Test–retest reliability of cardinal plane isokinetic hip torque and EMG

Available online at www.sciencedirect.com

Journal of Electromyography and Kinesiology 19 (2009) e345–e352 www.elsevier.com/locate/jelekin

Test–retest reliability of cardinal plane isokinetic hip torque and EMG Tina L. Claiborne a, Mark K. Timmons b, Danny M. Pincivero b,* b

a Department of Athletic Training, Adrian College, Adrian, MI, United States Human Performance and Fatigue Laboratory, Department of Kinesiology, The University of Toledo, Mail Stop 119, Toledo, OH 43606, United States

Received 13 April 2008; received in revised form 16 July 2008; accepted 16 July 2008

Abstract The objective of the present study was to establish test–retest reliability of isokinetic hip torque and prime mover electromyogram (EMG) through the three cardinal planes of motion. Thirteen healthy young adults participated in two experimental sessions, separated by approximately one week. During each session, isokinetic hip torque was evaluated on the Biodex Isokinetic Dynamometer at a velocity of 60 deg/s. Subjects performed three maximal-effort concentric and eccentric contractions, separately, for right and left hip abduction/adduction, flexion/extension, and internal/external rotation. Surface EMGs were sampled from the gluteus maximus, gluteus medius, adductor, medial and lateral hamstring, and rectus femoris muscles during all contractions. Intraclass correlation coefficients (ICC – 2,1) and standard errors of measurement (SEM) were calculated for peak torque for each movement direction and contraction mode, while ICCs were only computed for the EMG data. Motions that demonstrated high torque reliability included concentric hip abduction (right and left), flexion (right and left), extension (right) and internal rotation (right and left), and eccentric hip abduction (left), adduction (left), flexion (right), and extension (right and left) (ICC range = 0.81–0.91). Motions with moderate torque reliability included concentric hip adduction (right), extension (left), internal rotation (left), and external rotation (right), and eccentric hip abduction and adduction (right), flexion (left), internal rotation (right and left), and external rotation (right and left) (ICC range = 0.49–0.79). The majority of the EMG sampled muscles (n = 12 and n = 11 for concentric and eccentric contractions, respectively) demonstrated high reliability (ICC = 0.81–0.95). Instances of low, or unacceptable, EMG reliability values occurred for the medial hamstring muscle of the left leg (both contraction modes) and the adductor muscle of the right leg during eccentric internal rotation. The major finding revealed high and moderate levels of between-day reliability of isokinetic hip peak torque and prime mover EMG. It is recommended that the dayto-day variability estimates concomitant with acceptable levels of reliability be considered when attempting to objectify intervention effects on hip muscle performance. Ó 2008 Elsevier Ltd. All rights reserved. Keywords: Abduction; Adduction; Flexion; Extension; Electromyography

1. Introduction Isokinetic exercise has consistently been used as a reliable measure for evaluating voluntary strength of several muscle groups in a variety of conditions. Specific to the hip joint, a number of studies have assessed the reliability of voluntary torque generation with subjects in different positions (Calmels et al., 1997; Donatelli et al., 1991; Emery et al., 1999; Kea et al., 2001; Poulmedis, 1985), *

Corresponding author. Tel.: +1 419 530 5291; fax: +1 419 530 2477. E-mail address: [email protected] (D.M. Pincivero).

1050-6411/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.jelekin.2008.07.005

and from a variety of populations (Arokoshi et al., 2002; Eng et al., 2002; Hsu et al., 2002). Cahalan et al. (1988) reported high test–retest reliability (intraclass correlation (ICC) coefficient = 0.96) of standing isokinetic concentric and eccentric hip abduction, adduction, flexion, and extension strength. Emery et al. (1999) determined reliability coefficients of 0.64 and 0.85 for side-lying concentric and eccentric hip adduction, respectively. Similarly, Kea et al. (2001) observed moderate to high test–retest reliability coefficients ranging from 0.59 to 0.85 for isokinetic concentric and eccentric hip adduction and abduction. High test– retest reliability coefficients have also been observed for

e346

T.L. Claiborne et al. / Journal of Electromyography and Kinesiology 19 (2009) e345–e352

isokinetic hip strength for patients with hemiparesis (ICC = 0.88–0.98; Eng et al., 2002; Hsu et al., 2002), and osteoarthritis (ICC = 0.84–0.98; Arokoshi et al., 2002). Although such studies have provided preliminary evidence of adequate test–retest reliability, an estimate of the between-day error has yet to be addressed. Furthermore, with few exceptions (Gupta et al., 2004), isokinetic assessment of transverse plane motion (i.e., hip internal and external rotation) does not appear to have been investigated thoroughly in the scientific literature. The ability to differentiate between inherent day-to-day variability and the natural variability commonly generated by injury or training effects, may provide a valid basis for clinically relevant activities or drawing inferences from research data (Croisier et al., 2007). Measurement of the surface electromyogram (EMG) from contracting muscles during tasks such as maximaleffort contractions have often been used as a normalization procedure for sub-maximal or complex, dynamic activities (Yang and Winter, 1984). The value in doing so is predicated on the assumption that such EMG values retain acceptable levels of reliability and low day-to-day variability. Although previous studies have examined hip muscle EMG activity during different voluntary tasks, the focus of such examinations have been limited to isometric contractions and few muscles (Bolgla and Uhl, 2007; Nyland et al., 2004; Worrell et al., 2001). As a result, much less attention has been directed at maximal-effort, full range of motion planar contractions. Furthermore, as many lower extremity dynamic activities, such as walking, involve a serial pattern of muscle lengthening and shortening, a considerable level of variability may be imposed on the obtained EMG measures, as a result of contraction mode. In order to lend credence to EMG signal interpretation during concentric and eccentric contractions, it is necessary to investigate the degree of repeatability of this measure under a relatively standardized condition, such as a single-joint maximal-effort contraction. To date, little evidence exists within the scientific literature regarding EMG signal reliability of the superficial hip muscles, as a function of contraction mode. Therefore, the objective of the present study was to evaluate test–retest reliability of concentric and eccentric isokinetic hip torque and surface EMG of the prime movers, through the three cardinal planes of motion. 2. Methods 2.1. Subjects Subjects for this study included 13 healthy volunteers (7 men, 6 women, mean ± standard deviation, age = 26.54 ± 3.80 years, height = 172.15 ± 8.21 cm, body mass = 76.38 ± 17.78 kg). To assess past medical history and physical activity levels, the Physical Activity Readiness Questionnaire (PAR-Q) (American College of Sports Medicine, 2000) was completed by all subjects prior to data

collection. Subjects with a history of orthopedic injury to the lower extremity within the past year, cardiovascular, pulmonary, neurological, or systemic conditions that limited activity level were excluded from the study. Additionally, subjects who had undergone surgery, had a diagnosed low back injury, or had been diagnosed with a previous ligament injury of the knee, ankle or hip were also excluded from this investigation. Written informed consent was obtained from all subjects as approved by the Human Subjects Research and Review Committee at The University of Toledo. 2.2. Isokinetic hip torque Subjects participated in the experimental procedures over two days, separated by approximately one week. The testing procedures for each day of testing were identical. During the testing sessions, subjects’ hip abduction, adduction, flexion, extension, internal rotation, and external rotation strength for both legs were evaluated using the Biodex Isokinetic Dynamometer (System 2, Biodex Medical Systems, Inc, Shirley, NY) at a pre-set angular velocity of 60 deg/s. Following a five-minute sub-maximal warm-up on a stationary cycle, and 2–3 sub-maximal and maximal familiarization repetitions, each subject performed three maximal-effort reciprocal contractions for each muscle group and contraction type. The single highest peak torque (N m) value obtained during each set of three repetitions was obtained and used for statistical analysis. Previously reported to enhance performance, concurrent visual feedback from a computer monitor and verbal encouragement were provided to all subjects to promote maximal efforts during all trials (Hald and Bottjen, 1987; Kellis and Baltzopolous, 1996; Kim and Kramer, 1997; Kimura et al., 1999; McNair et al., 1996). Two minutes of seated rest were provided between each set of three contractions. The cushion setting for all trials was in the ‘‘hard” position to minimize deceleration at the end of the range of the motion that would have adversely affected torque generation during the subsequent contraction (Ling et al., 1999; Taylor et al., 1991). The order of strength testing for all subjects was concentric followed by eccentric right leg abduction/adduction, right leg flexion/extension, left leg abduction/adduction, left leg flexion/extension, right leg internal/external rotation, and left leg internal/external rotation. Each set of three contractions within each plane of movement was reciprocal, such that motion in one direction was immediately preceded by motion into the opposite direction. This order of testing was chosen for purposes of experimental convenience, and was assumed to have negligible effects on test–retest reliability. The hip motions of abduction, adduction, flexion and extension were evaluated with subjects in a standing position. For these standing tests, the tested leg was attached to the dynamometer resistance adapter with a Velcro strap slightly above the knee, while the subjects stood on a three-inch block to eliminate possi-

T.L. Claiborne et al. / Journal of Electromyography and Kinesiology 19 (2009) e345–e352

ble tested leg contact with the floor. A stable handhold was provided to the subjects to ensure stability. The greater trochanter of the femur was used as the anatomical reference at which the axis of the dynamometer was aligned. The range of motion for all movements was set using a manual goniometer by the same investigator throughout testing. The abduction/adduction range of motion was set from 10 deg of hip adduction to 30 deg of hip abduction, and the flexion/extension range of motion was set from 0 deg in the standing position to 60 deg of hip flexion. Gravity correction was performed automatically by the Biodex Advantage software program using a static torque measure obtained at 30 deg of hip abduction for the abduction/ adduction test and 60 deg of hip flexion for the flexion/ extension test. To evaluate internal/external rotation, subjects sat on a separate chair facing the dynamometer and Velcro straps were used to secure the lower leg just above the malleoli to the resistance adapter, and the thigh to the chair. The longitudinal axis, through the distal end of the thigh, was aligned with the axis of rotation of the dynamometer. The range of motion was set from 10 deg of internal rotation to 10 deg of external rotation. Gravity correction was obtained by measuring the static torque produced while the hip was in a position of 10 deg of internal rotation. 2.3. Measurement of hip muscle EMG Following adequate skin preparation (i.e., shaving, if necessary, and cleaning with isopropryl alcohol), electromyograms of the gluteus maximus, gluteus medius, adductor, rectus femoris, and the medial and lateral hamstring muscles were sampled, via disposable surface electrodes (Ag/AgCl; 0.8 cm diameter, center-to-center inter-electrode distance = 1.5 cm; Noraxon USA, Inc, Scottsdale, AZ). Electrode placement for the muscles were as follows – gluteus maximus: one-half of the distance between the greater trochanter and the ischial tuberosity; gluteus medius: one-half of the distance between the iliac crest and the greater trochanter; adductor muscle: one third of the distance between the pubic symphysis and the adductor tubercle; rectus femoris: one-half of the distance from the anterior superior iliac spine to the superior pole of the patella; medial and lateral hamstring: one-half of the distance between the ischial tuberosity and the medial and lateral femoral epicondyles (Delagi et al., 1981; Zipp, 1982 ). All electrodes were placed in a manner such that the bipolar recording sites were aligned with an approximated muscle fiber direction. The reference electrode was placed over the medial shaft of the tibia, 6– 8 cm below the inferior pole of the patella. Electromyographic activity was pre-amplified (100) 3 cm from the electrodes, and sampled at 2000 Hz by a differential amplifier (Bioengineering Technology Systems, Milan, Italy) set to a gain of 1000 for each muscle (common mode rejection ratio, 100 dB). Raw EMG signals were digitized and stored on computer disk for subsequent analysis by the Acknowl-

e347

Gluteus medius

Adductor

Position

Velocity

Torque

Fig. 1. Sample EMG, torque and position tracing obtained during a reciprocal, concentric abduction/adduction contraction. The highlighted section identifies the analysis window over a region of constant velocity in which the EMG signals were analyzed.

edge software program, version 3.2.6 (Biopac Systems, Santa Barbara, CA). The raw EMG signals were bandpass filtered (20–500 Hz), full-wave rectified, and integrated (IEMG), over the constant velocity phase of each contraction (Fig. 1). Each IEMG value was subsequently divided by the duration of the analysis window over the period of constant angular velocity. Average IEMG values were then determined by calculating the mean of the three consecutive contractions of each muscle during each contraction mode (concentric and eccentric) and motion. 2.4. Statistical analysis Descriptive data (mean, standard deviation, and 95% confidence intervals) were calculated for the single highest torque value of each plane of motion, direction, and contraction mode on both days of testing. Test–retest reliability of isokinetic peak torque was assessed via the intraclass correlation (ICC) coefficient (2, 1) for the right and left legs, separately (Shrout and Fleiss, 1979). The standard error of measurement (SEM) was calculated and expressed in absolute (N m) units for each variable. The ICCs were calculated for the EMG values of those muscles considered as prime movers during each contraction mode and direction (Norkin and Levangie, 1992). Specifically, the EMG reliability statistics are presented for the following muscles and movement directions: adduction: adductor muscles; abduction: gluteus medius; flexion: rectus femoris and adductors; extension: gluteus maximus, lateral hamstring, medial hamstring; internal rotation: adductors; external rotation: gluteus maximus and gluteus medius; and, internal rotation: adductors. 3. Results 3.1. Isokinetic hip torque Descriptive data for concentric and eccentric isokinetic peak torque of each motion on both days of testing are presented in Tables 1 and 2, respectively. Test–retest reliability

e348

T.L. Claiborne et al. / Journal of Electromyography and Kinesiology 19 (2009) e345–e352

coefficients and SEMs (absolute and relative units) are presented in Table 3. The results demonstrated high test–retest reliability coefficients (ICC > 0.80), bi-laterally, for concentric hip abduction, flexion, and extension (range ICC = 0.80–0.90; SEM range = 10.40–14.84 N m). High uni-lateral reliability coefficients were observed for concentric adduction (left), internal rotation (right) and external rotation (left), and eccentric hip abduction (left), adduction (left), flexion (right), and extension (left) (range ICC = 0.80–0.91; SEM range = 7.80–14.68 N m). Motions with moderate reliability (ICC: 0.50–0.79) included concentric hip adduction (right), internal rotation (left), and external

Table 1 Descriptive data [means, standard deviation (SD), standard error (SE) and 95% confidence intervals (CI)] for concentric hip abduction (ABD), adduction (ADD), flexion (FLEX), extension (EXT), internal rotation (IR) and external rotation (ER) isokinetic peak torque (N m) for the left and right legs, on both days Left leg

Right leg

Mean

SD

95% C.I.

Mean

SD

95% C.I.

Day 1 ABD ADD FLEX EXT IR ER

108.88 117.48 124.83 122.66 55.08 74.02

32.79 35.06 31.37 32.22 17.14 19.49

89.07–128.70 96.29–138.66 105.88–143.79 103.19–142.13 44.20–65.97 62.24–85.80

120.61 122.70 133.74 139.65 83.95 51.96

38.53 35.58 37.49 33.78 25.31 16.52

97.33–143.90 101.20–144.20 111.08–156.39 119.24–160.06 68.66–99.25 41.71–61.67

Day 2 ABD ADD FLEX EXT IR ER

114.12 127.76 126.20 136.59 61.72 69.29

35.98 40.56 34.05 34.55 17.98 15.17

92.37–135.86 103.25–152.27 105.62–146.78 115.71–157.47 51.80–72.75 60.13–78.46

126.59 134.30 126.78 140.04 88.19 51.95

41.39 47.27 25.89 31.77 28.94 11.93

101.58–151.61 105.73–162.87 111.14–142.43 120.84–159.24 70.71–105.68 44.74–59.15

Table 2 Descriptive data (means, standard deviation (SD), standard error (SE) and 95% confidence intervals (C.I.)) for eccentric hip abduction (ABD), adduction (ADD), flexion (FLEX), extension (EXT), internal rotation (IR) and external rotation (ER) isokinetic peak torque (N m) for the left and right legs, on both days Left leg

Right leg

Mean

SD

95% C.I.

Mean

SD

95% C.I.

Day 1 ABD ADD FLEX EXT IR ER

109.48 106.73 132.34 120.32 103.82 76.15

27.89 32.53 35.38 30.91 17.05 31.38

92.62–126.33 87.07–126.39 110.96–153.72 101.64–138.99 93.52–114.12 57.18–95.11

123.92 115.25 134.98 140.21 70.63 120.85

39.61 43.35 28.34 45.50 26.44 36.41

99.98–147.86 89.06–141.45 117.85–152.10 112.71–167.70 54.65–86.61 98.85–142.85

Day 2 ABD ADD FLEX EXT IR ER

115.25 111.31 132.05 129.75 100.51 71.00

28.42 29.54 34.89 35.44 22.64 21.22

98.07–132.42 93.46–129.16 110.96–153.13 108.33–151.17 86.82–114.19 58.18–83.82

131.02 111.54 135.71 132.48 72.68 114.20

32.41 29.96 34.90 34.26 30.58 38.33

111.44–150.61 93.43–129.64 114.62–156.80 111.78–153.19 54.20–91.17 91.03–137.37

Table 3 Intraclass correlation (ICC) coefficients and standard errors of measurement (SEM) for concentric and eccentric isokinetic hip abduction (ABD), adduction (ADD), flexion (FLEX), extension (EXT), internal rotation (IR) and external rotation (ER) isokinetic peak torque (N m) of the left and right legs Left ICC

Left SEM (N m)

Right ICC

Right SEM (N m)

Concentric ABD 0.87 ADD 0.87 FLEX 0.82 EXT 0.80 IR 0.72a ER 0.80

12.30 13.83 13.92 14.84 9.24 7.80

0.89 0.66a 0.83 0.90 0.87 0.62a

13.55 24.11 13.16 10.40 9.64 8.77

Eccentric ABD 0.82 ADD 0.90 FLEX 0.74a EXT 0.80 IR 0.49a ER 0.64a

12.10 10.03 18.06 14.68 14.18 15.80

0.78a 0.79a 0.91 0.76a 0.68a 0.79a

16.87 16.99 9.42 19.49 16.18 17.08

a

Indicates low to moderate reliability estimates.

rotation (right), and eccentric hip abduction and adduction (right), flexion (left), extension (right), internal rotation (right), and external rotation (right and left) (ICC range = 0.62–0.79; SEM range = 8.77–24.11 N m). Eccentric hip internal rotation (left) was observed to display low reliability (ICC = 0.49; SEM = 14.18 N m). 3.2. Hip muscle EMG Test–retest reliability coefficients and SEMs for hip muscle EMG during the concentric and eccentric contractions are presented in Tables 4 and 5, respectively. The results demonstrated high reliability estimates for 12 muscles when acting as prime movers (ICC = 0.81–0.95) during the concentric contractions, while six muscles revealed moderate reliability (ICC = 0.50–0.76). Similar findings were observed for hip EMG during the eccentric contractions as most muscles (n = 11) displayed high reliability measures (ICC = 0.82–0.95). A low, and an aberrant, reliability coefficient was observed for the medial hamstring muscle of the left leg (ICC = 0.42) and the adductor muscle (ICC = 1.46) during the eccentric internal rotation contraction, respectively. The only discernible pattern observed between the two contraction modes occurred for the medial hamstring muscle of the left leg in which a low and moderate reliability coefficient was observed. 4. Discussion The major findings of the present study demonstrated that between-day measures of isokinetic hip peak torque and muscle EMG are highly reliable for selected motions of the right and left legs during maximal-effort contractions. With respect to peak torque, high reliability estimates were observed for all concentric and eccentric

T.L. Claiborne et al. / Journal of Electromyography and Kinesiology 19 (2009) e345–e352 Table 4 Intraclass correlation (ICC) coefficients for surface EMG of the hip prime movers for the following motions: concentric isokinetic hip abduction, adduction, flexion, extension, internal rotation and external rotation Motion Abduction Adduction Flexion Extension

Internal rotation External rotation a

Muscle Gluteus medius Adductors Rectus femoris Gluteus maximus Lateral hamstring Medial hamstring Adductors Gluteus maximus Gluteus medius

Right ICC a

0.59 0.59a 0.76 0.83 0.93 0.75 0.89 0.83 0.84

Left ICC 0.88 0.81 0.65a 0.95 0.90 0.50a 0.81 0.82 0.81

Indicates low reliability estimates.

Table 5 Intraclass correlation (ICC) coefficients for surface EMG of the hip prime movers for the following motions: eccentric isokinetic hip abduction, adduction, flexion, extension, internal rotation and external rotation Motion

Muscle

Right ICC

Left ICC

Abduction Adduction Flexion

Adductors Gluteus medius Gluteus maximus Lateral hamstring Medial hamstring Rectus femoris Gluteus maximus Gluteus medius Adductors

0.85 0.89 0.73 0.88 0.85 0.93 0.54a 0.51a 1.46a

0.94 0.77 0.94 0.82 0.42a 0.77 0.83 0.86 0.95

Extension Internal rotation External rotation a

Indicates low, or unacceptable, reliability estimates.

measures (uni-lateral or bilateral), with the only exceptions being eccentric internal and external rotation. The findings also illustrated that the EMG measures retained high levels of reliability for most of the observed muscles. Particular attention, however, should be focused on the few isokinetic and EMG measurements that yielded relatively low levels of reliability, as future application should be done so with caution. 4.1. Isokinetic hip torque reliability Voluntary torque measurement, as assessed through isokinetic dynamometry, has been common practice and performed reliably, particularly when evaluating concentric knee extensor and flexor muscle strength (Feiring et al., 1990; Larsson et al., 2003; Montgomery et al., 1989; Pincivero et al., 1997). However, conclusive and reliable data regarding isokinetic hip torque have been limited within the scientific literature. With respect to able-bodied adults, Cahalan et al. (1989) evaluated concentric and eccentric hip abduction/adduction, flexion/extension, and internal/ external rotation in subjects in standing and seated positions and presented a single reliability value (ICC = 0.96) from pilot work. In a study examining the relation between isokinetic hip flexor/extensor strength and sprint speed, Guskiewicz et al. (1993) observed within-day reliability

e349

coefficients ranging from 0.82–0.96 at 60 and 240 deg/s. Slightly lower reliability values (ICC = 0.59–0.85) were reported by Kea et al. (2001) for concentric and eccentric isokinetic hip abduction/adduction in amateur hockey players. Although this latter study reported the SEM values, they were expressed in absolute (N m) units rendering their generalizations limited. The expected error generated from repeated testing in the present study demonstrated a wide range of values that were specific to a particular plane, contraction mode, and direction. The importance of providing SEM data for future applications in which a pretest–posttest design is applied highlights the clinical and research benefits of an isokinetic evaluation. Although healthy young adults served as the cohort in the present investigation, previous studies have demonstrated the reliability of isokinetic hip strength testing in various populations. In 20 older adults that had experienced a cerebrovascular accident (CVA), Eng et al. (2002) reported high reliability coefficients for isokinetic hip extension and flexion peak torque for the non-involved (ICC = 0.98 and ICC = 0.95) and hemiparetic (ICC = 0.97 and ICC = 0.98) sides. Similar reliability data were observed by Hsu et al. (2002) in nine patients with hemiparesis resulting from a single CVA; specifically, the repeated assessment of isokinetic hip flexor peak torque for the hemiparetic and non-involved limbs demonstrated ICC values of 0.91 and 0.89, respectively at 30 deg/s, and 0.93 and 0.88, respectively, at 90 deg/s. With respect to older adults with hip osteoarthritis, Arokoshi et al. (2002) reported high reliability values ranging from 0.84 – 0.98 for hip flexion/ extension and abduction/adduction peak torque at 60 and 120 deg/s. Rossi et al. (2006) observed high test–retest reliability coefficients (ICC = 0.88–0.94) in older adults prior to, and following, total hip arthroplasty, for hip extension/flexion peak torque at 60 deg/s. The data reported in the present investigation are somewhat consistent with the findings of the cited studies examining ablebodied young adults and specific patient populations. However, the instances in which moderate reliability torque values were observed suggest limited use of those particular actions may be warranted. Although isokinetic hip strength testing has received relatively less attention in the scientific literature than the knee joint, investigations have revealed the potential importance of this measurement to lower extremity function. Farrar and Thorland (1987) first observed low correlations between isokinetic hip flexion/extension torque at two different isokinetic velocities (60 and 300 deg/s) and sprint speed. However, Guskiewicz et al. (1993) later reported higher correlations between isokinetic hip flexion/extension torque and 40-yard sprint speed (r = 0.41– 0.57) in collegiate athletes, and surmised that testing position (i.e., standing versus seated or lying) played a significant role in their findings. Low correlation coefficients were reported by Kea et al. (2001) between isokinetic concentric and eccentric hip abduction/adduction peak torque and a lateral single-leg hop for distance in amateur hockey

e350

T.L. Claiborne et al. / Journal of Electromyography and Kinesiology 19 (2009) e345–e352

players. A subsequent study demonstrated that the mean ball velocity of a soccer kick was significantly related to swing leg hip adductor (r = 0.55–0.68), flexor (r = 0.61), and abductor (r = 0.53–0.57) peak torque as well as support leg hip extensor (r = 0.53–0.70) and abductor (r = 0.53) peak torque (Masuda et al., 2005). Upon examining the results of these studies, it is apparent that a meaningful application of isokinetic hip strength measures remains largely dependent on the specific dynamic task performed. Claiborne et al. (2006) recently demonstrated a significant relationship (r = 0.37) between isokinetic concentric hip abduction peak torque and frontal plane knee motion (i.e., valgus knee rotation), which has been implicated as mechanism of knee injury (Markolf et al., 1995), during a single-leg squat. The objectivity afforded by an isokinetic peak torque evaluation, as illustrated by the data in the present study and supported by others, demonstrates that this mode of strength assessment is highly reliable. 4.2. Hip muscle EMG reliability The generation of electrical signals (i.e., the EMG) from contracting muscles has long been established to have a close correspondence to force production under nonfatiguing conditions (Lawrence and DeLuca, 1983; Pincivero and Coelho, 2000). However, such examinations have been limited to isometric conditions and may not be reflective of muscle contractions involving joint motion. In order to gain further understanding of the contributory muscle actions that drives human movement, capturing reliable EMG data becomes critically important in order to draw reasonable research or clinical conclusions. The results from the present study demonstrated that most superficial hip muscles, when acting as prime movers, generated reliable EMG signals with various magnitudes. Similar observations were reported by Bolgla and Uhl (2007) in 13 healthy young adults performing a variety of leg abduction and single-leg weight bearing tasks while gluteus medius EMG activity was recorded. The findings from this previous study yielded ICC values greater than 0.84, with the exception of one exercise, while intra-subject variability ranged from 9–22%. However, the Bolgla and Uhl (2007) study only examined the EMG reliability from consecutively performed contractions and did not evaluate repeatability between separate testing sessions. Nyland et al. (2004) observed significantly lower gluteus medius EMG estimates in athletically-active women who displayed a relatively greater amount of femoral anteversion, while performing an isometric hip abduction-external rotation maneuver. In addition, an ICC coefficient of 0.97 (SEM = 0.7%) was calculated but the precise inter-trial comparison was not specified. With respect to high-intensity concentric and eccentric muscle contractions, Finucane et al. (1998) reported moderate to high knee extensor EMG reliability coefficients ranging from 0.62 to 0.97 in 10 healthy, young adults. Although test–retest reliability coef-

ficients were not reported, Croisier et al. (2007) observed between-day coefficients of variation ranging from 0.5% to 25% for the quadriceps femoris and hamstring muscle EMG during varying range of motion, maximal-effort isokinetic concentric and eccentric contractions. As the cited studies addressing EMG signal reproducibility present results from varied methodology and different muscles, there appears to be a similar pattern to the current investigation where moderate to high reliability values can be expected for the majority of the sampled muscles. To date, however, few data exist regarding hip muscle EMG during maximal-effort isokinetic contractions that would allow a comparison with the results from the present study; as a result, it is recommended that the presented reliability values be used as an index to gauge future applications of EMG analysis of the superficial hip muscles. It was observed in the present investigation that the generated reliability coefficients were, in some cases, considerably different between the right and left legs. Based upon inspection of Tables 3–5, there appeared to be no discernible pattern regarding the leg that generated higher reliability coefficients. A limitation to the present investigation includes the fact that limb dominance was not ascertained from the subjects, and could not be evaluated for its’ repeatability characteristics. With respect to the measures of muscle recruitment, only surface EMGs were obtained thereby limiting the examination to few muscles that contributed to some of the given motions. This limitation was best exemplified by the reliability coefficients for the right leg muscle during eccentric internal and external hip rotation (Table 5), as the primary movers for these motions lie deep to the gluteus maximus (Norkin and Levangie, 1992). Particular note should be made to the discrepant number of ICC values that fall within the low to moderate category. Evident from Tables 3 and 4, isokinetic peak torque reliability was low to moderate for eight measures (five right and three left) during the eccentric mode while only three measures were considered low to moderate for the concentric contractions. Furthermore, the majority of both isokinetic and EMG measures from the internal/external rotation tests were categorized as low to moderate, thereby raising doubt as to the applicability of evaluating this particular motion. It is recommended that additional exposures and practice to isokinetic hip internal/external rotation should be implemented in order to obtain more reliable estimates of torque recordings. 5. Conclusion The measurement of hip muscle strength has objectively demonstrated distinct advantages in terms of quantifying bi-lateral strength deficits resulting from stroke (Eng et al., 2002; Hsu et al., 2002) and osteoarthritis (Arokoshi et al., 2002), as well as following total hip arthroplasty (Rossi et al., 2006). Evaluating hip muscle strength as a predictor of lower extremity functional performance, however, has been shown to be task specific and moderate. In

T.L. Claiborne et al. / Journal of Electromyography and Kinesiology 19 (2009) e345–e352

order to extend research to clinically relevant applications isokinetic exercise must generate reliable data that concomitantly provides an indication of test–retest error. The results from the present investigation outline a useful guideline of interpreting isokinetic hip strength and muscle EMG assessment of the prime movers. It is, therefore, recommended that the ICC and relative SEM values reported in the present study be reviewed and applied to future clinical or research approaches that involve the serial assessment of hip muscle strength and myoelectrical activity. Acknowledgement The authors wish to thank Dr. Will G. Hopkins (Professor of Exercise Science, Auckland University of Technology) for his statistical advice and assistance. References American College of Sports Medicine. In: Franklin BA, Whaley MH, Howley ET, editors. ACSM’s guidelines for exercise testing and prescription. 6th ed. Baltimore: Lippincott Williams & Wilkins; 2000. p. 22–32. Arokoshi MH, Arokoshi JP, Haara M, Kankaanpaa M, Vesterinen M, Niemitukia LH, et al. Hip muscle strength and muscle cross sectional area in men with and without osteoarthritis. Journal of Rheumatology 2002;29(10):2185–95. Bolgla LA, Uhl TL. Reliability of electromyographic normalization methods for evaluating the hip musculature. Journal of Electromyography and Kinesiology 2007;17:102–11. Cahalan TD, Johnson ME, Liu S, Chao EYS. Quantitative measurements of hip strength in different age groups. Clinical Orthopaedics and Related Research 1989;246:136–45. Calmels PM, Nellen M, Inge van der Borne BS, Jourdin P, Minaire P. Concentric and eccentric isokinetic assessment of flexor–extensor torque ratios at the hip, knee, and ankle in a sample population of healthy subjects. Archives of Physical Medicine and Rehabilitation 1997;78:1224–30. Claiborne TL, Armstrong CW, Gandhi V, Pincivero DM. Relationship between hip and knee strength and knee valgus during a single leg squat. Journal of Applied Biomechanics 2006;22:41–50. Croisier JL, Malnati M, Reichard LB, Peretz C, Dvir Z. Quadriceps and hamstring isokinetic strength and electromyographic activity measured at different ranges of motion: a reproducibility study. Journal of Electromyography and Kinesiology 2007;17:484–92. Delagi EF, Perotto A, Iazzetti J, Morrison D. Anatomical guide for the electromyographer – the limbs. 2nd ed. Springfield, IL: Thomas; 1981. Donatelli R, Catin PA, Backer GS, Drane DL, Slater SM. Isokinetic hip abductor to adductor torque ratio in normals. Isokinetics and Exercise Science 1991;1:103–11. Emery CA, Maitland ME, Meeuwisse WH. Test–retest reliability of isokinetic hip adductor and flexor muscle strength. Clinical Journal of Sports Medicine 1999;9(2):79–85. Eng JJ, Kim CM, MacIntyre DL. Reliability of lower extremity strength measures in persons with chronic stroke. Archives of Physical Medicine and Rehabilitation 2002;83:322–8. Farrar M, Thorland W. Relationship between isokinetic strength and sprint times in college-age men. Journal of Sports Medicine and Physical Fitness 1987;27:368–72. Feiring DC, Ellenbecker TS, Derscheid GL. Test–retest reliability of the Biodex Isokinetic Dynamometer. Journal of Orthopaedic and Sports Physical Therapy 1990;11:298–300. Finucane SD, Rafeei T, Kues J, Lamb RL, Mayhew TP. Reproducibility of electromyographic recordings of submaximal concentric and

e351

eccentric muscle contractions in humans. Electroencephalography and Clinical Neurophysiology 1998;109:290–6. Gupta A, Fernihough B, Bailey G, Bombeck P, Clarke A, Hopper D. An evaluation of differences in hip external rotation strength and range of motion between female dancers and non-dancers. British Journal of Sports Medicine 2004;38:778–83. Guskiewicz K, Lephart S, Burkholder R. The relationship between sprint speed and hip flexion/extension strength in collegiate athletes. Isokinetics and Exercise Science 1993;3(2):111–6. Hald RD, Bottjen EJ. Effect of visual feedback on maximal and submaximal isokinetic test measurements of normal quadriceps and hamstrings. Journal of Orthopaedic and Sports Physical Therapy 1987;9:86–93. Hsu AL, Tang PF, Jan MH. Test–retest reliability of isokinetic muscle strength of the lower extremities in patients with stroke. Archives of Physical Medicine and Rehabilitation 2002;83:1130–7. Kea J, Kramer J, Forwell L, Birmingham T. Hip abduction–adduction strength and one-leg hop tests: test–retest reliability and relationship to function in elite ice hockey players. Journal of Orthopaedic and Sports Physical Therapy 2001;31(8):446–55. Kellis E, Baltzopolous V. Resistive eccentric exercise: effects of visual feedback on maximal moment of knee extensors and flexors. Journal of Orthopaedic and Sports Physical Therapy 1996;23(2):120–4. Kim HJ, Kramer JF. Effectiveness of visual feedback during isokinetic exercise. Journal of Orthopaedic and Sports Physical Therapy 1997;26:318–23. Kimura EF, Gulick DT, Lukasiewicz WC. Effect of visual feedback and verbal encouragement on eccentric quadriceps and hamstrings peak torque. Sports Medicine, Training and Rehabilitation 1999;9(1):61–70. Larsson B, Karlsson S, Eriksson M, Gerdle B. Test–retest reliability of EMG and peak torque during repetitive maximum concentric knee extensions. Journal of Electromyography and Kinesiology 2003;13:281–7. Lawrence JH, DeLuca CJ. Myoelectric signal versus force relationship in different human muscles. Journal of Applied Physiology 1983;54:1653–9. Ling W, Chen FC, McDonough AL. Evaluation of the cushion setting on performance of a Biodex II dynamometer. Archives of Physical Medicine and Rehabilitation 1999;80:350–4. Markolf KL, Burchfield DM, Shapiro MM, Shepard MF, Finerman GAM, Slauterbeck JL. Combined knee loading states that generate high anterior cruciate ligament forces. Journal of Orthopedic Research 1995;13:930–5. Masuda K, Kikuhara N, Demura S, Katsuta S, Yamanaka K. Relationship between muscle strength in various isokinetic movements and kick performance among soccer players. Journal of Sports Medicine and Physical Fitness 2005;45(1):44–52. McNair PJ, Depledge J, Brettkelly M, Stanley SN. Verbal encouragement: effects of maximum effort voluntary muscle activation. British Journal of Sports Medicine 1996;30:243–5. Montgomery LC, Douglass LW, Deuster PA. Reliability of an isokinetic test of muscle strength and endurance. Journal of Orthopaedic and Sports Physical Therapy 1989;10:315–22. Norkin CC, Levangie PK. Joint structure and function: a comprehensive analysis. 2nd ed. Philadelphia, PA: F.A. Davis Company; 1992. Nyland J, Kuzemchek S, Parks M, Caborn DN. Femoral anteversion influences vastus medialis and gluteus medius EMG amplitude: composite hip abductor EMG amplitude ratios during isometric combined hip abduction-external rotation. Journal of Electromyography and Kinesiology 2004;14(2):255–61. Pincivero DM, Lephart SM, Karunakara RA. Reliability and precision of isokinetic strength and musclar endurance for the quadriceps and hamstrings. International Journal of Sports Medicine 1997;18:113–7. Pincivero DM, Coelho AJ. Activation linearity and parallelism of the superficial quadriceps across the isometric intensity spectrum. Muscle and Nerve 2000;23(3):393–8.

e352

T.L. Claiborne et al. / Journal of Electromyography and Kinesiology 19 (2009) e345–e352

Poulmedis P. Isokinetic maximal torque power of Greek elite soccer players. Journal of Orthopaedic and Sports Physical Therapy 1985;6:293–5. Rossi MD, Brown LE, Whitehurst MA. Assessment of hip extensor and flexor strength two months after unilateral total hip arthroplasty. Journal of Strength and Conditioning Research 2006;20(2):262–7. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychological Bulletin 1979;86:420–8. Taylor NAS, Sanders RH, Howick EI, Stanley SN. Static and dynamic assessment of the Biodex dynamometer. European Journal of Applied Physiology 1991;62:180–8. Worrell TW, Karst G, Adamczyk D, Moore R, Stanley C, Steimel B, et al. Influence of joint position on electromyographic and torque generation during maximal voluntary isometric contractions of the hamstrings and gluteus maximus muscles. Journal of Orthopaedic and Sports Physical Therapy 2001;31(12):730–40. Yang JF, Winter DA. Electromyographic amplitude normalization methods: improving their sensitivity as diagnostic tools in gait analysis. Archives of Physical Medicine and Rehabilitation 1984;65:517–21. Zipp P. Recommendations for the standardization of lead positions in surface electromyography. European Journal of Applied Physiology 1982;50:41–54.

Tina L. Claiborne Ph.D., ATC, CSCS has been a certified athletic trainer since 1995. With her graduate work in kinesiology and biomechanics, she transitioned to academia in 2003, where she directed the athletic training education program at the University of Southern Maine. Recently, Dr. Claiborne moved to Adrian, MI where she is an Assistant Professor and serving as the Director of Athletic Training.

Mark K. Timmons completed his Ph.D. in Biomechanics in the Department of Kinesiology at The University of Toledo in 2005. Prior to his doctoral studies, Dr. Timmons completed his M.S. in Biomechanics at the University of Michigan and has worked numerous years as an Athletic Training in the university and clinical settings.

Danny M. Pincivero completed his B.A. (Physical Education) from York University (1992), M.Ed. (Athletic Training) from the University of Virginia (1993), Ph.D. (Exercise Physiology) from the University of Pittsburgh (1997), and B.S.E. (Mechanical Engineering) from The University of Toledo (2007). Currently, Dr. Pincivero is an Associate Professor in the Department of Kinesiology at The University of Toledo.