|
|
||||||||
Division of Applied Biomedical Research School of Biomedical and Health Sciences, King's College London, London, United Kingdom
Submitted 19 May 2006 ; accepted in final form 14 September 2006
| ABSTRACT |
|---|
|
|
|---|
skeletal muscle; motor control; skill
One of the hallmarks of long-lasting fatigue is, unlike metabolic and late (12+ h) eccentric fatigue, that it is not accompanied by pain. Indeed the term "insensible fatigue" would be a more characteristic name. In his review of long-lasting fatigue, Jones (19) noted that the severe decrement of mechanical response, at action potential frequencies of 1020 Hz typical of natural activation, might be expected to give rise to substantive errors in motor performance after fatiguing exercise. Although errors in proprioception have been noted (4), motor errors have not been reported, and Jones questioned whether this was because there was an efficient neural correction for the decreased response or simply that an effective test of skill had not yet been utilized. Subsequently, a decrease was shown (17) in lower limb-tracking performance immediately after exercising to 50% exhaustion, i.e., short-term metabolic fatigue. However, this type of error can largely be compensated for visually (27, 43), and there is indeed a proportionate enhancement of central motor effort during short-term (18) and long-term (6) fatigue. In addition, there is a slower reaction time to auditory stimulus during postexercise balance control (42), suggesting an additional attentional demand during fatigue.
Eccentric, compared with concentric, contractions are characterized by the development of high forces but a considerable lower metabolic turnover (1). Differences in subsequent fatigue development should thus highlight either metabolic causes following concentric contractions or the effect of high cross-bridge forces following eccentric contractions, such as damage by overstretch of unstable sarcomeres (39, 26). Although the long-term recovery over several days from eccentric fatigue has been described (e.g., Refs. 20, 34), the immediate differences in type and time course of fatigue are not clear. The intentions in this paper are thus 1) to compare the size and time course of long-lasting fatigue for equal concentric and eccentric contractions in human muscle at high temporal resolution and 2) to determine what affect the fatigue has on task-dependent performance.
| METHODS |
|---|
|
|
|---|
2 min. In a third series of experiments with the same protocols, either two sets (20 contractions) or eight sets (80 contractions) were required. Test responses were started
2 min afterward, the time required to unstrap the Kin-Com wrist cuff and set the response recording system. Each subject undertook concentric contractions on the first experimental day and eccentric contractions on the second experimental day, 2 -7 days later, using the nondominant arm.
Electrical stimulation and recording of response.
Single electrical shocks were applied at the top third of the biceps via two 5 x 5-cm conductive rubber electrodes using constant-current 0.1-ms stimuli (model DS7, Digitimer, Welwyn Garden City, Hertfordshire, UK). EMG signals were recorded from the belly of biceps with a MA-110 active electrode probe (Motion Lab Systems, Baton Rouge, LA) in all experiments except the skill series. Subjects were seated with the elbow joint at 90°, and the auxotonic twitch response was recorded as the inertial flexion of the forearm using a potentiometer connected via a lightly loaded (compliance
10°/N) wrist strap. Maximal twitch responses were ensured by recording stimulus-response curves for each subject before and at hourly intervals during the experiment. Responses were recorded every minute for 15 min before and 240 min after the fatiguing exercise. To compare these twitch measurements with conventional low-frequency fatigue, tetanic stimuli with 5 shocks at 1 Hz and 10 Hz and then 10 shocks at 50 Hz were given in a second series of experiments each minute for 15 min before and 20 min after the exercise. Results for these are given as the average of 10 tests during minutes 6-15 of the control period and for minutes 1120 after exercise. All mechanical records were measured at the peak response. In all cases, the timing of the stimuli and of the skill test was computer controlled throughout the recording period.
Skill test. In the series of experiments using tetanic stimuli, a 10-s tracking test was also performed every minute during the 15-min preexercise period and the 20-min recovery period. The subjects were asked to point at a moving target on a computer screen using a laser pointer that was strapped to the lower arm, the position of which was measured by the same potentiometer system as for recording the twitch response. The vertically moving target was synthesized for each test from 10 harmonic sine waves with random amplitude and base frequency of 1 wave each 10 s. The movement range, referred to the elbow joint, was 10°. Tracking performance was quantified in several ways. First, the lag between the target and response trace was calculated by finding the time difference for maximum coherence between the two signals. In all subsequent analyses, the target signal was shifted forwarded by this lag so that, for an ideal tracking performance, the signals would exactly superimpose. The movement error was calculated as the standard deviation of the difference between the new traces. The movement error was further analyzed to test the hypothesis that errors should be larger during activation of the fatigued biceps muscle than during activation of the unfatigued triceps muscle or while allowing the action of gravity. Because it is muscle force that accelerates the arm, the tracking signal was double differentiated and used to separated the tracking errors into those occurring during positive acceleration (biceps activation) and negative acceleration (triceps activation or gravity). The ratio of the standard deviation of each of these error types was used to estimate the symmetry of error between biceps and triceps. Results given are for the average of 10 tests during minutes 615 of the control period and for minutes 1120 after exercise.
Subjects. Subjects were healthy untrained men and women with a median age of 22 yr (range 2053 yr). Experiments were conducted in three series with different, but overlapping, subject groups. The first series of experiments, using 40 eccentric or concentric contractions followed for 4 h of test twitch stimulation, had 7 subjects. The second series with six subjects involved the skill test and tetanic stimulation, although one subject chose not to complete the tetanic stimulation. A third series with 5 subjects compared 20 or 80 contractions. The experimental protocols were approved by the College Research Ethics Committee, and all subjects provided written informed consent. Data are given as means ± SE where n refers to the number of subjects. Significance was set at P < 0.05 using paired or unpaired t-tests.
| RESULTS |
|---|
|
|
|---|
2 h. The response at the first measurement,
2 min after the end of the exercise, was depressed to only 78% control with a half-maximal decline in 67 min. Subjects, when asked about this delayed decline, reported no sense of fatigue; indeed, some were incredulous of the observed decline in their response.
|
2 h. By the end of the experiment at 4 h, recovery was still just less than half complete. Half-maximal decline occurred in 12 min, but there was an early rapid decline, similar in time scale to that for concentric contractions, followed by a much slower decline. The maximal decline was smaller following eccentric contractions (40.0 ± 5.1% of control at 120 min). The force exerted during the isokinetic concentric contractions, as might be expected from the Hill relationship, was about half the force produced during eccentric contractions (Fig. 1B) (58% for the first contractions falling to 40% for the final contractions). Thus, despite having developed a lower force, concentric contractions produced a greater long-lasting fatigue. To test whether the long-lasting nature of the fatigue observed is due to low-frequency fatigue (13), the second series of experiments were performed where the fatigue was measured with stimuli at 1, 10, and 50 Hz. Figure 2 shows that the initial fatigue (1120 min) was relieved at 50 Hz for both concentric and eccentric fatiguing routines, thus defining the observed fatigue as low frequency. The 10-Hz-to-50-Hz ratio showed significant declines to 0.56 ± 0.09 of its initial value for concentric and 0.80 ± 0.09 for eccentric contractions, values in similar proportion to the twitch decline (Fig. 1) but of smaller amplitude. To investigate the relationship between exercise intensity and fatigue, a third series of experiments were conducted using either 20 or 80 maximal concentric contractions in the fatiguing period (i.e., 2 or 8 sets instead of 4). The twitch response at 20 min after 20 contractions was 72 ± 2.2%, and after 80 contractions it was 50 ± 6.0% control (n = 5). There was no significant difference between 40 and 80 contractions, but the fatigue was about halved for 20 contractions.
|
50% after 60 min. Figure 3A shows that in the present experiments, there was also a decline in the M wave but that this was <20% for both concentric and eccentric contractions (Fig. 3C). The time scale of decline and recovery of the M wave was similar to the decline of the mechanical response, although there was evidence of an initial potentiation of the M wave during the first few minutes after the fatiguing exercise.
|
200 ms (Fig. 4B) and showed no significant change after exercise or between exercise types. Paired differences before and after exercise were 11 ± 15 ms for concentric exercise and 6 ± 6 ms for eccentric exercise (n = 6 subjects). The corrected error of movement, measured as the standard deviation of the difference between the target trace and the lag-shifted tracking trace, also showed no significant change after exercise (106.2 ± 4.6% of the control period error for concentric exercise and 94.6 ± 2.5% for eccentric exercise). Finally, these corrected errors were separated, point by point, according to the direction of acceleration of the tracking movement. The ratio of these errors (Fig. 4C), which are assumed to reflect activity of biceps (positive acceleration) and triceps or gravity (negative acceleration) muscle, again showed no significant difference either before or after either exercise type.
|
| DISCUSSION |
|---|
|
|
|---|
2 min after the end of exercise, by which time there is an almost complete recovery of creatine phosphate levels (38), although an early fatigue was seen during the exercise period (Fig. 1B). Other metabolite changes, such as lactate and pH, can be expected to be more moderate following our brief exercise protocol (41). A large long-lasting fatigue was indeed observed, with the test response more than halved for both concentric and eccentric fatiguing exercise. However, the time scales of fatigue were quite different for the two types of exercise: there was full recovery from the fatigue due to concentric exercise within 2 h, but with eccentric exercise not only was the recovery less than half complete at 4 h but the onset of fatigue was slower. These recoveries roughly match those observed for quadriceps muscle (30). The protocol used to monitor long-lasting fatigue in these studies, maximal single-shock electrical stimulation, differs from conventional studies of low-frequency fatigue that use the ratio of responses to 10- and 50-Hz stimulation, with the intention of emphasizing changes related to activation per shock (calcium release or calcium sensitivity). Although this index (see Fig. 2) gave similar degrees of fatigue, there are important differences in twitch force. Peak twitch measurements are affected by postactivation potentiation and by changes in the twitch time scale. Postactivation potentiation can indeed produce large increases in twitch force (15, 32), but there is a rapid recovery largely complete within 3 min. The twitch time scale is well known to be slowed immediately after fatigue, but this also recovers within 3 min and is replaced by a very small speeding of the twitch for the duration of the low-frequency fatigue (6, 11). Finally, the 50-Hz response itself is slightly depressed during long-lasting fatigue (11, 13). Taken together, this suggests that the twitch response is an effective and sensitive index of long-term fatigue (22). One reason for differences in fatigue between concentric and eccentric fatigue is that there is a rightward shift in the ascending force-length relationship following eccentric exercise, probably from the presence of damaged sarcomeres (35). However, this effect enhances the apparent eccentric fatigue, so it cannot explain the differences observed here. It has recently been shown (6) that long-lasting fatigue can be induced by low-force as well as maximal voluntary contractions. The investigators found that 10 min of isometric contractions at 10% maximal induced a 34% reduction in twitch force at 30 min, comparable to the fatigue we saw from the 20, rather than the 40, maximal contractions we used to develop a near maximal long-lasting fatigue.
Previous studies (e.g., Refs. 6, 13, 30, 25, 31, 32 but not 11) have not shown the delayed onset of long-lasting fatigue observed here. This may be related to a difference in protocols. In the present study, the fatiguing exercise was completed within 2 min, whereas exercise periods
20 min are common in other studies, thus masking the early decline. However, one recent study (11), which had an isometric exercise period of 1 min and measurements at 3, 9, 15, 21, and 27 min afterward, showed a delayed decline in force with a time scale not dissimilar from those here. The time scale of onset differed for eccentric and concentric exercise; both showed an early rapid decline in force but a continued and slower decline with eccentric exercise. The much smaller rapid decline following eccentric contractions (
30%) compared with following concentric contractions (
70%) may be related to the much smaller energy use with eccentric contractions (1, 33, 32). If so, the time scale for the development of long-lasting fatigue (half-time 6 min) should be linked to the presence of metabolic changes following intense short-term exercise, such as inorganic phosphate (2). There have been no convincing explanations of what governs the time scale of recovery from long-lasting fatigue following nondamaging (concentric) contractions, although the implicit suggestion (3, 12) remains that it reflects the time scale to dissolve the calcium phosphate precipitation in the sarcoplasmic reticulum. One important, but not previously stated, prediction of this hypothesis depends on the much higher solubility of calcium phosphate under acid conditions. Thus recovery from long-lasting fatigue should be considerably faster during anaerobic exercise than in equally intense but aerobic exercise. This might underlie why the intermittent contractions (whether concentric or eccentric) used in this study gave a more profound fatigue than those seen in studies using continuous isometric (and thus more ischemic) contractions.
The use of electrical stimulation during the test contractions in these studies defines the fatigue investigated as peripheral. Two tests were used to see whether the fatigue depended on changes in neuromuscular or in excitation-contraction coupling. First, the size of the M wave was measured, and second stimulus-responses curves were measured hourly during the fatigue: the small changes in both suggest that the fatigue measured here is largely of muscular origin. Coupled with the observation that the fatigue was overcome by using 50-Hz stimulation, it is reasonable to conclude that the fatigue we observed was dominated by low-frequency muscle fatigue. However, it should be noted that substantial decreases of the M wave, reaching a minimum of 42% of control levels with almost full recovery by 8 h, have been reported (25). A slower recovery following eccentric compared with concentric contractions has also been seen (32). Although our measurements confirm a long-lasting depression, the M wave fell only to 84% of control and was preceded by a short period of potentiation (Fig. 3A), which may be due to electrogenic hyperpolarization (16).
There have been a variety of suggestions for the mechanisms involved in long-lasting fatigue, linked to the central observation of decreased calcium release. In their review, Allen and Westerblad (3) propose elevated inorganic phosphate to be the key factor by leading to the precipitation of calcium phosphate. Alternatively excitation coupling may be disrupted by calcium-activated proteases (40). Our experimental findings provide some evidence for supporting the former because both the concentric and eccentric exercises can be expected to result in roughly similar profiles of activation (24) and calcium release (there is a moderate decrease of EMG activity in eccentric contractions) (21) but with very different use of metabolic energy. Thus the first phase of long-lasting fatigue (20-min peak and much larger in concentric contractions) is unlikely to be due to activation alone but to depend on the amount of metabolic activity. This gives support for the calcium phosphate hypothesis in this phase of fatigue. In contrast, the second phase of long-lasting fatigue (2-h peak and largely limited to eccentric exercise) was associated with greater mechanical forces, supporting the model of a response to muscle damage induced by the stretch (26, 39). Eccentric contractions are well documented to cause physical damage (10, 28), but the early time course and mechanisms involved remain uncertain. A prevalent model suggests that calcium entry leads to protease activation (35), although a reduction in calcium entry and protease action does not lead to a reduced fatigue (5). Gross cellular damage, measured as lactate dehydrogenase release from isolated rat muscle, occurs within 60 min following eccentric contractions, and increases with the number of eccentric contractions (44) and it might thus be related to the progressive decline in force seen 2090 min after the eccentric exercise.
It has been suggested (30) that the subjective sense of weakness and instability that immediately follows intense eccentric exercise might reflect the intense long-lasting fatigue. The ability of subjects to make judgments of force decreases with muscle fatigue (14), with the sensation of force closely matching the fractional effort rather than the actual outcome (7). One might thus expect motor performance to be substantially altered in fatigue. However, it has been argued (7) that the concurrent increases in EMG activities (11) imply a premotor awareness of fatigue that functionally resets sensation so as to allow proportionate use of motor groups. The observation here of no decrement in skilled performance supports this hypothesis of an accurate knowledge of fatigue in the premotor regions. It is not easy to predict what functional tasks would be affected by low-frequency fatigue (19, 35). The fatigue observed here was dominated by large-force motor units with incomplete (i.e., low frequency) activation. In this study, a rapidly but randomly moving visual target was chosen as the task as this required only brief activation. A further issue is that damage induced by eccentric contractions disturbs proprioception (35), although in our studies the use of visual feedback lessens this factor. The motor task in our study required only moderate forces to be used to accelerate the limb mass: further studies requiring larger forces but of brief duration remain to be undertaken.
In summary, we have shown that short bursts of maximal exercise cause a delayed, insensible, and long-lasting decline in muscle force that is more profound after concentric exercise but more long-lasting after eccentric exercise. The decline is compensated for in low-force motor tasks. This compounds the lack of personal knowledge of the fatigue and may thus lead to motor instability using large-force movements.
| ACKNOWLEDGMENTS |
|---|
|
|
|---|
| FOOTNOTES |
|---|
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.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. M. Dundon, J. Cirillo, and J. G. Semmler Low-frequency fatigue and neuromuscular performance after exercise-induced damage to elbow flexor muscles J Appl Physiol, October 1, 2008; 105(4): 1146 - 1155. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. S. Turner, K. J. Tucker, N. C. Rogasch, and J. G. Semmler Impaired neuromuscular function during isometric, shortening, and lengthening contractions after exercise-induced damage to elbow flexor muscles J Appl Physiol, August 1, 2008; 105(2): 502 - 509. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |