Journal Logo

Basic Sciences: Original Investigations

Effects of Recovery Modes after Knee Extensor Muscles Eccentric Contractions


Author Information
Medicine & Science in Sports & Exercise: November 2004 - Volume 36 - Issue 11 - p 1907-1915
doi: 10.1249/01.MSS.0000145526.43208.08
  • Free


Exercise-induced muscle injury frequently occurs after unaccustomed exercise, particularly if the exercise involves a large amount of eccentric actions, that is, muscle-lengthening contractions. The symptoms of exercise-induced muscle damage include delayed onset muscle soreness (DOMS), increase in volume and circumference of the limb, decrease in the range of motion, decrease in muscular strength, leakage of intracellular proteins into the blood, fiber swelling, and damage to the sarcomere structures (7). The decreased ability of eccentric-exercised muscles to generate force can last for several days or even weeks after the muscle damaging exercise. Because the loss of strength can have functional implications in daily life or in sport, it seems important to devise a means of speeding up the recovery process.

Many researchers have investigated various treatments in an attempt to reduce the symptoms of exercise-induced damage. These interventions have included, but are not limited to, nutritional supplementation with vitamin E (1), pharmacological treatments (23), as well as physical therapies (27). Although some success has been reported by a few authors, the majority of studies indicate that no effective method has yet been found to reduce the symptoms of exercise-induced muscle injury. To date, less attention has been paid to the effect of further physical activity on muscle function. Muscle contractions are effective in increasing local blood flow via the “muscle pump effect” and the release of vasoactive substances. The increased blood flow to an injured area may help remove cellular debris, increase nutrient delivery, and hence tissue repair (13). On the other hand, one can suppose that further exercise could act as an additional damaging stimulus.

Few studies have examined the role of muscle activity on the recovery of muscle function after eccentric exercise. Some authors demonstrated that additional eccentric contractions in the days after high-force eccentric exercise had no effect on recovery of muscle function, if one excludes the immediate torque decrement observed after additional eccentric bouts (5,6,11,19). Stretch-shortening cycle (SSC)-type exercises induce muscle damage and torque decrement (26). Yet running, an SSC-type exercise, is widely used on the athletic field to alleviate symptoms of exercise-induced damage. Surprisingly, there is a lack of information regarding the effect of running on the recovery from exercise-induced muscle damage. To the best of our knowledge, only one study (25) used a repetition of running bouts in the days after a marathon run and demonstrated that exercise impaired recovery of maximal dynamic peak torque. These results are surprising because damaging eccentric exercises do not impair recovery from exercise-induced damage (5,6,11,19). However, although marathon running induces muscle damage, it has other consequences (e.g., glycogen depletion) that might have confounding influences.

Electromyostimulation (EMS) is another means of increasing muscle blood flow, especially when stimulation at low frequencies (9) is used. The hemodynamic changes elicited by voluntary and electrically evoked contractions are similar in magnitude but different in duration, with voluntary contractions demonstrating the shorter-living vasodilation (15). EMS is also widely used in clinical settings as an effective nonpharmacological modality for the treatment of pain in various etiologies (22). This hypoalgesic effect seems specific to low frequencies (28). Thus, low-frequency EMS could potentially promote the recovery process. Despite the fact that EMS has gained popularity on the athletic field as an efficient recovery method, there is no scientific evidence regarding its ability to enhance the recovery process from exercise-induced muscle damage.

The aim of this experiment was therefore to further assess the effect of running and EMS upon the recovery of muscle function and DOMS. We hypothesized that running would have no effect on recovery of muscle function, whereas EMS would prove an efficient tool to alleviate symptoms of muscle damage. Muscle function was examined using voluntary and electrically evoked contractions which allowed us to detect the origin, that is, central and/or peripheral, of the exercise-induced voluntary torque decrement.


Approach to the problem and experimental design.

Dependent variables in the study included muscle function measurements of maximal isometric voluntary torque (MVC), maximal voluntary activation level (%VA), DOMS, and evoked contractile properties of the muscle. Percent VA identifies any central fatigue that might be responsible for MVC decrement. DOMS is a commonly used marker of exercise-induced damage. From a practical point of view, it is also the simplest and most used indicator of the athlete’s physical impairment. Evoked contraction measurements involved single twitch (Pt), 20-Hz tetanus (P20), and 80-Hz tetanus (P80) peak torques. These mechanical signals were further derived to obtain maximal rates of torque development (MRFD) and relaxation (MRFR) that give insight into the contractile properties of the muscle. Finally, a P20·P80−1 ratio was calculated to identify any high- or low-frequency fatigue. We measured these variables in one group of subjects who performed a one-legged eccentric exercise on day 0 and underwent a recovery intervention during the four following days. Variable measurements were performed before and 30 min, 24 h, 48 h, and 96 h after the exercise. Dependent variable data were analyzed as a function of independent variables, that is, time and recovery interventions.

We designed this experiment in a practical way. Our aim was to reproduce the conditions of the athletic field. As a consequence, we selected physically active subjects and the recovery interventions matched the athletes’ commonly observed practices. Recovery interventions consisted in a daily (i) 30-min run at 50% maximal oxygen uptake associated velocity, (ii) 30-min stimulation session of the lower limb muscles using low-frequency EMS at a self-chosen intensity, and (iii) passive recovery intervention. As the group underwent each recovery intervention, each subject exercised three times, 3 wk apart. Moreover, we had to free our measurements from the confounding influence of the repeated bout effect (4). For that reason, each subject exercised with one lower limb in the first exercise bout and the contralateral limb in the second bout. The first limb exercised was finally used for the third bout. As a result, only one limb exercised two times, 6 wk apart.

MVC were performed on each testing day to examine the pattern of torque recovery. Percent VA and evoked contractions enabled us to determine central and peripheral contributions to these recovery kinetics. We expected that running would not influence the recovery process, whereas EMS might enhance it.


Eight healthy male subjects took part in this study. None of the subjects had a history of neuromuscular or vascular disease. All were physically active (see Table 1) and were asked to refrain from strenuous exercise during the week preceding the protocol. Despite their trained status, participants were nonweight-trained individuals who had not been involved in any resistance-training program for at least 6 months before participation in the present study. The experiment was conducted according to the Declaration of Helsinki. The participants were fully informed of the procedure and the risks involved in this study, and gave their written consent. They were also allowed to withdraw from the study at will. Approval for the project was obtained from the local committee on human research.

Individual physical characteristics and training habits.

Experimental protocol.

The fatiguing exercise consisted of intermittent one-legged downhill running on a motorized treadmill (EF 1800, Tecmachine, Andrézieux-Bouthéon, France) at a speed of 7 km·h−1 with a 12% negative incline. The subject had to hop forward on one foot and “brake” himself during each contact phase. The contralateral limb, which was not involved in the exercise, was flexed and swayed freely. Each subject performed 15 bouts of 1-min duration followed by a 30-s passive recovery period. The fatiguing exercise designed in this study was more demanding than the commonly used maximal eccentric contractions. Because our subjects were active, we had to choose a highly strenuous exercise to induce large MVC decrease. The one-legged running task was chosen because in preliminary experiments we failed to induce great impairment of knee extensor muscles (KE) torque when performing maximal eccentric contractions on an isokinetic ergometer. As the subjects exercised roughly at a frequency of 2 Hz and carried out 15 repetitions of 1-min duration, each subject achieved about 1800 submaximal eccentric contractions. During the run, each subject wore a chest safety harness. The rope fastened to the harness was not tight so that the subject had to bear his weight but could nevertheless be securely retained in case of a fall.

MVC and neuromuscular measurements were performed before (PRE), 30 min (POST), 24 h, 48 h, and 96 h after the exercise. Pain measurements were also performed using a visual analog scale. The fatiguing exercise was performed on day 0; from day 1 to day 4, each subject performed a daily 30-min recovery intervention consisting either in (i) running (RUN), (ii) stimulating their KE and plantarflexor (PF) muscles using large surface electrodes (EMS), or (iii) no recovery intervention (CONTR). The exercise and testing sessions were performed in the morning and in the afternoon, respectively. With each participant acting as his own control, they randomly exercised three times to undergo all three recovery interventions. One lower limb was involved in the first exercise bout and the contralateral limb in the second bout. The first limb exercised was finally used for the third bout. As a result, only one limb exercised two times, 6 wk apart. Moreover, they underwent all the recovery interventions in a randomized order. After each bout, the involved limb recovered for 6 wk to limit the confounding influence of the repeated bout effect that can last for up to 6 wk in KE after downhill running (4). Individual MVC data are presented in Table 2. Two-way ANOVA, with repeated measures performed on data of repeated bouts, revealed no time × bout interaction, suggesting that 6 wk was an adequate time period to avoid any repeated bout effect. The lack of repeated bout effect was likely due to the fact that (i) KE muscles regularly perform eccentric contractions during everyday activities, and (ii) our population especially regularly practiced physical activities involving repetitive SSC-contractions of the KE muscles.

Individual values for MVC (N·m) for bouts 1, 2, 3 at before (PRE), 30 min (POST), 24 h, 48 h, and 96 h after the exercise; the subjects exercised alternatively either with the right (R) or the left (L) leg.

Recovery interventions.

In the RUN condition, the participants exercised for 30 min on a motorized treadmill (EF 1800, Tecmachine) at 50% of the speed associated with maximal oxygen uptake, which was determined using a continuous, incremental test during a preliminary visit with simultaneous measurement of oxygen uptake. The test was performed on the motorized treadmill and began at a speed of 9 km·h−1 followed by a gradual increase of 1 km·h−1 every 2 min until exhaustion. The last stage completely performed was considered as the speed associated with maximal oxygen uptake. The EMS intervention consisted of low-frequency stimulation of KE and PF (8 Hz, pulse width 400 μs) at a self-chosen, strong but comfortable intensity (i.e, 20–30 mA) using a commercially available stimulator (Compex Sport-P, Compex SA, Ecublens, Switzerland) for 30 min. From POST to 96 h, the subjects were asked to refrain from stretching or any other recovery intervention or medication.

Measurements of muscle contractile function.

Before each testing session, the subjects performed a 10-min warm-up on a cycle ergometer (Excalibur, Lode, Groningen, The Netherlands) at a self-selected power output. All muscle contractile measurements were conducted on KE in isometric conditions using a Biodex isokinetic ergometer (Biodex System 3, Biodex Corporation, Shirley, NY). Subjects were seated with the trunk vertical and the knee and hip fixed at a 90° angle (0° = full extension). Velcro straps were applied tightly across the thorax and around the ankle so as to fix it to a foam pad placed on the lever arm.

Torque measurements.

MVC measurement involved two trials. The subjects were strongly encouraged, and the best result was used for further analysis. Maximal voluntary activation (%VA) was estimated using a technique based on the interpolated-twitch method. Briefly, an electrically evoked twitch was superimposed on the isometric plateau. The ratio of the amplitude of the superimposed twitch over the size of the twitch in the relaxed muscle (control twitch) was then calculated to obtain %VA as suggested by Strojnik and Komi (26):

where MVC is the maximal torque level and Tb the torque level just before the superimposed twitch.

The mean control twitch was a potentiated twitch that was measured as explained below. The difficulty of the interpolated-twitch method is applying the superimposed twitch at the maximal torque level. To bypass this limit, Strojnik and Komi (26) have included a correction in the original equation for the cases where the twitch is applied when the torque level is already slightly declining. The amplitude of the superimposed twitch is reevaluated, taking into account the torque level before the superimposed twitch and the maximal torque level. The theoretical basis of this correction stems from the fact that the relation between the torque level and the superimposed twitch is linear. According to Norregaard et al. (18), this is the case for KE at torque levels above 25% of MVC. The correction in the original equation was used only for trials where Tb ≥ 95% Tmax. Other trials were discarded from the analysis.

Electrical stimulation.

Electrical stimulation was applied percutaneously to the femoral nerve using a ball probe cathode pressed in the femoral triangle, 3–5 cm below the inguinal ligament. The anode, a 10- by 5-cm self-adhesive stimulation electrode (Compex SA, Ecublens, Switzerland), was located in the gluteal fold. A high-voltage stimulator (Digitimer DS7, Hertfordshire, UK) was used to deliver square-wave stimulus of 500-μs duration, 400-V maximal voltage, and intensity ranging from 50 to 110 mA. The optimal intensity of stimulation was set by progressively increasing the stimulus intensity until the maximal isometric twitch torque was reached. The same intensity was used for the other testing sessions. The electrically evoked force measurements comprised two 0.5-s tetani at a frequency of 80 and 20 Hz, then three single twitches. A 30-s recovery period was allowed between each stimulation train. The single twitches were separated by 3 s. The tetani were applied only at PRE, POST, and 48 h. In pilot experiments, we observed that the 0.5-s stimulation train duration was sufficient to obtain torque values similar to those obtained from stimulation trains of 1-s duration. We chose the shorter stimulation duration, sufficient to maximally activate KE, in order to limit the discomfort induced by high-frequency tetanus. The last three twitches were averaged to obtain a mean twitch. This potentiated twitch was considered as the control twitch for the calculation of %VA and for the twitch comparison between the different sessions.

The following parameters were obtained from the mechanical response of the evoked twitch: (i) peak twitch torque (Pt), that is, the maximal isometric twitch torque; (ii) maximal rate of twitch force development (MRFDtw), that is, the maximal value of the first derivative of the torque signal; and (iii) maximal rate of twitch force relaxation (MRFRtw), that is, the lowest value of the first derivative of the mechanical signal. The following parameters were obtained from the mechanical response of the evoked tetani: (i) peak torque (P80 and P20, respectively), that is, the highest value of tetanus torque production; (ii) maximal rate of tetanus force development (MRFD80 and MRFD20, respectively), that is, the maximal value of the first derivative of the mechanical signal; and (iii) maximal rate of tetanus force relaxation (MRFR80 and MRFR20, respectively), that is, the lowest value of the first derivative of the torque signal. Finally, a P20·P80−1 ratio was calculated to identify any high- or low-frequency fatigue. All mechanical data were stored using commercially available software (Tida 4.11, Heka Electronik, Lambrecht/Pfalz, Germany).

Pain measurement.

KE soreness was evaluated using a visual analog scale consisting in a 100-mm horizontal line with an item at each extremity: from “no pain” to “very, very sore.” Subjects were asked to put a vertical mark on the horizontal line to describe the pain experienced during daily locomotion. The distance between the origin of the scale and the vertical mark was used as the pain score.

Statistical analysis.

The data, except the soreness values, were normalized to rest values. All descriptive statistics presented are mean values ± SD. Normal distribution was checked using a Shapiro-Wilk test of normality. Each study variable was then compared between the different testing conditions using two-way ANOVA with repeated measures. Newman-Keuls post hoc tests were applied to determine between-means differences if the analysis of variance revealed a significant main effect for time or interaction of recovery intervention × time. For all statistical analyses, a P value of 0.05 was accepted as the level of significance. Statistical power values were also calculated for various significant differences and ranged from 0.1 to 0.9 (see Table 3). All the statistical analyses were performed with the Statistica 6.0 software for Windows.

Statistical power associated with the two-way ANOVA with repeated measures; size effect corresponds to the number of subjects needed to obtain respective power values at 0.80 (commonly accepted) for each variable; data are presented when statistical significant differences (P < 0.05) were found (N = 8).


The recovery time course for the different variables did not differ significantly among RUN, EMS, and CONTR conditions. However, all parameters demonstrated significant time effects.

Voluntary contractions and muscle soreness.

MVC decreased 30 min after the exercise (P < 0.001) and did not recover during the 4 d after the fatiguing bout (see Fig. 1, upper panel). Postexercise MVC decrements were not significantly different between the RUN, EMS, and CONTR conditions (−17.0 ± 9.8%, −18.0 ± 5.8%, and −14.4 ± 12.9%, respectively).

FIGURE 1—Recovery time courses for isometric maximal voluntary contraction (MVC; upper panel), voluntary activation level (%VA; middle panel), and delayed onset muscle soreness (DOMS; lower panel) observed when using running (RUN), electromyostimulation (EMS) or passive (CONTR) recovery modes. Values are means ± SE. Significantly different from baseline values: *:
P < 0.05; ** P < 0.01; *** P < 0.001. Significantly different from 24-h values: §§§ P < 0.001. Significantly different from 48-h values: ††† P < 0.001.

The voluntary activation level displayed a significant time effect (P < 0.05): it slightly decreased at POST for the three conditions, but had significantly recovered 24 h after the damaging bout (see Fig. 1, middle panel). Postexercise activation level decrements were not significantly different between the RUN, EMS, and CONTR conditions (−5.0 ± 5.6%, −9.5 ± 13.1%, and −4.0 ± 7.5%, respectively).

Muscle soreness showed a significant time effect (P < 0.001). DOMS was significantly elevated above baseline values from 24 to 96 h (see Fig. 1, lower panel), and peaked at 48 h (5.4 ± 2.5; P < 0.001). At 96 h, muscles were still sore, but DOMS had significantly decreased as compared with 24 h and 48 h postexercise (P < 0.001).

Electrically evoked contractions.

P20, P80, and the P20·P80−1 ratio showed significant differences over time (P < 0.001, P < 0.01, and P < 0.001, respectively). Figure 2 demonstrates that all these variables were reduced 30 min after the downhill run. The mechanical responses to tetanic stimulation displayed different recovery patterns. P80 was still below baseline values at 48 h (P < 0.01), whereas P20 was also reduced 48 h postexercise, but significantly higher than POST values (P < 0.001). Finally, the P20·P80−1 ratio had completely recovered 48 h after the damaging exercise. Interestingly, MVC, P20, and P80 decrements were not statistically different (−9.6 ± 14.5%, −13.2 ± 14.2%, and −12.3 ± 11.3%, respectively) at 48 h, whereas P20 was significantly more altered at POST than MVC and P80 (−44.6 ± 19.9%; P < 0.001 vs −16.4 ± 8.3% and −16.3 ± 15.1%, respectively).

FIGURE 2—Recovery time courses for mechanical responses to high- (P80; upper panel) and low-frequency tetani (P20; middle panel) observed when using running (RUN), electromyostimulation (EMS), or passive (CONTR) recovery modes. Lower panel displays the same information for the P20·P80−1 ratio. Values are means ± SE. Significantly different from baseline values: *:
P < 0.05; ** P < 0.01; *** P < 0.001. Significantly different from POST values: §§§ P < 0.001.

There was a significant difference in the measurement of Pt over time (P < 0.001; see Fig. 3). Pt decreased at POST (P < 0.001), and had already recovered at 24 h.

FIGURE 3—Recovery time course for single twitch for the different recovery interventions: running (RUN), electromyostimulation (EMS), or passive (CONTR) recovery. Values are means ± SE. Significantly different from baseline values: ***:
P < 0.001.

Finally, changes in maximal rates of torque development and relaxation are summarized in Table 4. Maximal rates of torque development and relaxation of the single twitch and stimulation trains demonstrated significant time effects (P < 0.001).

Maximal rates of force development (MRFD) and relaxation (MRFR) of the mechanical responses to tetanic and single stimulations measured 30 min (POST), 1 d (24 h), 2 d (48 h), and 4 d (96 h) after the fatiguing exercise.


The results reveal that the recovery modes tested in the present study had no effect on voluntary and electrically evoked torque recovery time courses in trained individuals. Four days after exercise, MVC was still reduced. This prolonged torque loss is mainly due to peripheral alterations, namely muscle damage and low-frequency fatigue. Interestingly, the latter phenomenon is certainly not involved from 48 to 96 h because the P20·P80−1 ratio and Pt showed complete recovery at 48 h.

Lack of effect of recovery interventions.

Sherman et al. (25) previously demonstrated significant impairment of maximal KE dynamic peak torque recovery in subjects performing a 20- to 45-min daily running exercise during the 7 d after a marathon run. Conversely, our results suggest that a 30-min daily run did not impair MVC recovery from eccentric exercise. In the present experiment, the lack of effect of SSC contractions as an additional damaging stimulus is rather consistent with studies demonstrating that further eccentric contractions are not detrimental to muscle function recovery (5,6,11,19). Thus, it seems that the effects of running on recovery differ according to whether muscle damage is due to pure eccentric exercise or to marathon running.

In the present study, we found no benefit in performing EMS contractions during recovery. Denegar et al. (10) previously reported a significant reduction of perceived pain and increased range of motion when applying low-frequency EMS on damaged elbow flexors. Conversely, Craig et al. (8) did not find any measurable effect of EMS on DOMS and range of motion. The inefficiency of EMS intervention could, at least partly, arise from methodological problems. As shown in Figure 1 (upper panel), MVC displayed a high variability in the EMS condition, especially at 96 h. This might be due to the fact that the subjects chose the stimulation intensity themselves and/or to the low statistical power of MVC at 96 h. Although they were instructed to choose a “strong but comfortable” intensity, we cannot be certain that every participant evoked optimal hemodynamic changes within the damaged muscles. Excessive intensity might have led to partial ischemia and additional damage, whereas insufficient intensity might not have been adequate to significantly increase local muscle blood flow. Further studies are needed to accurately determine individual optimal stimulation parameters effective in maximizing the muscle pump effect before evaluating EMS efficiency in eccentric-exercise symptom relief.

Torque recovery time course.

The fact that MVC decrement observed at POST was still partially present at 96 h is in line with previous work on KE response to eccentric exercise. Many studies have demonstrated that KE MVC can be reduced for 5–7 d after the fatiguing exercise (17). Other studies demonstrated that MVC had returned to rest values 4 d after the damaging bout (27). Such discrepancies may arise from differences in exercise characteristics (i.e., number of repetitions, contraction intensity and velocity, initial muscle length, and mode of exercise).

In the present experiment, the long-lasting MVC decrease cannot be attributed to a central limitation. Central fatigue contributed to the torque loss observed 30 min after the exercise. However, activation level was not different from rest values at 24, 48, and 96 h. Similar results have been reported by Newham et al. (16). These authors showed that subjects were able to fully activate their sore muscles during the days after a damaging exercise. Michaut et al. (14) observed the same phenomenon in the elbow flexors. In the absence of any central limitation, peripheral alterations are responsible for voluntary torque decrement. These might include low-frequency fatigue, muscle damage, and impaired plasmalemmal action potential conduction.

The latter phenomenon is generally considered as a minor factor in eccentric-contraction-induced injury (30). Accordingly, plantarflexor muscle M-wave was unchanged both immediately and 24 h after a 20-min downhill run (3). Conversely, Sayers et al. (24) reported reduced M-wave responses up to 9 d after eccentric-exercise involving elbow flexors of untrained individuals. Discrepancy between these results and ours may arise from the muscle group involved in the exercise (that is, lower- vs upper-limb muscle) and physical status of the subjects. Determining which of the two, muscle damage or low-frequency fatigue, represents the major contributor to torque decrement still remains a point of controversy in eccentric exercise literature (20). According to our results, low-frequency fatigue significantly contributes to torque decrement from POST to 48 h after the fatiguing exercise, as shown by the preferential alterations in low-frequency mechanical response, single-twitch, and P20·P80−1 ratio. Conversely, Brown et al. (2) reported that the low- to high-frequency ratio was depressed for 3 d after a stimulated eccentric exercise involving KE muscles. The finding that electrically induced exercise produced more pronounced low-frequency fatigue compared with voluntary exercise (21) may explain the discrepancy between our results and those of Brown et al. (2). Newham et al. (17) reported that the low-to-high frequency ratio had recovered 24 h after a bench-stepping exercise. In the present study, this ratio was not measured 24 h after the exercise, but the single twitch recovery time course suggests that low-frequency fatigue was absent at this point in time. Low-frequency fatigue is usually associated with failure of excitation-contraction (E-C) coupling (12). This alteration of intracellular Ca2+ kinetics may be partly responsible for the observed alterations of maximal rates of torque development and relaxation. Nevertheless, owing to the low statistical power of these variables, these results should be interpreted with caution.

At 48 h, P20, P80 and MVC displayed a uniform torque loss, with concurrent peak of DOMS. This general alteration of muscle function is also apparent when looking at rates of torque development recovery time courses. This might represent the implication of a single and powerful factor, that is, that damage to the sarcomere structures and the accompanying inflammation process could be mainly involved at this point in time.

A clear sequence of events contributing to strength loss after voluntary eccentric-contraction-induced injury has never been described in trained humans. Our results suggest that an alteration of the E-C coupling might be involved during the first 2 d after the damaging bout. From 2 to 4 d after the exercise, damage to force-generating structures could account for the remaining torque deficit. Using animal models, Warren et al. (29) stated that over the first 3 d of injury, most (∼75%) of the strength loss is ascribed to a failure of the E-C coupling pathway. The remainder of the strength loss is attributed to damage to the force-bearing elements within the muscle. From 5 to 14 d postinjury, the proportion of the strength deficit accounted for by contractile protein loss increases from 40 to 45% to almost 100%. The authors hypothesized that this model may differ qualitatively from that elicited by submaximal, voluntary eccentric contractions performed by humans. Our results suggest that the sequence of events is similar in trained individuals but demonstrates time differences. These findings are strictly limited to this specific population and may not be representative of the typical response to eccentric exercise in untrained subjects, who may demonstrate longer time courses of recovery. Moreover, the statistical power values of MVC (at 48 and 96 h), %VA (at POST), P80 (at POST and 48 h), and P20 (at 48 h) did not reach a level (0.80) commonly accepted as adequate to achieve statistical power. Thus, some of these results should be interpreted with caution.


This study was aimed at evaluating the benefit of performing running or EMS to hasten the recovery process from knee extensor muscle eccentric-contraction–induced injury in trained individuals. The results reveal that the recovery interventions tested here have no effect on voluntary and electrically evoked torque recovery time courses. Further studies are needed to optimize electrical stimulation patterns before evaluating its efficiency as a recovery means. MVC did not recover completely during the 4 d after the damaging bout. This prolonged torque loss is mainly due to peripheral alterations, that is, muscle damage and low-frequency fatigue. Our results suggest that an alteration of the E-C coupling might be involved during the first 2 d after the eccentric exercise. From 2 to 4 d after the muscle insult, damage to force-generating structures could account for the remaining torque deficit. These findings are in line with previous work using animal models but may be strictly limited to the specific population studied here. Generalization of these findings to the eccentric exercise literature requires further studies involving untrained subjects.


1. Beaton, L. J., D. A. Allan, M. A. Tarnopolsky, P. M. Tiidus, and S. M. Phillips. Contraction-induced muscle damage is unaffected by vitamin E supplementation. Med. Sci. Sports Exerc. 34:798–805, 2002.
2. Brown, S. J., R. B. Child, A. E. Donnelly, J. M. Saxton, and S. H. Day. Changes in human skeletal muscle contractile function following stimulated eccentric exercise. Eur. J. Appl. Physiol. Occup. Physiol. 72:515–521, 1996.
3. Bulbulian, R., and D. K. Bowles. Effect of downhill running on motoneuron pool excitability. J. Appl. Physiol. 73:968–973, 1992.
4. Byrnes, W. C., P. M. Clarkson, J. S. White, S. S. Hsieh, P. N. Frykman, and R. J. Maughan. Delayed onset muscle soreness following repeated bouts of downhill running. J. Appl. Physiol. 59:710–715, 1985.
5. Chen, T. C., and S. S. Hsieh. Effects of a 7-day eccentric training period on muscle damage and inflammation. Med. Sci. Sports Exerc. 33:1732–1738, 2001.
6. Chen, T. C., and S. S. Hsieh. The effects of repeated maximal voluntary isokinetic eccentric exercise on recovery from muscle damage. Res. Q. Exerc. Sport 71:260–266, 2000.
7. Clarkson, P. M., and M. J. Hubal. Exercise-induced muscle damage in humans. Am. J. Phys. Med. Rehabil. 81:S52–S69, 2002.
8. Craig, J. A., M. B. Cunningham, D. M. Walsh, G. D. Baxter, and J. M. Allen. Lack of effect of transcutaneous electrical nerve stimulation upon experimentally induced delayed onset muscle soreness in humans. Pain 67:285–289, 1996.
9. Cramp, A. F., C. Gilsenan, A. S. Lowe, and D. M. Walsh. The effect of high- and low-frequency transcutaneous electrical nerve stimulation upon cutaneous blood flow and skin temperature in healthy subjects. Clin. Physiol. 20:150–157, 2000.
10. Denegar, C. R., D. H. Perrine, A. D. Rogol, and R. Rutt. Influence of transcutaneous nerve stimulation on pain, range of motion, and serum cortisol concentration in females experiencing delayed onset muscle soreness. J. Orthop. Sports Phys. Ther. 11:100–103, 1989.
11. Ebbeling, C. B., and P. M. Clarkson. Muscle adaptation prior to recovery following eccentric exercise. Eur. J. Appl. Physiol. Occup. Physiol. 60:26–31, 1990.
12. Jones, D. A. High-and low-frequency fatigue revisited. Acta Physiol. Scand. 156:265–270, 1996.
13. MacAuley, D. C. Ice therapy: how good is the evidence? Int. J. Sports Med. 22:379–384, 2001.
14. Michaut, A., M. Pousson, Y. Ballay, and J. Van Hoecke. Eccentric exercise session effects on muscle contractility short term recovery. J. Soc. Biol. 194:171–176, 2000.
15. Miller, B. F., K. G. Gruben, and B. J. Morgan. Circulatory responses to voluntary and electrically induced muscle contractions in humans. Phys. Ther. 80:53–60, 2000.
16. Newham, D. J., D. A. Jones, and P. M. Clarkson. Repeated high-force eccentric exercise: effects on muscle pain and damage. J. Appl. Physiol. 63:1381–1386, 1987.
17. Newham, D. J., K. R. Mills, B. M. Quigley, and R. H. Edwards. Pain and fatigue after concentric and eccentric muscle contractions. Clin. Sci. (Lond.) 64:55–62, 1983.
18. Norregaard, J., J. J. Lykkegaard, P. M. Bulow, and B. Danneskiold-Samsoe. The twitch interpolation technique for the estimation of true quadriceps muscle strength. Clin. Physiol. 17:523–532, 1997.
19. Nosaka, K., and M. Newton. Repeated eccentric exercise bouts do not exacerbate muscle damage and repair. J. Strength Cond. Res. 16:117–122, 2002.
20. Proske, U., and D. L. Morgan. Muscle damage from eccentric exercise: mechanism, mechanical signs, adaptation and clinical applications. J. Physiol. 537:333–345, 2001.
21. Ratkevicius, A., A. Skurvydas, E. Povilonis, B. Quistorff, and J. Lexell. Effects of contraction duration on low-frequency fatigue in voluntary and electrically induced exercise of quadriceps muscle in humans. Eur. J. Appl. Physiol. Occup. Physiol. 77:462–468, 1998.
22. Rushton, D. N. Electrical stimulation in the treatment of pain. Disabil. Rehabil. 24:407–415, 2002.
23. Sayers, S. P., C. A. Knight, P. M. Clarkson, E. H. Van Wegen, and G. Kamen. Effect of ketoprofen on muscle function and sEMG activity after eccentric exercise. Med. Sci. Sports Exerc. 33:702–710, 2001.
24. Sayers, S. P., B. T. Peters, C. A. Knight, M. L. Urso, J. Parkington, and P. M. Clarkson. Short-term immobilization after eccentric exercise. Part I: contractile properties. Med. Sci. Sports Exerc. 35:753–761, 2003.
25. Sherman, W. M., L. E. Armstrong, T. M. Murray, et al. Effect of a 42.2-km footrace and subsequent rest or exercise on muscular strength and work capacity. J. Appl. Physiol. 57:1668–1673, 1984.
26. Strojnik, V., and P. V. Komi. Neuromuscular fatigue after maximal stretch-shortening cycle exercise. J. Appl. Physiol. 84:344–350, 1998.
27. Tiidus, P. M., and J. K. Shoemaker. Effleurage massage, muscle blood flow and long-term post-exercise strength recovery. Int. J. Sports Med. 16:478–483, 1995.
28. Walsh, D. M., C. Liggett, D. Baxter, and J. M. Allen. A double-blind investigation of the hypoalgesic effects of transcutaneous electrical nerve stimulation upon experimentally induced ischaemic pain. Pain 61:39–45, 1995.
29. Warren, G. L., C. P. Ingalls, D. A. Lowe, and R. B. Armstrong. Excitation-contraction uncoupling: major role in contraction-induced muscle injury. Exerc. Sport Sci. Rev. 29:82–87, 2001.
30. Warren, G. L., C. P. Ingalls, S. J. Shah, and R. B. Armstrong. Uncoupling of in vivo torque production from EMG in mouse muscles injured by eccentric contractions. J. Physiol. 515(Pt 2):609–619, 1999.


©2004The American College of Sports Medicine