This study aimed to investigate central and peripheral contributions to fatigue during repeated maximal voluntary isometric plantar flexions (MVCs). Changes in joint torque, level of activation (LOA), resting twitch amplitude (RT), electromyographic signals (EMG), and presynaptic inhibition of Ia afferents were investigated during 9 bouts of 10 MVCs. MVCs lasted for 2 s and were separated by 1 s. The interval between bouts was 10 s. Electrical stimulation was applied to the tibial nerve; at rest to evoke RTs, M waves, and two (1.5-s interval) H reflexes; with the soleus EMG at 30% of that during MVC to evoke M waves and two H reflexes; and during MVCs to measure LOA. Over the nine bouts, LOA decreased by 12.6% and RT by 16.2%. EMG root mean square during MVCs remained unchanged for the soleus and tibialis anterior muscles, but it decreased for medial gastrocnemius. Peripheral fatigue (decrease in RT) was positively correlated to LOA, whereas central fatigue (decrease in LOA) was not. Depression of both H reflexes suggests that presynaptic inhibition after the first bout was partly induced by homosynaptic postactivation depression of the Ia terminal. The H-reflex-to-M-wave ratio increased with fatigue in both passive and active states, with no change in the ratio of the second H reflex to the first, thereby indicating a decrease of presynaptic inhibition during fatigue. The results indicate that both central and peripheral mechanisms contributed to the fatigue observed during repeated MVCs and that the development of peripheral fatigue was influenced by the level of voluntary activation and initial plantar flexor torque.
- H reflex
- presynaptic inhibition
fatigability is known to vary substantially between muscle groups, muscle action types, and individuals. Fatigue has both central and peripheral components, whose relative contribution to fatigue appears to be task dependent (10). Central fatigue refers to an activity-induced inability to fully activate a muscle voluntarily, whereas peripheral fatigue implies that the ability of the muscle to produce force is reduced.
During sustained muscle actions, central and peripheral fatigue have been described to develop during maximal (3), as well as submaximal (23), voluntary efforts. During intermittent muscle actions, both central and peripheral fatigue develop when the effort is maximal (39), whereas, when the effort is submaximal or there is sufficient rest between each muscle action, fatigue has been shown to be caused mainly by peripheral mechanisms (2). Commonly discussed factors that affect peripheral fatigue include energy supply (38), muscle fiber-type distribution (40, 41), muscle strength before fatigue (19, 20), and the length of the muscle (12). Some of the possible mechanisms of central fatigue include suboptimal facilitation from the motor cortex (39), decreased facilitation from muscle spindles (27), increased inhibition from group III and IV afferents (13, 14), and desensitization of the motoneurons (22). Moreover, the ability of the central nervous system to fully activate a muscle during maximal efforts has been described to vary substantially between muscle groups, muscle action types, and individuals (1, 26, 44). Some of the differences in fatigability could, therefore, be explained by the varying levels of voluntary activation that subjects are able to achieve. The level of activation (LOA) of a muscle during maximal voluntary effort can range between 80 and 100% (1, 21). If a muscle is only activated to 80% of its full capacity, fatigue is likely to occur more slowly than if it were fully activated. Intermittent maximal voluntary activations should constitute a useful model for investigating the importance of the ability to fully activate a muscle for the relative contributions of central and peripheral fatigue.
Decreased excitation from Ia afferents, due to decreased firing frequencies of muscle spindles, has been suggested to be one mechanism that could cause central fatigue (27, 28). It is also possible that the size of the excitatory postsynaptic potential induced by each Ia afferent action potential decreases during fatigue. The strength of the excitation achieved by a same-sized compound Ia afferent action potential on its homonymous motoneuron pool can be measured by means of the Hoffmann reflex (H reflex) (17, 31, 47). A decrease in the H reflex can be attributed to decreased motoneuron excitability and/or to increased presynaptic inhibition of Ia afferents (17). Presynaptic inhibition of Ia afferents could arise either from GABA-mediated primary afferent depolarization (PAD) (37), which lasts for up to 500 ms, or from homosynaptic postactivation depression (HPAD) (18, 36), with a duration of up to 15 s. Whereas the PAD interneurons are controlled by both afferent and descending circuits, HPAD is believed to be caused by an intrinsic mechanism within the Ia terminal itself and only occurs when the muscle spindle is, or has recently been, discharging (6, 18, 42, 45). It is possible that reduced facilitation from Ia afferents via presynaptic inhibition could be influential enough to weaken the net excitatory input to the motoneuron pool and thereby contribute to an impaired ability to fully activate a muscle.
The first aim of the present study was to investigate the peripheral and central contributions to fatigue during isometric intermittent maximal voluntary plantar flexions. The second aim was to examine whether LOA during an initial unfatigued maximal voluntary plantar flexion can influence the relative contribution of central and peripheral fatigue. The third aim was to assess whether there is a change in presynaptic inhibition of Ia afferents during intermittent maximal voluntary plantar flexor actions and, if so, whether it is possible to distinguish PAD-mediated inhibition from HPAD.
Subjects. Experiments were performed on 16 healthy, habitually active male subjects who volunteered to participate in the study. Six of the subjects were subsequently excluded from analysis due to methodological considerations (see Exclusion criteria and methodological considerations below). Consequently, data from 10 subjects [age 28 ± 5 (SD) yr, body height 1.79 ± 0.05 m, body mass 79 ± 5 kg] were used in the analysis. Subjects gave their informed consent and were asked not to exercise 24 h before the experiment. The local ethics committee approved all procedures, and the study was conducted according to the Declaration of Helsinki.
Experimental setup. Subjects were lying prone on a bench with their right ankle at 90° and their right foot strapped tightly to a vertical footplate, instrumented with a force transducer (KRG-4, Thermo Nobel). Subjects were instructed to keep their body still throughout the whole experiment, and their pelvis was strapped to the bench to keep the knee in a straight position and prevent adverse movements during the protocol. Before the onset of the fatigue protocol, resting H-reflex and M-wave recruitment curves were established (see Stimulation procedures below), and then both submaximal and maximal plantar flexor efforts were practiced. The practice period was followed by one 2-s maximal voluntary isometric plantar flexion (MVC) and, in six of the subjects, one 2-s maximal voluntary isometric dorsiflexion, after which at least 7 min of rest were given before the fatigue protocol was initiated.
Fatigue protocol. Subjects performed nine bouts of 10 MVCs. Each MVC was held for 2 s. An audio signal was presented to guide the timing of the MVCs. The time between each MVC was 1 s, and the pause between bouts was ∼10 s (Fig. 1A). The subjects were strongly verbally encouraged to rapidly produce and maintain maximal effort for 2 s for all 90 MVCs.
Stimulation procedures. Before the initiation of the fatigue protocol and after each bout, sets of electrical stimulations were applied to the tibial nerve. The total time for each set of stimulations was ∼10 s. Initially, subjects relaxed their muscles completely (passive state) for ∼5 s, during which a pair (1.5-s interval) of H reflexes and a pair (0.02-s interval) of supramaximal M waves were evoked (2H and 2M in Fig. 1, A and B). Stimulations started ∼1.5 s after the end of the 10th action of each bout. After this 5-s period, subjects were asked to match 30% of their soleus surface electromyographic (EMG) signal (EMGs) root mean square (RMS) (EMGs-RMS) (see EMG recordings below) achieved during the prefatigue MVC (active state). This activation was guided by visual feedback and held for another 5 s, during which the same electrical stimulations were applied as in the passive state.
All electrical stimulations were applied through a cathode (4-mm diameter, Ag-AgCl, Medicotest) placed in the popliteal fossa and an anode (100 × 50-mm coal-rubber electrode, Cefar Medical) placed on the anterior aspect of the thigh, just proximal to the knee (35). Stimulations consisting of two 0.5-ms supramaximal pulses with a 0.02-s interval were delivered by a Grass constant-voltage somato-sensory stimulator (SI0DSCM A, Grass Instrument) and a stimulus isolation unit (SIU8T B, Grass Instrument) to induce twitches in the torque signal and M waves in the EMG. Supramaximal intensity was based on the size of the M wave in the soleus and medial gastrocnemius. When there was no increase in M-wave amplitude with an increase in stimulus intensity, the intensity was further increased by 50%. It was carefully monitored so that this increase did not cause a reduction in the resting twitch (RT) due to activation of the antagonist muscles. Each H reflex was induced in the soleus muscle by a single 1-ms pulse delivered to the tibial nerve by a constant-current stimulator (DS7A, Digitimer). From the H-reflex and M-wave recruitment curves, the stimulation intensity resulting in the highest soleus H-reflex amplitude in the passive muscle was used for evoking all H reflexes. This stimulus intensity also resulted in a small M wave that was used to assess any changes that might occur, e.g., in the interface between the stimulus electrode and the nerve, while still affording sufficient sensitivity to detect changes in the H reflex. At this intensity, an adequate H reflex cannot be obtained in the gastrocnemius muscles.
Torque recordings. Force signals were low-pass filtered at 50 Hz (NL 125, Digitimer) and analog-to-digital converted at a sampling frequency of 5 kHz by using a 14-bit Power 1401 and Spike2 data collection software (Cambridge Electronic Design). The force measured by the transducer was multiplied by its lever arm (0.30 m) to achieve a measurement of plantar flexor torque (strength). Plantar flexor torque during the MVCs of the fatigue protocol was measured as the mean torque during a 1-s period in the middle of the 2-s period of each MVC. In the statistical analysis, the average of the third, fourth, sixth, and seventh MVC in each bout was used to represent the torque of that bout.
LOA was assessed by means of the twitch interpolation technique (30). Supramaximal electrical stimulations were applied to the tibial nerve during the fifth MVC of each bout in the fatigue protocol and during the rest period between each bout. RTs were measured as the peak torque occurring within 200 ms from when the supramaximal stimulation was applied (Fig. 1C). To achieve the superimposed twitch (ST) during MVC, the mean torque measured during the 50 ms preceding the supramaximal stimulation was subtracted from the peak torque within 200 ms from when the supramaximal stimulation was applied (Fig. 1C). The LOA during each bout was then calculated as LOA (%) = [1 - (ST/RT induced after the same bout)] × 100.
Definitions of fatigue. Fatigue was defined as a decrease in maximal voluntary plantar flexor torque over bouts. Central fatigue was defined as a decrease in the LOA from the first bout, and peripheral fatigue as a decrease of the amplitude of the RT from after the first bout.
EMG recordings. Intramuscular EMG (EMGi) were collected from the soleus, medial gastrocnemius, and tibialis anterior muscles. Bipolar fine-wire electrodes (soft AG7/40T, Medwire) with a recording length of 2 mm (7-mm interelectrode distance) were inserted into the muscles by means of sterile hypodermic needles (Terumo, 0.4 × 30 mm). The EMGi were amplified (1,000 times, Myosystem 2000, Noraxon) and band-pass filtered between 15 and 1,000 Hz (NL 125, Digitimer). The signal was corrected for direct current offset, and the RMS of the EMGi (EMGi-RMS) was measured with the muscles completely relaxed and then subtracted from all other measurements. During the MVCs, the EMGi-RMS was calculated over the same 1-s period as the torque. These EMGi-RMS values were then normalized to the peak-to-peak amplitude of the M wave in the respective recording. In the six subjects who performed a maximal voluntary dorsiflexion before the onset of the fatigue protocol, the tibialis anterior EMGi-RMS measured over a 1-s period in the middle of the 2-s maximal dorsiflexion was used to normalize the tibialis anterior EMGi-RMS during the plantar flexor MVCs of the fatigue protocol.
For matching of the soleus activation (in the active state, Fig. 1A), EMGs was recorded from soleus by a pair of electrodes (2.5-mm diameter, Ag-AgCl, Sensor Medics) placed over the center of the muscle with an interelectrode distance of 2 cm. The signal was amplified 2,000 times (NL 824, Digitimer) and band-pass filtered (NL 125, Digitimer) between 15 and 300 Hz. Evoked potentials (M waves and H reflexes) were recorded via an additional pair of surface electrodes placed over the soleus muscle and Achilles tendon. The signal was amplified 100 times (NL 824, Digitimer) and band-pass filtered between 15 and 1,000 Hz (NL 125, Digitimer). All EMG signals were analog-to-digital converted at a sampling frequency of 5 kHz by using a 14-bit Power 1401 and Spike2 data collection software (Cambridge Electronic Design).
M-wave and H-reflex measurements. The H reflex was measured as the peak-to-peak amplitude between 20 and 50 ms after stimulation of the tibial nerve, and the supramaximal M wave was measured as the peak-to-peak amplitude between 5 and 15 ms (Fig. 1B). The peak-to-peak amplitude of the greater of the two M waves was used as the maximum M wave (usually the second of the two). The two reflex parameters evaluated were the ratio of the first H reflex to the maximum M wave (H/M) and the ratio of the second H reflex to the first H reflex (H2/H1). H reflexes evoked in a passive soleus muscle are influenced by both motoneuron excitability and presynaptic inhibition, whereas those evoked with the soleus EMGs-RMS controlled at 30% of MVC are assumed to be mainly affected by presynaptic inhibition. Not considering motoneuron excitability, H/M can be affected by both HPAD- and PAD-mediated inhibition. Changes in H2/H1, on the other hand, are assumed to be caused by changes in HPAD only, because the second H reflex is evoked after the time during which it is possible for PAD-mediated presynaptic inhibition from the first stimulation to affect the size of the second H reflex. Therefore, a decreased H/M without a change in the H2/H1 would, for example, indicate increased PAD-mediated inhibition with no change in HPAD, whereas an increased H2/H1 with an unchanged H/M would suggest an increase in HPAD with a concomitant decrease in PAD-mediated inhibition.
Exclusion criteria and methodological considerations. Activation of tibialis anterior muscle spindles is known to presynaptically inhibit Ia afferents of the plantar flexors. Three subjects were, therefore, excluded due to consistent activation of tibialis anterior as the H reflexes were evoked. Activation of tibialis anterior also caused the H/M and the H2/H1 evoked after one of the nine bouts to be excluded for two of the subjects. Another three subjects were excluded due to inconsistency of the M wave when the H reflex was evoked in the passive state. Ten subjects remained who were included in all analyses, except that of the H/M and the H2/H1 in the active state. When evoking H reflexes at an activation of 30%, it is difficult to keep the effective stimulus intensity to the tibial nerve constant. In the active state, 5 of the 10 subjects did not fulfill the criteria of consistent M waves and were excluded from analysis. H reflexes evoked with the soleus EMGs-RMS at 30% of that during an MVC could only be affected by changes in presynaptic inhibition, because the overall motoneuron excitability was held constant. In the active state, there was no significant difference between the EMG-RMS before the first and the second H reflex (100-ms window) for either the soleus or tibialis anterior. This means that any difference between the first and the second H reflex is caused by presynaptic mechanisms.
Statistical analysis. Normality of the data was confirmed by using the Shapiro-Wilk W test. Changes in RT and reflex data from before the protocol to after the first bout were tested by means of a paired Student's t-test. Repeated-measures ANOVAs were used for detecting changes from the first to the ninth bout for voluntary torque, LOA, RT, EMGi-RMS from the three muscles, and reflex data. If the data did not conform to the assumption of sphericity, the P value was Huynh and Feldt corrected. The level of significance was set to P ≤ 0.05, and trends were considered at 0.05 < P ≤ 0.1.
Where a significant main effect or a trend was found in the ANOVA, data were normalized to measurements taken during the first bout (for torque, EMGi-RMS and LOA) or after the first bout (for RT, H/M, and H2/H1), and a line was fitted to the data. A regression was then performed to assess how well this line described the data. Correlations were performed between the decrease in voluntary torque and the extent of central and peripheral fatigue. Correlations and multiple regressions were also employed to investigate to what extent the changes in RT and LOA could be explained by the subject's MVC torque and LOA during the first bout. ANOVAs and t-tests were performed by using Statistica (StatSoft), and correlations and regressions by using Origin (Microcal Software). Unless elsewhere stated, all data are presented as means ± SE.
Voluntary plantar flexor torque. The mean maximal voluntary plantar flexor torque during the first bout was 148.4 ± 10.2 N·m (range 114.4–206.7 N·m). There was a main effect of bouts on the mean voluntary plantar flexor torque [F(8,72) = 11.5], and the normalized plantar flexor torques of the 10 subjects decreased in a curvilinear manner over the nine bouts (Fig. 2A). The mean decrease in voluntary torque was 29.0 ± 4.9%, and the range of the decrease among individuals was between 0.7 and 55.8%.
LOA. During the first bout, the mean LOA was 85.1 ± 4.0% (range 67.9–99.9%), and by the ninth bout it had decreased to 74.4 ± 4.4% (range 56.9–99.6%). There was a main effect of bouts on the LOA [F(8,72) = 4.0]. The normalized LOA decreased in a curvilinear fashion over the nine bouts (Fig. 2B). The mean decrease in LOA from the first to the ninth bout was 12.6 ± 2.8%, and the range of the decrease was from 0.4 to 24.5%.
RT amplitudes. RT was potentiated from 75.1 ± 5.7 to 89.4 ± 4.5 N·m after the first bout. During the fatigue protocol, there was a main effect of bouts [F(8,72) = 7.7] on RT. The normalized RT decreased in a curvilinear manner over the nine bouts (Fig. 2C). The mean decrease in RT from after the first to after the ninth bout was 16.2 ± 5.9%, ranging from 57.9 to -6.9%.
Relationship among plantar flexor torque, LOA, and central and peripheral fatigue. Central fatigue was not significantly correlated to the decrease in mean torque, whereas peripheral fatigue was (r2 = 0.57). The sum of central and peripheral fatigue was only slightly better correlated to the decrease in torque (r2 = 0.60) than peripheral fatigue alone.
Regressions revealed that no significant relationship existed between decrease in LOA (central fatigue) from bouts 1 to 9 and either torque (strength) or LOA during the first bout. The decrease in RT (peripheral fatigue) from the first to the last bout could by 55% (r2 = 0.55) be explained by the maximal voluntary plantar flexor torque during the first bout (Fig. 3A). The LOA during the first bout could explain 39% (r2 = 0.39) of the decrease in the RT (Fig. 3B). A multiple regression revealed that LOA and plantar flexor torque during the first bout could together explain 58% of the decrease in RT (r2 = 0.58). This means that the higher the initial plantar flexor torque and LOA, the greater the decrease in RT, i.e., the progression of peripheral fatigue.
Although the development of central fatigue was not dependent on the initial strength (torque) or the LOA during the first bout, some subjects developed more central fatigue than others (Fig. 2B). The mean LOA over all bouts was found to explain 60% (r2 = 0.60) of the decrease in RT (Fig. 3C). A multiple regression showed that the initial strength, the initial LOA, and the mean LOA, together, could explain 74% (r2 = 0.74) of the variation in peripheral fatigue developed over the nine bouts.
EMG. No significant change was observed in the supramaximal M wave over time. There was a main effect of bouts on the EMGi-RMS of the medial gastrocnemius [F(8,72) = 5.5] (Fig. 4A), whereas no effect was found for the EMGi-RMS of the soleus muscle (Fig. 4B). The EMGi-RMS of medial gastrocnemius decreased in a curvilinear manner during the nine bouts. The mean decrease over the nine bouts was 17.3 ± 3.8%, and the range of the decrease was between 2.5 and 36.4%. There was no change over the nine bouts in the EMGi-RMS of the tibialis anterior. In the six subjects who performed a maximal voluntary dorsiflexion, the mean coactivation of tibialis anterior during the nine bouts of maximal voluntary plantar flexions ranged from 1.4 to 5.9% of that during a maximal voluntary dorsiflexion.
H/M and H2/H1. H/M evoked in the passive state was significantly reduced after the first bout (from 0.56 ± 0.05 to 0.34 ± 0.05) (Fig. 5A). At the same time, the passive H2/H1 increased from 0.75 ± 0.03 before to 1.01 ± 0.06 after the first bout. This implies an increase in HPAD, presumably as a result of increased soleus Ia afferent activity during the contractions. There was a trend for H/M to change during fatigue [F(8,64) = 2.1]. The regression revealed that the passive H/M increased slightly in a linear manner from after the first to after the last bout (Fig. 5A), whereas no changes occurred over bouts for the passive H2/H1. This indicates that, whereas HPAD did not change over bouts, either excitability of the passive motoneuron pool increased and/or PAD-mediated inhibition decreased.
In the active state (30% soleus EMGs-RMS), H/M was reduced after the first bout (from 0.52 ± 0.05 to 0.47 ± 0.05) (Fig. 5B), whereas there was no change in active H2/H1 from before (0.97 ± 0.03) to after (1.03 ± 0.04) the first bout. This implies that HPAD was induced by the 30% muscle action and that, after the first bout of MVCs, presynaptic inhibition had increased further. There was a main effect of bouts on active H/M [F(8,32) = 2.2]. The active H/M was best described by a linear increase over the nine bouts (Fig. 5B). There were no changes over bouts for active H2/H1. It was not possible to deduce whether HPAD increased or decreased with fatigue, because H2/H1 approached 1 already at an activation of 30%. The increase in active H/M is unlikely to be caused by changes in soleus motoneuronal excitability, but rather by decreases in either PAD-mediated inhibition or HPAD.
The main findings of this study were that both central and peripheral mechanisms contributed to fatigue during repeated MVCs and that the level of voluntary activation and the initial plantar flexor strength influenced the development of peripheral, but not central, fatigue. Presynaptic inhibition of Ia afferents initially increased and then decreased slightly over the fatigue protocol. Because our method of measuring HPAD may have saturated early, it cannot be concluded from this whether presynaptic inhibition of Ia afferents decreased due to decreased PAD-mediated inhibition or due to reduced firing frequencies of the muscle spindles decreasing HPAD.
Central vs. peripheral fatigue. With the current activity-rest intervals within and between bouts, it was found that MVCs induced both peripheral (16.2%) and central fatigue (12.6%), with only peripheral fatigue being correlated to the decrease in plantar flexor strength (r2 = 0.57). The extent of peripheral fatigue was similar to that found by Kawakami et al. (21) after 100 MVCs. In their study, however, subjects developed more than twice as much central fatigue, which might have been due to their subjects starting at a relatively higher LOA (96% vs. our 85%). However, as we found no relationship between the achieved LOA and the extent of central fatigue, the difference in central fatigue between the two studies is more likely to be due to the given rest interval (10 s between bouts in this study and 1 s between MVCs in the study of Kawakami et al.). It has been shown that central fatigue recovers relatively quickly (23), and even a 10-s rest might be enough for the central nervous system to partly recover (39).
The putative mechanisms of central fatigue are numerous. Decreased cortical drive to the motoneuron pool (7), increased activity of group III and IV afferents (4, 13), and increased recurrent inhibition (25) are some of the main candidates whose relative contributions may be task dependent. The decrease of both the active and passive H/M after the first bout confirms previous investigators' findings that the H reflex decreases when measured after the very onset of exercise, due to either pre- or postsynaptic mechanisms (9, 11, 29). This study demonstrates that the initial reduction in the H reflex is, at least partly, caused by presynaptic mechanisms, because the H/M decreased also in the active state, where it was attempted to maintain motoneuron excitability constant via the same soleus EMG. The increase in H2/H1 in the passive state after the first bout suggests that increased muscle spindle firing during the first bout of MVCs increased HPAD and was thereby partly responsible for the decrease in H/M. Because H2/H1 did not change in the active state, either the test was not sufficiently sensitive or the 30% MVC induced Ia firing high enough for the HPAD to attain a level that did not change over the fatigue protocol.
H/M has been described to increase with fatigue during a sustained submaximal constant-torque task (24), probably due to a gradual increase in neural drive to the motoneuron pool. However, when the H reflex is measured in a passive muscle after the end of a sustained maximal plantar flexion, it has been found to be reduced (15). This decrease has been interpreted to indicate a decrease in motoneuron excitability due to increased inhibition from group III and IV afferents. However, because H reflexes were evoked without a controlled underlying activation, the decrease of the H reflex in those studies may also have been caused by presynaptic mechanisms. The small increase in H/M over bouts in the relaxed and the active state suggests a small but significant decrease in presynaptic inhibition over the fatigue protocol. Because H2/H1 exceeded 1 and remained unchanged throughout the protocol, it is difficult to deduce whether the decrease in presynaptic inhibition was due to decreased PAD-mediated inhibition or decreased HPAD. The distinction between the two presynaptic mechanisms is important for the interpretation of whether excitation from muscle spindles increases or decreases during fatiguing exercise. Whereas decreased HPAD would probably be caused by decreased muscle spindle firing, decreased PAD-mediated inhibition with sustained muscle spindle firing would imply an increased excitation from the Ia afferents onto the motoneuron pool.
The extent of peripheral fatigue has been discussed to be dependent on the targeted LOA (expressed in percentage of voluntary maximum) or on the absolute muscle force used during the fatigue protocol (10). Although, to our knowledge, no one has investigated this before, many have speculated that a low ability to completely, voluntarily activate a muscle might lead to prolonged endurance (34, 46). A novel finding of the present study was that, during this specific protocol of intermittent MVCs, subjects who had a poor ability to achieve full voluntary activation before fatigue developed less peripheral fatigue. Less peripheral fatigue during incomplete activation could be brought about via redistributing muscle activation within or between agonist muscles. Such behavior has been observed during voluntary submaximal muscle actions (5, 19). On the other hand, in a group of agonist muscles that are fully or close to fully activated, intermittent resting of parts of the muscle or muscle group would not be possible without lowering the LOA.
In addition, the higher the level of voluntary activation, the faster the energy depletion (38), and the rapid accumulation of metabolites, such as inorganic phosphate, is likely to impair the release and reuptake of Ca2+ from the sarcoplasmic reticulum (43). Furthermore, difference in fiber-type distribution has long since been put forward as a factor explaining variations in peripheral fatigue between individuals (41). According to the “size principle,” slow-twitch motor units (type I) are recruited before fast twitch (type II) as force is increased (16). This would imply that subjects who cannot achieve full activation of their muscles in a given task would not have recruited all of their fast-twitch motor units at the optimal firing frequency, thereby decreasing the rate of fatigue. The medial gastrocnemius consists of a higher proportion of fast-twitch fibers than the soleus muscle (8) and is, therefore, expected to develop peripheral fatigue at a higher rate than the soleus muscle. In this study, the decline in torque output was not measured separately for the soleus and medial gastrocnemius, and the extent of peripheral and central fatigue of the individual muscles can, therefore, not be elaborated on. However, it can be noted that the EMG of the medial gastrocnemius was found to decrease over the nine bouts, whereas no change was observed for soleus. This is in accordance with previous studies that described a greater decrease in the EMG of the gastrocnemius compared with the soleus muscle during a sustained maximal voluntary plantar flexion (32), as well as during hopping (33). These results might infer that central fatigue develops faster in the medial gastrocnemius than in the soleus muscle.
This study was supported by grants from the Swedish Center for Sport Research and the Swedish Research Council.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society