Research Report
Construct Validity of Muscle Force Tests of the Rotator Cuff Muscles: An Electromyographic Investigation Rebecca L. Brookham, Linda McLean, Clark R. Dickerson R.L. Brookham, MSc, is a doctoral student in the Department of Kinesiology, University of Waterloo, Waterloo, Ontario, Canada. L. McLean, PhD, is Associate Professor, School of Rehabilitation Therapy, Queens University, Kingston, Ontario, Canada. C.R. Dickerson, PhD, is Assistant Professor, Department of Kinesiology, University of Waterloo, 200 University Ave W, Waterloo, Ontario, N2L 3G1 Canada. Address all correspondence to Dr Dickerson at:
[email protected]. [Brookham RL, McLean L, Dickerson CR. Construct validity of muscle force tests of the rotator cuff muscles: an electromyographic investigation. Phys Ther. 2010;90: 572–580.] © 2010 American Physical Therapy Association
Background. Manual muscle tests (MMTs) are used in clinical settings to evaluate the function and strength (force-generating capacity) of a specific muscle in a position at which the muscle is believed to be most isolated from other synergists and antagonists. Despite frequent use of MMTs, few electromyographic evaluations exist to confirm the ability of MMTs to isolate rotator cuff muscles. Objective. This study examined rotator cuff isolation during 29 shoulder muscle force tests (9 clinical and 20 generic tests).
Design. An experimental design was used in this study. Participants and Measurements. Electromyographic data were recorded from 4 rotator cuff muscles and 10 additional shoulder muscles of 12 male participants. Maximal isolation ratios (mean specific rotator cuff muscle activation to mean activation of the other 13 recorded muscles) defined which of these tests most isolated the rotator cuff muscles.
Results. Three rotator cuff muscles were maximally isolated (obtained highest isolation ratios) within their respective clinical test groups (lateral rotator test group for the infraspinatus and teres minor muscles and abduction test group for the supraspinatus muscle). The subscapularis muscle was maximally isolated equally as effectively within the generic ulnar force and clinical medial rotation groups. Similarly, the supraspinatus and teres minor muscles were isolated equally as effectively in some generic test groups as they were in their respective clinical test groups.
Limitations. Postural artifact in the wire electrodes caused exclusion of some channels from calculations. The grouping of muscle force tests based on test criteria (clinical or generic tests and muscle action) may have influenced which groups most isolated the muscle of interest. Conclusions. The results confirmed the appropriateness of 9 commonly used clinical tests for isolating rotator cuff muscles, but suggested that several other muscle force tests were equally appropriate for isolating these muscles.
Post a Rapid Response or find The Bottom Line: www.ptjournal.org 572
f
Physical Therapy
Volume 90
Number 4
April 2010
Construct Validity of Muscle Force Tests
M
uscle force tests that isolate the rotator cuff muscles allow for uncompromised interpretation of muscle function and strength (force-generating capacity). Identification of these muscle force tests will promote accurate assessment of weakness, which in turn could direct strategies for strengthening and injury prevention. Manual muscle tests (MMTs) are used to evaluate a muscle’s strength in a position in which it is believed to be most isolated from other muscle contributions.1 During an MMT, individuals exert maximal effort in defined postures against static manual resistance provided by a clinician. Because one cooperative function of the rotator cuff muscles is to maintain the humeral head within the glenoid cavity,2 true isolation (when all other surrounding muscles are inactive) is an unlikely state. Relative isolation of the rotator cuff is more physiologically realistic and is defined as occurring in a muscle force test for which the muscle of interest is most activated and when all other surrounding muscles are least activated (relative to each muscle’s maximum). The definition of relative isolation does not require maximal activation of the muscle of interest, but rather greater activation of that muscle relative to the mean activation of the other surrounding muscles. Few evaluations have confirmed the ability of MMTs to isolate rotator cuff muscles. Studies have identified muscle force tests that produce maximal activation of the rotator cuff muscles, but frequently omitted consideration of contributions from surrounding muscles and, therefore, did not confirm isolation.3–7 Kelly et al8 did consider contributions from surrounding synergistic muscles when performing an electromyographic (EMG) examination of the rotator cuff muscles (excluding the teres minor muscle) to identify isolation muscle force tests. Optimal MMTs were April 2010
determined based on 4 criteria: maximal activation of the cuff muscle, minimal activation from involved synergists, good test-retest reliability, and minimal positional pain provocation. Twenty-nine isometric muscle force tests were performed—27 core muscle force tests (3 exertion tests [elevation, lateral rotation, and medial rotation], 3 scapular elevation tests [0°, 45°, and 90°], and 3 humeral rotation tests [45°, 0°, and ⫺45°]) and 2 other tests (Gerber push-off test and Gerber push-off with force test). The authors8 concluded that the optimal isolation tests were: elevation at 90 degrees of scapular elevation and 45 degrees of lateral rotation for the supraspinatus muscle, lateral rotation at 0 degrees of scapular elevation and 45 degrees of medial rotation for the infraspinatus muscle, and Gerber push with force test for the subscapularis muscle. One limitation of the study by Kelly et al8 was that they gave no evidence for the assumed synergists of the rotator cuff, ignoring the possibility that shoulder muscle function changes with posture9 and, therefore, that rotator cuff synergists may change as posture changes. Furthermore, the study was limited to recording the EMG activity of 8 muscles, so it was possible that key synergists were not measured (eg, the EMG activity of the teres minor muscle was not recorded, and it has been found to act in synergy with the infraspinatus muscle in lateral rotation3,10). The Gerber push with force test was identified as the optimal MMT for the subscapularis muscle. This conclusion may be flawed because this test was assessed independently from the core muscle force tests; it was excluded from the analysis of variance (ANOVA), as it did not fit the format of the other tests. Because this test was excluded from the ANOVA and the integrated EMG activity was assessed only by
rank order, it is not known whether the lift-off test produced a significantly higher integrated EMG signal in the subscapularis muscle than in the other 28 muscle force tests. Jenp et al11 used a different technique to isolate the rotator cuff muscles: from 29 test postures, those muscle force tests that produced the largest EMG activity for the rotator cuff muscles of interest were identified as potential postures of isolation. Minimal activation of surrounding muscles (pectoralis major; anterior, middle, and posterior deltoid; and the other 3 rotator cuff muscles) then was assessed from only these potential postures to identify postures of isolation. The authors’ initial selection of postures that maximally activated the rotator cuff muscles may have eliminated other potential isolation postures because a muscle may not have to be in a state of maximal activation to be isolated. Further research is needed to evaluate the ability of MMTs to relatively isolate the rotator cuff muscles. The primary objective of this research was to evaluate rotator cuff relative isolation during 29 muscle force tests. We hypothesized that
Available With This Article at ptjournal.apta.org • eAppendix 1: Descriptions of the Muscle Force Test Groups (Part A), & Maximal Voluntary Contractor Testing Positions (Part B) • eAppendix 2: Mean (SD) Percentage of Muscle Activation for Each Muscle Force Test for All Participants • Audio Abstracts Podcast This article was published ahead of print on February 4, 2010, at ptjournal.apta.org.
Volume 90
Number 4
Physical Therapy f
573
Construct Validity of Muscle Force Tests muscle force tests commonly used by clinicians would be most effective in isolating the rotator cuff muscles and that these muscles would be most isolated in muscle force tests based on their respective primary actions (lateral rotation for the teres minor and infraspinatus muscles, medial rotation for the subscapularis muscle, and abduction for the supraspinatus muscle).
Method Participants Students were recruited using poster advertisements. Twelve right-hand– dominant male university students (mean age⫽20.7 years, range⫽18 – 29; mean height⫽180.6 cm, range⫽165.1–193.1; and mean weight⫽76.7 kg, range⫽49.0 – 88.2) participated in the experiment after providing informed consent. All participants exhibited full range of motion in the shoulder. Exclusion criteria included a history of upper-limb or low back injury within the previous 6 months and known neuromuscular, cardiovascular, or metabolic conditions that might affect the participants’ safety during muscle force tests. Electromyography After cleaning the skin surface with Betadine solution containing 10% povidone-iodine,* 4 bipolar intramuscular electrodes† were inserted into the supraspinatus, infraspinatus, and teres minor muscles (needle: 27 gauge, 30-mm length; placements similar to those described by Delagi and Perotto12) and into the subscapularis muscle (needle: 25 gauge, 50-mm length; placement similar to that described by Nemeth et al13) on the right side. The needles were removed, and the wires (44 gauge) remained in the muscle during testing. After cleaning the skin surface with * Purdue Products LP, One Stamford Forum, Stamford, CT 06901-3441. † CareFusion (formerly Viasys Healthcare Inc), 3750 Torrey View Ct, San Diego, CA 92130.
574
f
Physical Therapy
Volume 90
isopropyl alcohol, 10 silver-silver chloride bipolar surface electrodes (product #272)‡ were placed on the right side over the latissimus dorsi; triceps brachii (long head); biceps brachii; anterior, middle, and posterior fibers of deltoid; pectoralis major (sternal and clavicular insertions); and middle and upper trapezius muscles using the placements described by Cram and Kasman.14 Electrodes were interfaced with a Noraxon Telemyo 2400T G2 wireless transmitter.‡ The EMG signals were preamplified close to the source (gain of 500) and band-pass filtered (10 –1,500 Hz). The sampling rate was 4,000 Hz. Functional tests (maximal voluntary contraction [MVC] muscle force tests described in eAppendix 1, part B; available at ptjournal.apta.org) were performed to ensure proper electrode placement. Testing Protocol Prior to experimental trials, muscle-specific (2 each) MVCs were performed in recommended postures14 (eAppendix 1, part B; available at ptjournal.apta.org). Participants then performed 29 ramped, maximal-effort, isometric contractions with their right arm. These 29 muscle force tests were divided into 7 groups containing functionally similar muscle force tests (eAppendix 1, part A; available at ptjournal.apta. org). Nine of these muscle force tests were clinically used tests, which were divided into lateral rotation, medial rotation, and abduction muscle force test groups. Twenty additional generic muscle force tests were performed to assess further isolation possibilities and were organized into groups based on hand force direction (palmar, dorsal, radial, or ulnar). Resistance to muscle force tests was provided by a stationary steel beam. Muscle force tests ‡ Noraxon USA Inc, 13430 N Scottsdale Rd, Suite 104, Scottsdale, AZ 85254.
Number 4
were performed in a randomized order (within and between groups). Trials were 6 seconds in duration (participants were instructed to ramp up to maximal activation during the second to fourth seconds of each trial and then relax for the remaining time), and 2 minutes of rest was enforced between muscle force tests, as recommended by De Luca.15 EMG Analysis Electromyographic data were processed using Matlab 7.0.1 software.§ All raw EMG data were full-wave rectified and filtered using a single-pass, second-order, low-pass Butterworth filter (3-Hz cutoff frequency). Peak activation levels of filtered MVC trials were chosen as 100% activation, and experimental trials (29 muscle force tests) were normalized to these peaks by participant and muscle (expressed as %MVC). Mean muscle activations for each muscle for each muscle force test were calculated during the second to fourth seconds (during maximal activation) of each 6-second trial. Isolation Ratios To identify which muscle force tests most isolated the muscles of interest, we determined during what muscle force test there was a maximal amount of EMG activity in the muscle of interest, relative to a minimal EMG activation of all of the other recorded muscles. This was determined by an isolation ratio (IR) (equation 1), which was used to define muscular isolation for each rotator cuff muscle in each muscle force test. (1)
Isolation Ratio (IR) ⫽ (% MVC activity of rotator cuff muscle of interest/100) (兺%MVC of all other 13 recorded muscles/1,300)
§
The MathWorks Inc, 3 Apple Hill Dr, Natick, MA 01760-2098.
April 2010
Construct Validity of Muscle Force Tests To illustrate the meaning of an IR, consider IRs equal to zero, 1, and infinity: an IR of zero indicates that the rotator cuff muscle of interest was not activated (turned off); an IR of 1 indicates that the rotator cuff muscle of interest was activated equally as much as the mean activation of the other 13 recorded muscles; and an IR of infinity indicates that the rotator cuff muscle of interest was active when all the other 13 recorded muscles were not activated (turned off), representing true isolation. Furthermore, an IR greater than 1 (eg, 1.5) indicates the rotator cuff muscle of interest is activated more (eg, 1.5 times more) than the mean activation of the other 13 recorded muscles. Isolation ratios are affected most by other active muscles, such as synergistic muscles that contribute to the main action of the rotator cuff muscle of interest. Antagonistic muscles, which act in opposition to the main action of the rotator cuff muscles of interest, would be expected to be minimally active during rotator cuff MMTs. Therefore, antagonistic muscles would be expected to contribute very little to the mean activation of the muscles in the denominator of the IRs. Participants most often reached a maximal plateau between the second and fourth seconds of each trial. Therefore, mean IR values were calculated over this 2-second window. The sampled window was adjusted to accommodate for anomalies in the time-series EMG data of individual muscle force tests. Due to the sensitivity of intramuscular electrodes, visually identifiable artifacts (large, rapid spikes in amplitude) occasionally occurred within the 2-second window. Often it was possible to shift the window slightly to avoid the artifact (a total of 116 trials were adjusted). When this was not possible, data for that muscle on that trial were excluded from the analysis (excluded trials per total trials for the April 2010
infraspinatus, supraspinatus, teres minor, and subscapularis muscles were 14/348, 99/348, 68/348, and 54/348, respectively). Adjustments were made in the IR calculation for the other muscles to reflect these missing data (removal of the muscle with artifact and division by the correct number of remaining muscles in the equation). There were 175 total adjusted IRs, of which 77.7% removed only 1 muscle, 17.7% removed 2 muscles, and 4.6% removed 3 muscles from the denominator. Statistical Analysis Statistical analysis was performed using JMP IN 5.1.2 software.㛳 Four oneway, repeated-measures ANOVAs (one for each rotator cuff muscle) compared IRs across the muscle force test groups. The groups were first confirmed with a one-way ANOVA, which confirmed that there were no statistical differences (P⬍.05) among mean IRs for the muscle force tests within the functional groups. Homogenous variance was seen among groups when mean IR data were transformed to a natural logarithm. When the ANOVAs indicated significant differences among mean IRs, post hoc analyses (Tukey honestly significant difference test) were performed to determine significant differences among muscle force test groups (P⬍.05). A sample size estimate was not calculated a priori, but rather was driven by convenience and expense. Fatigue Analysis The first 2 muscle force tests performed by each participant were repeated at the end of testing. The EMG data were down-sampled to 2,048 Hz, fast Fourier analysis was performed, and mean and median power frequency values (MnPF, MdPF) were calculated. Paired t tests (one-tailed) assessed whether signif㛳
SAS Institute Inc, PO Box 8000, Cary, NC 27513.
icant changes in MdPF or MnPF were evident between the initial and final trials (P⬍.05). Percent differences in MdPF and MnPF values were calculated, as follows: (2) % Difference ⫽ Final Value ⫺ Initial Value Initial Value
冋
册
䡠 100
Role of the Funding Source This study was supported by funds from the Department of Kinesiology, University of Waterloo. The department had no role in the design, conduct, or reporting of this research.
Results Isolation of the Infraspinatus Muscle Differences in IRs for the defined muscle force test groups existed (P⬍.0001), and the mean maximal IR occurred in the clinical lateral rotation group (Fig. 1). There were significant differences among the IRs of the following groups: lateral rotation ⬎ medial rotation and ulnar, dorsal, and palmar force ⬎ radial force and abduction. Isolation of the Supraspinatus Muscle Differences in IRs for the defined muscle force test groups existed (P⬍.0001), and the mean maximal IR occurred within the clinical abduction muscle force test group (Fig. 2). There were significant differences among the IRs of the following groups: (1) abduction ⬎ medial rotation and palmar and ulnar force and (2) radial and dorsal force ⬎ ulnar force. Isolation of the Teres Minor Muscle Differences in IRs for the defined muscle force test groups existed (P⬍.0001), and the mean maximal IR occurred within the clinical lateral rotation muscle force test group (Fig. 3). There were significant differences between IRs in the follow-
Volume 90
Number 4
Physical Therapy f
575
Construct Validity of Muscle Force Tests ing groups: (1) lateral rotation and ulnar force ⬎ radial force and abduction and (2) medial rotation and palmar and dorsal force ⬎ abduction.
Figure 1. Mean isolation of the infraspinatus muscle using the isolation ratio (IR) among muscle force test groups. The clinical muscle force tests are within the medial rotation, lateral rotation, and abduction muscle force test groups (white bars). The generic muscle force tests are within the palmar, dorsal, radial, and ulnar force groups (blue bars). The results of the Tukey honestly significant difference test are shown; levels not connected by the same letter are significantly different. The error bars represent ⫺1 standard deviation. The highest mean (SD) IR (maximal isolation) was found in the lateral rotation group (2.29⫾1.17); the lowest mean IR was found in the abduction group (0.68⫾0.33).
Figure 2. Mean isolation of the supraspinatus muscle using the isolation ratio (IR) among muscle force test groups. The clinical muscle force tests are within the medial rotation, lateral rotation, and abduction muscle force test groups (white bars). The generic muscle force tests are within the palmar, dorsal, radial, and ulnar force groups (blue bars). The results of the Tukey honestly significant difference test are shown; levels not connected by the same letter are significantly different. The error bars represent ⫺1 standard deviation. The highest mean (SD) IR (maximal isolation) was found in the abduction group (1.84⫾1.27); the lowest highest mean IR was found in the ulnar group (0.96⫾0.60).
576
f
Physical Therapy
Volume 90
Number 4
Isolation of the Subscapularis Muscle Differences in IRs for the defined muscle force test groups existed (P⬍.0001), and the mean maximal IR occurred within the generic ulnar force muscle force test group (Fig. 4). There were significant differences between IRs in the following groups: (1) ulnar force ⬎ palmar force, lateral rotation, abduction, and dorsal and radial force; (2) medial rotation ⬎ lateral rotation, abduction, and dorsal and radial force; and (3) palmar force ⬎ abduction and dorsal and radial force. Mean Muscle Activation Mean percentage of muscle activation of all participants in each muscle force test is shown in eAppendix 2 (available at ptjournal.apta.org). On average, the maximum activation levels were: 83% for the infraspinatus muscle during prone lateral rotation (muscle force test 13), 74% for the supraspinatus muscle during the empty can test (muscle force test 1), 64% for the teres minor muscle during prone lateral rotation (muscle force test 13), and 58% for the subscapularis muscle during neutral arm posture with ulnar resistance (muscle force test 28). Fatigue Significant decreases in both MnPF and MdPF existed only in the infraspinatus muscle (P⫽.016 and P⫽.022, respectively), with mean percent decreases of 8.3% and 11.4%, respectively. Significant decreases in MdPF occurred for the biceps brachii muscle (P⫽.026), which decreased an average of 5.3%.
Discussion The purpose of this study was to identify muscle force tests that relaApril 2010
Construct Validity of Muscle Force Tests tively isolated the rotator cuff muscles. The supraspinatus, infraspinatus, and teres minor muscles were maximally relatively isolated (produced maximal IRs) within their respective functional clinical test groups (lateral rotation for the infraspinatus and teres minor muscles and abduction for the supraspinatus muscle), although other generic muscle force tests were found to be equally effective in isolating these muscles. The subscapularis muscle was relatively isolated within its respective clinical medial rotation group; however, this level of isolation was second to that of the ulnar force group, although not statistically different. These findings establish traditional MMTs as effective for the relative maximal isolation of individual rotator cuff muscles. Considerable population variability (as indicated by large data standard deviations) suggests that specific muscle force tests may not relatively isolate rotator cuff muscles for all individuals. Therefore, we recommend that the rotator cuff muscles be tested in a number of isolation muscle force tests to verify isolation and promote accurate muscle assessment. Isolation and Activation of the Infraspinatus Muscle The highest mean IR for the infraspinatus muscle occurred within the lateral rotation group (X⫽2.29, SD⫽0.53) and was significantly higher than mean IRs in all other groups. Therefore, the infraspinatus muscle was most isolated (activated 2.29 times more than the other 13 recorded muscles) within the clinical lateral rotation muscle force tests. These findings support the hypothesis that clinical tests in the lateral rotation group (prone and sitting lateral rotation) are appropriate in isolating and assessing the strength and function of the infraspinatus muscle. The findings matched expectations, as the infraspinatus April 2010
Figure 3. Mean isolation of the teres minor muscle using the isolation ratio (IR) among muscle force test groups. The clinical muscle force tests are within the medial rotation, lateral rotation, and abduction muscle force test groups (white bars). The generic muscle force tests are within the palmar, dorsal, radial, and ulnar force groups (blue bars). The results of the Tukey honestly significant difference test are shown; levels not connected by the same letter are significantly different. The error bars represent ⫺1 standard deviation. The highest mean (SD) IR (maximal isolation) was found in the lateral rotation group (1.76⫾0.94); the lowest mean IR was found in the abduction group (0.84⫾0.49).
Figure 4. Mean isolation of the subscapularis muscle using the isolation ratio (IR) among muscle force test groups. The clinical muscle force tests are within the medial rotation, lateral rotation, and abduction muscle force test groups (white bars). The generic muscle force tests are within the palmar, dorsal, radial, and ulnar force groups (blue bars). The results of the Tukey honestly significant difference test are shown; levels not connected by the same letter are significantly different. The error bars represent ⫺1 standard deviation. The highest mean (SD) IR (maximal isolation) was found in the ulnar force group (1.95⫾1.20); the lowest mean IR was found in the abduction group (0.39⫾0.38).
Volume 90
Number 4
Physical Therapy f
577
Construct Validity of Muscle Force Tests muscle is primarily a lateral rotator and thus should be significantly activated in lateral rotation muscle force tests. Previous findings also showed the infraspinatus muscle to be most isolated from synergists (supraspinatus and posterior deltoid muscles) during sitting lateral rotation.8 The infraspinatus muscle has been found to be significantly activated during side-lying lateral rotation (peak of 85% MVC),3 and previous literature10 also has shown the infraspinatus and teres minor muscles to be equally activated during prone and side-lying lateral rotation. In the present study, the maximal activation of the infraspinatus muscle was, on average, 83% and occurred during prone lateral rotation (muscle force test 13). These findings suggest that the infraspinatus muscle may be isolated in some positions during which it is maximally activated, which complies with previous methods of determining isolation muscle force tests.3,10,16 However, the sitting lateral rotation muscle force test, which was found to be equally effective in isolating the infraspinatus muscle, produced lower activations (average of 58% MVC), demonstrating that muscle force tests of maximal activation are not always indicative of maximum isolation. Isolation and Activation of the Supraspinatus Muscle The supraspinatus muscle was most isolated from the other 13 recorded muscles in the abduction, lateral rotation, radial force, and dorsal force groups. These results indicate that the clinical abduction muscle force tests (empty can, full can, Blackburn, and supraspinatus muscle neutral abduction muscle force tests) and muscle force tests within the radial force group (abduction and flexion), dorsal force group (abduction and horizontal abduction), and lateral rotation group (prone lateral rotation in an abducted humeral posture and sit578
f
Physical Therapy
Volume 90
ting lateral rotation) are effective in isolating the supraspinatus muscle from surrounding muscles and are appropriate to use in assessing the strength and function of the supraspinatus muscle. Because the supraspinatus muscle aids the deltoid muscle in shoulder abduction, it is not surprising that the supraspinatus muscle is isolated in muscle force tests of abduction. It was surprising to see the supraspinatus muscle isolated in muscle force tests of flexion and lateral rotation (within the radial force and lateral rotation groups, respectively). This isolation could be a result of grouping muscle force tests containing combinations of abduction and flexion or abduction and lateral rotation postures within the same group. Perhaps if muscle force tests contained only one primary posture and were grouped separately, a clearer distinction would be made in isolating the supraspinatus muscle in abduction muscle force tests. On average, the maximal activation of the supraspinatus muscle was 74%, which occurred during the empty can test (muscle force test 1). Other muscle force tests identified to equally isolate the supraspinatus muscle produced only submaximal activations (average of ⱖ29% MVC). Previous literature has recommended empty can and full can tests be used to assess the integrity of the supraspinatus muscle.8,17,18 Empty can and full can tests were found to be equivalent in accuracy for detecting supraspinatus muscle tears.18 These tests are very similar to the abduction (90°) muscle force test within the radial force group. Kelly et al8 considered the infraspinatus muscle as a synergist of the supraspinatus muscle and compared supraspinatus muscle isolation during 6 muscle force tests (including the empty can and full can tests) and concluded that the full can test best isolated the supraspinatus muscle.
Number 4
Results from the present study showed that there was no significant difference between IRs of the empty can and full can tests, indicating the comparable ability of both muscle force tests to isolate the supraspinatus muscle. Isolation and Activation of the Teres Minor Muscle The teres minor muscle was most isolated from the other 13 recorded muscles within the lateral rotation, medial rotation, palmar force, dorsal force, and ulnar force groups. These results confirm that clinical tests (prone and sitting lateral rotation) are appropriate in assessing the teres minor muscle. However, other muscle force tests within the medial rotation, ulnar force, palmar force, and dorsal force groups (medial rotation test; shoulder extension, adduction, and abduction tests; and horizontal abduction and adduction tests) are equally effective in isolating the teres minor muscle. The teres minor muscle is primarily a lateral rotator, so it was expected that it would be isolated in muscle force tests of lateral rotation. The teres minor muscle originates on the infraspinous fossa of the scapula and inserts on the greater tubercle of the humerus, so it also was not surprising for the teres minor muscle to be isolated in muscle force tests of extension and adduction. It was surprising, however, to find that the teres minor muscle also was isolated in muscle force tests of medial rotation and abduction, during which the muscle would not be expected to be very active. It is possible that the teres minor muscle is activated in these muscle force tests of medial rotation and abduction to help stabilize the glenohumeral joint and prevent anterior dislocation. The teres minor muscle was maximally activated an average of 64%, and this occurred during prone lateral rotation (muscle force test 13). April 2010
Construct Validity of Muscle Force Tests Previous literature10 showed the teres minor muscle to be significantly activated in prone and sitting lateral rotation muscle force tests similar to those of the present study. However, the teres minor muscle was not always maximally activated (average of ⱖ24% MVC) in other isolation muscle force tests. Isolation of the Subscapularis Muscle The subscapularis muscle was most isolated (1.95 times more activated) from the other 13 recorded muscles within the muscle force tests of the ulnar force group. However, these levels of isolation were not significantly different from those displayed during muscle force tests within the clinical medial rotation group (1.60). These results confirm that clinical tests (prone medial rotation, lift-off, and belly press muscle force tests) are appropriate in assessing the subscapularis muscle. However, other muscle force tests from the ulnar force group (shoulder extension and adduction) are equally effective in isolating the subscapularis muscle and can be used in the assessment of the strength and function of the subscapularis muscle. The present findings indicate no difference in isolation of the subscapularis muscle between the lift-off and belly press muscle force tests, and both muscle force tests have been shown to maximally activate the subscapularis muscle.7 However, some authors have concluded that the lift-off muscle force test was superior compared with the belly press muscle force test, as the subscapularis muscle was more isolated from the pectoralis major muscle4 and from the pectoralis major and latissimus dorsi muscles8 during the lift-off muscle force test. The highest activation of the subscapularis muscle was, on average, 58% and occurred during the neutral arm posture with ulnar resistance (muscle force test 28). Although maximal activation occurred April 2010
within an identified isolation muscle force test, other defined isolation muscle force tests produced submaximal activations (ⱖ34% MVC) of the subscapularis muscle. Assessment of Fatigue Although significant decreases in the frequency content of the EMG signal existed for the infraspinatus and biceps muscles, there was no indication (frequency decrease) that any of the other 12 muscles were fatigued. Recommended rest levels15 were provided, and the number of maximal muscle force tests in the present study was the same as those done in previous studies.8,11 We conclude that fatigue had minimal, if any, impact on the results. Limitations If channels were excluded from calculations due to motion artifact, or if motion artifact was missed and these data were included in the analysis, inaccuracy would enter the IR calculations, and thus the identification of isolation muscle force tests using that criterion may be incorrect. In addition, no data were collected to support the reliability of these IRs. The grouping of muscle force tests based on specific muscle force test criteria may have inflated or deflated group mean IRs, which could have influenced which groups most isolated the muscle of interest. Due to the instrumentation used in this study, the methods do not exactly mimic clinical application of the tests. There is limited generalizability of these results due to the small sample size (of participants who were asymptomatic).
Conclusion This study demonstrated that attempts to isolate the rotator cuff muscles fully are problematic. However, it also revealed that muscle force tests exist that relatively isolate the rotator cuff muscles and can detect large changes in rotator cuff
muscle activation. The rotator cuff muscles were maximally relatively isolated within the following groups: lateral rotation group (infraspinatus muscle); abduction, lateral rotation, and radial and dorsal force groups (supraspinatus muscle); lateral rotation, medial rotation, and palmar, dorsal, and ulnar force groups (teres minor muscle); and medial rotation and ulnar force groups (subscapularis muscle). We confirmed both study hypotheses. First, the clinical tests were generally most effective in achieving maximum isolation of the rotator cuff muscles, although other muscle force tests also were equally effective. Second, although rotator cuff muscles were relatively isolated within muscle force tests based on their primary action, in some instances, alternative muscle force tests (and actions) similarly isolated some muscles. The results confirmed the appropriateness of currently applied clinical tests in assessing individual rotator cuff muscle status and identified additional muscle force tests that similarly relatively isolated these muscles. These findings may provide clinicians with more confidence, qualitative accuracy, and flexibility in their assessment of musclespecific strength. Future studies of rotator cuff isolation should continue to investigate the contributions of these surrounding muscles during these and other muscle force tests and should include a larger population sample. All authors provided concept/idea/research design, writing, data collection and analysis, and consultation (including review of manuscript before submission). Ms Brookham and Dr Dickerson provided project management. Dr Dickerson provided fund procurement and facilities/equipment. Ms Brookham provided participants. Dr McLean and Dr Dickerson provided institutional liaisons.
Volume 90
Number 4
Physical Therapy f
579
Construct Validity of Muscle Force Tests Study approval was provided by the University of Waterloo Office of Research Ethics. This study was supported by funds from the Department of Kinesiology, University of Waterloo. This article was received January 23, 2009, and was accepted October 30, 2009. DOI: 10.2522/ptj.20090024
References 1 Daniels L, Worthingham C. Muscle Testing Techniques of Manual Examination. Philadelphia, PA: WB Saunders Co; 1986. 2 Moore L, Dalley AF. Clinically Oriented Anatomy. 4th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 1999. 3 Townsend H, Jobe FW, Pink M, Perry J. Electromyographic analysis of the glenohumeral muscles during a baseball rehabilitation program. Am J Sports Med, 1991; 19:264 –272. 4 Greis PE, Kuhn JE, Schultheis J, et al. Validation of the lift-off test and analysis of subscapularis activity during maximal internal rotation. Am J Sports Med. 1996;24: 589 –593. 5 Decker M, Tokish J, Ellis H, et al. Subscapularis muscle activity during selected rehabilitation exercises. Am J Sports Med. 2003;31:126 –134. 6 Suenaga N, Minami A, Fujisawa H. Electromyographic analysis of internal rotation motion of the shoulder in various arm positions. J Shoulder Elbow Surg. 2003;12: 501–505.
580
f
Physical Therapy
Volume 90
7 Tokish J, Decker M, Ellis H, et al. The bellypress test for the physical examination of the subscapularis muscle: electromyographic validation and comparison to the lift-off test. J Shoulder Elbow Surg. 2003; 12:427– 430. 8 Kelly BT, Kadrmas WR, Speer KP. The manual muscle examination for rotator cuff strength; an electromyographic investigation. Am J Sports Med. 1996;24:581– 588. 9 Liu J, Hughes RE, Smutz WP, et al. Roles of deltoid and rotator cuff muscles in shoulder elevation. Clin Biomech. 1997;12:32– 38. 10 Ballantyne BT, O’Hare SJ, Paschall JL, et al. Electromyographic activity of selected shoulder muscles in commonly used therapeutic exercises. Phys Ther. 1993;73: 668 – 682. 11 Jenp Y, Malanga GA, Growney ES, An K. Activation of the rotator cuff in generating isometric shoulder rotation torque. Am J Sports Med. 1996;24:477– 485. 12 Delagi E, Perotto A. Anatomic Guide for the Electromyographer. 2nd ed. Springfield, IL: Charles C Thomas, Publisher; 1980. 13 Nemeth G, Kronberg M, Brostrom L. EMG recordings from the subscapularis muscle: description of a technique. J Orthop Res. 1990;8:151–153. 14 Cram JR, Kasman GS. Introduction to Surface Electromyography. Gaithersburg, MD: Aspen Publishers Inc; 1998. 15 De Luca CJ. The use of surface electromyography in biomechanics. J Appl Biomech. 1997;13:135–163.
Number 4
16 Dark A, Ginn KA, Halaki M. Shoulder muscle recruitment patterns during commonly used rotator cuff exercises: an electromyography study. Phys Ther. 2007;87:1039 – 1046. 17 Jobe FW, Moynes DR. Delineation of diagnostic criteria and a rehabilitation program for rotator cuff injuries. Am J Sports Med. 1982;10:336 –339. 18 Itoi E, Tadato K, Sano A, et al. Which is more useful, the “full can test” or the “empty can test,” in detecting the torn supraspinatus tendon? Am J Sports Med. 1999;27:65– 68. 19 Malanga GA, Jenp Y-N, Growney ES, KaiNan A. EMG analysis of shoulder positioning in testing and strengthening the supraspinatus. Med Sci Sports Exerc. 1996; 28:661– 664. 20 Blackburn TA, McLeod WD, White B, Wofford L. EMG analysis of posterior rotator cuff exercise. Athl Train. 1990;25:40 – 45. 21 Clarkson MH, Gilewich GB. Musculoskeletal Assessment: Joint Range of Motion and Manual Muscle Strength. Baltimore, MD: Lippincott Williams & Wilkins; 1989. 22 Janda V. Muscle Function Testing. London, United Kingdom: Butterworths; 1983. 23 Gerber C, Krushell RJ. Isolated rupture of the tendon of the subscapularis muscle. J Bone Joint Surg Br. 1991;73:389 –394. 24 Gerber C, Hersche O, Farron A. Isolated rupture of the subscapularis tendon: results of operative repair. J Bone Joint Surg Am. 1996;78:1015–1023.
April 2010
