Shoulder Muscle Activation Levels During Four Closed Kinetic Chain Exercises With and Without Redcord Slings

Abstract

De Mey, K, Danneels, L, Cagnie, B, Borms, D, T'Jonck, Z, Van Damme, E, and Cools, AM. Shoulder muscle activation levels during four closed kinetic chain exercises with and without Redcord slings. J Strength Cond Res 28(6): 1626–1635, 2014—During resistance training protocols, people are often encouraged to target the scapular stabilizing musculature (middle and lower trapezius and serratus anterior) while minimizing shoulder prime mover activation (upper trapezius and large glenohumeral muscles) in their training regime, especially in overhead athletes with scapular dyskinesis. To increase the activation levels in the stabilizing muscles without drastically increasing the activation in the prime movers, unstable surfaces are frequently used during closed kinetic chain (CKC) exercises. However, the specific influence of Redcord slings (RS) as an unstable surface tool on the shoulder muscle activation levels has rarely been investigated, despite these results may be used for adequate exercise selection. Therefore, a controlled laboratory study was performed on 47 healthy subjects (age, 22 ± 4.31 years; height, 176 ± 0.083 cm; weight, 69 ± 8.57 kg) during 4 CKC exercises without and with RS: half push-up (HPU), knee push-up (KPU), knee prone bridging plus (KPBP), and pull-up. When using RS, serratus anterior muscle activation decreased during the KPU and KPBP exercise. In addition, a drastic increase in pectoralis major muscle activation was found during the HPU and KPBP exercise. Consequently, the use of RS does not necessarily imply that higher levels of scapular stabilizer muscle activation will be attained. These findings suggest that RS might be an appropriate training tool when used within a general strengthening program but should not be preferred over a stable base of support when training for specific scapular stabilization purposes.

Introduction

The shoulder complex accounts for a considerable proportion of injuries both in the general resistance training population and in overhead athletes in particular. In these individuals, specific muscle imbalances have been implicated in the etiology of shoulder disorders such as scapular dyskinesis and shoulder impingement syndrome. Their scapular stabilizing muscles have been found to be impaired with respect to the scapular and glenohumeral prime movers (^5,28,35). Additionally, large muscle groups such as the upper trapezius (UT), pectoralis major (PM), anterior (AD) and posterior deltoid (PD), and latissimus dorsi (LD) are often targeted in their training regime with the objective to produce gains in strength and hypertrophy subsequently neglecting the stabilizing muscles such as the middle and lower trapezius (MT, LT) and serratus anterior (SA). Therefore, exercises to strengthen the MT, LT, and SA while avoiding high activation levels in the large muscle groups (UT, PM, AD, PD, LD) should be incorporated into their training programs (^20,22). From a scapular point of view, exercises with low UT/MT, UT/LT and UT/SA muscle ratios are preferred (⁶).

Shoulder exercises are often prescribed to restore upper extremity function (¹⁴) or to prevent injury (³⁸) and enhance athletic performance (³⁷), especially in overhead athletes. With the objective to select the most appropriate exercises, the recruitment patterns of the shoulder girdle muscles have been studied during both open kinetic chain (OKC) and closed kinetic chain (CKC) exercises (¹⁷). Although OKC exercises are commonly employed in the training of throwing athletes, they often put considerable stress on the shoulder joint, especially in the “high five” position (²¹). In contrast, CKC exercises have been shown to stimulate mechanoreceptors and have been found to recruit the stabilizing muscles around the shoulder girdle, so contributing to proper shoulder stabilization (^18,31,33,40). Research suggests this stimulus can be enlarged by adding an unstable surface, possibly leading to higher levels of muscle activation (^3,15). As a result, an unstable surface is often used to facilitate neuromuscular adaptations in response to strength training (^{26,27,34,36,37,42,46–48}). Generally, an activation level of >60% of maximum voluntary isometric contraction (MVIC) is needed in terms of muscle strengthening purposes with an increase of ≥10% MVIC required to prefer 1 exercise over another (^1,2,23). Considering scapular muscle balance restoration, ratios <60% are preferred (⁶), although an exact threshold in terms of clinically relevant improvements between conditions has not yet been defined.

A novel training device providing an unstable base of support, Redcord slings (RS), has been described in recent research (^8,16,37). Some authors found it useful in the treatment of various musculoskeletal pathologies (^49,50) to improve proprioception (⁴⁵) and to enhance the athlete's athletic performance (^{16,37,41,43,44}). However, whether RS activate the scapular stabilizing muscle components (MT, LT, and SA) while limiting the activation in the larger muscle groups (UT, PM, AD, PD, LD) is currently unclear, despite this information could be used for adequate exercise selection. Therefore, the main purpose of this study was to examine scapular and glenohumeral muscle amplitude levels and scapular muscle ratios during 4 selected CKC exercises without and with RS. We hypothesized that each scapular and glenohumeral muscle component and each scapular muscle ratio would be altered when comparing between both conditions. The results of this study could make clear whether coaches should use RS as an unstable device during particular CKC exercises.

Methods

Experimental Approach to the Problem

Surface electromyography (EMG) was used to measure muscle activation amplitude levels of scapulothoracic (UT, MT, LT, SA) and glenohumeral (PM, AD, PD, LD) musculature. Before data collection, the exercises were demonstrated by one of the researchers. Then, the participant carried out the exercises to become familiar with the exercise receiving corrective feedback when needed. Subsequently, MVIC's were determined for normalization purposes. Then, the 4 exercises were performed with and without RS to investigate the influence of the slings on the muscle amplitude levels: half push-up (HPU), knee push-up (KPU), knee prone bridging plus (KPBP), and pull-up (PU) (Figures 1–4). Exercises requiring extreme levels of core stability were not chosen, because some subjects might not have been able to maintain their neutral spinal alignment during each exercise, possibly influencing the results of our investigation (³²). To prevent order biasing, the exercise sequence was randomized. In addition to the measurements of the amplitude levels of the scapular and glenohumeral muscles, the UT/MT, UT/LT, and UT/SA ratios were calculated. This was done because from an injury prevention perspective, it is particularly interesting to select exercises with low scapular muscle ratios (⁶).

Figure 1: Half push-up exercise without and with Redcord slings.

Figure 2: Knee push-up exercise without and with Redcord slings.

Figure 3: Knee prone bridging plus exercise without and with Redcord slings.

Figure 4: Pull-up exercise without and with Redcord slings.

Subjects

Forty-seven recreational athletes (26 men, 21 women; mean ± SD; age, 22 ± 4.31 years; weight, 69 ± 8.57 kg; height 176 ± 0.083 cm; body mass index, 22 ± 2.05 kg·m⁻²) were recruited through private physical training practices and fellow students during 2 years (2011–2012). Subjects were between 18 and 30 years old. There were no subjects <18 years of age. Participants were included if they were in good general health, had no complaints of shoulder pain or instability in the past 12 months, and had no history of orthopedic surgery of the shoulder or surrounding region. Moreover, each subject needed to be able to perform the exercises with proper proximal stability. Some experience with RS was allowed, yet no long-term training experience was tolerated to prevent the level of training to be of influence. All subjects gave written informed consent for this investigation, which was approved by the Ethical Committee of the Ghent University Hospital (Belgium).

Procedures

Electromyography

For registration of EMG activation, a Noraxon Myosystem 1400 electromyographic receiver (Noraxon USA Inc., Scottsdale, AZ, USA) was used. In all participants, the dominant side was tested, which was prepared by shaving and cleaning the skin surface to reduce skin impedance (<10 kΩ). Bipolar Ag-Cl surface electrodes (Blue sensor; Medicotest, Ølstykke, Denmark) were placed over the tested muscles, and a reference electrode was placed over the homolateral clavicle (^{6,10,12,25,30}). The researcher confirmed that the electrodes were correctly placed by inspecting the EMG signals on a computer screen during specific muscle testing. The sampling rate was 1000 Hz. All raw myoelectric signals were preamplified (overall gain, 1000; common rate rejection ratio, 115 dB, signal-to-noise ratio, <1 μV root mean square baseline noise). The Myoresearch XP Master Edition 1.07.41 Software Program was used for signal processing. All raw EMG signals were analog/digital-converted (12-bit resolution) with a sampling rate of 1000 Hz, and after rectification, cardiac artifact reduction, and smoothing (root mean square = 100 Hz), the results were normalized to the maximal activity measured during the MVIC trials. The EMG data for each muscle and each subject were averaged for each phase across the 3 intermediate repetitions of the 5 completed trials, as the first and last repetitions were dismissed for further analysis to avoid the influence of habituation and fatigue.

Exercise Testing

Maximum voluntary isometric contraction were determined for normalization by performing 5-second isometric contractions against manual resistance (^7,10,12,30). A metronome was used to control duration of phases, and subjects were encouraged by verbal feedback. After MVIC testing, participants performed the 4 exercises with and without RS. During each exercise, one of the examiners encouraged the participants verbally and, if necessary, corrected their performance. Subjects completed 5 repetitions of each exercise with 5 seconds of intermediate rest, while a resting period of 2 minutes was held between exercises. Each exercise consisted of a 3 seconds concentric and 3 seconds eccentric phase. During the EMG registration, simultaneous video recordings (Sony Handycam, DCR-HC 37, Sony Europe Limited, Zaventem, Belgium) were made, and a metronome was used to control movement speed (60 beeps per minute).

During the HPU, the subject was positioned in 45° above a metal bar. Hands were placed in a pronated position slightly wider than shoulder width. Then, the subject performed a push-up until the elbows were flexed 90° to prevent excessive anterior translations in the glenohumeral joint. The same exercise description was given when performing the exercise with the RS, so replacing the metal bar with the slings (Figure 1). During the KPU, the subject was positioned in a push-up position on the knees. Feet were elevated, and hands were placed slightly wider than shoulders width. Then, a push-up was performed until the elbows were flexed 90°. The same exercise was performed with the RS while gripping the slings 10 cm above the ground (⁸) (Figure 2). During the KPBP, the subject was positioned while leaning on the elbows with the feet slightly elevated, performing scapular protraction and retraction movements. The same exercise was performed with the RS, elbows positioned 10 cm above the ground (Figure 3). During the PU exercise, the subject was positioned supine grasping the bar with both hands. While maintaining neutral spinal alignment and maintaining contact with the heels to the floor, the subject performed a PU until the elbows were flexed 90°. The same exercise was performed while grasping the RS. During each exercise, subjects were instructed to maintain neutral spinal alignment (Figure 4).

Statistical Analyses

SPSS Statistics 19 for Windows (SPSS Science, Chicago, IL, USA) was used for statistical analysis and started with a Kolmogorov-Smirnov test, showing normal distribution of the data. Our goal was to determine differences in individual muscle activation patterns and scapular muscle ratios when performing the same exercise with and without RS. Therefore, paired t-tests were performed for both normalized means (UT, MT, LT, SA, PM, AD, PD, LD) and muscle ratios (UT/MT, UT/LT, UT/SA). Statistical significance was accepted at α < 0.05.

Results

The Effect of Redcord Slings on Shoulder Electromyographic Activation

The results of the comparative analysis of the 4 exercises without and with RS on the normalized activation levels can be found in Tables 1 and 2. Significantly increased activation in the UT (4.49; p = 0.026), LT (5.78; p = 0.028), PM (27.16; p = 0.005), and PD (9.22; p = 0.003) was observed when the HPU with RS was compared with the normal HPU, whereas other muscles showed no significant differences. The KPU showed no significant differences except for the SA (−15.19; p = 0.003) and AD (−13.97, p = 0.045), which showed decreased activation levels with RS. For the KPBP exercise, significant influences were noted for all muscles with the exception of the UT, LT, and LD. In this exercise, the PM (33.91; p = 0.001) was the only muscle showing a significantly increased activation level with RS. The MT (−5.70; p = 0.039), SA (−12.73; p = 0.014), AD (−42.82; p = 0.0001), and PD (−10.42; p = 0.035) significantly decreased their activation level when using RS. During the PU, a significantly decreased MT (−9.57; p = 0.002) and LT (−7.06; p = 0.014) muscle activation was found. On the contrary, AD (10.81; p = 0.024) and LD (14.10; p = 0.006) showed significantly increased activation levels during this exercise.

Table 1: Mean and SD for normalized EMG activity of each muscle during the half push-up exercise.

Table 2: Mean and SD for normalized EMG activity of each muscle during the knee push-up exercise.

The Effect of Redcord Slings on Scapular Muscle Ratios

The results of the comparative analysis of the 4 exercises without and with RS on the scapular muscle ratios can be found in Tables 3 and 4. During the HPU exercise, a significantly increased UT/MT (38.10; p = 0.01) and UT/SA (30.80; p = 0.001) ratio was noted when using RS. The UT/MT (20.82; p = 0.036) and UT/SA (16.13; p = 0.015) ratios were also significantly increased during the KPU with RS, whereas the UT/LT (−8.84; p = 0.401) ratio did not significantly change (Tables 5 and 6). When the KPBP was performed with RS, a significant increase could be observed for the UT/SA (30.78; p = 0.031) ratio (Tables 7 and 8). During the PU, UT/MT (11.61; p = 0.014) and UT/LT (15.74; p = 0.008) ratios significantly increased, whereas the UT/SA (−71.95; p = 0.015) ratio significantly decreased.

Table 3: Mean and SD for normalized EMG activity of each muscle during the knee prone bridging plus exercise.

Table 4: Mean and SD for normalized EMG activity of each muscle during the pull-up exercise.

Table 5: Mean and SD for scapular muscle ratios during the half push-up exercise without and with Redcord slings.

Table 6: Mean and SD for scapular muscle ratios during the knee push-up exercise without and with Redcord slings.

Table 7: SD for scapular muscle ratios during the knee prone bridging exercise without and with Redcord slings.

Table 8: Mean and SD for scapular muscle ratios during the pull-up exercise without and with Redcord slings.

Discussion

Athletes in general and overhead athletes in particular are often recommended to include scapular strengthening exercises in their training regime (^19,21). Exercises that highly activate MT, LT, and SA muscle components while minimizing activation in the large muscle groups are often preferred (^20,22). Slings are frequently used to provoke increased activation levels in these muscles. Therefore, analyzing the extent to which a muscle is recruited while performing certain activities with and without RS can help trainers and physical therapists to select the most appropriate exercises for a particular case. To our knowledge, this study is the first to investigate the effect of RS on scapulothoracic and glenohumeral muscle activation during the 4 selected exercises. It was hypothesized that each muscle component and each scapular muscle ratio would have been altered when comparing the same exercise without and with RS. However, the main finding was that scapular muscle activation decreased, whereas glenohumeral muscle activation increased; although, this was only the case for all muscles during all exercises. These findings are in accordance with the findings of previous research on this topic showing that not all muscles increase or decrease their muscle activation levels in response to an unstable surface (^9,11,26,27).

Concerning the changes in scapular muscle activation, the SA decreased during the KPU (−15.19% MVIC; p = 0.003) and KPBP exercise. The significantly decreased SA activation during the KPU exercise was also found by Lehman et al. (²⁶) and Maenhout et al. (²⁹). They suggested that a higher position of the hands, placing more weight on lower and less on upper extremities, could lead to the decreased SA activation. Our study (placing the hands 10 cm above the floor when using RS) confirms this hypothesis. The same was found for the KPBP exercise. Nevertheless, some studies did not observe a difference in SA activation, which may be explained by the way the unstable surface is created in these various investigations (^9,36,42). Generally, the amplitude levels of the SA were found high during the 2 stable push-up exercises, but decreased to a moderate level when using RS (¹³).

Concerning glenohumeral muscle activation, the 2 push-up exercises (without and with RS) and the KPBP exercise (with RS) showed high activation in the PM. The use of RS increased PM muscle activation during the HPU (27.16% MVIC; p = 0.005) and KPBP (33.91% MVIC; p = 0.001) exercise, which is in accordance with the results by Sandhu et al. (⁴²). The increased activation levels may be caused by the way shoulder abduction needs to be controlled when RS are applied. The RS may have created an unstable condition in multiple directions because there was no contact with the floor when using the slings. Indeed, some authors who used unstable surfaces as a Swiss ball, so maintaining some contact with the floor, could not observe any difference (²⁷) or even found a decreased PM activation (¹¹), whereas we found an increased activation. During the KPU exercise, the AD showed decreased muscle activation with the use of RS. The AD and PD also significantly decreased their activation level, mainly in the AD (−42% MVIC; p = 0.0001), for which the reason currently remains undefined. During the PU exercise, a significantly higher AD (10.81% MVIC; p = 0.024) activation was found, which could be explained by the position of the hands. These were placed slightly wider than shoulder width. Under this condition, the AD muscle is at its greatest length and under its greatest tension, especially when combined with a horizontal abduction position (at the end of the concentric phase). Latissimus dorsi muscle activation is also increased (14.10% MVIC; p = 0.006). This muscle contributes to core stabilization, and possibly the unstable surface increased its demand of stability, explaining the significant increase when using slings.

Concerning the scapular muscle ratios, 5 exercises could be selected on the basis of a low UT/SA ratio: HPU with and without RS, KPU with and without RS, and KPBP without RS. There was not a single exercise that could be selected based on a low UT/MT and UT/LT ratio. Furthermore, the results demonstrated that for all ratios and all exercises tested, RS were not preferred over a stable condition, except for the UT/LT ratio during the KPU exercise. However, no statistical differences could be found for the UT/LT ratio between both conditions.

Using the Redcord device as an unstable surface has some practical advantages compared with other exercise material such as gym equipment. It is portable and time efficient, which eliminates the time factor that is a barrier for many people trying to perform their prescribed exercises (^4,8,39). However, the need for further individualization is reflected by the high SD values found in this study. Lehman (²⁴) already pointed to the inconsistent responses across participants during certain exercises and among specific muscles with the addition of unstable surfaces. This suggests that other factors than the equipment that is used, such as the participant's characteristics, may influence muscle recruitment to a greater extent than surface instability. Indeed, large variability in the interindividual responses is generally found in papers supporting the use of surface instability as a training tool (²⁴). The results of our investigation largely confirm these findings. Thus, clinicians should be aware that individual factors may play a large role in how muscle activation levels are recruited during a certain exercise. Individuals can present with markedly varied responses different from the mean when performing a certain exercise, especially when an unstable surface is used. An argument can even be made that in some cases, the muscle activation is decreased, also decreasing the stress on the muscle (²⁴). This is relevant to the training professional when designing exercise programs that attempt to create a training effect over time.

Some limitations should be taken into account when interpreting the results of this study. First, the results of this trial only provide evidence for the influence of RS during the selected exercises. The results do not imply that RS influence the specific recruitment patterns of the shoulder musculature in a similar way during other exercises than the ones studied. It does neither provide the necessary information to comment on the effectiveness of those exercises. Second, it should be noted that extrapolation of the results to athletes with shoulder pain should be performed with caution. For example, Tucker et al. (⁴⁶) found that, during closed chain exercises, the MT activity differs in overhead athletes with a history of secondary shoulder impingement compared with those who lack this history. Probably, the influence of RS on the muscle activation is also dependent on the athlete's current level of symptoms when performing closed chain exercises. Third, the individual degree of difficulty during each exercise was not taken into account in our study. In contrast, Huang et al. (¹⁶) made an individual progression in the level of difficulty, which was determined by the subject's ability to execute the exercise comfortably and correctly. Although we selected exercises based on their limited need for proximal stability, with all subjects receiving constant feedback, this may be the major limitation from a clinical point of view.

Based on the results of our study, some further investigations might be interesting to perform. First, the influence of RS on the activation levels of other muscles, such as those of the rotator cuff, could be of interest. Because these muscles act as the main glenohumeral stabilizers, investigating their muscle activation levels and study the influence of various unstable conditions during CKC exercises might be relevant. Second, the study could be repeated with a group of athletes suffering from shoulder dysfunction like scapular dyskinesis or impingement syndrome. Subsequently, comparing the results of that study with the current investigation could be relevant because this would give more insight into how these aspects impact the muscle recruitment during each exercise. Third, a randomized controlled trial could be performed to study the effect of exercises with RS in athletes suffering from mild impingement symptoms. Finally, another point of interest would be to examine the effect of an individual progression in degree of difficulty during the exercises and to accompany the EMG measurements by a synchronized kinematic evaluation while using the full Redcord workstation.

Practical Applications

Athletes involved in general resistance training and overhead athletes training for specific muscle strengthening purposes are recommended to include exercises for the scapular stabilizing muscles while minimizing the activation in the larger muscles around the shoulder. Closed kinetic chain exercises on an unstable surface are frequently used for proper strengthening of the shoulder, often incorporating RS as a training device. Therefore, knowing how RS influence the activation levels in particular shoulder girdle muscles, especially the MT, LT, and SA, is of interest. This study investigated 4 CKC exercises with and without RS. Based on the results, the coach is encouraged to use RS within general strengthening training programs because they highly activate the large prime mover muscles of the shoulder girdle. However, coaches should be aware that using RS does not necessarily imply that higher activation will be obtained in the muscles responsible for scapular stabilization. Consequently, the use of RS is not always preferred over a stable surface when the goal is to strengthen the MT, LT, and SA while minimizing the activation in the scapular and glenohumeral prime movers.

Acknowledgments

As there has been no financial support for this study, the authors are grateful to the volunteers who participated in this study. The authors disclose they have no professional relationship with the equipment used during this study. The results of the present study do not constitute endorsement of the device by the authors or the National Strength and Conditioning Association.