European Journal of Applied Physiology

, Volume 106, Issue 6, pp 815–825 | Cite as

Electromyographic analysis of hip adductor muscles during incremental fatiguing pedaling exercise

Original Article

Abstract

The purpose of this study was to investigate activity of hip adductor muscles over time and during a representative crank cycle in fatiguing pedaling. Sixteen healthy men performed incremental pedaling exercise until exhaustion. During the exercise, surface electromyogram (EMG) was detected from adductor magnus (AM), adductor longus (AL), and selected thigh muscles. Temporal changes to normalized EMG in AM muscle resembled those in vastus lateralis (VL) muscle, whereas those in AL muscle showed later onset of increase from baseline compared with AM and VL muscles. During a representative crank cycle, the same level of normalized EMG was found between propulsive and pulling phases for AM muscle, whereas muscle activation of AL muscle during the pulling phase was statistically significant higher than that during the propulsive phase. We concluded that AM and AL muscles were gradually recruited over time during fatiguing pedaling exercise, but their temporal change and activation phases were not completely the same.

Keywords

Adductor magnus Adductor longus Cycling Surface electromyogram Needle electromyogram Ultrasonography 

Introduction

Pedaling exercise on a bicycle ergometer has been widely used in hundreds of studies for physiological and/or biomechanical research (Baum and Li 2003; Chapman et al. 2008; Ericson et al. 1985; Gregor et al. 1985; Hug et al. 2004a; Li 2004; Neptune et al. 1997; Ryan and Gregor 1992). To evaluate activation levels, neuromuscular fatigue, and the functional role of muscles during pedaling, surface electromyogram (EMG) was applied to working muscles such as quadriceps femoris and the hamstrings (Ericson et al. 1985; Glass et al. 1998; Gregor et al. 1985; Helal et al. 1987; Hug et al. 2003; Li and Caldwell 1998; Moritani et al. 1993; Neptune and Hull 1998; Raasch et al. 1997; van Ingen Schenau et al. 1992). These two muscle groups have been analyzed well regarding recruitment patterns over time, showing that muscle activation increases significantly with exercise time during incremental pedaling exercise (Glass et al. 1998; Hug et al. 2003, 2004b; Laplaud et al. 2006). Muscle activation during a representative crank cycle has also been analyzed in pedaling exercise using surface EMG technique (Chapman et al. 2008; Ericson et al. 1985; Li 2004; MacIntosh et al. 2000; Ryan and Gregor 1992). During a representative crank cycle in pedaling exercise, activation phase, e.g., propulsive and pulling phases, has been clarified to estimate the functional role of working muscles (Hakansson and Hull 2005; Neptune et al. 1997; Raasch et al. 1997; Ting et al. 1999).

Recent studies have suggested that some muscles that have not been reported by surface EMG techniques are likely to contribute greatly to the performance of pedaling exercise based on muscle functional magnetic resonance imaging (mfMRI), which is able to estimate muscle activity level of working muscles during exercise semiquantitatively (Meyer and Prior 2000). For example, Akima et al. (2005) showed that activation pattern of the adductor magnus (AM) muscle resembled that of the knee extensors, and that an index of activation, i.e., change in T2 time, of AM muscle was highly correlated with power output during sprint cycling (r = 0.688, p < 0.0001) (Akima et al. 2005). Richardson et al. (1998) and Endo et al. (2007) also reported that activation pattern of AM muscle did not differ from those of the knee extensors and flexors during moderate to very heavy pedaling exercise. From an anatomical point of view, professional road cyclists have been shown to display selective hypertrophy in the AM, vastus lateralis (VL), biceps femoris (BF), and sartorius muscles (Hug et al. 2006). Given these findings, AM muscle has been speculated to represent one of the major working muscles during pedaling. However, to the best of our knowledge, neither the activation profile in the adductor muscle group over time nor the phases in which this muscle group activates during a representative crank cycle have been clarified.

The adductor muscles occupy approximately 25% of the volume in the thigh segment for healthy individuals (Akima et al. 2007). By comparison, the knee extensors and flexors occupy approximately 50 and 25% in the thigh segment, respectively (Akima et al. 2007). As muscle volume represents the major determinant of force-generating capacity (Fukunaga et al. 2001), adductor muscles would have a relatively high potential of force-generating capacity in the thigh muscle groups. From this evidence, we infer that the adductor muscles play an important functional role in human movements such as pedaling.

The purpose of this study was to investigate activation profiles of adductor muscle, i.e., AM and adductor longus (AL) muscles during exhaustive pedaling exercise. To clarify temporal recruitment and activation patterns during a representative crank cycle of adductor muscle, an exhaustive incremental pedaling exercise was used. This task is one of the most popular tests to measure aerobic capacity for healthy individuals as well as highly motivated sports athletes and would be available to characterize activation pattern of working muscle. To acquire activation pattern in adductor muscles depending on time, we could estimate contribution of these muscles during exhaustive pedaling exercise (Glass et al. 1998; Hug et al. 2003, 2004b; Laplaud et al. 2006). We hypothesized that adductor muscles would also recruit with time same as knee extensors which is recognized as major working muscle and would activate in inconsistent pattern during a representative crank cycle compared with the muscle which contributes to other joint motion, i.e., knee extensor, since adductor muscles contribute to adduction, extension, or flexion of hip joint (Green and Morris 1970; Pressel and Lengsfeld 1998).

Methods

Subjects

Subject comprised 16 healthy men. Before this experiment, the procedure, purposes, and risks associated with this study were explained and written informed consent was obtained. Mean (± standard deviation (SD)) physical characteristics of subjects were as follows: age, 21.5 ±1.5 years; height, 172.5 ±4.5 cm; and weight, 66.3 ± 7.0 kg. The experimental protocol was approved by the Ethics Committee of the Research Center of Health, Physical Fitness & Sports at Nagoya University.

Experimental protocol

Subjects performed the incremental pedaling exercise until exhaustion on an electrically braked cycle ergometer (Aerobike 800; Combi Wellness, Tokyo, Japan). Saddle height was set at the trochanter height of individual. Consequently, the knee of the subject was slightly flexed (i.e., knee joint angle approximately 170° compared to full extension of 180°) at the bottom of the crank cycle. Feet were fixed to the pedals by the pedal strap. The handle bar was in an upright position. The trunk of the subject was approximately vertical to horizontal line during exercise. During the pedaling exercise, subjects remained seated on the saddle of the ergometer. Workload in the test was increased by 15 W every 1 min starting from 60 W until a pedaling cadence of 60 rpm could not be maintained. During the pedaling exercise, surface EMG was measured from four thigh muscles including the adductor muscles.

EMG recording

Surface EMG was collected with active electrodes from the AM, AL, VL, and BF muscles of the left leg. Electrode specifications in this study were as follows: amplification, differential; interelectrode distance, 1 cm; contact sensors, two 0.1 × 1 cm2 silver bars; preamplifier gain, 10-fold; input impedance, >1015 Ω // 0.2 pF; and common mode rejection ratio, 92 dB. The main amplifier unit feature was a gain of 100-fold and frequency response of 20 ± 5 to 450 ± 50 Hz (DE-2.1 sensor and Bagnoli-8 main amplifier unit; Delsys, Boston, MA). The system for surface EMGs was the same as in our previous study (Watanabe and Akima 2008). EMG signals amplified by EMG system were sampled at 1000 Hz (16 bits) by an A-D converter (Power Lab; AD Instruments, Melbourne, Australia) and stored on personal computer (PCG-6LTN; Sony Corp., Tokyo, Japan) using Chart 5.3 software (AD Instruments). Prior to attaching the electrodes, skin was shaved, abraded, and cleaned with alcohol. The electrodes for VL and BF muscles were placed at the midpoint between the head of the great trochanter and inferior edge of the patella and the midpoint between the ischial tuberosity and lateral epicondyle of the tibia, respectively. These electrodes were placed parallel to estimated muscle fibers of muscles. As AM and AL muscles could partly appear at the superficial region of the skin, defining the location of electrodes using conventional techniques such as visual or palpation methods was difficult. To overcome this problem, we used ultrasonography equipment (Logiq 5; GE Healthcare, Duluth, GA) to define anatomical properties of the superficial region of individual adductor muscles for electrode placement. The subjects sat on the edge of chair with abduction of the hip joint (Fig. 1). The probe for the ultrasonography was applied to the medial side of the proximal thigh to obtain axial images. The superficial region of the AM muscle appeared between the gracilis and semimembranosus muscles and that of the AL muscle was identified between the sartorius and gracilis muscles. Edges of AM and AL muscles, i.e., borders of adjacent muscles, were defined from ultrasound images (Fig. 1) and marked using felt-tip pen. Electrodes were positioned at the center of the superficial region of each muscle. The electrodes for AM and AL muscles were attached approximately on the longitudinal axis of femur as the estimated line of muscle fibers. A common reference electrode was attached on the iliac crest. After placement of all electrodes, we checked whether appropriate EMG signal was obtained using manual muscle testing.
Fig. 1

Procedure for defining the superficial region of the adductor magnus and longus muscles with ultrasonography. RF rectus femoris, VM vastus medialis, Sar sartorius, AL adductor longus, Gr gracilis, AM adductor magnus, SM semimembranosus, ST semitendinosus, GM gluteus maximus

Kinematic parameter

During pedaling exercise, crank angle and cadence was recorded by potentiometer (JT22-E; Nidec Copal Electronics, Tokyo, Japan) mounted on ergometer, and knee joint angle was recorded with electro-goniometer (SG150, Biometrics, Ltd., Gwent, UK). These data were sampled at 1000 Hz using an A-D converter and stored on a personal computer synchronized with EMG data.

Analysis

To investigate recruitment patterns over time, root mean square (RMS) was calculated during 10 crank cycles at given times during the test, at every 10% of exhaustion time. If sampling time cross over different workload, data during 10 cycles before increase of workload was collected. To evaluate recruitment patterns of four tested muscles during incremental exercise, RMS at each time was normalized by the value of the first stage, i.e., 10% of exhaustion time, for each muscle (Helal et al. 1987; Hug et al. 2004b; Viitasalo et al. 1985). Behavior of normalized EMG from the initial stage to exhaustion stage would be an index of temporal recruitment patterns through the exercise among four tested muscles.

To identify the activation pattern during a representative crank cycle, normalized EMGs at 10, 40, 70, and 100% of exhaustion time were analyzed to divide a representative crank cycle into “propulsive” and “pulling” phases. Since saddle shaft incline was 78° to the horizontal line, the pedal axis was furthest from the seat at a crank angle of 162° and nearest at 342° (crank angle at the top of the crank cycle was defined as 0°). From this geometry of cycle ergometer, propulsive and pulling phases were determined as 342–162° and 162–342°, respectively. Muscle activation during the propulsive and pulling phases was also normalized by the value of a representative crank cycle at 10% of exhaustion time. At the four selected times, muscle activation pattern depending on crank angle was calculated as follows: each RMS of 10 crank cycles at four selected times was filtered with a fourth-order Butterworth filter (cut-off frequency was 8 Hz; zero-phase lag). The filtered RMS was averaged every 1° of the crank cycle and normalized by the peak value among all four selected times for each muscle. Crank angle at which peak muscle activation occurred (peak timing) and duration of muscle activation (duration) were calculated from each muscle at each of the four selected times. The threshold value to determine duration was 10% of the peak muscle activation level for each muscle during exercise (Baum and Li 2003).

To collect kinematic parameters during exercise, maximal and minimum knee joint angles and cadence were calculated at each crank cycle after filtering with a low pass filter (cut-off frequency was 5 Hz). During 10 crank cycles at 10 given stages which is identical with the sampling time to calculate RMS, maximal and minimum knee joint angles and cadence were averaged for each subject.

Reliability test for surface EMG recorded from the AM and AL muscles

We had confirmed the reliability to record surface EMG from the AM and AL muscles using needle EMG in the same experimental materials. Six healthy men performed pedaling exercise for 2 min at 100 W and 2 min at 150 W. Cadence was set at 30–40 rpm to avoid pain for the subjects and a noise on needle EMG signal, which was determined from the result of preliminary experiments. Needle EMG signal was recorded using disposable concentric electrode with a recording surface of 0.03 mm2, a shaft diameter of 0.3 mm, and an electrode length of 25 mm (TECA DFC25, VIASYS Healthcare, Manor Way, England). The electrodes were connected to an amplifier having a bandwidth of 20–500 Hz (AB-611J, Nihon Kohden, Tokyo, Japan). Two needle electrodes were inserted into the AM and AL muscles near surface electrodes of these muscles with guidance of ultrasonography. The holders of inserted needles and needle electrodes were fixed on the skin by an elastic tape during exercise. Surface EMG signals and two needle EMG signals for each AM and AL muscles were recorded simultaneously during exercise. Recorded two needle EMG signals in each muscle were averaged.

Statistics

All data are provided as mean ± standard deviation (SD). Normalized EMGs depending on time, duration, and peak timing of a representative crank cycle were analyzed by two-way (muscle-by-time) repeated-measures analysis of variance (ANOVA). In the case of a two-factor interaction or main effects, the analysis was broken down into one-way repeated-measures ANOVA. When a significant change with time in each muscle was obtained, the Dunnett test for the value at 20% of exhaustion time was performed for normalized EMGs and Tukey’s method was performed for comparison of normalized EMG at the stage with that at the next stage and peak timing and duration of a representative crank cycle. Normalized EMGs during the propulsive and pulling phases were analyzed by two-way repeated-measures ANOVA (phase-by-time). In the case of a two-way interaction, the analysis was broken down into one-way repeated-measure ANOVA. Knee joint angles and cadence was analyzed by one-way repeated-measures ANOVA. When a significant difference between phases at each given time was apparent, Tukey’s method was performed as a post-hoc test. In cases where a significant change with time in each phase was obtained, a Dunnett test for the value at 10% of exhaustion time was performed as a post-hoc test. Prior to ANOVA analysis, a Mauchly’s sphericity test was performed to assess the validity to use ANOVA test, i.e., to assess homogeneity of variance and covariance. If sphericity assumption was violated, the Greenhouse-Geisser correction was used to adjust the degrees of freedom. The level of statistical significance was set at p < 0.05. Statistical analyses were performed using SPSS software (version 15.0J, SPSS, Tokyo, Japan).

Results

Mean exhaustion time and power output at exhaustion time were 10.9 ± 1.9 min and 208.1 ± 36.6 W, respectively. Workloads at each given time are shown in Table 1.
Table 1

Work load and percentage of maximal workload at given stages

% Exhaustion time

Work load (W)

% Maximal work load

10

66.6 ± 7.7

32.1 ± 3.1

20

78.8 ± 14.0

37.6 ± 4.1

30

98.4 ± 10.9

47.4 ± 3.7

40

110.6 ± 16.3

52.9 ± 3.4

50

133.1 ± 15.4

63.9 ± 3.1

60

144.4 ± 21.1

69.1 ± 3.2

70

163.1 ± 24.4

78.1 ± 5.5

80

178.1 ± 25.6

85.2 ± 2.4

90

195.9 ± 28.7

93.8 ± 5.5

100

209.1 ± 29.7

100.0 ± 0.0

For maximal and minimum knee joint angles and cadence, there was no significant change in the pedaling exercise as a result of ANOVA analysis (maximal knee joint, p = 0.07; minimum knee joint, p = 0.178; cadence, p = 0.102) (Table 2). These results represent knee joint kinematics and cadence was constant throughout the incremental pedaling exercise.
Table 2

Maximal and minimum knee joint angles and cadence at given stages

% Exhaustion time

Knee joint angle

Cadence

Max

Min

10

165.6 ± 7.0

60.6 ± 6.5

60.7 ± 1.1

20

166.6 ± 7.0

60.4 ± 7.4

60.3 ± 1.0

30

168.1 ± 7.2

60.4 ± 7.7

60.4 ± 1.1

40

170.4 ± 8.1

60.7 ± 8.2

60.0 ± 0.8

50

171.0 ± 7.7

61.0 ± 8.5

60.4 ± 1.2

60

171.3 ± 7.9

61.0 ± 7.8

59.9 ± 1.0

70

171.6 ± 7.1

61.8 ± 8.0

60.2 ± 0.8

80

173.2 ± 8.3

62.4 ± 7.4

59.4 ± 1.2

90

172.7 ± 7.8

62.1 ± 7.6

60.2 ± 1.2

100

169.1 ± 7.1

61.6 ± 8.5 (degrees)

60.9 ± 1.5 (rpm)

In the reliability test for surface EMG recorded from the AM and AL muscle, similar muscle activation pattern was observed between needle EMG and surface EMG for each muscle. In the AL muscle, the data was detected from four subjects since needle electrode in the AL muscle for two of the subjects unfastened during exercise. Activation pattern obtained with surface electrode in the AM and AL muscles parallels that with needle electrode (Fig. 2). There was no statistically significant difference in peak timing during a representative crank cycle between needle EMG and surface EMG in the AM muscle at 100 W (needle, 78.5 ± 29.3°; surface, 81.3 ± 30.4°; p = 0.433) and at 150 W (needle, 76.8 ± 32.7°; surface, 79.5 ± 35.0°; p = 0.443) and in the AL muscle at 100 W (needle, 282.0 ± 67.1°; surface, 285.5 ± 60.6°; p = 0.470) and at 150 W (needle, 277.3 ± 54.0°; surface, 266.0 ± 59.4°; p = 0.394) as a result of a paired t-test.
Fig. 2

Representative electromyography signals detected with needle and surface electrodes from the adductor magnus and adductor longus muscles. The vertical lines indicate the top of a crank cycle during pedaling. AM adductor magnus, AL adductor longus

For normalized EMG during incremental fatiguing exercise, a significant muscle-by-time interaction was identified (p < 0.001) using two-way ANOVA, indicating that recruitment patterns depending on time were not consistent among the selected muscles. Then, the analysis was broken down into one-way ANOVA within each muscle comparison and the result showed the initiation of significant increase in normalized EMG for the AM, AL, VL, and BF muscles compared to 20% of exhaustion time started at 50, 80, 50, and 70% of exhaustion time, respectively (all p < 0.05) (Fig. 3). A significant difference was identified between normalized EMG of VL muscle at 90% of exhaustion time and that at 100% of exhaustion time (p < 0.05) (Fig. 3).
Fig. 3

Normalized electromyographic activity until exhaustion. Values represent mean and standard deviation (SD). *p < 0.05 versus 20% of exhaustion time. #p < 0.05 versus next stage. AM adductor magnus, AL adductor longus, VL vastus lateralis, BF biceps femoris

For AM and BF muscles, no significant phase-by-time interaction was found at selected four stages (Fig. 4). While a significant interaction was obtained in AL and VL muscles (AL, p = 0.028; VL, p < 0.001). In AL muscle, normalized EMG during the pulling phase was significantly higher than that during the propulsive phase at all four stages (p < 0.05), and normalized EMG during the propulsive and pulling phases at 100% of the exhaustion time was significantly higher than that at 10% of exhaustion time (p < 0.05). Normalized EMG of VL muscle during the propulsive phase was significantly higher than that during the pulling phase at all four stages (p < 0.05) and significant increase was initiated from 70% of the exhaustion time compared to that at 10% of the exhaustion time (p < 0.05).
Fig. 4

Normalized electromyographic activity during the propulsive and pulling phases at 10, 40, 70, and 100% of exhaustion time. Values represent mean and standard deviation (SD). *p < 0.05 versus 10% of exhaustion time. #p < 0.05 versus propulsive phase or pulling phase. AM adductor magnus, AL adductor longus, VL vastus lateralis, BF biceps femoris

Normalized EMG during a representative crank cycle at 10, 40, 70, and 100% of the exhaustion time was presented in Fig. 5. For peak timing of muscle activation during a representative crank cycle, no significant muscle-by-time interaction was seen; however, a main effect in muscle factor was obtained (p < 0.001) (Table 3). Thus, peak timing was consistent during incremental exercise for each muscle, but they were significantly different among the muscles. For AM muscle, the peak timing appeared at significantly earlier timing during a representative crank cycle compared to that of AL and at later timing during a representative crank cycle compared to that of VL at all four selected stages (p < 0.05). While peak timing of AL muscle occurred at a significantly later timing during a representative crank cycle than that of VL and BF muscles throughout exercise time (p < 0.05) (Table 3).
Fig. 5

Normalized electromyographic activity depending on crank angle at 10, 40, 70, and 100% of exhaustion time. Error bar indicates standard deviation (SD) at 100% of exhaustion time. 0° is top of the crank cycle. AM adductor magnus, AL adductor longus, VL vastus lateralis, BF biceps femoris

Table 3

Peak timing and duration of muscle activation during a representative crank cycle at 10, 40, 70, and 100% of exhaustion time

% Exhaustion time

AM

AL

VL

BF

Peak timing

 10

148.5 ± 68.6

218.1 ± 61.8*

50.5 ± 27.3*#

137.0 ± 75.6#+

 40

141.9 ± 69.3

211.3 ± 86.0*

47.2 ± 26.0*#

129.9 ± 57.5#+

 70

146.3 ± 80.5

218.0 ± 75.7*

42.9 ± 23.0*#

130.0 ± 50.0#+

 100

114.4 ± 57.4

192.9 ± 74.5*

44.8 ± 17.4*#

146.8 ± 57.4#+

Duration

 10

61.1 ± 76.7

109.7 ± 103.2

131.5 ± 31.0

66.2 ± 91.8

 40

146.9 ± 116.3a

130.9 ± 123.6

158.6 ± 30.4a

95.3 ± 96.5

 70

202.4 ± 97.5ab

188.5 ± 136.5ab

157.1 ± 18.2a

157.8 ± 86.6ab

 100

246.4 ± 58.5ab

260.0 ± 88.0abc

165.9 ± 14.3a

220.4 ± 67.4abc (degrees)

AM adductor magnus, AL adductor longus, VL vastus lateralis, BF biceps femoris

p < 0.05 versus AM. p < 0.05 versus AL. + p < 0.05 versus VL. p < 0.05 versus 10% of exhaustion time. p < 0.05 versus 40% of exhaustion time. p < 0.05 versus 70% of exhaustion time

In the duration of muscle activation, a significant muscle-by-time interaction (p < 0.001) and time main effect (p < 0.001) were obtained. This means that duration of muscle activation alters with progress in time in different manner among the muscles. For AM, AL, and BF muscles, the duration of muscle activation was significantly longer with increase in exercise stage (p < 0.05). On the contrary, significant increases in the duration of muscle activation in the VL muscle began at 40% of exhaustion time, then it was constant until exhaustion (p < 0.05) (Table 3).

For muscle activation during a representative crank cycle, two distinct patterns were seen among subjects in AM and AL muscles. In AM muscle, 8 of the 16 subjects showed a pattern similar to Subject 3, 5 subjects had a pattern similar to Subject 7 (Fig. 6), and the remaining 3 subjects could not be separated into either of these two patterns. Muscle activation of AM muscle mainly occurred during the propulsive phase for Subject 3, but was seen in both phases for Subject 7. In AL muscle, 9 of the 16 subjects showed a pattern similar to Subject 1, 5 subjects had a pattern similar to Subject 6 (Fig. 6), and the remaining 2 subjects could not be separated into either of these two patterns. AL muscle in Subject 1 displayed peak activation during the pulling phase, whereas that in Subject 6 exhibited additional activation near the top of crank cycle with peak activation during the pulling phase.
Fig. 6

Example of two distinct patterns of normalized electromyographic activity depending on crank angle for AM (Subjects 3 and 7) and AL muscles (Subjects 1 and 6) at 40 and 100% of exhaustion time. 0° is top of the crank cycle. AM adductor magnus, AL adductor longus

Discussion

The purpose of this study was to investigate recruitment patterns over time and activation patterns during a representative crank cycle of the adductor muscles in incremental fatiguing pedaling exercise. To the best of our knowledge, this is the first study to have recorded activation of the adductor muscles using surface EMG techniques during pedaling. The primary results of this study were that recruitment patterns depending on time for AM muscle were similar compared with that for VL muscle. Conversely, the onset of increases in normalized EMG in the AL muscle was later than that of VL muscle (Fig. 3). In a representative crank cycle analysis, no statistically significant difference in normalized EMGs was seen between propulsive and pulling phases for AM muscle, whereas a significant difference was noted for AL muscle (Fig. 5).

The individual hip adductor muscles are quite close together at medial side of proximal thigh. Thus, it was suspected that surface EMG signal detected from the AM and AL muscles include the activation of adjacent muscles. In this study, reliability test for surface EMG recorded from the AM and AL muscles were performed using needle EMG technique. From the results of this reliability test, it was confirmed that surface EMG recorded from the AM and AL muscles adequately reflected the activation of each muscle (Fig. 2).

In the incremental fatiguing pedaling exercise (Hug et al. 2003, 2004b; Laplaud et al. 2006; Lucia et al. 1999), surface EMG of working muscle is thought to be influenced by two main factors: increased workload and muscle fatigue. To estimate fatigue of the working muscle, intramuscular lactate concentration is a good indicator that is determined by biopsy technique (Green et al. 1983, 1989; Hollidge-Horvat et al. 2000). Green et al. (1983) reported that during exhaustive incremental pedaling exercise, intramuscular lactate concentration in VL muscle, as the major working muscle for pedaling exercise, increased mildly with increased workload and increased substantially after 86% of maximal power output. From this time point of the transition in muscle lactate concentration, reductions in muscle pH caused by increased protons released from lactic acid would occur in VL muscle. In this study, 85.2% of maximal workload was given at 80% of exhaustion time (Table 1), so, after this stage (i.e., 90 and 100% of exhaustion time), muscle fatigue through falling muscle pH may have been initiated in VL muscles. We, therefore, speculated that muscle activation in this study was mainly influenced by increased workload throughout the incremental fatiguing exercise, but the effect of muscle fatigue could be greater at the stages after 80% of exhaustion time.

Before exhaustion time, specific temporal changes in normalized EMG were seen in VL muscle (Fig. 3). This increase in activation of VL muscle could be reflected in higher excitation of neuromuscular activity to continue the given work of the major working muscle under conditions of lower muscle pH. From this result, activation of VL muscle was estimated to be strongly related to pedaling exercise and may represent one of the major limiting factors to continued incremental fatiguing pedaling exercise. VL muscle has been used as an index of muscle activation during the pedaling exercise in previous studies (Green et al. 1983, 1989; Housh et al. 1995; Lucia et al. 1999; Moritani et al. 1993) because this muscle has been recognized as the major working muscle. Our result support the idea of VL muscle as the major working muscle for pedaling exercise, as suggested in previous studies (Akima et al. 2005; Ericson et al. 1986).

Temporal changes in activation of the AM muscle, i.e., the onset of increase in normalized EMG from baseline, were similar to those for VL muscle (Fig. 3). Similar recruitment pattern over time between AM and VL muscles may indicate that the central nervous system controls recruitment and firing rate of motor units to meet required tasks in similar ways during incremental fatiguing pedaling exercise. AM muscle may thus also play an important role in pedaling motion. This speculation could be supported by the clear correlation (r = 0.688, p < 0.0001) between activation of AM muscle as evaluated by mfMRI and power output during sprint cycling in a previous study (Akima et al. 2005). Moreover, earlier studies have suggested higher activation in AM muscle induced by pedaling exercise, along with knee extensors and flexors (Akima et al. 2005; Endo et al. 2007; Richardson et al. 1998). As muscle volume is the major determinant of force-generating capacity (Fukunaga et al. 2001), the large muscle volume of AM muscle (17% of thigh segment) (Akima et al. 2007) may also support our speculation.

Conversely, for AL muscle, the beginning of increases in normalized EMG from baseline was slower than those of VL muscle (Fig. 3). As AL muscle has a small volume (4% of thigh segment) (Akima et al. 2007), it was assumed that contribution of this muscle to pedaling exercise would be small. This assumption would explain the no correlation (p = 0.132, n.s.) between the percent change in activation of AL muscle and power output during sprint cycling (Akima et al. 2005). Furthermore, Fleckenstein et al. (1991) reported no statistically significant change in activity in AL muscle during pedaling exercise in a mfMRI study (Fleckenstein et al. 1991). In this study, however, activation of this muscle increased over time from 80% of exhaustion time during incremental fatiguing exercise (Fig. 3). This result suggests that AL muscle may also be one of the working muscles, but activation in pedaling exercise was limited to the latter half of the incremental fatiguing pedaling.

The profile of muscle activation during a representative crank cycle has been used to estimate the functional roles of working muscles in pedaling exercise (Li and Caldwell 1998; Mileva and Turner 2003; Neptune et al. 1997; Ting et al. 1999). For example, we can easily understand VL muscle engaged in the propulsive phase of pedaling from the EMG activity using this analysis (Fig. 4). Activation of AM muscle was found during both propulsive and pulling phases, whereas that of AL muscle was found mainly during the pulling phase (Figs. 4, 5). The peak of muscle activation was presented at the middle of the propulsive phase in AM muscle and the early part of the pulling phase in AL muscle (Fig. 5, Table 3). Even though these muscles are both categorized as adductor muscles, AM and AL muscles showed a distinct difference in activation patterns. Anatomically, both AM and AL muscles appear to play a role in hip adduction, in addition to contributing to extension and flexion of the hip joint, respectively (Green and Morris 1970; Pressel and Lengsfeld 1998). The variance of muscle activation patterns during a representative crank cycle between AM and AL muscles may be associated with the function of hip joint motion in the sagittal plane. In biomechanical studies, hip extension torque is reportedly generated during the propulsive phase of pedaling and hip flexion torque is generated during the pulling phase of pedaling (Ericson et al. 1986; Gregor et al. 1985; van Ingen Schenau et al. 1992). AM muscle could thus generate extension torque at the hip joint, whereas AL muscle could generate flexion torque at the hip joint during pedaling. In these discussions, it should be noted that large interindividual variability, which indicated as SD at 100% of exhaustion time in Fig. 5, was obtained in activation pattern during a representative crank cycle for AM and AL muscles compared with VL muscle (0.27, 0.26, and 0.12 of mean SD during a representative crank cycle for AM, AL, and VL muscles, respectively). Interestingly, two distinct patterns of muscle activation during one crank cycle were found among subjects for both AM and AL muscles (Fig. 6). In AM muscle, muscle activation of the pattern for Subject 3 was mainly engaged in the propulsive phase, where the lower joints are extending. In contrast, the pattern in Subject 7 showed two peaks during the propulsive phase and first half of the pulling phase. In this case, AM muscle could contribute to both extension and flexion movements of the lower joints as a hip adductor or extensor. Two distinct patterns in AL muscle showed a common peak during the pulling phase, so these patterns would commonly contribute to the movement during pulling phase as a hip adductor or flexor. However, the pattern in Subject 6 had an additional peak near the top of the crank cycle. In this pattern, AL muscle may be primarily engaged in hip extension exceptionally since Dostal et al. (1986) reported that AL muscle contributes to hip extension when the hip joint is flexed (Dostal et al. 1986) such as when the pedal is located at the top of the crank cycle. Distinct muscle activation patterns among subjects were reported in BF (Li and Caldwell 1998; Ryan and Gregor 1992) and lower leg muscles (Hug et al. 2008; Ryan and Gregor 1992). Ryan and Gregor (1992) suggested that muscles possessing simple functions like monoarticular muscles reveal smaller interindividual variability in activation pattern, whereas muscles possessing multiple functions like biarticular muscle show larger interindividual variability in activation during constant pedaling exercise. AM and AL muscles are monoarticular muscles crossing the hip joint, but contribute various joint motion at the hip joint (Green and Morris 1970; Pressel and Lengsfeld 1998). Large interindividual variability in activation pattern such as distinct patterns among subjects during a representative crank cycle for AM and AL muscles may be due to the multiple functional roles during pedaling exercise.

In the analysis of peak timing and duration of muscle activation in a representative crank cycle, the four tested muscles clearly showed different EMG activity patterns during fatiguing pedaling (Table 3). Peak timing of EMG activity may represent the crank angle where peak force generation for pedaling action depends on their functional roles and/or anatomical properties. No statistically significant change in peak timing was seen with exercise time, demonstrating optimum crank position was steady in individual muscle throughout fatiguing pedaling. A statistically significant difference was found in peak timing between AM and AL muscles (Table 3), suggesting that the optimum crank positions during pedaling movement would clearly differ between AM and AL muscles.

Conversely, the duration of muscle activation in AM, AL, and BF muscles increased with exercise time during fatiguing pedaling. Interestingly, VL muscle did not display any change in duration of EMG activity, unlike the other three tested muscles (Table 3). This implies that only increasing level of muscle activation, not duration, act as the factor controlling activation in VL muscle during incremental fatiguing pedaling. For AM, AL, and BF muscles, however, both increasing level and duration were factors regulating muscle activation for increase of workload with progress in time or muscle fatigue near exhaustion during pedaling. This activation pattern in AM, AL, and BF muscles could be one of the coping mechanisms for incremental fatigue pedaling.

In conclusion, we investigated muscle activation of the hip adductor muscles over time and crank phase during incremental fatiguing pedaling exercise using surface EMG. The AM muscle displayed similar recruitment patterns to knee extensor muscles over time and activated during both propulsive and pulling phases. The AL muscle was recruited from a later time and after heavier workload compared with knee extensor muscle. Muscle activation of AL muscle mainly arose during the pulling phase. Variance was seen in muscle activation patterns between AM and AL muscles during incremental fatiguing pedaling exercise.

Notes

Acknowledgments

This research was supported in part by a Grant-in-Aid for Scientific Research (#17300207) from the Japanese Ministry of Education, Science, Sports and Culture. The authors thank Dr. Masaaki Hirayama of Department of Neurology, Nagoya University Graduate School of Medicine and Dr. Teruhiko Koike of Research Center of Health, Physical Fitness & Sports, Nagoya University for their technical support and helpful suggestion of needle EMG recording.

References

  1. Akima H, Kinugasa R, Kuno S (2005) Recruitment of the thigh muscles during sprint cycling by muscle functional magnetic resonance imaging. Int J Sports Med 26:245–252. doi:10.1055/s-2004-821000 PubMedCrossRefGoogle Scholar
  2. Akima H, Ushiyama J, Kubo J, Fukuoka H, Kanehisa H, Fukunaga T (2007) Effect of unloading on muscle volume with and without resistance training. Acta Astronaut 60:728–736. doi:10.1016/j.actaastro.2006.10.006 CrossRefGoogle Scholar
  3. Baum BS, Li L (2003) Lower extremity muscle activities during cycling are influenced by load and frequency. J Electromyogr Kinesiol 13:181–190. doi:10.1016/S1050-6411(02)00110-4 PubMedCrossRefGoogle Scholar
  4. Chapman AR, Vicenzino B, Blanch P, Hodges PW (2008) Patterns of leg muscle recruitment vary between novice and highly trained cyclists. J Electromyogr Kinesiol 18:359–371. doi:10.1016/j.jelekin.2005.12.007 PubMedCrossRefGoogle Scholar
  5. Dostal WF, Soderberg GL, Andrews JG (1986) Actions of hip muscles. Phys Ther 66:351–361PubMedGoogle Scholar
  6. Endo MY, Kobayakawa M, Kinugasa R, Kuno S, Akima H, Rossiter HB, Miura A, Fukuba Y (2007) Thigh muscle activation distribution and pulmonary VO2 kinetics during moderate, heavy, and very heavy intensity cycling exercise in humans. Am J Physiol Regul Integr Comp Physiol 293:R812–R820. doi:10.1152/ajpregu.00028.2007 PubMedGoogle Scholar
  7. Ericson MO, Nisell R, Arborelius UP, Ekholm J (1985) Muscular activity during ergometer cycling. Scand J Rehabil Med 17:53–61PubMedGoogle Scholar
  8. Ericson MO, Bratt A, Nisell R, Arborelius UP, Ekholm J (1986) Power output and work in different muscle groups during ergometer cycling. Eur J Appl Physiol Occup Physiol 55:229–235. doi:10.1007/BF02343792 PubMedCrossRefGoogle Scholar
  9. Fleckenstein JL, Haller RG, Lewis SF, Archer BT, Barker BR, Payne J, Parkey RW, Peshock RM (1991) Absence of exercise-induced MRI enhancement of skeletal muscle in McArdle’s disease. J Appl Physiol 71:961–969PubMedGoogle Scholar
  10. Fukunaga T, Miyatani M, Tachi M, Kouzaki M, Kawakami Y, Kanehisa H (2001) Muscle volume is a major determinant of joint torque in humans. Acta Physiol Scand 172:249–255. doi:10.1046/j.1365-201x.2001.00867.x PubMedCrossRefGoogle Scholar
  11. Glass SC, Knowlton RG, Sanjabi PB, Sullivan JJ (1998) Identifying the integrated electromyographic threshold using different muscles during incremental cycling exercise. J Sports Med Phys Fitness 38:47–52PubMedGoogle Scholar
  12. Green DL, Morris JM (1970) Role of adductor longus and adductor magnus in postural movements and in ambulation. Am J Phys Med 49:223–240PubMedGoogle Scholar
  13. Green HJ, Hughson RL, Orr GW, Ranney DA (1983) Anaerobic threshold, blood lactate, and muscle metabolites in progressive exercise. J Appl Physiol 54:1032–1038PubMedGoogle Scholar
  14. Green HJ, Jones LL, Houston ME, Ball-Burnett ME, Farrance BW (1989) Muscle energetics during prolonged cycling after exercise hypervolemia. J Appl Physiol 66:622–631PubMedGoogle Scholar
  15. Gregor RJ, Cavanagh PR, LaFortune M (1985) Knee flexor moments during propulsion in cycling—a creative solution to Lombard’s Paradox. J Biomech 18:307–316. doi:10.1016/0021-9290(85)90286-6 PubMedCrossRefGoogle Scholar
  16. Hakansson NA, Hull ML (2005) Functional roles of the leg muscles when pedaling in the recumbent versus the upright position. J Biomech Eng 127:301–310. doi:10.1115/1.1865192 PubMedCrossRefGoogle Scholar
  17. Helal JN, Guezennec CY, Goubel F (1987) The aerobic-anaerobic transition: re-examination of the threshold concept including an electromyographic approach. Eur J Appl Physiol Occup Physiol 56:643–649. doi:10.1007/BF00424804 PubMedCrossRefGoogle Scholar
  18. Hollidge-Horvat MG, Parolin ML, Wong D, Jones NL, Heigenhauser GJ (2000) Effect of induced metabolic alkalosis on human skeletal muscle metabolism during exercise. Am J Physiol Endocrinol Metab 278:E316–E329PubMedGoogle Scholar
  19. Housh TJ, deVries HA, Johnson GO, Housh DJ, Evans SA, Stout JR, Evetovich TK, Bradway RM (1995) Electromyographic fatigue thresholds of the superficial muscles of the quadriceps femoris. Eur J Appl Physiol Occup Physiol 71:131–136. doi:10.1007/BF00854969 PubMedCrossRefGoogle Scholar
  20. Hug F, Faucher M, Kipson N, Jammes Y (2003) EMG signs of neuromuscular fatigue related to the ventilatory threshold during cycling exercise. Clin Physiol Funct Imaging 23:208–214. doi:10.1046/j.1475-097X.2003.00497.x PubMedCrossRefGoogle Scholar
  21. Hug F, Bendahan D, Le Fur Y, Cozzone PJ, Grelot L (2004a) Heterogeneity of muscle recruitment pattern during pedaling in professional road cyclists: a magnetic resonance imaging and electromyography study. Eur J Appl Physiol 92:334–342. doi:10.1007/s00421-004-1096-3 PubMedCrossRefGoogle Scholar
  22. Hug F, Decherchi P, Marqueste T, Jammes Y (2004b) EMG versus oxygen uptake during cycling exercise in trained and untrained subjects. J Electromyogr Kinesiol 14:187–195. doi:10.1016/S1050-6411(03)00081-6 PubMedCrossRefGoogle Scholar
  23. Hug F, Marqueste T, Le Fur Y, Cozzone PJ, Grelot L, Bendahan D (2006) Selective training-induced thigh muscles hypertrophy in professional road cyclists. Eur J Appl Physiol 97:591–597. doi:10.1007/s00421-006-0218-5 PubMedCrossRefGoogle Scholar
  24. Hug F, Drouet JM, Champoux Y, Couturier A, Dorel S (2008) Interindividual variability of electromyographic patterns and pedal force profiles in trained cyclists. Eur J Appl Physiol 104:667–678. doi:10.1007/s00421-008-0810-y PubMedCrossRefGoogle Scholar
  25. Laplaud D, Hug F, Grelot L (2006) Reproducibility of eight lower limb muscles activity level in the course of an incremental pedaling exercise. J Electromyogr Kinesiol 16:158–166. doi:10.1016/j.jelekin.2005.04.002 PubMedCrossRefGoogle Scholar
  26. Li L (2004) Neuromuscular control and coordination during cycling. Res Q Exerc Sport 75:16–22PubMedGoogle Scholar
  27. Li L, Caldwell GE (1998) Muscle coordination in cycling: effect of surface incline and posture. J Appl Physiol 85:927–934PubMedGoogle Scholar
  28. Lucia A, Sanchez O, Carvajal A, Chicharro JL (1999) Analysis of the aerobic-anaerobic transition in elite cyclists during incremental exercise with the use of electromyography. Br J Sports Med 33:178–185. doi:10.1136/bjsm.33.3.178 PubMedCrossRefGoogle Scholar
  29. MacIntosh BR, Neptune RR, Horton JF (2000) Cadence, power, and muscle activation in cycle ergometry. Med Sci Sports Exerc 32:1281–1287. doi:10.1097/00005768-200007000-00015 PubMedCrossRefGoogle Scholar
  30. Meyer RA, Prior BM (2000) Functional magnetic resonance imaging of muscle. Exerc Sport Sci Rev 28:89–92PubMedGoogle Scholar
  31. Mileva K, Turner D (2003) Neuromuscular and biomechanical coupling in human cycling: adaptations to changes in crank length. Exp Brain Res 152:393–403. doi:10.1007/s00221-003-1561-y PubMedCrossRefGoogle Scholar
  32. Moritani T, Takaishi T, Matsumoto T (1993) Determination of maximal power output at neuromuscular fatigue threshold. J Appl Physiol 74:1729–1734PubMedGoogle Scholar
  33. Neptune RR, Hull ML (1998) Evaluation of performance criteria for simulation of submaximal steady-state cycling using a forward dynamic model. J Biomech Eng 120:334–341. doi:10.1115/1.2797999 PubMedCrossRefGoogle Scholar
  34. Neptune RR, Kautz SA, Hull ML (1997) The effect of pedaling rate on coordination in cycling. J Biomech 30:1051–1058. doi:10.1016/S0021-9290(97)00071-7 PubMedCrossRefGoogle Scholar
  35. Pressel T, Lengsfeld M (1998) Functions of hip joint muscles. Med Eng Phys 20:50–56. doi:10.1016/S1350-4533(97)00040-4 PubMedCrossRefGoogle Scholar
  36. Raasch CC, Zajac FE, Ma B, Levine WS (1997) Muscle coordination of maximum-speed pedaling. J Biomech 30:595–602. doi:10.1016/S0021-9290(96)00188-1 PubMedCrossRefGoogle Scholar
  37. Richardson RS, Frank LR, Haseler LJ (1998) Dynamic knee-extensor and cycle exercise: functional MRI of muscular activity. Int J Sports Med 19:182–187. doi:10.1055/s-2007-971901 PubMedCrossRefGoogle Scholar
  38. Ryan MM, Gregor RJ (1992) EMG profiles of lower extremity muscles during cycling at constant workload and cadence. J Electromyogr Kinesiol 2:69–80. doi:10.1016/1050-6411(92)90018-E CrossRefGoogle Scholar
  39. Ting LH, Kautz SA, Brown DA, Zajac FE (1999) Phase reversal of biomechanical functions and muscle activity in backward pedaling. J Neurophysiol 81:544–551PubMedGoogle Scholar
  40. van Ingen Schenau GJ, Boots PJ, de Groot G, Snackers RJ, van Woensel WW (1992) The constrained control of force and position in multi-joint movements. Neuroscience 46:197–207. doi:10.1016/0306-4522(92)90019-X PubMedCrossRefGoogle Scholar
  41. Viitasalo JT, Luhtanen P, Rahkila P, Rusko H (1985) Electromyographic activity related to aerobic and anaerobic threshold in ergometer bicycling. Acta Physiol Scand 124:287–293. doi:10.1111/j.1748-1716.1985.tb07663.x PubMedCrossRefGoogle Scholar
  42. Watanabe K, Akima H (2008) Cross-talk from adjacent muscle has a negligible effect on surface electromyographic activity of vastus intermedius muscle during isometric contraction. J Electromyogr Kinesiol. doi:10.1016/j.jekin.2008.06.002

Copyright information

© Springer-Verlag 2009

Authors and Affiliations

  1. 1.Graduate School of Education and Human DevelopmentNagoya UniversityChikusaJapan
  2. 2.Research Center of Health, Physical Fitness & SportsNagoya UniversityChikusaJapan

Personalised recommendations