Neuromuscular electrical stimulation (NMES) has been frequently employed as a modality for strength training in healthy humans (11,19) and athletes (18). It has indeed been demonstrated that maximal voluntary strength can be effectively increased after multiple sessions of NMES resistance training, especially for the quadriceps femoris muscle (3). Strength gains associated with an NMES training program have been recently attributed to an increase in electromyographic (EMG) activity and activation level (11,19) as well as to muscle hypertrophy (11).
Most studies have focused on the effects of NMES training on voluntary strength development. To our knowledge, only a few studies (6,15,20) have investigated the effects of multiple sessions of NMES on muscle fatigue, and their results remain contradictory. For example, Duchateau and Hainaut (6), using an electrical fatigue test, reported that muscle fatigue remained unchanged after 6 wk of NMES training. In contrast, Kim et al. (15) showed that the endurance time for a cycling exercise increased after NMES programs lasting 4 wk. These discrepancies could be ascribed to the criteria used to assess muscle fatigue. Indeed, muscle fatigue can refer to a failure to maintain the required torque (8) as well as any exercise-induced reduction in the ability of a muscle to generate maximal voluntary force (10). By using a submaximal sustained isometric contraction as a tool to investigate muscle fatigue, it is therefore possible to obtain two indexes of muscle fatigue: endurance time and maximal voluntary contraction (MVC) loss induced by the exercise.
Endurance time for a submaximal isometric contraction may be influenced by gender (4,12,13), muscle length (23), or limb immobilization (27). For example, the endurance time for a contraction maintained at 15-20% of MVC was longer in women than in men (13) and in short compared with longer muscle length (23). Such differences in endurance time have been explained, at least in part, by the difference in maximal absolute force and thus by the different absolute target force for the sustained contraction performed at a given relative intensity of MVC (4,12,13,23). Although it is not a universal finding (27), several authors (4,12,13) have reported that the endurance time for a submaximal isometric task was inversely related to the absolute contraction intensity. It can, therefore, be hypothesized that the NMES training-induced strength gains will reduce the endurance time for a submaximal isometric contraction.
Despite differences in endurance time, many studies reported that the decline in MVC torque recorded at the end of the fatiguing contraction, which represents a second index for the amount of fatigue experienced during exercise (10), was not affected by gender or muscle length (13,23). For example, Place et al. (23) observed a similar MVC loss at both 35 and 75° knee joint angles despite a greater endurance time at the shorter muscle length (i.e., 35°). These authors also reported a similar central fatigue but a different magnitude of changes in the behavior of contractile properties for the two joint angles after the sustained contraction, thus indicating that the central and peripheral contributions to fatigue were different in the two testing conditions. Furthermore, Nordlund et al. (21) recently demonstrated, using an intermittent fatiguing protocol, that the higher the plantar flexor torque and the level of maximal voluntary activation, the higher the peripheral fatigue. In contrast, the development of central fatigue was not influenced by these two factors. By using a NMES training model in which the strength gains have been ascribed to neural adaptations in the first 4 wk of training and to muscular changes between weeks 4 and 8 of training (11), it would therefore be possible to distinguish whether NMES training-induced neuromuscular adaptations affect the relative contributions of central and peripheral factors to fatigue during a submaximal sustained effort.
The first purpose of the present study was therefore to investigate the effects of 4 and 8 wk of NMES training on the endurance time of a submaximal isometric fatiguing contraction performed at the same relative intensity by the knee extensor muscles. A secondary aim was to examine the effects of NMES training on mechanisms (i.e., central vs peripheral) contributing to task failure.
Approach to the Problem and Experimental Design
This study used data collected, but not analyzed, in an earlier study of neuromuscular adaptations to NMES training (11). The present experiment was conducted to examine the effects of 4 and 8 wk of NMES training on the endurance time for a submaximal isometric fatiguing contraction performed at the same relative intensity. The fatiguing contraction of the knee extensor muscles was performed at a target value of 20% MVC torque. The EMG activity and muscle activation obtained under MVC were recorded before and after the fatiguing task to assess central fatigue. Torque and EMG responses obtained under electrically evoked contractions were examined before and after the fatiguing contraction to analyze peripheral fatigue. During the fatiguing contraction, the EMG activity was quantified at consecutive sampling intervals that represented 10% of the endurance time. Eighteen subjects were tested 2 wk (WK-2) prior to baseline and before the training program to assess interreliability measurements for the main parameters. Then, as 10 subjects underwent a NMES training program, the other eight were not retested for the fatiguing contraction but served as control for a previous investigation (11). The 10 subjects were tested before (B), after 4 (WK4), and after 8 (WK8) wk of NMES training. The training program consisted of 32 × 18-min sessions of isometric (bilateral) NMES over an 8-wk period, with four sessions per week. Three to four days of rest separated the 16th and the 32nd training sessions from the WK4 and WK8 testing sessions, respectively. All measurements were carried out on the dominant leg, which was the right leg for all subjects. The independent variables were the session (i.e., B, WK4, and WK8) and the time (before and after the fatiguing contraction) at which the measurement was taken. Dependant variables were MVC torque, EMG activity and muscle activation obtained during MVC, evoked contractions (twitch stimulations) and associated maximal M wave, endurance time, and EMG activity recorded during the fatiguing contraction.
Data on 10 subjects (mean ±SD: age 24 ± 5 yr, body height 178 ± 7 cm, body mass 74 ± 9 kg) were used in this study. Before participation, each subject was informed of the purpose and potential risks of the study and gave written voluntary consent. None of them had engaged in systematic strength training or NMES in the 12 months preceding the beginning of the experiments, but some were active in recreational sports. Approval for the project was obtained from the University of Burgundy committee on human research. All procedures used in this study were in conformity with the Declaration of Helsinki.
Eighteen subjects were tested at WK-2 and B to assess interreliability measurements for the main parameters. Then, only the 10 subjects were required to perform the same protocol with the right knee extensor muscles at B, at WK4, and at WK8. The experimental protocol was as follows (Fig. 1): 1) four electrically evoked twitches separated by 8 s, during which subjects were asked to relax; 2) knee extensor MVC torque assessment with doublets superimposed over the isometric plateau (superimposed doublet) and 3 s after the contraction (potentiated doublet) to assess maximal voluntary activation level (VAL) according to the twitch interpolation technique (1); 3) knee flexor MVC; 4) four electrically evoked twitches separated by 8 s in resting conditions (considered as PRE); and 5) an isometric contraction of the knee extensor muscles sustained at 20% of the corresponding MVC torque until failure. Four single twitches were delivered as described above immediately after the sustained contraction (i.e., POST), then steps 2 and 3 were also performed at the end of the fatiguing contraction. The POST MVC was performed approximately 30 s after the end of the sustained submaximal contraction.
Instantaneous isometric torque at the knee joint was recorded using a Biodex isokinetic dynamometer (Shirley, NY). Subjects were placed in a seated posture with the trunk-thigh angle at 90° and the knee flexed at 60° (where 0° corresponds to the full extension of the knee). Each subject was securely strapped to the test chair by two crossover-shoulder harnesses and a belt across the hip joint. The axis of the dynamometer was aligned with the anatomical knee joint axis, and the lever arm was attached 2-3 cm above the lateral malleolus with straps. To allow biceps femoris (BF) EMG recordings, a board (thickness ∽ 3 cm) was placed underneath the subject with a hole where the electrodes were placed to avoid any compression between the surface electrodes and wire on the seat. Subjects were asked to cross their arms during the testing procedure. Gravity correction was obtained by measuring the torque exerted on the dynamometer lever arm by the weight of the limb. Once measured, the effect of gravity was automatically incorporated by the Biodex Advantage Software program in the output provided (30). Torque signals were digitized online (sampling frequency: 2 kHz) by using a digital computer (IPC 486).
Each subject performed two maximal 5-s isometric contractions of the knee extensor muscles before the fatiguing contraction, separated by a 60-s rest period between each trial. Only one trial was performed after the fatiguing contraction to avoid fatigue recovery from the fatigued state. Strong verbal encouragement was given to the subjects during each MVC. Subjects were also provided with visual feedback of the knee extensor muscle torque on a monitor that was placed 1 m in front of them. The greatest torque achieved by the subject was taken as the MVC torque and was used for calculation of submaximal target torques. The same procedure was adopted for assessment of the knee flexor MVC to obtain the peak EMG from the BF muscle and to assess coactivation during the fatiguing contraction.
Each subject performed an isometric fatiguing contraction at a target value of 20% MVC torque as determined from MVC performed on that day. This intensity was chosen in accordance with previous studies performed in our laboratory (23,25). Visual feedback of the torque exerted was displayed on the monitor. The fatiguing contraction was terminated when the torque fell below the required target for a 3-s period despite strong verbal encouragement by the investigators. This time was recorded as the endurance time. Subjects were not informed of the endurance time until completion of the third session.
Transcutaneous electrically evoked contractions were induced by using a high-voltage (maximal voltage: 400 V) constant-current stimulator (model DS7, Digitimer, Hertfordshire, UK). The femoral nerve was stimulated using a monopolar cathode ball electrode (0.5-cm diameter) pressed into the femoral triangle by the same experimenter. The anode was a 50-cm2 (10 × 5 cm) rectangular electrode (Compex Medical SA, Ecublens, Switzerland) located in the gluteal fold opposite the cathode. The individual stimulation intensity was set by progressively increasing the stimulus intensity until there was no further increase in peak twitch torque (i.e., the highest value of the knee extensor twitch torque) nor in concomitant peak-to-peak M-wave amplitudes. This intensity was further increased by 20% (i.e., supramaximal intensity) and then maintained for single and paired stimulations. Individual supramaximal intensities were between 45 and 100 mA. The stimulus duration was 1 ms, and the interval of the stimuli in the doublet was 10 ms.
The EMG activity of the vastus lateralis (VL), vastus medialis (VM), rectus femoris (RF), and BF was recorded with pairs of circular (recording diameter: 10 mm) silver chloride surface electrodes positioned lengthwise over the middle of the muscle belly with an interelectrode (center to center) distance of 20 mm. This site was determined in pilot testing by eliciting at a given intensity the greatest M-wave amplitude for each muscle via femoral nerve stimulation. This procedure was performed to avoid the innervation zone and therefore to obtain the optimal amplitude of EMG response. The reference electrode was attached to the patella of the contralateral leg. The placement of each electrode was marked on the skin with indelible ink so that it could be exactly repositioned from session to session. Low resistance between the two electrodes (< 5 kΩ) was obtained by abrading the skin with emery paper and cleaning with alcohol. Myoelectrical signals were first amplified (custom-made amplifier: common mode rejection ratio = 90 dB, impedance input = 100 MΩ; gain = 1000) and then filtered with a bandwidth frequency ranging from 15 Hz to 5.0 kHz and simultaneously digitized online (Tida, Heka Elektronik, Lambrecht/Pfalz, Germany; sampling frequency: 2000 Hz) using a computer (IPC 486). All EMG data were stored with commercially available software (Tida 5.0, Heka Elektronik).
Mechanical parameters for single stimuli were analyzed, and the average of the four stimuli was used for further analysis. Only trials recorded at PRE and POST were considered because twitches potentiated by MVC (i.e., step 4) are more sensitive than unpotentiated ones (i.e., step 1) to detect early fatigue (16). The following twitch contractile properties were analyzed: i) peak twitch (PT), the highest value of twitch torque production; ii) time to peak twitch (TPT), the time needed to obtain twitch maximal torque calculated from the origin of the mechanical signal; and iii) half relaxation time (HRT), the time needed to obtain half of the decline in twitch maximal torque. MVC was considered as the peak torque attained during the contraction.
Twitch interpolation technique.
The twitch interpolation technique involves the delivery of one or more supramaximal electrical stimuli to the appropriate motor nerve during an attempted MVC (1) and assumes that the motor units that are not recruited or not discharging at their maximal frequency (incomplete activation) should yield a detectable force increment over the isometric plateau as a consequence of the stimulation of their axons. To improve the signal-to-noise ratio (14) and/or reduce measurement variability (28), some studies have used multiple rather than single stimuli. In the present study, each trial consisted in doublets evoked over the isometric plateau (superimposed doublet) and 3 s after the contraction (potentiated doublet). Maximal VAL was estimated according the following formula: VAL = [1 − (superimposed doublet amplitude/potentiated doublet amplitude)] × 100 (1).
M-wave peak-to-peak amplitude and duration were analyzed for VL, VM, and RF muscles at PRE and POST, with the average of the four trials used for analysis. EMG values of VL, VM, RF, and BF associated with MVC were analyzed over a 500-ms period once the torque had reached a plateau. The root mean square (RMS) EMG values of VL, VM, and RF during MVC were calculated and then normalized to the peak-to-peak amplitude of the maximal M wave (i.e., RMS/M ratio) for respective muscles. For the RMS values associated with MVC performed after the fatiguing contraction, the M wave recorded at the end of the fatiguing contraction was considered for normalization to account for peripheral influences (neuromuscular propagation failure and/or changes in impedance). The level of coactivation during MVC was calculated by normalizing the RMS values of the BF when this muscle was acting as an antagonist to the RMS obtained when this muscle was acting as an agonist, that is, during knee flexion, and was expressed as a percentage. During the fatiguing contraction, the RMS activity was quantified at consecutive sampling intervals that represented 10% of the endurance time for the VL, VM, RF, and BF muscles and then normalized to the respective RMS values determined during the MVC performed before the fatiguing task.
The training program has been previously described in detail elsewhere (11). Briefly, it consisted of 32 18-min sessions of isometric NMES over an 8-wk period, with four sessions per week. Forty isometric contractions were carried out during each training session. During the stimulation, subjects were seated on a machine typically used for strength training of the quadriceps muscle (Multiform, La Roque D'Anthéron, France) with the knee joint fixed at a 60° angle (where 0° corresponds to full extension of the knee). Three 2-mm-thick self-adhesive electrodes were placed over each thigh. The positive electrodes, measuring 25 cm2 (5 × 5 cm), which had membrane depolarizing properties, were placed as close as possible to the motor point of the VL and VM muscles. The negative electrode, measuring 50 cm2 (10 × 5 cm), was placed 5-7 cm below the inguinal ligament. Rectangular wave-pulsed currents (75 Hz) lasting 400 μs were delivered by a battery-powered stimulator (Compex Medical SA, Sport P, Ecublens, Switzerland) with a rise time of 1.5 s, a steady tetanic stimulation time of 4 s, and a fall time of 0.75 s. Each stimulation was followed by a 20-s pause. Stimulation intensity was monitored online and was gradually increased throughout the training session to a level of maximally tolerated intensity, which varied between 30 and 120 mA, according to the pain threshold of each subject. No subject reported serious discomfort. Each session was preceded by a standardized warm-up, consisting of 5 min of submaximal NMES at a freely chosen intensity (5 Hz, pulses lasting 200 μs). The average torque produced by NMES was 67 ± 13% of the MVC.
To assess the interday (i.e., WK2 and B) variability of the main measurements, coefficients of variation (CV) were calculated for each subject from the equation: SD/mean × 100. The interday variability of the present measurements was quantified with an intracorrelation coefficient based on repeated-measures ANOVA with the testing session as the independent variable. A one-factor ANOVA with repeated measures was used to assess the effect of the training program over B, WK4, and WK8 on the time to task failure and target torque. Separate two-factor (session × time) ANOVA with repeated measures on session and time were performed to compare the dependent variables measured before and after the fatiguing protocol. Three-factor ANOVA (session × time × muscle) with repeated measures on session and time was used to compare RMS during the fatiguing contraction. One-way ANOVA was consistently adopted to compare preexercise with postexercise changes in dependent variables and the mean rates of increase in EMG activity over the three sessions. Post hoc analyses (Newman-Keuls) were used to test for differences among pairs of means when appropriate. Linear regression analysis (Pearson's product-moment correlation) was used to compare the degree of association between variables. A significance level of P < 0.05 was used to identify statistical significance. The statistical analyses were performed by using Statistica software for Windows (Statsoft, version 6.1 Statistica, Tulsa, OK). All data are expressed as means ± SD in the text and table and as means ± SE in the figures.
Table 1 shows the interday reproducibility and variability between the two testing sessions (i.e., WK-2 and B) for the endurance time and the respective dependent variables recorded before and after the isometric fatiguing contraction. ICC and CV were 0.960 and 7.53% for endurance time and ranged from 0.738 to 0.930 and from 2.63 to 12.99% for the dependent variables, respectively.
Effects of NMES Strength Training
Knee extensor MVC torque increased significantly between B and WK4 (+16 ± 12%, P < 0.001), between WK4 and WK8 (+10 ± 6.0%, P < 0.001), and between B and WK8 (+26 ± 12%, P < 0.001) (Fig. 2), which meant that the average target torque (20% MVC) sustained during the fatiguing contraction increased over the testing sessions (52.7 ± 4.8 N·m at B, 60.9 ± 8.2 N·m at WK4, and 66.3 ± 7.1 N·m at WK8, P < 0.001, respectively). The relative increase in target torque was similar (P >0.05) between B and WK4 and between WK4 and WK8.
Muscle activation was higher at WK4 (88.4 ± 13.6%, P < 0.05) and at WK8 (90.3 ± 9.7, P < 0.01) than at B (85.5 ± 12.6%), whereas no significant difference was noted between WK4 and WK8 (P > 0.05). VL and VM RMS/M values associated with MVC increased significantly between B and WK4 (+27 ± 40 and +35 ± 23%, P < 0.01, respectively) and between B and WK8 (+39 ± 49 and +39 ± 25%, P < 0.001, respectively), whereas no significant changes occurred between WK4 and WK8 (P >0.05). RF RMS/M ratio was not affected (P > 0.05) by the training program.
No significant changes were noted in knee flexion MVC (P >0.05), BF EMG activity (P > 0.05), or the level of coactivation during knee extension MVC (P > 0.05), nor in contractile properties (P > 0.05) or M-wave duration or amplitude (P >0.05) over the three testing sessions.
Endurance time decreased significantly between B and WK4 (−17 ± 25%, P < 0.05), between WK4 and WK8 (−15 ± 17%, P < 0.05) and between B and WK8 (−32 ± 18%, P < 0.01; Fig. 2). This reduction was similar (P > 0.05) between B and WK4 and between WK4 and WK8. Significant negative correlations, fitted with a linear function, were found between the endurance time absolute changes and the target torque absolute gains between B and WK4 (R2 = 0.504, P < 0.05) (Fig. 3A), between WK4 and WK8 (R2 = 0.543, P < 0.05) (Fig. 3B), and between B and WK8 (R2 = 0.418, P < 0.05) (Fig. 3C).
Effects of Fatiguing Contraction
MVC and neural activation.
After the fatiguing contraction, the MVC torque of the knee extensors decreased significantly (from 263 ± 24 to 202 ± 39 N·m at B, from 305 ± 41 to 228 ± 60 N·m at WK4, and from 330 ± 36 to 251 ± 51 N·m at WK8, P < 0.001), and the relative reduction (P < 0.001) was similar over the three sessions (P > 0.05; Fig. 4A). Maximal VAL decreased significantly after the fatiguing contraction (P < 0.01), and this reduction was similar (P >0.05) over the three sessions (Fig. 4A). RMS/M during MVC diminished significantly after the fatiguing task in all three muscles and for the three sessions (P < 0.001; Fig. 4C). The reduction in RMS/M values was similar over the three sessions (P >0.05) and in the three muscles (P > 0.05).
Knee flexor MVC and the associated BF EMG activity decreased significantly (−12 ± 11 and −9 ± 20%; P < 0.01 and P < 0.05, respectively) after the fatiguing contraction, and these reductions were similar (P > 0.05) over the three sessions. The level of coactivation during knee extensor MVC diminished significantly (−18 ± 35%, P < 0.05) after the fatiguing contraction, and the relative change was similar (P > 0.05) over all three sessions.
M-wave and contractile properties.
No significant session × time interactions or main effects were noted for PT, TPT, or HRT (P > 0.05, data not shown) nor for M-wave amplitude in the three muscles (P > 0.05, data not shown). VL and VM M-wave durations were significantly reduced between PRE and POST (from 8.88 ± 2.49 to 8.09 ± 2.48 ms for VL and from 9.45 ± 3.70 to 8.58 ± 3.39 ms for VM, P < 0.05, respectively). The reduction in M-wave duration was similar over the three testing sessions (P > 0.05). No significant changes were noted in RF duration (P > 0.05).
EMG Activity during the Sustained Contraction
The EMG activity of the VL, VM, and RF muscles increased significantly (P < 0.001) during the fatiguing task, and no differences (P > 0.05) were found among the three muscles at the start and at the end of the fatiguing contraction. The average EMG activity of the knee extensor muscles was similar (P > 0.05) at the start of the contractions over the three sessions (14.7 ± 5.2% at B, 13.8 ± 6.9% at WK4, and 13.3 ± 5.1% at WK8; Fig. 5). However, the mean knee extensor EMG activity was significantly lower at WK4 (P < 0.05) and at WK8 (P < 0.01) than at baseline, whereas no significant differences were noted between WK4 and WK8 (P > 0.05; Fig. 5). These differences were found to be significant starting at 70% of the endurance time between B and WK4 and starting at 50% of the endurance time between B and WK8 (Fig. 5). Nonetheless, the average rates of increase in knee extensor EMG activity between the start and the end of the contractions were similar (P >0.05) over the three sessions (0.44 ± 0.30%·s−1 at B, 0.57 ± 0.35%·s−1 at WK4, and 0.57 ± 0.31%·s−1 at WK8). The EMG activity of the BF during the fatiguing contractions was lower (P < 0.001) compared with the knee extensor muscles and increased significantly (from 3.1 ± 1.5 to 4.6 ± 1.9%, P < 0.01) during the fatiguing contraction. The BF EMG activity remained similar up to 60% of the endurance time (P > 0.05), after which it increased significantly. BF EMG activity was similar at the beginning and the end of the fatiguing contraction over the three sessions (P >0.05) and increased at a similar rate for all three sessions (P >0.05).
Although NMES has been largely employed as a means of strength training, the effects of multiple sessions of NMES on muscle fatigue have been poorly investigated. The aim of the present study was to examine the effects of 4 and 8 wk of NMES training programs on endurance time and on mechanisms (i.e., central vs peripheral) contributing to task failure for a fatiguing submaximal isometric contraction performed with the knee extensor muscles at the same relative intensity. This study demonstrated that the endurance time decreased significantly after NMES training and that this reduction was directly related to the target torque absolute gains. The main finding of the present study was that the relative contributions of central and peripheral factors to fatigue were not affected by the NMES training-induced neuromuscular adaptations, as attested by the similar MVC torque and neural drive reductions at the end of the fatiguing contraction over the three sessions, whereas twitch contractile properties were not altered by the fatiguing task.
The variables measured in the present study demonstrated a low variability (CV < 15%) and a high degree of reproducibility (ranging from 0.738 to 0.960) between two different testing sessions (i.e., WK-2 and B). According to Stokes (29), CV ≤ 15% can be considered acceptable in biological systems. Moreover, a previous study from our laboratory demonstrated that the endurance time for a submaximal isometric contraction performed with the knee extensor muscles was reproducible across sessions (25).
Effects of NMES strength training.
The effects of an 8-wk NMES training program on neuromuscular properties of the knee extensor muscles have been described in detail elsewhere (11). Briefly, the increase in quadriceps muscle MVC (~30%) after the training program was accompanied by a significant increase in both muscle activation and EMG activity, that is, neural activation, but also by an increase in quadriceps muscle cross-sectional area (~6%).
Endurance time is reduced after NMES training.
NMES training-induced target torque enhancement (26%) was associated with a 32% decrease in endurance time. These findings are in agreement with several studies that showed that the stronger the subject is, the shorter the endurance time (4,12,13,23). Such findings reveal that the endurance time for a sustained contraction is inversely related to the absolute target torque exerted during the task. Because we found negative correlations between endurance time absolute changes and target torque absolute gains over the three sessions, it could be speculated that the greater absolute target torque after the training program might have induced an increase in intramuscular pressures (26), a greater occlusion of blood flow, and thus a greater mechanical compression of the local vasculature (2), which enhanced the accumulation of metabolites (2,26), therefore contributing to task failure. Consequently, the lower endurance time after NMES training could be ascribed to a greater absolute target torque.
Similar level of fatigue.
In the present study, we observed a magnitude of MVC loss (~23%) that was consistent with findings from previous studies performed in our laboratory (23,25) using the same fatiguing protocol. Because the decline in MVC torque recorded at exhaustion represents the amount of fatigue experienced during exercise (10), it could thus be suggested that muscle fatigue was similar after NMES training despite a shorter endurance time. However, the effects of NMES training on muscle fatigue are still equivocal. Duchateau and Hainaut (6) reported that 6 wk of NMES training-induced strength gains were associated with a similar decline in evoked force during an intermittent electrical fatigue task, which is line with our findings. In contrast, Kim et al. (15) observed that 4 wk of NMES training have no effects on maximal voluntary strength but increased significantly the endurance time for a cycling exercise. These discrepancies may be explained, at least in part, by the criteria used to assess muscle fatigue (i.e., endurance time vs MVC loss). Moreover, none of these studies investigated whether NMES training-induced neuromuscular adaptations affect the relative contributions of central and peripheral factors to fatigue.
In the present study, we found the same muscle activation loss and comparable RMS/M alterations at the end of the fatiguing contraction over the three sessions, which was in line with a recent study performed in our laboratory (23). These results indicate that a decreased descending command from supraspinal and/or spinal centers was probably the major cause of MVC loss and that the extent of central fatigue remained similar after the NMES training program.
We also observed a similar rate of rise in the average knee extensor EMG over the three sessions, although the EMG values at exhaustion were lower after 4 and 8 wk of training than at baseline. Despite the insensitivity of the surface EMG signal in detecting modest changes in motor unit activity during a sustained contraction (9), we assume that signal cancellation occurred similarly over the three sessions. It can thus be suggested that even though training resulted in an increase in neural drive during maximal effort (i.e., MVC), the behavior of the CNS during the submaximal fatiguing contraction was not affected by our NMES training program.
During a submaximal isometric fatiguing contraction, the progressive increase in EMG activity has been ascribed to the recruitment of larger motor units with minimal contribution from discharge rate (5). It appears, therefore, that the difference in EMG activity at exhaustion observed over the three sessions was primarily due to a difference in the number of motor units that were recruited. Nonetheless, the similarity in the rates of increase in EMG activity indicates that the balance of excitatory and inhibitory input received by the motor neurons was comparable over the three sessions. It could thus be suggested that different proportions of the motor unit population were active at exhaustion. In contrast, Semmler et al. (27) reported that subjects who increased the endurance time of a submaximal contraction after 4 wk of immobilization did so by means of a lower rate of increase in the EMG activity. Moreover, these authors (27) also reported that immobilization induced an alteration in the distribution of EMG activity among the elbow flexor muscles. In our study, the activation among the synergists of the knee extensor muscles was similar over the three sessions, thus suggesting that training and immobilization differently affect CNS behavior during a fatiguing task.
Antagonist EMG activity progressively increased during the fatiguing contractions, a finding that is in line with previous studies (7,25). Moreover, the reduction in both knee flexor MVC and BF EMG activity after the fatiguing contraction indicated that central fatigue also affects the antagonist muscles. Whatever the neural mechanisms involved in the control of antagonist muscle (17), the NMES training program did not influence the behavior of the BF muscle during the sustained contraction. Consequently, the shorter endurance time cannot be attributed to changes in antagonist muscle activity.
Because twitches potentiated by MVC are more sensitive than unpotentiated ones in detecting early fatigue (16), we compared the twitch properties evoked before and after the fatiguing task and found no significant differences between them, thus suggesting that peripheral fatigue did not contribute to the MVC reductions over the three sessions. Place et al. (23) recently demonstrated that peak twitch and peak doublet, recorded using the same fatiguing protocol, were significantly potentiated at knee joint angle of 35° (short length) but not at 75° (long length). In the present study, the fatiguing task was performed with the knee flexed at 60°. Consequently, it is not surprising to observe that contractile properties were not affected by the submaximal isometric contraction because our testing position was very close to the long muscle length adopted by Place et al. (23)
Practical applications of NMES training.
The present study demonstrated that maximal voluntary strength can be significantly increased by multiple sessions of NMES resistance training in healthy humans. In the context of rehabilitation, NMES may be useful in minimizing the loss of quadriceps muscle strength before, during, and after an immobilization period. However, our data also showed that muscle fatigue was unchanged after the 8-wk NMES training program. These results are in line with those reported in patients with refractory heart failure (24) after a similar training protocol, thus indicating that NMES training protocols should be performed in concert with aerobic programs or with other stimulation patterns (e.g., low-frequency NMES, 10-20 Hz) (22) to improve both muscle strength and resistance to fatigue.
In conclusion, the endurance time for a submaximal isometric contraction performed at the same relative intensity of knee extensor MVC was shorter after 4 and 8 wk of an NMES training program, and this was associated with the higher absolute target torque sustained during the fatiguing task. Despite a shorter endurance time, the NMES training program did not affect the magnitude of fatigue at exhaustion (assessed by MVC reductions) and the relative contributions of central and peripheral factors to fatigue.
We gratefully acknowledge the cooperation of all our subjects. We also express our gratitude to Nicolas Place and Dr. Nicola Maffiuletti for useful comments on a previous version of the manuscript, to Gilles Cometti for providing the Multiform device, and to Mary Bouley for the English revision of the manuscript.
1. Allen, G. M., S. C. Gandevia, and D. K. McKenzie. Reliability of measurements of muscle strength and voluntary activation using twitch interpolation. Muscle Nerve
2. Barnes, W. S. The relationship between maximum isometric strength and intramuscular circulatory occlusion. Ergonomics
3. Bax, L., F. Staes, and A. Verhagen. Does neuromuscular electrical stimulation strengthen the quadriceps femoris? A systematic review of randomised controlled trials. Sports Med
. 35:191-212, 2005.
4. Clark, B. C., T. M. Manini, D. J. The, N. A. Doldo, and L. L. Ploutz-Snyder. Gender differences in skeletal muscle fatigability are related to contraction type and EMG spectral compression. J. Appl. Physiol
. 94:2263-2272, 2003.
5. de Ruiter, C. J., M. J. Elzinga, P. W. Verdijk, W. van Mechelen, and A. de Haan. Voluntary drive-dependent changes in vastus lateralis motor unit firing rates during a sustained isometric contraction at 50% of maximum knee extension force. Pflugers Arch
. 447:436-444, 2004.
6. Duchateau, J., and K. Hainaut. Training effects of sub-maximal electrostimulation in a human muscle. Med. Sci. Sports Exerc
. 20:99-104, 1988.
7. Ebenbichler, G. R., J. Kollmitzer, L. Glockler, T. Bochdansky, A. Kopf, and V. Fialka. The role of the biarticular agonist and cocontracting antagonist pair in isometric muscle fatigue. Muscle Nerve
8. Edwards, R. H. Human muscle function and fatigue. Ciba Found. Symp
. 82:1-18, 1981.
9. Farina, D., R. Merletti, and R. M. Enoka. The extraction of neural strategies from the surface EMG. J. Appl. Physiol.
10. Gandevia, S. C. Spinal and supraspinal factors in human muscle fatigue. Physiol. Rev
. 81:1725-1789, 2001.
11. Gondin, J., M. Guette, Y. Ballay, and A. Martin. Electromyostimulation training effects on neural drive and muscle architecture. Med. Sci. Sports Exerc
. 37:1291-1299, 2005.
12. Hicks, A. L., J. Kent-Braun, and D. S. Ditor. Sex differences in human skeletal muscle fatigue. Exerc. Sport Sci. Rev
. 29:109-112, 2001.
13. Hunter, S. K., and R. M. Enoka. Sex differences in the fatigability of arm muscles depends on absolute force during isometric contractions. J. Appl. Physiol
. 91:2686-2694, 2001.
14. Kent-Braun, J. A., and R. Le Blanc. Quantitation of central activation failure during maximal voluntary contractions in humans. Muscle Nerve
15. Kim, C. K., T. E. Takala, J. Seger, and J. Karpakka. Training effects of electrically induced dynamic contractions in human quadriceps muscle. Aviat. Space Environ. Med
. 66:251-255, 1995.
16. Kufel, T. J., L. A. Pineda, and M. J. Mador. Comparison of potentiated and unpotentiated twitches as an index of muscle fatigue. Muscle Nerve
17. Levenez, M., C. Kotzamanidis, A. Carpentier, and J. Duchateau. Spinal reflexes and coactivation of ankle muscles during a submaximal fatiguing contraction. J. Appl. Physiol
. 99:1182-1188, 2005.
18. Maffiuletti, N. A., S. Dugnani, M. Folz, E. Di Pierno, and F. Mauro. Effect of combined electrostimulation and plyometric training on vertical jump height. Med. Sci. Sports Exerc
. 34:1638-1644, 2002.
19. Maffiuletti, N. A., M. Pensini, and A. Martin. Activation of human plantar flexor muscles increases after electromyostimulation training. J. Appl. Physiol
. 92:1383-1392, 2002.
20. Marqueste, T., F. Hug, P. Decherchi, and Y. Jammes. Changes in neuromuscular function after training by functional electrical stimulation. Muscle Nerve
21. Nordlund, M. M., A. Thorstensson, and A. G. Cresswell. Central and peripheral contributions to fatigue in relation to level of activation during repeated maximal voluntary isometric plantar flexions. J. Appl. Physiol
. 96:218-225, 2004.
22. Nuhr, M., R. Crevenna, B. Gohlsch, et al. Functional and biochemical properties of chronically stimulated human skeletal muscle. Eur. J. Appl. Physiol
. 89:202-208, 2003.
23. Place, N., N. A. Maffiuletti, Y. Ballay, and R. Lepers. Twitch potentiation is greater after a fatiguing submaximal isometric contraction
performed at short vs. long quadriceps muscle length. J. Appl. Physiol
. 98:429-436, 2005.
24. Quittan, M., G. F. Wiesinger, B. Sturm, et al. Improvement of thigh muscles by neuromuscular electrical stimulation in patients with refractory heart failure: a single-blind, randomized, controlled trial. Am. J. Phys. Med. Rehabil
. 80:206-214, 2001.
25. Rochette, L., S. K. Hunter, N. Place, and R. Lepers. Activation varies among the knee extensor muscles
during a submaximal fatiguing contraction in the seated and supine postures. J. Appl. Physiol
. 95:1515-1522, 2003.
26. Sejersted, O. M., A. R. Hargens, K. R. Kardel, P. Blom, O. Jensen, and L. Hermansen. Intramuscular fluid pressure during isometric contraction of human skeletal muscle. J. Appl. Physiol
. 56:287-295, 1984.
27. Semmler, J. G., D. V. Kutzscher, and R. M. Enoka. Limb immobilization alters muscle activation patterns during a fatiguing isometric contraction. Muscle Nerve
28. Suter, E., and W. Herzog. Effect of number of stimuli and timing of twitch application on variability in interpolated twitch torque. J. Appl. Physiol
. 90:1036-1040, 2001.
29. Stokes, M. J. Reliability and repeatability of methods for measuring muscle in physiotherapy. Physiother. Pract
. 1:71-76, 1985.
30. Taylor, N. A., R. H. Sanders, E. I. Howick, and S. N. Stanley. Static and dynamic assessment of the Biodex dynamometer. Eur. J. Appl. Physiol. Occup. Physiol
. 62:180-188, 1991.