms of twitch interpolation (Hales & Gandevia 1988) have not con®rmed the original prediction of Merton (1954) that moto units are fully recruited and discharging at rates suf®cient for fusion at the onset of maximal voluntary isometric contractions. For several muscles in which identical methods have been used voluntary activation is usually above 85±90% during attempted maximal voluntary isometric contractions (including for the diaphragm) (e.g. Allen et al. 1993, Allen et al. 1995, Herbert & Gandevia 1996, for review see Gandevia et al. 1995). Averaged across one fatiguing `maximal' contraction in several subjects, or several contractions in one subject, voluntary activation declines signi®cantly. This is de®ned as `central fatigue' (Gandevia et al. 1995). Its development should contribute to the decline in discharge rate of moto neurones during fatigue. For maximal voluntary isometric contractions the moto neurone discharge rates after the initial peak force are about 20±50 Hz for upper limb muscles with variation between subjects and muscles (e.g. Bellemare et al. 1983). If the voluntary contraction continues for more than a few seconds, discharge rates decline and approach a plateau towards 60 s (e.g. Grimby et al. 1981, Bigland-Ritchie et al. 1983a, Gandevia et al. 1990).

Correspondence: Professor S. C. Gandevia, Prince of Wales Medical Research Institute, High Street, Randwick, Sydney, NSW Australia 2031. Ó 1998 Scandinavian Physiological Society

275

Neural control and muscle fatigue á S C Gandevia

This decline in rate assists the output from the muscle because relaxation times for whole muscle usually lengthen during strong isometric contractions and thus lower discharge frequencies produce the same fraction of maximal force (Bigland-Ritchie et al. 1983b). The decline in moto unit rate has been called muscle `wisdom' (Marsden et al. 1983) but given that the ®ring may fall too low to maintain full activation of the muscle and that muscle relaxation rate sometimes increases in fatigue (e.g. McKenzie & Gandevia 1991), the term is not appropriate. More irregular ®ring of moto neurones develops with fatigue (Gandevia et al. 1990, Garland et al. 1994). In fatiguing anisometric contractions moto units show variable discharge patterns (Miller et al. 1996). An extreme example of the predominance of central factors in fatigue occurs when a task cannot be continued (`task failure') but the muscles can produce the required force when stimulated electrically. This possibility was recognized by Mosso, Waller, Lombard and others last century and a recent report supports the proposition: when subjects could no longer voluntarily plantar¯ex the ankle at 30% maximum, the force could be maintained by electrical stimulation of the nerve innervating the plantar¯exors (Loescher et al. 1996a). This phenomenon is likely to depend on the subjects, the degree of maximal voluntary drive for the muscle group, the peripheral muscle properties, and the intensity of the fatiguing task. In this brief review I consider the effects of feedback from muscle and the role of supraspinal sites on the changing behaviour of moto neurones during fatigue. Much of the data derive from isometric exercise in which electrophysiological measures can be made without interference from movement. De®ciencies in techniques used to assess human muscle fatigue are mentioned.

Acta Physiol Scand 1998, 162, 275±283

M U S CL E A F F E RE N T I N P UT S D U R I N G M U S CL E F A T I G U E During maximal voluntary contractions of a human muscle, the input from specialized muscle mechanoreceptors (innervated by large-diameter afferents) and from chemosensors and nociceptors (innervated by small-diameter afferents) will change. Based largely on data from animal studies, the probable changes in muscle afferent input are shown in Figure 1. Muscle spindle afferents During isometric contractions the fusimotor system recruits human muscle spindle endings but their discharge frequency declines with time if the contraction is sustained beyond 1±2 s and fatigue develops (e.g. Mace®eld et al. 1991). Neglecting the effect of presynaptic inhibition, spindle-mediated facilitation of moto neurones will decline during a sustained isometric contraction, i.e. disfacilitation develops. However, in concentric contractions the spindle afferents' role will differ because fusimotor drive may be insuf®cient to overcome muscle shortening and spindle afferents in the contracting muscle may fall silent (Burke et al. 1978, Al Falahe et al. 1990). In eccentric contractions spindle input may be accentuated as animal studies indicate that spindle afferents increase their responsiveness to stretch during fatigue (e.g. Nelson & Hutton 1985, Hayward et al. 1991). Golgi tendon organ afferents Tendon organs respond to active muscle forces generated by the moto units which insert into them (Jami 1992, cf. Hulliger et al. 1995). However, their discharge adapts during the ®rst seconds of force production. Thereafter, the ensemble input from tendon organs is

Figure 1 Schematic representation of the

likely changes in muscle afferent input during a sustained maximal voluntary contraction of a limb muscle. Data compiled mostly from studies in animals.

276

Ó 1998 Scandinavian Physiological Society

Acta Physiol Scand 1998, 162, 275±283

well suited to signal force produced by the muscle. The stretch sensitivity of individual tendon organs varies with fatigue (Hutton & Nelson 1986, Thompson et al. 1990, Hayward et al. 1991). The practical consequences of changes in stretch sensitivity of muscle spindle and tendon organ afferents during muscle fatigue are uncertain, particularly as their re¯ex effects may be attenuated at a spinal level, and because the effects are small compared with those in the more numerous group III and IV afferents (Hayward et al. 1991). Non-spindle group II and groups III and IV muscle afferents During muscle fatigue the discharge of small-diameter muscle afferents (groups III and IV) increases according to the temperature, chemical and the mechanical environment of their free nerve endings (e.g. Paintal 1960, Kniffki et al. 1978, Kaufman et al. 1983, Hayward et al. 1991). They have little or no background discharge and, because they are so numerous, small changes in their discharge will produce massive increases in their ensemble input to the central nervous system (CNS). Following fatiguing contractions most group II nonspindle afferents and group III mechanosensitive afferents have increased discharge rates, increased sensitivity to stretch and palpation but reduced responsiveness to additional contractions (Hayward et al. 1991). Group III afferents increase their discharge by about 1 Hz within the ®rst 10 s of an isometric contraction but this is not sustained (Sinoway et al. 1993). Metabolites produced by prolonged muscle contraction reduce the mechanical thresholds of group III and IV afferents (e.g. Rotto et al. 1988). Some group III afferents discharge during slow locomotion in decerebrate cats (Pickar et al. 1994), but there are no data on the discharge of human group III and IV afferents in voluntary contractions. Some recordings in animals may be of limited guidance because contractions are produced by electrical stimulation without distributing it in a more natural asynchronous way to moto units. C HA N G E S A T TH E MO T O N E U RO N AL L E VE L D UR I N G M US CL E FA T I G UE Intrinsic moto neuronal properties, re¯ex inhibition and disfacilitation, Renshaw cell inhibition, and insuf®cient drive from supraspinal sites may all contribute to the decline in moto unit ®ring rate in fatigue. Theoretically, it would be optimal if each moto neurone had its output individually controlled to produce the highest force from its muscle ®bres depending on their contractile state. However, a compartmentalized re¯ex machinery which modulates each moto neurone's output based on local re¯ex inputs from group III and IV afferents surrounding its muscle ®bres is highly unÓ 1998 Scandinavian Physiological Society

S C Gandevia á Neural control and muscle fatigue

likely. Systematic changes in the intrinsic properties of moto neurones with increasing threshold and size (Kernell & Monster 1982, Binder & Mendell 1990), along with nonhomogeneous distribution of re¯ex inputs within the moto neurone pool, will aid optimization of the whole pool's output. An intrinsic property of moto neurones which could mediate the decline in moto neuronal discharge rates during fatigue is that ®ring rates adapt when moto neurones are stimulated with intracellular or extracellular current (e.g. Kernell & Monster 1982, Spielmann et al. 1993, Sawcuk et al. 1995). The reduction in ®ring rate (`late adaptation') is appropriately more pronounced for type F than type S moto units. Additional changes develop with repeated bursts of stimulation (Spielmann et al. 1993). The precise pattern of moto neuronal discharge can in¯uence the force produced (see Bevan et al. 1992), but whether the observed increase in irregularity of ®ring actually increases force production in fatigue is unclear. In intact preparations, moto neuronal behaviour during fatigue will be modulated by: (i) supraspinal commands to interneurones and moto neurones; (ii) control by brainstem serotonergic pathways of the `gain' and possible bistable behaviour of moto neurones (e.g. Hultborn & Kiehn 1992); (iii) classical re¯ex and recurrent inputs to a- and c-moto neurones; (iv) presynaptic modulation speci®c classes of re¯ex inputs to moto neurones (e.g. Jankowska 1992). Some of these factors are shown in Figure 2. To determine how the whole system will behave during fatigue is daunting because of the many inputs to the moto neurone pool, their combined monosynaptic and nonmonosynaptic effects on moto neurones, and their changes in size with time and with fatigue. Some re¯ex effects on moto neurones are considered brie¯y below (see Windhorst & Boorman 1995). The connection between Ia afferents and the moto neurone is well studied but its behaviour during fatigue is not established. Indirect evidence reveals that muscle afferents provide signi®cant excitation to the moto neurone pool (up to 30%) during sustained fatiguing isometric contractions (e.g. Gandevia et al. 1990, Mace®eld et al. 1993). However, presynaptic inhibition will attenuate the re¯ex effectiveness of any Ia input in some movements, at least in the lower limb (e.g. Capaday & Stein 1987, Meunier & Pierrot-Deseilligny 1989). Effective transmitter depletion at the Ia-moto neurone synapse (homosynaptic depression) may also develop (e.g. Curtis & Eccles 1960, Bongiovanni & Hagbarth 1990, Lev-tov & Pinco 1992). As indicated earlier, muscle spindle discharge decreases during fatiguing static contractions in human subjects (Mace®eld et al. 1991), irrespective of any re¯ex excitation of fusimoto neurones (e.g. Ljubisavljevic 277

Neural control and muscle fatigue á S C Gandevia

Acta Physiol Scand 1998, 162, 275±283

Figure 2 A simpli®ed representation of some re¯ex pathways affecting a- and c-moto neurones. Renshaw cells and presynaptic inhibition are included. Filled cells are inhibitory. No afferent input reaches moto neurones exclusively by a monosynaptic path and all interneurones in the re¯ex circuits are probably in¯uenced by descending commands.

et al. 1992, Ljubisavljenc & Anastasijevic 1994, Ljubisavljevic et al. 1995). It is not established whether late increases in fusimotor ®ring in the cat are relevant to voluntary fatigue of human muscles. As the unloading re¯ex seems reduced in fatigue of human ®nger extensor muscles, any excitation of homonymous fusimotor neurones is insuf®cient to maintain the prefatigue levels of spindle-mediated facilitation. Any attenuation of `stretch' re¯exes will reduce the ability of spinal re¯exes to oppose length changes (e.g. Hagbarth et al. 1995). However, methodological dif®culties abound in the interpretation of studies of human re¯exes in fatigue (see below). Golgi tendon organ afferents are traditionally believed to re¯exly reduce moto neurone output (Jami 1992). The picture is more complex in fatigue because of spindle afferent convergence on Ib inhibitory interneurones (see Jankowska 1992), such that Ib effects can be excitatory (e.g. Pratt 1995). In a sustained contraction, the Ib inhibitory postsynaptic potentials in moto neurones decline so that the gain of force feedback is reduced (Zytnicki et al. 1990, see also Kirsch & Rymer 1987). While there seems no doubt that inputs from smalldiameter muscle afferents re¯exly alter moto outputs, the way it is achieved in fatiguing voluntary contractions is unresolved. The inputs will act at spinal and supraspinal sites. In the anaesthetized cat the spinal machinery exists to mediate inhibition among synergist muscles during fatigue (Hayward et al. 1988). Many have implied that these afferents exert their inhibitory effects at a spinal level (e.g. Bigland-Ritchie et al. 1986, see also Garland & McComas 1990) but the evidence is circumstantial. The decline in moto neurone discharge in strong contractions is present when muscles contract voluntarily after fatigue produced by electrical stimulation (Garland & McComas 1990), but absent when 278

muscle afferents are blocked by local anaesthesia (Gandevia et al. 1990, Mace®eld et al. 1993). Additional inhibition of moto neurones derives from Renshaw cells which receive excitatory inputs from moto neurones, particularly those of high threshold (Hultborn et al. 1988) as well as descending and peripheral inputs. In the cat the rapid discharge of Renshaw cells declines nonlinearly when moto neurone frequency declines so that their output can contribute to the adaptation of moto neurone rates during fatigue (Windhorst & Boorman 1995). Studies using indirect H-re¯ex methods in humans suggest that recurrent inhibition (and/or prolongation of the after-hyperpolarization) increases in maximal efforts (Kukulka et al. 1986), but declines in submaximal ones (Loescher et al. 1996b). M O TO R CO R T I CA L A N D S U PR A S PI N A L F A CT O R S I N M US CL E FA T I G U E In fatiguing exercise, there are changes in central enkephalinergic, dopaminergic and serotonergic systems (e.g. Hoffman et al. 1990, Bailey et al. 1993, Persson et al. 1993, cf. van Hall et al. 1995). These presumably control vigilance and motivation, pain and its tolerance, while other neuroendocrine changes alter availability of substrates for muscle contraction. Manipulation of these factors should alter centrally mediated components of fatigue. A challenge is to isolate direct roles for these systems in both central fatigue and subsequent task failure. Given the behavioural changes that accompany muscle fatigue (including changes in subjective effort, attention and pain), there must be many associated supraspinal changes measurable at an electrophysiological and biochemical level, but, as for events at a spinal level, it is dif®cult to determine which are secÓ 1998 Scandinavian Physiological Society

Acta Physiol Scand 1998, 162, 275±283

ondary to peripheral fatigue and which contribute to central fatigue and eventual task failure. As indicated earlier, intrinsic moto neuronal, spinal and supraspinal factors could all generate central fatigue. However, we must examine the roles the moto cortex and supraspinal drive play in the decline in moto neurone ®ring frequency in fatigue. Particularly in humans, cortical output monosynaptically excites most spinal moto nuclei (e.g. Porter & Lemon 1993) without the effects of presynaptic inhibition (Nielsen & Petersen 1994). Hence supraspinal centres which alter corticomoto neuronal output will affect moto neurones in a most direct way. Some recent studies have focused on the development of changes in the corticospinal system during fatigue. In monkeys trained to fatigue their elbow ¯exors there were variable changes in the discharge of corticomotor neuronal cells (Maton et al. 1994). In humans performing repetitive submaximal contractions, the size and area of electromyographic (EMG) responses to transcranial magnetic stimulation of the cortex decline after cessation of exercise (Brasil Neto et al. 1993, McKay et al. 1995, Zanette et al. 1995). We recently examined the changes in the EMG and force responses to transcranial stimulation of the motor cortex during sustained maximal voluntary isometric contractions of the elbow ¯exors lasting 1±3 min (Figure 3). Voluntary activation was monitored during exercise with twitch interpolation from the moto point and by measurement of force twitches evoked by transcranial stimulation (Gandevia et al. 1996, Taylor et al. 1996). The silent period in the EMG following high-intensity cortical stimulation lengthened by about 50 ms from its control value (see also McKay et al. 1996) within 30 s of a maximal voluntary contraction or when

S C Gandevia á Neural control and muscle fatigue

an initially submaximal contraction later required maximal effort. The short-latency excitatory response to cortical stimulation appeared to increase in maximal contractions by about 50% (Figure 3a). One factor which could underlie this is the recruitment of additional corticospinal cells, not simply those with a projection to the elbow ¯exors, but others with a divergent projection to several arm muscles including the elbow ¯exors (Mills and Thomson 1995). If present, such progressive spread of recruitment would lead to the contraction of additional related muscles during the task, a phenomenon commonly observed with strong isometric efforts (Figure 4b). Three pieces of evidence suggest that these changes were focal cortical events. Firstly, the changes could not be prevented by vibration of the muscles designed to make up for diminished spindle input. Secondly, if the sustained maximal contraction involved one muscle group, the cortical `silent period' lengthened only in that group and not in nearby muscles. Finally, the growth of the excitatory response to cortical stimulation was not present when the corticospinal axons were stimulated distal to the cortex at the cervicomedullary junction. Despite their robustness, changes in moto cortical excitability probed by arti®cial stimulation do not directly affect the central fatigue during sustained efforts. First, these changes in cortical excitability recovered within 30 s, even when the contracting elbow ¯exors were maintained ischaemic, a manoeuvre assumed to maintain the chemically mediated input via group III and IV afferents (see above). However, during ischaemia maintained after a sustained maximal contraction with a sphygmomanometer cuff, central fatigue persisted. Thus, moto cortical stimulation at the start of

Figure 3 (a) Biceps brachii EMG during maximal

voluntary contractions (MVCs) of elbow ¯exors. Upper ®ve traces during brief control MVCs. Middle traces during a sustained 2 min MVC and lower traces during recovery. The ®rst vertical dashed line shows time of the transcranial magnetic stimulation. Second vertical dotted line shows the time of return of continuous EMG and marks the end of the silent period in control trials. This lengthens during the sustained MVC and recovers quickly. The moto evoked potential (MEP, just right of dashed line) is variable but grows during the sustained contraction and recovers rapidly. (From Taylor et al. 1996.) (b) Force during the sustained MVC shown in panel (a). Arrows indicates the time of a cortical stimulus. Corresponding force increments are shown sequentially in a raster below. (Panel (b), unpublished records from Gandevia et al. 1996. Ó 1998 Scandinavian Physiological Society

279

Neural control and muscle fatigue á S C Gandevia

Acta Physiol Scand 1998, 162, 275±283 Figure 4 Schematic representation of sites at

which changes occur during maximal isometric contractions. Filled cells are inhibitory. (a) (1) indicates apparently increased excitability of the moto cortex, (2) increased intracortical inhibition responsible for silent period prolongation, and (3) the corticospinal synapses and a-moto neurones. (b) Extended view of moto cortex to show the progressive recruitment of additional corticospinal outputs to elbow ¯exor muscles and related muscles during fatigue, that is, from the `central' region to the peripheral region within the box enclosing moto cortical output cells. This would lead to progressive contraction of `synergist' muscles during fatigue.

the maximal contraction increased ongoing voluntary force by about 1%, but as central fatigue evolved, the same cortical stimulus produced progressively more force (Figure 3b). Hence, the voluntary moto output became progressively less than optimal. The evoked force increments were followed immediately by a decrement: its duration increased with that of the silent period but its amplitude diminished due to the slowing in muscle relaxation rate. The presence of cortically evoked `twitches' late in the prolonged contractions show that no site distal to the moto cortex was working at its absolute maximum. The most parsimonious explanation for the development of central fatigue is not that it re¯ects changes in moto cortical excitability, but that drive effectively `upstream' of cortical output declined (such that it could be augmented by superimposed arti®cial stimulation). This reduction in drive may be mediated by supraspinal effects of group III and IV afferents. If so, a crucial question is not why moto neurones fail to ®re fast enough but why the moto cortex does not work as hard as it could. The failure of moto cortical drive to remain suf®cient during fatigue raises questions about the role of the cortical and subcortical areas projecting to the primary moto cortex and other cortical areas with direct corticospinal projections. T E CH N I C AL LI MI T A T I O N S TO T H E S T UD Y O F HU MA N MU S C LE F AT I G U E Experimental methods are no better than their weakest link. Because it is dif®cult to make the `cleanest' measurements to deduce changes in neural control during strong contractions, experimenters have often accepted conditions which are not ideal. Three examples are given to highlight problems that occur in measurements of force, EMG, and moto neuronal `excitability'. Any conclusions about central control of muscle during 280

fatigue must survive rigorous assessment of the methods on which they are based. (i) Twitch responses to electrical stimulation are used to monitor peripheral fatigue. The twitch is very susceptible to the `history' of contraction with fatigue usually reducing the twitch more than tetanic force (e.g. Edwards et al. 1977). If submaximal stimulus intensities are used (e.g. with knee extensors) the same moto units are not tested with each stimulus. Furthermore, with fatigue, such a stimulus will activate fewer motor axons because of activity-dependent reductions in axonal excitability (Bergmans 1973). Hence, for force measurements in fatigue stimulus intensities must be supramaximal. (ii) Levels of EMG are commonly used as surrogate measurements of neural drive to the fatiguing muscle. Problems arise because the size of the intracellular action potential is not constant with fatigue (Sandercock et al. 1985): maximal compound muscle action potentials initially grow and then decline during a sustained contraction (Hicks et al. 1989, see also Fuglevand et al. 1993). The latter change re¯ects slowing in the conduction velocity of muscle ®bres. Furthermore, surface EMG will not necessarily measure `drive' to one muscle as it is easily contaminated by activity in nearby muscles. (iii) Tests of moto neuronal excitability can only be conducted indirectly with H-re¯exes and stretch and unloading responses. None of these re¯exes is purely monosynaptic in humans (Burke et al. 1984) and re¯ex size must therefore depend on the activity of interneurones. These re¯exes could only provide valid tests of moto neurone behaviour if exactly the same afferent volley were evoked and all other re¯ex and descending inputs were unaltered. These conditions are not easily achieved. Ideally the re¯ex behaviour of single moto units should be tested, as has been done for ®ring properties during fatigue (e.g. Enoka et al. 1989, GarÓ 1998 Scandinavian Physiological Society

Acta Physiol Scand 1998, 162, 275±283

land et al. 1994). Finally, any apparent change in re¯ex behaviour must be interpreted in the light of the known reduction in moto neurone ®ring rate during sustained isometric contractions. C O N CL U S I O N S With intense brief exercise, force is likely to depend largely on muscle and biomechanical factors. If maximal exercise continues beyond about 10 s, central factors including de®cient drive `upstream' of moto cortical output attenuate performance and eventually stop it. If contractions are suf®ciently intense, the threshold of the moto cortex appears reduced for both inhibitory and excitatory processes. There is a need to study in experimental animals the neural processes controlling activity of moto neurones and moto cortical neurones, and to apply rigorous methods to assess these processes in voluntary exercise in humans. The author's laboratory is supported by the National Health and Medical Research Council and the Asthma Foundation of New South Wales. Comments on the manuscript by Janet Taylor, Jane Butler and Gabrielle Allen are gratefully acknowledged. I regret that space limitations prevented citation of additional references.

R E FE RE N CE S Ê .B. 1990. Response Al-Falahe, N.A., Nagaoka, M. & Vallbo, A pro®les of human muscle afferents during active ®nger movements. Brain 113, 325±346. Allen, G.M, Gandevia, S.C. & McKenzie, D.K. 1995. Reliability of measurements of muscle strength and voluntary activation using twitch interpolation. Muscle Nerve 18, 593±600. Allen, G.M., McKenzie, D.K., Gandevia, S.C. & Bass, S. 1993. Reduced voluntary drive to breathe in asthmatic subjects. Resp Physiol 93, 29±40. Bailey, S.P., Davis, J.M., Ahlborn, E.N. 1993. Neuroendocrine and substrate responses to altered brain 5-HT activity during prolonged exercise to fatigue. J Appl Physiol 74, 3006±3012. Bellemare, R., Woods, J.J., Johansson, R. & Bigland-Ritchie, B. 1983. Motor-unit discharge rates in maximal voluntary contractions of three human muscles. J Neurophysiol 50, 1380±1392. Bergmans, J. 1973. Physiological observations on single human nerve ®bres. In: J.E. Desmedt (Ed) New Developments in Electromyography and Clinical Neurophysiology, Vol. 2, pp 89±127. Karger, Basel. Bevan, L., Laouris, Y., Reinking, R.M. & Stuart, D.G. 1992. The effect of the stimulation pattern on the fatigue of single moto units in adult cats. J Physiol (Lond) 449, 85±108. Bigland-Ritchie, B., Dawson, N.J., Johansson, R.S. & Lippold, O.C. 1986. Re¯ex origin for the slowing of motoneurone ®ring rates in fatigue of human voluntary contractions. J Physiol (Lond) 379, 451±459. Ó 1998 Scandinavian Physiological Society

S C Gandevia á Neural control and muscle fatigue

Bigland-Ritchie, B., Johansson, R., Lippold, O.C.J., Smith, S. & Woods, J.J. 1983a. Changes in motoneurone ®ring rates during sustained maximal voluntary contractions. J Physiol (Lond) 340, 335±346. Bigland-Ritchie, B., Johansson, R., Lippold, O.C. & Woods, J.J. 1983b. Contractile speed and EMG changes during fatigue of sustained maximal voluntary contractions. J Neurophysiol 50, 313±324. Binder, M.D. & Mendell, L.M. (Eds). 1990. The Segmental Motor System. Oxford University Press, New York. Bongiovanni, L.G. & Hagbarth, K.-E. 1990. Tonic vibration re¯exes elicited during fatigue from maximal voluntary contractions in man. J Physiol (Lond) 423, 1±14. Brasil-Neto, J.P., Pascual-Leone, A., Valls-SoleÂ, J., Cammarota, A., Cohen, L.G. & Hallet, M. 1993. Post exercise depression of moto evoked potentials: a measure of central nervous system fatigue. Exp Brain Res 93, 181±184. Burke, D., Hagbarth, K.-E. & LoÈfstedt, L. 1978. Muscle spindle activity in man during shortening and lengthening contractions. J Physiol (Lond) 277, 131±142. Burke, D., Gandevia, S.C. & McKeon, B. 1984. Monosynaptic and oligosynaptic contributions to human ankle jerk and H-re¯ex. J Neurophysiol 52, 435±448. Capaday, C. & Stein, R.B. 1987. Difference in the amplitude of the human soleus H re¯ex during walking and running. J Physiol (Lond) 392, 513±522. Curtis, D.R. & Eccles, J.C. 1960. Synaptic action during and after repetitive stimulation. J Physiol (Lond) 150, 374±398. Edwards, R.H., Hill, D.K., Jones, D.A. & Merton, P.A. 1977. Fatigue of long duration in human skeletal muscle after exercise. J Physiol (Lond) 272, 769±778. Enoka, R.M., Robinson, G.A. & Kossev, A.R. 1989. Task and fatigue effects on low-threshold moto units in a human hand muscle. J Neurophysiol 62, 1344±1359. Enoka, R.M. & Stuart, D.G. 1992. Neurobiology of muscle fatigue. J Appl Physiol 72, 1631±1648. Fuglevand, A.J. 1996. Neural aspects of fatigue. Neuroscientist 2, 203±206. Fuglevand, A.J., Zackowski, K.M., Huey, K.A. & Enoka, R.M. 1993. Impairment of neuromuscular propagation during human fatiguing contractions at submaximal forces. J Physiol (Lond) 460, 549±572. Gandevia, S.C. 1996. Kinaesthesia: roles for afferent signals and moto commands. In: Handbook of Physiology Section 12, pp 128±172. American Physiological Society. Gandevia, S.C., Mace®eld, G., Burke, D. & McKenzie, D.K. 1990. Voluntary activation of human moto axons in the absence of muscle afferent feedback. Brain 113, 1563±1581. Gandevia S.C., Allen G.M., Butter J.-E., Taylor J.L. 1996. Supraspinal factors in human muscle fatiques. Evidence for subspinal output from the motor cortex. J Physiol (Lond) 490, 529±536. Gandevia, S.C., Allen, G.M. & McKenzie, D.K. 1995. Central fatigue: critical issues, quanti®cation and practical implications. Adv Exp Med Biol 384, 281±294. Garland, S.J., Enoka, R.M., Serrano, L.P. & Robinson, G.A. 1994. Behavior of moto units in human biceps brachii during a submaximal fatiguing contraction. J Appl Physiol 76, 2411±2419. 281

Neural control and muscle fatigue á S C Gandevia

Garland, S.J. & McComas, A.J. 1990. Re¯ex inhibition of human soleus muscle during human muscle fatigue. J Physiol (Lond) 401, 547±558. Grimby, L., Hannerz, J. & Hedman, B. 1981. The fatigue and voluntary discharge properties of single moto units in man. J Physiol (Lond) 316, 545±554. Hagbarth, K.-E., Bongiovanni, L.G. & Nordin, M. 1995. Reduced servo-control of fatigued human ®nger extensor and ¯exor muscles. J Physiol (Lond) 485, 865±872. Hales, J.P. & Gandevia, S.C. 1988. Assessment of maximal voluntary contraction with twitch interpolation: an instrument to measure twitch responses. J Neurosci Meth 25, 97±102. Hayward, L., Breitbach, D. & Rymer, W.Z. 1988. Increased inhibitory effects on close synergists during muscle fatigue in the decerebrate cat. Brain Res 440, 199±203. Hayward, L., Wesselmann, U. & Rymer, Z. 1991. Effects of muscle fatigue on mechanically sensitive afferents of slow conduction velocity in the cat triceps surae. J Neurophysiol 65, 360±370. Herbert, R.D. & Gandevia, S.C. 1996. Muscle activation in unilateral and bilateral efforts assessed by moto nerve and cortical stimulation. J Appl Physiol 80, 1351±1356. Hicks, A., Fenton, J., Garner, S. & McComas, A.J. 1989. M wave potentiation during and after muscle activity. J Appl Physiol 66, 2606±2610. Hoffman, P., Terenius, L. & ThoreÂn, P. 1990. Cerebrospinal ¯uid immunoreactive b-endorphin concentration is increased by voluntary exercise in the spontaneously hypertensive rat. Regul Pept 28, 233±239. Hulliger, M., SjoÈlander, P., Windhorst, U.R. & Otten, E. 1995. Force coding by populations of cat golgi tendon organ afferents. In: A. Taylor, M.H. Gladden & R. Durbaba (Eds) Alpha and Gamma Motor Systems, pp. 302±308. Plenum Press, New York. Hultborn, H., & Kiehn, O. 1992. Neuromodulation of vertebrate moto neuron membrane properties. Curr Opin Neurobiol 2, 770±775. Hultborn, H., Lipski, J., Mackel, R. & Wigstrom, H. 1988. Distribution of recurrent inhibition within a moto nucleus. I. Contribution from slow and fast moto units to the excitation of Renshaw cells. Acta Physiol Scand 134, 347±361. Hutton, R.S. & Nelson, D.L. 1986. Stretch sensitivity of Golgi tendon organs in fatigued gastrocnemius muscle. Med Sci Sports Exerc 18, 69±74. Jami, L. 1992. Golgi tendon organs in mammalian skeletal muscle: functional properties and central actions. Physiol Rev 72, 623±666. Jankowska, E. 1992. Interneuronal relay in spinal pathways from proprioceptors. Progr Neurobiol 38, 335±387. Jones, L.A. 1995. The senses of effort and force during fatiguing contractions. In: Gandevia, S.C., Enoka, R.M., McComas, A.J., Stuart, D.G. & Thomas, C.K (Eds) Fatigue: Neural and Muscular Mechanisms, Plenum Press, New York. Adv Exp Med Biol 384, pp 305±313. Kaufman, M.P., Longhurst, J.C., Rybicki, K.J., Wallach, J.H. & Mitchell, J.H. 1983. Effects of static muscular contraction on impulse activity of groups III and IV afferents in cats. J Appl Physiol 55, 105±112.

282

Acta Physiol Scand 1998, 162, 275±283

Kernell, D. & Monster, A.W. 1982. Time course and properties of late adaptation in spinal motoneurones of the cat. Exp Brain Res 46, 191±196. Kirsch, R.F. & Rymer, W.Z. 1987. Neural compensation for muscular fatigue evidence for signi®cant force regulation in man. J Neurophysiol 57, 1893±1910. Kniffki, K.D., Mense, S. & Schmidt, R.F. 1978. Responses of group IV afferent units from skeletal muscle to stretch, contraction and chemical stimuli. Exp Brain Res 31, 511±522. Kukulka, C.G., Moore, M.A. & Russell, A.G. 1986. Changes in human a-motoneuron excitability during sustained maximum isometric contractions. Neurosci Lett 68, 327±333. Lev-Tov, A. & Pinco, M. 1992. In vitro studies of prolonged synaptic depression in the neonatal rat spinal cord. J Physiol (Lond) 447, 149±169. Ljubisavljevic, M., Jovanovic, K. & Anastasijevic, R. 1992. Changes in discharge rate of fusimoto neurones provoked by fatiguing contractions of cat triceps surae muscles. J Physiol (Lond) 445, 499±513. Ljubisavljevic, M. & Anastasijevic, R. 1994. Fusimotoinduced changes in muscle spindle out¯ow and responsiveness in muscle fatigue in decerebrate cats. Neurosci 63, 339±348. Ljubisavljevic, M., Anastasijevic, R. & Trifunjagic, D. 1995. Changes in fusimoto discharge rate provoked by isotonic fatiguing muscle contractions in decerebrate cats. Brain Res 673, 126±132. LoÈescher, W.N., Cresswell, A.G. & Thorstensson, A. 1996a. Central fatigue during a long-lasting submaximal contraction of the triceps surae. Exp Brain Res 108, 305±314. LoÈescher, W.N., Cresswell, A.G. & Thorstensson, A. 1996b. Recurrent inhibition of soleus alpha-motoneurons during a sustained submaximal plantar ¯exion. Electroencephalogr clin Neurophysiol 101, 334±338. Mace®eld, V.G., Hagbarth, K.-E., Gorman, R., Gandevia, S.C. & Burke, D. 1991. Decline in spindle support to a motoneurones during sustained voluntary contractions. J Physiol (Lond) 440, 497±512. Mace®eld, V.G., Gandevia, S.C., Bigland-Ritchie, B., Gorman, R.B. & Burke, D. 1993. The ®ring rates of human motoneurones voluntarily activated in the absence of muscle afferent feedback. J Physiol (Lond) 471, 429±443. Marsden, C.D., Meadows, J.C. & Merton, P.A. 1983. `Muscular wisdom' that minimized fatigue during prolonged effort in man: peak rates of motoneuron discharge and slowing of discharge during fatigue. Adv Neurol 39, 169±211. Maton, B., Fourment, A. & Belhaj-Saif, A. 1994. Changes in precentral moto cortical cells activity during fatigue induced by submaximal isometric intermittent contractions. Neural and Neuromuscular Aspects of Muscle Fatigue: Satellite Symposium to the Society for Neuroscience B13. McKenzie, D.K. & Gandevia, S.C. 1991. Recovery from fatigue of human diaphragm and limb muscles. Respir Physiol 84, 49±60. Merton, P.A. 1954. Voluntary strength and fatigue. J Physiol (Lond) 123, 553±564.

Ó 1998 Scandinavian Physiological Society

Acta Physiol Scand 1998, 162, 275±283

Meunier, S. & Pierrot-Deseilligny, E. 1989. Gating of the afferent volley of the monosynaptic stretch re¯ex during movement in man. J Physiol (London) 419, 753±763. Miller, K.J., Garland, S.J., Ivanova, T. & Ohtsuki, T. 1996. Motor-unit behavior in humans during fatiguing arm movements. J Neurophysiol 75, 1629±1636. Mills, K.R. & Thomson, C.C.B. 1995. Human muscle fatigue investigated by transcranial magnetic stimulation. Neuroreport 6, 1966±68. McKay, W.B., Stokic, D.S., Sherwood, A.M., Vrbova, G. & Dimitrijevic, M.R. 1996. Effect of fatiguing maximal voluntary contraction on excitatory and inhibitory responses elicited by transcranial magnetic moto cortex stimulation. Muscle Nerve 19, 1017±1024. McKay, W.B., Tuel, S.M., Sherwood, A.M., Stokic, D.S. & Dimitrijevic, M.R. 1995. Focal depression of cortical excitability induced by fatiguing muscle contraction: a transcranial magnetic stimulation study. Exp Brain Res 105, 276±282. Nelson, D.L. & Hutton, R.S. 1985. Dynamic and static stretch responses in muscle spindle receptors in fatigued muscle. Med Sci Sports Exerc 17, 445±450. Nielsen, J. & Petersen, N. 1994. Is presynaptic inhibition distributed to corticospinal ®bres in man? J Physiol (Lond) 477, 47±58. Paintal, A.S. 1960. Functional analysis of group III afferent ®bres of mammalian muscle. J Physiol (Lond) 152, 250±270. Persson, S., Jonsdottir, I., ThoreÂn, P., Post, D., Nyberg, F. & Hoffmann, P. 1993. Cerebrospinal ¯uid dynorphinconverting enzyme activity is increased by voluntary exercise in the spontaneously hypertensive rat. Life Sci 53, 643±652. Pickar, J.G., Hill, J.M. & Kaufman, M.P. 1994. Dynamic exercise stimulates group III muscle afferents. J Neurophysiol 71, 753±760. Porter, R. & Lemon, R. 1993. Corticospinal Function and Voluntary Movement. Clarendon Press, Oxford. Pratt, C.A. 1995. Evidence of positive force feedback among hindlimb extensors in the intact standing cat. J Neurophysiol 73, 2578±2583.

Ó 1998 Scandinavian Physiological Society

S C Gandevia á Neural control and muscle fatigue

Rotto, D.M. & Kaufman, M.P. 1988. Effect of metabolic products of muscular contraction on discharge of group III and IV afferents. J Appl Physiol 64, 2306±2313. Sandercock, T.G., Faulkner, J.A., Albers, J.W. & Abbrecht, P.H. 1985. Single moto unit and ®ber action potentials during fatigue. J Appl Physiol 58, 1073±1079. Sawcuk, A., Powers, R.K. & Binder, M.D. 1995. Spike frequency adaptation studied in hypoglossal motoneurons of the rat. J Neurophysiol 73, 1799±1810. Sinoway, L.I., Hill, J.M., Pickar, J.G. & Kaufman, P. 1993. Effects of contraction and lactic acid on the discharge of group III muscle afferents in cats. J Neurophysiol 69, 1053±1059. Spielmann., J.M., Laouris, Y., Nordstrom, M.A., Robinson, G.A., Reinking, R.M. & Stuart, D.G. 1993. Adaptations of cat motoneurons to sustained and intermittent extracellular activation. J Physiol (Lond) 464, 75±120. Taylor, J.L., Butler, J.E., Allen, G.M. & Gandevia, S.C. 1996. Changes in moto cortical excitability during human muscle fatigue. J Physiol (Lond) 490, 519±528. Thompson, S., Gregory, J.E. & Proske, U. 1990. Errors in force estimation can be explained by tendon organ desensitization. Exp Brain Res 79, 365±372. van Hall, G., Raaymakers, J.S., Saris, W.H. & Wagenmakers, A.J. 1995. Ingestion of branched-chain amino acids and tryptophan during sustained exercise in man: failure to affect performance. J Physiol (Lond) 486, 789±794. Windhorst, U. & Boorman, G. 1995. Overview: potential role of segmental motor circuitry in muscle fatigue. Adv Exp Med Biol 384, 241±258. Plenum Press, New York. Zanette, G., Bonato, C., Polo, A., Tinazzi, M., Manganotti, P. & Fiaschi, A. 1995. Long-lasting depression of motorevoked potentials to transcranial magnetic stimulation following exercise. Exp Brain Res 107, 80±86. Zytnicki, D., La¯eur, J., Horcholle-Bossavit, G., Lamy, F. & Jami, J. 1990. Reduction of 1b autogenetic inhibition in motoneurons during contractions of an ankle extensor muscle in the cat. J Neurophysiol 64, 1380±1389.

283