Abstract
BACKGROUND:
Scapular dyskinesis has been associated with shoulder dysfunctions, and changes in electromyographic (EMG) activity have been reported during the execution of some exercises.
OBJECTIVE:
To compare upper limb muscles EMG of asymptomatic subjects with and without scapular dyskinesis during bench-press and dumbbell fly exercise performed on different surface stability conditions.
METHODS:
Twenty-eight physically active men were allocated into two groups: Control group (
RESULTS:
No differences were found between groups, regardless of exercise or surface type. Inserting the unstable surface in the supine exercise promoted a slight increase in upper trapezius EMG activity for the control [4.32 (95% CI: 1.04 to 7.60)] and dyskinesis [3.30 (95% CI: 0.34 to 6.27)] groups.
CONCLUSIONS:
There is no difference in upper limb muscle EMG activity between subjects with and without scapular dyskinesis. In addition, inserting unstable surfaces did not modify EMG activity.
Introduction
The scapula provides a stable basis for upper limb motor function, and the scapular muscles are responsible for promoting the stability and mobility of the shoulder joint complex [1, 2]. At the same time, the scapula acts as a connection between the trunk and the upper limbs, ensuring the necessary synergy for movement and force transmission [3].
Changes in scapular movements are called scapular dyskinesis and described as a sign of scapular instability due to weakness or imbalance of the periscapular muscles [4]. Several researchers have suggested that this “unstable” or “dyskinetic scapula” can be associated with shoulder disorders [2, 5]. It is suggested to strengthen the periscapular muscles in order to fix and improve scapular function [6].
Exercises with axial load on unstable bases have generally been utilized to strengthen periscapular muscles, especially for the serratus anterior [7, 8]. The use of unstable bases may increase neuromuscular demand without the need for high loads [9, 10, 11, 12]. However, these advantages remain uncertain, especially for subjects with scapular dyskinesis. Few studies have tested exercises with unstable bases in subjects with scapular dyskinesis, and the results suggest that this dysfunction changes the EMG activity of scapular stabilizers [11, 13]. In fact, insertion of an unstable base caused an increase in the EMG activity of the serratus anterior in individuals without scapular dyskinesis. On the other hand, subjects with scapular dyskinesis showed decreased EMG activity in the serratus anterior [11, 13]. However, such results are restricted to push-up exercises which require greater stability in the movement because only the distal segment is fixed, and this creates greater demand for stabilization of the shoulder girdle [7, 14]. Furthermore, it is important to highlight that the possibilities for prescribing push-ups are limited since intensity manipulation depends on the subject’s weight.
Unlike the push-up, strength exercises performed with support from the back and scapula on benches or balls may be more effective to strengthen the serratus anterior of subjects with scapular dyskinesis. The demand for stabilization of the scapula is probably lower, requiring less trapezius activity and enabling greater activity in the anterior serratus [15]. Thus, unlike what is observed in the push-up exercise, inserting unstable surfaces in the bench-press and dumbbell fly can generate an increase in EMG activity of serratus anterior in subjects with and without scapular dyskinesis.
We hypothesize that inserting an unstable surface will provide increased EMG activity of the periscapular and shoulder muscles. However, as these are exercises are performed with scapular support, no differences will be observed between groups. Thus, the present study aimed to analyze the muscular activity of subjects with scapular dyskinesis during the bench press and dumbbell fly exercises performed on a stable and unstable basis.
Methods
Participants
The sample included 28 men (14 Dyskinesis/14 Control); 24.6
Those who could not perform the proposed exercises would be excluded from the study, as well as those who: used medications, which positively or negatively interfered with performance such as creatinine or appetite stimulators; were affected by a limiting health condition; performed another physical activity program during collection. The F-test with repeated measures and interaction analysis within and between groups was considered to define the sample size with
Surface electromyography (EMG)
Seven channels of the EMG800C-1632 Signal Acquisition System (EMG System of Brazil) were used to obtain the electromyographic records for the EMG evaluation. The electromyographic signals were captured through different active and simple surface electrodes placed on the serratus anterior (SA), upper (UT) and lower (LT) trapezius, pectoralis major (PM), anterior deltoid (AD), biceps brachii (BB) and triceps brachii (TB). The reference electrode was positioned in the mid-distal region of the tibia interposed with conductive gel. The other electrodes were positioned on the dominant side of the volunteer, and they were specifically positioned according to the recommendation of the European Recommendations for Surface Electromyography of the SENIAM Project [17] for the trapezius (upper and lower fibers), anterior deltoid, biceps and triceps muscles, while positioning followed the guidelines of Nascimento et al. [9] and Reiser et al. [18] for the pectoralis major muscle. The serratus anterior was positioned as described in the study by Park and Yoo, [16] with the electrodes positioned on the 7th ribs.
The EMGLab V1.1 (EMG System Brazil, version 2012) software program was used for visualizing and processing the signals. The data were collected at a sampling frequency of 2000 Hz. A signal curve straightening procedure was used, which aims to eliminate high frequencies from the EMG register. This process was performed by using a low pass filter (10 Hz) and filtering with Butterworth 4th order, generating a processed signal called a linear envelope.
The EMG amplitude values were obtained by calculating the EMG integral using routines pre-established by the EMGLab V1.1 (EMG System Brazil) program. Finally, 500 Hz low pass and 15 Hz high pass digital filters were applied and the EMG signal normalization process was performed.
The EMG data collected during each exercise were normalized by the maximum value obtained in three Maximum Isometric Voluntary Contractions (MVICs) collected during manual muscular testing for each muscle. Integral EMG values were adjusted by the maximum value allowed in three of the corresponding MVIC muscles, i.e. the ratio between the average amplitude value recorded in the exercises and the maximum value of the MVIC records. Three MVICs for each muscle were collected for six seconds with a two-minute interval between them, according to Ekstrom et al. [19].
Design and procedures
The volunteers visited the laboratory three times at 48-hour intervals. The first visit aimed to perform an anthropometric assessment and evaluate dyskinesis. The dynamic observational method was used for scapular dyskinesis evaluation, in which the volunteer was asked to remain in the orthostatic position and perform eight repetitions of bilateral arm elevation movement (three seconds for each phase: concentric and eccentric) in the scapular plane until maximum range by holding a dumbbell weighing approximately 3% of their body weight. The movement execution was recorded in the posterior view by a digital camcorder (1.00 meters high and 2.00 meters away) with a 60 Hz sampling frequency (SONY model DCR-SX21).
Two physiotherapists with experience in the orthopedic area performed all evaluations independently. The type of scapular dyskinesis was categorized according to the guidelines of Kibler et al. [20]. A visualized prominence of the inferior angle of the scapula was interpreted as Type I; Increased prominence of the medial border – Type II; excessive elevation of the upper angle – Type III; absence of scapular dyskinesis – Type IV. Later, the assessors categorized the subjects into only two types (yes or no). All three dyskinesis categories (types I to III) were collapsed into a single category of “yes” (scapular dyskinesis) and type IV was relabeled “no” (normal scapular motion).
Performing the dumbbell fly and bench press exercises on a stable and unstable surface.
Mean and standard deviation (SD) of the normalized values of the EMG activity of the muscles analyzed during the dumbbell fly and bench press exercise conditions for both groups
Table note: AD
Mean and standard deviation (SD) of the anthropometric variables of the subjects who participated in the study
All the participants performed the one-repetition maximum (1RM) tests during the second visit and were familiarized with the bench-press and dumbbell fly exercises. The American Society of Exercise Physiologists [21] protocol was used to determine the maximum load supported for the 1RM test. All volunteers initially performed a specific warm-up with five repetitions at 50% of the volunteer’s estimated maximum load, and a series of three repetitions were then performed with approximately 80% of the estimated load after a two-minute interval. A three-minute interval was used to initiate the test. The subjects were instructed to perform two complete repetitions, but the maximum load was considered when the subject could only perform 1 repetition. The evaluators had between three and five attempts to obtain the maximum load, and the interval between each attempt was standardized at five minutes. An interval of ten minutes was stipulated between the exercises. The third session was intended for the experimental protocol, and the subjects from both groups were tested under the same conditions. The skin was shaved at the electrode sites, gently abraded and cleaned with alcohol to reduce skin impedance. Next, MVIC tests were performed for the evaluated muscles (pectoralis major, serratus anterior, inferior and superior trapezius, anterior and posterior deltoid, biceps and triceps).
The exercises and surfaces were subsequently randomized through a draw from an envelope after this step, with the bench press and crucifix exercises being performed on stable and unstable surfaces (Fig. 1). The subjects performed a series of 6 repetitions with a 50% load of 1RM and with controlled speed by metronome (one second for concentric phase and two seconds for eccentric phase). The unstable base condition was performed using a Swiss ball positioned in the dorsal region on the upper trunk.
Mixed linear models were used to analyze the dependent variables in each variation of the exercise. An autoregressive covariance matrix was considered. The groups and exercise conditions were considered as fixed factors, and the volunteers were considered random factors. A Q-Q plot was analyzed to attest to the normality of the residuals for all variables. The logarithmic transformation of the data was performed to adjust the model.
Bonferroni post-hoc was used for the isolated effect of the group and exercise or interaction between them. The analysis and interpretation of within and between-group differences were performed through the mean difference values and their respective 95% confidence intervals (95% CI). A statistical significance of 5% was accepted for all tests. All statistical procedures were performed by a blinded researcher using the SPSS version 22 software program (IBM Corp., New York, NY, USA).
Results
The descriptive characteristics of participants between experimental groups are shown (Table 1). Both groups had similar characteristics [Age: Control
Mean difference (95% CI) within and between groups of normalized EMG activity in the different exercises conditions
Mean difference (95% CI) within and between groups of normalized EMG activity in the different exercises conditions
Table note: AD
The descriptive values corresponding to muscle activity during the performance of the dumbbell fly and bench-press exercises in the stable and unstable bases are described (Table 2). Visual inspection of normalized muscle activity shows low UT activity and high AD activity during dumbbell fly and low UT and BB activity and high AD activity during bench press regardless of groups or surfaces.
The results of Table 3 present the values of the differences between the unstable and stable conditions in both exercises, as well as the differences between the evaluated groups.
The results indicated that there was an increase in muscle activity of the upper trapezius during the bench-press on the unstable base, with an increase of 4.32 (95% CI 1.04 to 7.60) for control (
The present study revealed that inserting the unstable device did not increase the EMG activity of most of the assessed muscles. On the other hand, subjects with and without scapular dyskinesis had similar EMG activity in both exercises with instability devices, partially confirming the initial hypothesis. Although upper trapezius EMG activity increased in the unstable bench press, it is noteworthy that the differences observed were small, and probably have no practical or clinical relevance.
When reviewing the literature, it is observed that the use of unstable surfaces is restricted to push-up exercises considering subjects with scapular dyskinesis [11, 13]. Some points about this exercise should be highlighted: 1) Its biomechanical characteristic imposes a greater need for stabilization since the distal segment is fixed and the direction of the load is axial [13, 22]. Thus, the periscapular muscles may have their activation levels impaired in high instability situations [8]; 2) It is an exercise with restricted possibilities regarding intensity manipulation, since this variable not only depends on the physical fitness and training level of the subject, but also on their weight; a factor which cannot be easily modified [22].
The findings of the present study showed that the bench press and dumbbell fly exercises did not present any difference between the groups, regardless of the surface. This suggests that individuals with scapular dyskinesis can perform the exercises on both bases (stable or unstable) without suffering any alteration in muscle activity. The support of the scapula on the bench or ball probably promoted a stable condition which did not require a high demand for stabilization.
This is in contrast to what was observed when the subjects with scapular dyskinesis performed the push-ups on an unstable basis, since this condition caused a decrease in the EMG activity of the serratus anterior, one of the main periscapular muscles [11, 13]. Recent systematic reviews [7, 8] verified that using unstable surfaces during the push-up exercise reduces serratus anterior activation when compared to the stable surface and recommend selecting exercises with scapular protraction and its variations for better recruitment of the serratus anterior.
Regarding our results, bench-press and dumbbell fly can be effective for strengthening the primary muscles (PM, AD, BB and TB) and the serratus anterior due to its role in the protraction movement of the scapula. On the other hand, the two portions of the trapezius muscle evaluated showed low EMG activity levels. Although not evaluated, the middle trapezius and rhomboids would probably present similar results considering the biomechanical characteristics of the tasks performed.
So, if the objective is to increase the neuromuscular demand of periscapular muscles for stabilization, other exercise options seem to be better options. Considering the wide mobility of the scapulothoracic joint, professionals should include different exercises which explore all degrees of freedom of the scapula in order to increase the demand of all the periscapular muscles.
Assuming that there was no difference between the groups analyzed in the present study, we will compare our findings considering both groups with previous studies which analyzed the effect of instability during the execution of both exercises.
Regarding the dumbbell fly exercise, Reiser et al. [23] confirmed similar results, with no difference in agonist muscle EMG activity when subjects performed the exercise on a Swiss ball or over an horizontal bench. However, Nascimento et al. [9] observed that the agonist muscles showed increased EMG activity, and the periscapular muscles were not influenced by the instability insertion. Our study corroborates the results found regarding the periscapular muscles, suggesting that the addition of instability in the dumbbell fly exercise (rotational load) is not sufficient to affect this muscle group.
Nevertheless, our findings differ from Nascimento et al. [9] regarding the agonist muscles. We believe that the two aspects can justify the difference between the results. The first is related to the devices used to generate instability, since Nascimento et al. [9] used a proprioceptive disc which has a smaller contact area than the Swiss ball, and possibly the instability level was higher. The second aspect refers to the EMG signal normalization process adopted in the study by Nascimento et al., [9] which was the Isometric Reference Contraction. In this sense, the statistical differences only reported by the p-value analysis may have been overestimated, which leads us to question whether the observed changes would be clinically relevant.
Concerning the bench press exercise, our results only showed higher EMG activity for the UT during the unstable condition in both groups. Despite the statistical difference found, it is important to emphasize that EMG activity levels below 20% are classified as low [24], a fact which was observed in all conditions tested for this muscle. Therefore, it is suggested that differences for UT are not clinically relevant. Regarding the other analyzed muscles, it is observed that there are studies in the literature which corroborate our results and also those which differ; a fact that occurs concerning the periscapular and agonist muscles.
Regarding the periscapular muscles, we only identified one study which analyzed inserting instability during the bench press, which was conducted by Nascimento et al. [9]. The authors found that instability caused an increase in EMG activity for SA and LT, with no difference for the other muscles; however, the differences in the present study were restricted to UT. The divergence between the results may once again be explained by the two arguments previously discussed in the context of the dumbbell fly exercise (the device used to generate instability and strategy for EMG signal normalization).
In relation to the agonist muscles, there are studies which also did not observe differences in EMG activity with instability insertion [25, 26, 27, 28], as well as others that verified an increase for some of the tested muscles [9, 29]. Furthermore, there was one study which observed a decrease in PM and TB activity [30]. It is noteworthy that there is a clear difference in the number of repetitions, execution speed and intensity in these studies, as well as a difference between the devices which generate instability. Considering the importance of these variables on the responses of EMG activity, any justification is compromised, since there is no minimum standardization between studies which enables coherent comparison.
In practical terms, the present study expanded possibilities in relation to the exercises available for subjects with scapular dyskinesis, especially manipulating the intensity and an analysis of muscle activity with instability insertion; a condition, which did not prove harmful in the tested exercises. Nonetheless, our results indicate that inserting the unstable surface did not increase the EMG demand for the assessed muscles. The use of the Swiss ball was not able to generate a significant degree of instability for the scapulothoracic joint. Although not tested, it is possible to speculate that this strategy increases the neuromuscular demand of the trunk and lower limbs due to the need to maintain postural control.
Lastly, this study has the following strengths: 1) provides a comparison between the muscular activity of the subjects with scapular dyskinesis with a control group; 2) use of exercises with axial and rotational loads on stable and unstable surfaces; and 3) assessment of agonist and stabilizer muscles of upper extremity. Regarding limitations of our study, we can highlight: 1) our findings are limited to two exercises tested at a single intensity; 2) both groups were formed by asymptomatic subjects, even differing concerning the presence of scapular dyskinesis; and 3) we did not assess the concentric and eccentric exercise phases separately, and this can help to understand muscle activity during each exercise.
In this sense, future studies need to investigate symptomatic subjects. It is additionally and equally important to investigate the muscle activity pattern by applying instability in different exercises (with a free or fixed distal segment, with axial and rotational loads), and also exploring more body segments.
Conclusions
Subjects with and without scapular dyskinesis have similar EMG activity during the free distal segment exercises, regardless of instability insertion. In practical terms, these exercises expand the possibilities of prescription regarding differing the training intensity. Finally, it is noteworthy that even though the instability caused an increase in upper trapezius muscle activity, this response has no clinical relevance. Therefore, from an EMG point of view there is no advantage to performing these exercises on unstable surfaces.
Footnotes
Acknowledgments
The authors thank the participants for their contribution to the study.
Conflict of interest
The authors declare no conflicts of interest.
Funding
No financial or material support of any kind was received for the work described in this article.