Quadriceps femoris muscle resistance to fatigue using an electrically elicited fatigue test following intense endurance exercise training.

Electrical stimulation of skeletal muscle has been used as an evaluation tool to examine the effects and potential sites of human muscle fatigue.[1-5] To date, however, there have been few attempts to validate electrically elicited fatigue tests (EEFTs) of human man skeletal muscle.

Fitts and Holloszy[6] first demonstrated that a prolonged, intense endurance exercise training program on a tread-mill alters the contractile and fatigue properties of rat soleus muscle. Using in situ stimulation of rat soleus muscle, these investigators reported significantly less decline in the soleus muscle's peak tetanic tension (compared with controls) after 30 minutes of direct nerve stimulation using 250-millisecond trains at 100 Hz, at a rate of 110 trains per minute. Duchateau and Hainaut[7] extended these findings to the human adductor pollicis muscle. They reported significantly less loss of force (after exercise training) using an electrical fatigue test consisting of 60 repeated tetanies of 1 second duration at 30 Hz delivered at a rate of 30 per minute. Eight healthy untrained subjects exercised daily for 3 months and were tested before and after training; however, no control group was reported.[7] Duchateau and Hainaut[8] subsequently reported that 6 weeks of submaximal training via electrostimulation does not improve human adductor pollicis muscle resistance to fatigue. They found, however, that an equivalent voluntary submaximal training program resulted in significantly less muscle fatigue, particularly in the last 30 seconds of a 1-minute electrical fatigue test.[8] Unfortunately, again no control group was described.

Based on the results of these studies, it appears that voluntary exercise training in both rat soleus and human adductor pollicis muscles improves the resistance to fatigue during an EEFT. Because these investigators reported different stimulation characteristics, the improvement in fatigue resistance on an EEFT appears to be independent of at least some of the stimulation characteristics (ie, pulse duration and stimulation frequency). A major limitation of these studies was the lack of control groups that either (1) were not performing some type of exercise training or (2) were not the same control group subjects studied prior to exercise training. Though these limitations fail to provide us with conclusive evidence that EEFTS can demonstrate improvements in resistance to fatigue, there may be some value in adapting EEFTs for clinical use.

In our study, we used a construct validation approach to determine whether an EEFT could reflect the adaptations in quadriceps femoris muscles induced by 12 weeks of intense endurance exercise training on a bicycle ergometer. We reasoned that endurance exercise training, at sufficiently high intensities, would recruit and exercise train most muscle fibers (irrespective of fiber type) of untrained trained quadriceps femoris muscles. Based on previous reports,[6,7,9,-11] we expected high-intensity exercise would induce improvements in the oxidative capacity of the quadriceps femoris muscles, thereby improving the fatigue resistance. Therefore, an EEFT, if valid, should be able to detect improvements in fatigue resistance resulting from 12 weeks of intense endurance exercise training of the quadriceps femoris muscles.

Our initial research hypothesis was that the percentage of decline in torque produced during an EEFT would be less after a 12-week program of intense endurance exercise training compared with the percentage of decline in torque of control group subjects who did not exercise. Additionally, we hypothesized that the percentage of torque decline after inducing "acute" fatigue by voluntary exercise would be greater than the percentage of torque decline without prior fatiguing exercise, but would be less after 12 weeks of intense endurance exercise training. The null hypothesis was that there would be no differences in the extent of quadriceps femoris muscle torque decline after 12 weeks of endurance exercise training or after inducing acute fatigue by maximal voluntary exercise.

The purpose of our study was to test the hypothesis that changes in quadriceps femoris muscle endurance, induced by 12 weeks of intense endurance exercise training, would alter the percentage of decline in quadriceps femoris muscle torque during an EEFT.

Method

Subjects

Seventeen subjects (9 men and 8 women) with no known neuromuscular impairment, ranging in age from 22 to 45 years years (X=28, SD=7.2), volunteered to participate in the study. Eleven of the subjects (6 men and 5 women) agreed to participate in 12 weeks of intense endurance exercise training. Six subjects (3 men and 3 women) agreed to serve as control subjects and did not perform any regular exercise for 12 weeks. Only subjects who were not currently or previously (within past 6 months) participating in regular endurance exercise training were eligible and assigned to a group. None of the subjects had a history of left hip or knee pathology, and none were taking steroidal medication. The control group subjects either were unable to exercise on a regular basis due to time constraints or did not wish to exercise on a bicycle ergometer. All control group subjects, however, agreed to participate in the fatigue tests and assessments of maximal oxygen consumption ([Vo.sub.2] max) before and after 12 weeks of training. Control group subjects also agreed to not engage in any regular endurance or resistance-type exercise for the 12-week duration.

All subjects agreed to participate and signed an informed consent form outlining the procedures. The characteristics of all the subjects prior to the start of the 12 weeks of the study are presented in Table 1.

[TABULAR DATA OMITTED]

Procedure

Determination of maximal oxygen consumption. Two days prior to the initial fatigue test, each subject performed a test of [Vo.sub.2] max on a Quinton model 845 ergometer.(*) Work rate was increased every 2 minutes by 150 kpm until subjects were unable to continue exercising or maintain a pedal cadence of 60 rpm. The criteria used for determining [Vo.sub.2]max were the following: (1) a clear leveling off of oxygen uptake ([Vo.sub.2]) with an increase in work rate and (2) a respiratory exchange ratio of greater than 1.15. During the last 3 to 4 minutes of each test, expired gases were collected in neoprene meteorological balloons at 30-second intervals. Fractions of oxygen and carbon dioxide in expired gases were analyzed with a mass spectrometer,[dagger] which was routinely calibrated with gases analyzed by the Scholander technique. Expired volumes were measured with a Tissot chain-compensated spirometer.

After the 12-week period, subjects were retested on the bicycle ergometer to determine any changes in [Vo.sub.2] max resulting from intense endurance exercise training. The [Vo.sub.2]max retests were all performed after 72 hours of each subject's last exercise training session.

Electrically elicited fatigue test. All subjects performed a total of four EEFTs of the left quadriceps femoris muscles. Two tests were performed before and the other two tests immediately after the 10 weeks of training. Prior to the initial fatigue test, all subjects were asked to record their diets and the time of their meals for the 2 days preceding the initial fatigue test. Because diet may dramatically influence the test results, subjects were encouraged to consume foods high in carbohydrates ([greater than or equal to] 500 g per day) for each of the 2 days prior to testing.[12] Dietary intake and times of meals were verified by a registered dietitian prior to a second fatigue test.

Electrically elicited fatigue has been described previously.[3] Each subject was seated at a Cybex[R] II isokinetic dynamometer[double dagger] with the hip at 80 to 90 degrees of flexion and the knee positioned at 45 degrees of flexion. The subject's knee position was maintained throughout the duration of the test by mechanical stops attached to the input shaft of the dynamometer. Calibration of the Cybex[R] II isokinetic dynamometer[double dagger] was performed routinely in the static position with known loads prior to each test.

We used a medium-frequency electrical cal stimulators[sections] with a 2,500-Hz sinusoidal carrier wave, with interruptions at 50 pulses per second (pps) to electrically stimulate the left knee extensor muscles. Each subject had flexible carbon-rubber, gelled electrodes (4.5 x 10 cm) placed over the left vastus lateralis muscle proximally and the vastus medialis muscle distally on the thigh. Electrodes were secured with tape and elastic wrappings. Each subject performed three maximal voluntary isometric contractions (MVICs) of the left knee extensor with a 1-minute rest interval between repetitions. Peak torque (in foot-pounds) was determined from the Cybex[R] dynamometer's dual-channel strip chart recorder using an undamped signal.

Sixty percent of each subject's highest MVIC was calculated and used as the initial target torque to begin the fatigue test. Current from the stimulator was adjusted until the torque of the electrically stimulated knee extension was equal to 60% of the highest MVIC. After a 1-minute rest interval, the left quadriceps femoris muscles were stimulated 50 times using a 5-second-on-2-second-off duty cycle with a 2-second ramp time. The same current intensity was used throughout the test. Torque was recorded continuously for the duration of the test. Immediately following the 50th knee extension, each subject performed an MVIC to examine the effect of stimulation on the recovery of volitional peak torque.

Five days later, all subjects were given an identical EEFT as just described. Each subject's MVIC was obtained, electrodes were placed, and current intensity eliciting 60% of MVIC was determined. All second tests were conducted within 30 minutes of the time of day of the initial fatigue test to reduce the variability in peak torque measurements. The only difference in administering the second fatigue test was that each subject performed a 1 1/2-minute bout of repeated maximal isokinetic knee extensions at 180 [degrees]/s immediately prior to the EEFT. To ensure that a consistent number of contractions were performed during the voluntary exercise, each subject was paced using a metronome set at a rate of 50 extensions per minute. immediately following the completion of the exercise bout, the subject's knee was repositioned in 45 degrees of flexion, the dynamometer's speed selector was reset to zero, and the stimulation current was started as previously described.

The rationale for administering the second fatigue test was to determine whether the percentage of decline in quadriceps femoris muscle peak torque changed (ie, increased) when the muscle was acutely "fatigued." We also wanted to examine whether the extent of fatigue resistance after acute fatigue might be altered after 12 weeks of intense endurance exercise training.

After exercise training or no exercise training, all subjects performed two more EEFTs as described previously. Again, subjects were asked to record all dietary intake and times of meals for the 2 days prior to initial test, which occurred 72 hours after stopping exercise training. The second tests were again performed after 5 days at approximately the same time of day. Immediately preceding the second test, subjects performed the same 1 1/2-minute bout of voluntary isokinetic exercise at 180 [degrees]/s. A single MVIC of the knee extensor was performed immediately after the last stimulated contraction while peak torque was recorded.

Description of Training Protocol

The training protocol consisted of exercising 6 days per week for 45 to 60 minutes per day at an intensity of 70% to 90% of [Vo.sub.2]max for 12 weeks. All subjects were untrained and unfamiliar with bicycle ergometry at the start of the 12-week training period. Subjects exercised the majority of time at work rates requiring 70% to 80% of [Vo.sub.2]max during the initial 3 weeks. During weeks 4 through 6, work rates were adjusted to intensities ties requiring 80% to 85% of the subjects, [Vo.sub.2]max, and during weeks 7 through 12, the majority of each exercise session was spent at work rates of 85% to 90% of [Vo.sub.2]max. In the last 6 weeks (ie, weeks 7-12), subjects also performed interval training during two to three of their weekly exercise sessions. Interval training consisted of performing four to five 3-minute bouts of exercise at an intensity that elicited near-maxima [Vo.sub.2] (based on achievement of near-maximal heart rates). The remainder of each exercise session was spent at work rates of 85% to 90% of [Vo.sub.2]max.

Work rates were adjusted weekly, based on monitoring heart rate-[Vo.sub.2] relationships during an exercise session. This same training protocol has previously been used and demonstrated to induce biochemical and histochemical adaptations in the vastus lateralis muscle consistent with increasing the endurance capacity.[10,11] All subjects were able to tolerate the exercise sessions well, as none of the subjects withdrew due to injury or excessive fatigue.

Data Analysis

The subjects, physical characteristics (ie, age, body weight, [Vo.sub.2]max, and resting heart rate) at the start of the study were analyzed using t tests for independent samples. The [Vo.sub.2]max values before and after 12 weeks of exercise training in the exercise and control groups were analyzed using a 2x2 mixed analysis of variance (ANOVA).

Based on previous work with this electrical stimulation protocol, we calculated each subject's percentage of decline in torque in order to compare individual responses. We calculated the percentage of decline in torque (based on the 60% of MVIC target torque) using The mean of contractions 1 through 5, 21 through 25, and 46 through 50 as representative values of fatigue over the test duration. The mean percentage of decline was analyzed with a four-way ANOVA using a mixed design (2x2x2x23). Factor A consisted of the between-groups factor, control group versus exercise training group. Factor B was a repeated-measures factor for time of testing (ie, before or after 12 weeks). Factor C was a repeated-measures factor with two levels of fatiguing exercise (ie, with or without preceding maximal voluntary fatiguing exercise), and factor D was a repeated-measures factor with three levels (ie, the percentage of decline in torque throughout the fatigue test). An analysis of the main effects and all interactions was performed. When significant F ratios were found, post hoc pair-wise comparisons were performed.

The percentage of recovery of peak torque during an MVIC performed immediately after the 50th stimulated contraction was also calculated. Comparisons were made between groups, before and after 12 weeks of exercise training, and with and without fatiguing voluntary exercise using a three-way (2x2x2) ANOVA, with post hoc pair-wise comparisons where appropriate. The criterion for significance testing was .05.

Results

Subjects in the exercise training and control groups were similar in age, [Vo.sub.2] max, and resting heart rates at the start of this study (Tab. 1). The exercise group subjects weighed more than the control group subjects at the start of the study. Maximal oxygen consumption increased an average of 18% in the exercise training group, whereas the control group subjects, [Vo.sub.2]max did not change over the 12-week study period (fig. 1).

The results of the four-way ANOVA are summarized in Table 2. The percentage of decline in peak torque of contractions 1 through 5, 21 through 25, and 46 through 50 in the control and exercise groups before and after endurance training on the initial EEFT are summarized in Table 3 and depicted n Figures 2a and 2c. No differences were found in the mean percentage of decline in muscle torque elicited by electrical stimulation on the initial (ie, without preceding voluntary isokinetic exercise) EEFT between groups of subjects (main group effects [factor A]) either before or after the 12-week study period.

[TABULAR DATA OMITTED]

The percentages of decline in peak torque of contractions 1 though 5, 21 through 25, and b through 50 after inducing quadriceps femoris muscle fatigue by a 1 1/2-minute bout of maximal voluntary isokinetic exercise are summarized in Table 4 and depicted in Figures 2b and 2d. Similarly, there were no group differences in the mean percentage of decline in electrically elicited muscle torque immediately after 1 1/2 minutes of voluntary exercise either before or after the 12-week study period. Though no group differences were demonstrated before or after the 12-week period, there was a greater percentage of decline in electrically elicited torque (main fatiguing exercise effect) at each mean contraction number for both groups after fatiguing voluntary exercise compared with no prior fatiguing exercise (compare panels a and b and panels c and d in Fig. 2). Therefore, a greater percentage of decline in electrically elicited torque was observed in both groups before and after the 12-week period on the EEFT following maximal volitional exercise (Tabs. 3 and 4).

[TABULAR DATA OMITTED]

As expected, greater fatigue was observed as the EEFT progressed. The percentages of decline in electrically elicited torque (main contraction number effect [factor D]) were all different from each other for both groups and both sets of fatigue tests.

After 12 weeks of intense endurance exercise training, the mean peak torque obtained during an MVIC prior to the fatigue tests were not different (data not shown). Therefore, the initial target torques of 60% of MVIC were not different before or after 12 weeks of endurance training in the exercise and control groups.

Immediately after the last electrically elicited muscle contraction, the percentage of torque recovered (as a percentage of the original MVIC) was not different between groups either before or after the 12-week program of exercise (Tab. 5). The percentage of torque recovered immediately following the EEFT performed after a fatiguing bout of voluntary exercise was modestly, though significantly, higher in the exercise group than in the control group (Tab. 5).

[TABULAR DATA OMITTED]

Discussion

The 12-week program of intense endurance exercise training on the bicycle ergometer resulted in substantial, though anticipated, improvements in [Vo.sub.2]max. We are confident the quadriceps femoris muscles of the exercise subjects underwent biochemical and histochemical alterations consistent with improvements in quadriceps femoris muscle oxidative capacity from endurance exercise. Because bicycle ergometry was unfamiliar to all the exercising subjects before training, quadriceps femoris muscle fiber contractions would likely predominate during such an intense lower-extremity activity and presumably result in large increases in oxidative as well as glycolytic enzymes. Furthermore, an identical training program used in previous reports[10,11] resulted in demonstrable increases in oxidative capacity of the same muscle groups.

The lack of decrease in the percentage of decline in torque (fatigue resistance) during electrically elicited muscle contractions after exercise training was somewhat surprising. We initially hypothesized that the percentage of decline would be less after exercise training, because the muscles would become more fatigue-resistant. This premise was not confirmed by the results of this study. These results are in contrast to the findings of Duchateau and Hainaut,[7,8] who demonstrated smaller decrements in force during an EEFT after exercise training. Several major differences in the methods of stimulation, stimulation characteristics, muscles studied, and training protocols make direct comparisons of these previous findings with our results difficult.

In addition, we believed initially that the mean percentage of decline in torque elicited via stimulation would be greater after fatiguing voluntary exercise, which was demonstrated by these results. These findings imply that the EEFT is able to detect acute muscle fatigue induced by volitional exercise.

The mean percentage of decline in electrically elicited torque after inducing acute fatigue, however, did not change as a result of endurance exercise training. Inducing acute fatigue resulted in virtually identical declines in torque as before training in both groups, suggesting a similar activation of a select type and amount of quadriceps femoris muscle fibers. Based on the known differences between voluntary and electrically elicited muscle contractions,[13-15] we believe these results are due to the selective activation of predominantly fast-twitch muscle fibers within the field of stimulation. In addition, the superficial arrangement of type II skeletal muscle fibers within the muscle[16,17] may increase the susceptibility of type II muscle fiber activation during high-frequency percutaneous electrical stimulation as used in this study.

Several possibilities may help explain the findings of this study. First, the endurance exercise training program may not have induced endurance training adaptations within the selectively activated type II muscle fibers. Second, not all the muscle fibers that underwent training adaptations were necessarily the same muscle fibers activated during the stimulation tests. It does seem likely, however, when a 60% MVIC is produced, at least some fibers that had undergone training-induced adaptations would be active. Support for this statement mainly comes from the work of Gollnick and colleagues[18,19] and Andersen and Sj[phi]gaard,[20] who have demonstrated that during cycling intensities that require from 60% of [Vo.sub.2]max to supramaximal efforts, the quadriceps femoris muscle fiber recruitment progressively involved type I, then type II (IIa then IIb) muscle fibers. These studies clearly demonstrate type II fiber activation at similar endurance exercise training intensities as used in our study.

Therefore, we favor the opinion that the EEFT assesses predominantly fast-twitch muscle fibers that either were not the muscle fibers that were endurance exercise trained or were not able to demonstrate any training-induced changes due to altered activation characteristics imposed by the stimulation.[13] The EEFT was able to demonstrate more decline in torque after inducing fatigue by repeated maximal voluntary contractions. The voluntary fatigue protocol used in our study has previously been demonstrated to activate all types of skeletal muscle fibers and reflect the percentage of fast-twitch muscle fibers performing the activity.[21] Because the EEFT demonstrated more fatigue when the type II muscle fiber pool was exhausted, we are confident the additional decline reflects fatigue mainly in the type II muscle fiber pool.

The percentage of muscle peak torque recovered during an MVIC performed immediately after the end of stimulation ranged from 68%, to 82% of the prestimulation MVIC torque level (Tab. 4). The percentage of muscle peak torque recovery did not appear to differ between the two groups either before or after exercise training. It is interesting to note that when acute fatigue (induced by repeated maximal voluntary exercise) was followed by 50 contractions induced by electrical stimulation, the percentage of torque recovered during an MVIC was modestly, but significantly, higher in the exercise group compared with the control group. This finding suggests that repeated maximal voluntary exercise may have fatigued the majority of muscle fibers (irrespective of fiber type) to an equal amount in both groups, but the exercise-trained group may have improved their ability to recover a larger portion of their isometric torque more quickly following training. Those muscle fibers exhausted by the 1 1/2-minute bout of repeated maximal isokinetic exercise, then activated by electrical stimulation, appeared to remain fatigued immediately after the stimulation, and therefore unable to obtain prestimulation isometric torque levels.

After exercise training, although the percentage of decline in electrically elicited torque was not altered, the percentage of isometric torque recovery was improved in the exercise group to an average of 90% (Tab. 4). We have observed similar improvements in peak torque recovery after repeated isokinetic contractions following intense exercise training.[10] In addition, some reports[22-24] indicate that the recovery of muscle force can be attributed to a greater level of oxidative enzymes in slow- and fasttwitch muscle fibers, which in turn help to replenish adenosine triphosphate and creatine phosphokinase stores. A greater level of oxidative enzymes in both type I and II quadriceps femoris muscle fibers is an expected adaptation of the endurance exercise training program.[11]

We believe the results of this study have several practical implications for the assessment of human muscle performance by physical therapists. First, high-frequency percutaneous electrical stimulation of skeletal muscles selectively activates muscle fiber types in a manner quite distinct from volitional recruitment. Therefore, changes in muscle fiber activation order should be considered when studying potential causes of "physiological" fatigue using electrical stimulation. Second, improvements in local quadriceps femoris muscle endurance induced by intense endurance exercise training are not reflected in decrements in torque (force) caused by electrical stimulation. The extent or rate of decline in torque with repeated contractions is not a sufficiently responsive measure of endurance exercise training-induced changes in the local muscle. Whether or not exercise programs consisting primarily of resistance training or anaerobic training would change the percentage of torque decline on an EEFT is currently unknown and awaits future study.

Based on the results of our study, a decline in peak torque is not altered by electrical stimulation, although the recovery of volitional muscle peak torque was improved. The recovery of force, therefore, may be a more responsive measure of endurance adaptations in healthy individuals following muscle contractions induced by electrical stimulation. Physical therapists interested in assessing various aspects of quadriceps femoris muscle performance may still find EEFTs useful. Presumably, EEFTs may assess changes in glycolytic pathways of the fast-twitch muscle fiber pool in response to resistance exercise programs.[25]

Limitations

Our sample consisted only of subjects with healthy quadriceps femoris muscles. Therefore, the results of this study should not be generalized to subjects with muscle weakness, particularly when peak torque output is initially diminished or likely to fluctuate between testing sessions. Our rationale for using the percentage of decline in peak torque (ie, as a percentage of initial peak torque) as our dependent measure is predicated on obtaining a consistent torque output during an MVIC. Before and after the 12-week period, the initial torque outputs were not significantly altered. Therefore, we are confident that the use of percentage of decline in torque does not represent differences in initial torque outputs and is unlikely to have masked any differences in absolute torque outputs between the groups.

We also used a stimulation unit commonly found in rehabilitation settings in an effort to improve the potential application of the test. Stimulation and current characteristics selected in studies using animal preparations (eg, Burke's fatigue test[26]) or other studies[6] were not an option on the stimulator used in our study. Therefore, we urge caution in generalizing our findings to other stimulation characteristics that may demonstrate endurance-related adaptations. In previous studies, only select contractile properties of the muscle have been found to change as a result of endurance exercise training, whereas peak rates of tension development during a maximal tetanic contraction remained unaltered.[6] Some other studies using animal preparations,[12] however, have demonstrated a decline in force similar to that found in our study.

Finally, our sample size was small, and there is a possibility of a type II error (ie, failure to reject the null hypothesis when it is false). In lieu of our decision to fail to reject the null hypothesis as a major finding, any type II error would argue for finding less percentage of decline in torque in the exercise training group as a result of exercise training, which is not supported by our findings (Figs. 2c and 2d). Nevertheless, we performed a power analysis of the mean percentage of decline in torque throughout the fatigue test after exercise training (data depicted in Fig. 2c). We selected this data set for our power analysis because the group means appeared most different and a conservative estimate would therefore be generated. Our analysis indicates that in order to raise the power of our findings to 80% (or decrease the probability of committing a type II error to 20%) 44, 96, and 82 subjects would be required, respectively, at each of the three representative mean contractions depicted in Figure 2c. With our limited sample size, there remains the possibility that the groups differed in the percentage of decline in torque after 12 weeks of exercise training. Our confidence, however, in not committing a type II error would be considerably reduced by a larger sample size.

To obtain significantly different group effects, a still larger sample size (ie, greater than 200 subjects per group) would be necessary. Though statistically significant differences may be possible when the sample size is dramatically increased, we do not believe these differences would prove clinically meaningful. Such a high number of subjects required confirms our decision to fail to reject the null hypothesis. In addition, it does not support the use of percentage of decline in torque during an EEFT to reflect changes in local quadriceps femoris muscle fatigue resistance.

Conclusions

The results of this validation study suggest that the percentage of decline in torque during an EEFF is not able to detect any improvements in quadriceps femoris muscle endurance induced by intense endurance exercise training. Because the EEFT was unable to distinguish any populations of muscle fibers that had undergone a training effect, the EEFT may not be an assessment of the entire quadriceps femoris muscle fiber pool, but rather a select representation of type II skeletal muscle fibers within the field of stimulation.

Only the recovery of volitional isometric torque after all (or a significant number) of the muscle fibers had been exhausted by a preceding bout of maximal voluntary exercise followed by electrical stimulation was improved by intense endurance exercise training.

We interpret these findings to imply that EEFTS are assessing predominantly type II muscle fibers, which are more apt to be activated with high. frequency percutaneous stimulation.

References

[1] Hultman E, Spriet LL. Skeletal muscle metabolism, contraction force, and glycogen utilization during prolonged electrical stimulation in humans. J Physiol (Lond). 1986;374:493-501. [2] Binder-Macleod SA, McDermond LR. Changes in the force-frequency relationship of the human quadriceps femoris muscle following electrically and voluntarily induced fatigue. Phys Ther. 1992-72:95-104. [3] McDonnell MK, Delitto A, Sinacore DR, Rose SJ. Electrically elicited fatigue test of the quadriceps femoris muscle: description and reliability. Phys Ther. 1987;67:941-945. [4] Edwards RHT, Human muscle function and fatigue, In: Porter R, Whelan J, eds. Human Muscle Fatigue: Physiological Mechanisms. London, England: Pitman Publishing Ltd; 1981: 1-19. [5] Snyder-Mackler L, Binder-Macleod SA, Williams PR. Fatigability of human quadriceps femoris muscle following anterior cruciate ligament reconstruction. Med Sci Sports Exerc. 1993;25:783-789. [6] Fitts RH, Holloszy JO. Contractile properties of rat soleus muscle: effects of training and fatigue. Am J Physiol. 1977;233:C86-C91. [7] Duchateau F, Hainaut K. Training effects on muscle fatigue in man. Eur J Appl Physiol. 1984;53:248-252. [8] Duchateau F, Hainaut K. Training effects of submaximal electrostimulation in a human muscle. Med Sci Sport Exerc. 1988;20:99-104. [9] Ikai M, Yabe K. Training effect of muscular endurance by means of voluntary and electrical stimulation. Int Z Angew Physiol. 1969;28: 55-60. [10] Sinacore DR, Coyle EF, Hagberg JM, Holloszy JO. Histochemical and physiological correlates of training- and detraining-induced changes in the recovery from a fatigue test. Phys Ther. 1993;73:661-667. [11] Meredith CN, Frontera WR, Fisher EC, et al. Peripheral effects of endurance training in young and old subjects. J Appl Physiol. 1989;72:2278-2284. [12] Barclay CJ, Loiselle DS. Dependence of muscle fatigue on stimulation protocol: effect of hypocaloric diet. J Appl Physiol. 1992;72:2278-2284. [13] Delitto A, Snyder-Mackler L. Two theories of muscle strength augmentation using percutaneous electrical stimulation. Phys Ther. 1990;70:158-164. [14] Sinacore DR, Delitto A, King DS, Rose SJ. Type II fiber activation with electrical stimulation: a preliminary report. Phys Ther. 1990;70: 416-422. [15] Halkjaer-Kristensen J, Ingemann-Hansen T. Fibre recruitment pattern in the quadriceps muscle group during percutaneous electrical stimulation of the femoral nerve in man. In: Morecki A, Fidelus K, Kedzior K, Wit A, eds. Biomechanics VII-A. Baltimore, Md: University Park Press; 1981:307-311. [16] Lexell J, Downham D, Sjostrom M. Distribution of different fibre types in human skeletal muscles: a statistical and computational study of the fibre type arrangement in m. vastus lateralis of young, healthy males. J Neurol Sci. 1984;65:353-365. [17] Pernus F, Erzen I. Arrangement of fiber types within fascicles of human vastus lateralis muscle. Muscle Nerve. 1991;14:304-309. [18] Gollnick PD, Karlsson J, Piehl K, et al. Selective glycogen depletion in skeletal muscle fibres of man following sustained contractions. J Physiol (Lond). 1974;241:59-67. [19] Gollnick PD, Piehl K, Saltin B. Selective glycogen depletion pattern in human muscle fibres after exercise of varying intensity and at varying pedalling rates. J Physiol (Lond). 1974; 241:45-57. [20] Andersen P, Sj[phi]gaard G. Selective glycogen depletion in the subgroups of type II muscle fibres during intense submaximal exercise in man. Acta Physiol Scand. 1976;98:318-322. [21] Thorstensson A, Karlsson J. Fatiguability and fibre composition of human skeletal muscle. Acta Physiol Scand. 1976;98:318-322. [22] Tesch PA, Thorsson A, Fujitsuka N. Creatine phosphate in fiber types of skeletal muscle before and after exhaustive exercise. J Appl Physiol. 1989;66:1756-1759. [23] Harris RC, Edwards RHT, Hultman E, et al. The time course of phosphorylcreatine resynthesis during recovery of the quadriceps muscle in man. Pflugers Arch. 1976:367:137-142. [24] Jansson E, Dudley GA, Norman B, Tesch PA. Relationship of recovery from intense exercise to the oxidative potential of skeletal muscle. Acta Physiol Scand. 1990;139:147-152. [25] Binder-Macleod SA, Snyder-Mackler L. Muscle fatigue: clinical implications for fatigue assessment and neuromuscular electrical stimulation. Phys Ther. 1993;73:902-910. [26] Burke RE, Levine DN, Tsairis P, Zajac FE. Physiological types and histochemical profiles in motor unit of cat gastrocnemius. J Physiol (Lond). 1973;234:723-748.