Physical Therapy Korea

The serratus anterior (SA) is a crucial muscle for protraction at the scapulothoracic joint and serves as a stabilizer for the scapula [1]. SA is particularly important for scapular upward rotation, as it maintains the scapulohumeral rhythm and optimizes the function of the shoulder complex during arm elevation [2-4]. Additionally, SA contributes to the posterior tilting and external rotation of the upwardly rotating scapula [1]. For overhead activities, such as shoulder elevation through flexion or abduction, the coracoid process must move posteriorly and cranially, while the inferior angle of the scapula moves anteriorly and caudally [5].

An imbalance or weakness in the SA can lead to abnormal scapular motion, scapular dysfunction, and altered muscle activation patterns [6]. Notably, decreased SA activation has been observed in patients with shoulder impingement, glenohumeral instability, and rotator cuff disease [6]. Scapular dyskinesis has also been reported in 57% of patients with shoulder impingement and 32% of patients with shoulder instability [7,8]. Therefore, selectively strengthening the SA is crucial for individuals with shoulder dysfunction such as impingement syndrome [9].

Several studies have been conducted to investigate effective methods for strengthening the SA [10]. Ludewig et al. [11] reported that selective strengthening to increase the activity of the SA while reducing the use of the upper trapezius (UT) muscle is required for effective SA strengthening. Among the push-up plus (PUP), elbow PUP, and wall PUP exercises, PUP was found to be the optimal exercise for selective maximal activation of the SA with minimal activation of the UT muscle [11]. However, SA activity was found to be lower in PUP exercises compared to other arm-raising exercises [12,13]. Additionally, excessive activation of the pectoralis major (PM) muscle can lead to problems such as reduction of subacromial space and subacromial impingement [14].

To compensate for this, a method for applying the abdominal drawing-in maneuver (ADIM) was developed. These exercise methods have been reported to be more effective than other core stabilization exercises for the optimal coactivation of the deep trunk muscles [15]. Core stability is essential for transferring forces and energy generated during movement or exercise to the upper extremities [16,17]. Performing PUP with ADIM has been found to improve core stabilization and subsequently improve SA activity [18]. However, proper execution of ADIM requires attention to breathing, and not many people perform ADIM properly, requiring additional training time [15,18]. Kim and Yoo [19] used a method involving an increase in dynamic stability during hip adduction (HA) exercise. This method reportedly increases the muscle activity of the external oblique (EO) and internal oblique (IO) muscles during isometric HA exercise, inducing intra-abdominal pressure and improving the dynamic stability of the core through coactivation [20-22]. This approach allowed for an easier route to reveal the effects of the exercise.

Therefore, the purpose of this study was to compare the activation levels of PM, SA, EO, and IO muscles during PUP exercises with isometric contraction of HA in healthy male participants. Furthermore, we aimed to compare the changes and effects observed in the activation of the four muscles during PUP, PUP with ADIM, and PUP with HA exercises. We hypothesized that the PUP with HA exercise would increase the activation of the SA, EO, and IO and reduce the PM activation and PM/SA ratio compared with the other PUP exercises.

1. Participants

The study was conducted on 19 healthy male participants (age: 39.0 ± 6.8 years; height: 173.3 ± 5.7 cm; weight: 75.2 ± 7.8 kg). The selection criteria included individuals without any pain in the neck or upper limbs for the past 6 months and those without any surgical or neurological pathologies. Participants who had difficulty in accurately performing the PUP movement for five or more repetitions were excluded from the study [12]. The experimental procedures were explained to all participants, and written informed consent was obtained prior to the study. This study was approved by the Yonsei University Mirae Institutional Review Board (IRB no. 1041849-202104-BM-058-01).

2. Instruments

Muscle activity was measured using the surface electromyography (EMG) system Noraxon TeleMyo 2400 (Noraxon Inc.), and the software Noraxon MyoResearch 1.06 XP (Noraxon Inc.) was used to collect the muscle activity data. The band pass filter was set at 10–450 Hz with a sampling rate of 1,000 Hz. Before attaching the disposable Ag/AgCl surface electrodes, the skin was prepared by removing the stratum corneum with a razor and gauze and then cleaning with alcohol [23]. The EMG electrode for the PM was attached at the mid-point of the muscle belly of the sternal part of the PM, approximately 2 cm inside the axillary fold wrinkles of the participant. EMG electrode for the SA was attached to the anterior surface of the latissimus dorsi, just below the inferior angle of the scapula. EMG electrode for EO was attached slightly obliquely to the midpoint between the lateral side of the rectus abdominis muscle and the iliac crest of anterior superior iliac spine (ASIS) between the ribs. EMG electrode for the IO was attached to a point 1 cm inside the ASIS and below the line connecting both ASISs [23,24]. Maximum voluntary isometric contraction (MVIC) of the hip adductor muscles was measured in the quadruped position using the Smart KEMA Pressure Sensor (KOREATECH, Inc.) (Figure 1). Measured data was transmitted to a tablet PC (Galaxy Tab S4; Samsung, Inc.) through a Bluetooth network and analyzed using the Smart KEMA application (KOREATECH, Inc.) [25].

Figure 1. Smart KEMA Pressure Sensor (KOREATECH, Inc.).

3. Procedures

The participants of this study were measured for the muscle activities of PM, SA, EO, and IO while performing exercises in three PUP positions, which were conventional PUP, PUP with ADIM, and PUP with HA. Each participants practiced each exercise for 10 minutes to become familiar with the three PUP exercises. After becoming familiar with the movements, they rested for 10 minutes. The MVIC of each muscle was measured by attaching electrodes [2,26]. To collect MVIC data, participants performed MVIC of the PM, SA, EO, and IO muscles based on the muscle manual testing position, with muscle activity signals recorded during a 5-second period of maintaining each contraction [27]. Signals were analyzed during the middle 2–4 seconds, excluding the first and last seconds of the movement [28]. The maximum pressure of HA was measured by performing MVIC with a Smart KEMA Pressure Sensor between the knees in a quadruped position. After the rest period, the three PUP exercises were performed randomly. Each exercise was performed thrice for 5 seconds using a metronome set at 60 beats per minute. To prevent the effect of muscle fatigue on the measurements, there was a 3-minute rest between each exercise. The order of the three exercises was randomly determined using Microsoft Excel (Microsoft).

The starting position is a quadruped position with hands at shoulder-width and knees hip-width apart. The wrists are aligned vertically under the acromioclavicular joint. The elbows are slightly flexed and the cervical spine is maintained in a neutral position [12,18]. The movement position involves maintaining scapular protraction for a maximum of 5 seconds before returning to the starting position (Figure 2A).

Figure 2. PUP exercises. (A) PUP, (B) PUP with ADIM, and (C) PUP with HA. PUP, push-up plus; PUP with ADIM, push-up plus with abdominal drawing-in maneuver; PUP with HA, push-up plus with hip adduction.

The starting position is in the same posture as that in the PUP exercise. Contract the abdomen so that the navel is pulled towards the spine without allowing any movement in the pelvis, ribs, or spine. The participant maintains normal respiration while contracting the abdominal muscles [29]. At the same time, hold the scapula in the maximum protraction position for 5 seconds and then return to the starting position (Figure 2B).

The starting position is in the same posture as that in the PUP exercise. After placing the Smart KEMA Pressure Sensor between the knees, the participant contracted the hip adductor muscles to 20% of maximum pressure while receiving visual feedback via a tablet PC (Galaxy Tab S4) [24]. Concurrently, the participant maintained scapular protraction for 5 seconds before returning to the starting position (Figure 2C).

4. Data Collection

During the exercise performed in the three PUP exercise postures, the EMG signals of PM, SA, EO, and IO were collected for 5 seconds while performing the exercise, and the average value of the muscle activation during the middle 3 seconds, excluding the first and last seconds, was processed using the root mean square. In addition, the muscle activation of each muscle was standardized using %MVIC, which measures the muscle activation during MVIC as described by Kendall et al. [26].

5. Statistical Analysis

The data in this study were analyzed using IBM SPSS ver. 25.0 for Windows (IBM Co.). All data values in this study were tested for normal distribution using the Shapiro–Wilk test, and assumptions for parametric tests were verified using Levene’s test for equal variances. To compare muscle activity and muscle activity ratio according to the three PUP exercises, repeated ANOVA was used with a significance level of 0.05. When a significant difference was found, post hoc tests were conducted using paired-sample t-tests with Bonferroni correction, and the significance level was set at 0.017 (0.05/3).

1. Pectoralis Major

A significant difference in the activity of the PM was observed between the PUP, PUP with HA, and PUP with ADIM exercises (p < 0.001) (Table 1). According to a post hoc t-test, PM activity during the PUP with HA exercise was significantly lower than that during the PUP and PUP with ADIM exercise (p < 0.001 and p < 0.001, respectively) (Figure 3). However, no significant difference was observed between the PM muscle activity when performing the PUP and PUP with ADIM exercises (p = 0.04).

Table 1 . Electromyographic activation for each exercise (%MVIC).

Muscle	PUP	PUP with ADIM	PUP with HA	F	p-value	η2
Pectoralis major	24.25 ± 12.60	20.53 ± 11.90	17.44 ± 9.45	10.40	< 0.001	0.06
Serratus anterior	36.47 ± 12.45	36.75 ± 11.68	46.68 ± 11.91	25.57	< 0.001	0.14
External oblique	13.68 ± 9.61	23.30 ± 10.68	20.23 ± 10.61	35.67	< 0.001	0.14
Internal oblique	20.91 ± 24.18	34.42 ± 22.24	26.20 ± 19.01	5.92	0.130	0.06

Values are presented as mean ± standard deviation. %MVIC, percentage of maximal voluntary isometric contraction; PUP, push-up plus; PUP with ADIM, push-up plus with abdominal drawing-in maneuver; PUP with HA, push-up plus with hip adduction..

Figure 3. Multiple comparisons of electromyographic activation among exercises. PUP, push-up plus; PUP with ADIM, push-up plus with abdominal drawing-in maneuver; PUP with HA, push-up plus with hip adduction; PM, pectoralis major; SA, serratus anterior; EO, external oblique; IO, internal oblique; %MVIC, percentage of maximal voluntary isometric contraction. ***p < 0.001.

2. Serratus Anterior

A significant difference in the muscle activity of the SA was observed between the PUP, PUP with ADIM, and PUP with HA exercises (p < 0.001). According to a post hoc t-test, SA muscle activity during the PUP with HA exercise was significantly higher than that during the PUP and PUP with ADIM exercise (p < 0.001 and p < 0.001, respectively). However, no significant difference was noted between the SA activity when performing the PUP and PUP with ADIM exercises (p = 0.88).

3. External Oblique

A significant difference in the activity of the EO was observed during the PUP, PUP with ADIM, and PUP with HA exercises (p < 0.001). According to a post hoc t-test, EO activity during the PUP with ADIM exercise was significantly higher than that during the PUP and PUP with HA exercises (p < 0.001 and p < 0.001, respectively). In addition, EO muscle activity during the PUP with HA exercise was significantly higher than that during the PUP exercise (p < 0.001).

4. Internal Oblique

No significant difference in the muscle activity of the IO was observed during the PUP, PUP with ADIM, and PUP with HA exercises (p = 0.13). According to a post hoc t-test, IO muscle activity during the PUP with ADIM exercise was significantly higher than that during the PUP and PUP with HA exercise (p < 0.001 and p < 0.001, respectively). However, no significant difference was noted between the IO activity when performing the PUP and PUP with HA exercises (p = 0.29).

5. Pectoralis Major/Serratus Anterior Activity Ratio

A significant difference in the PM/SA activity ratio was observed between the PUP, PUP with HA, and PUP with ADIM exercises (p < 0.001) (Table 2). According to a post hoc t-test, the PM/SA activity ratio during the PUP with HA exercise was significantly lower than that during the PUP and PUP with ADIM exercises (p < 0.001 and p < 0.001, respectively) (Figure 4). However, no significant difference was observed between the PM/SA activity ratio when performing the PUP and PUP with ADIM exercises (p = 0.046).

Table 2 . Ratio of muscle activity for each exercise.

Muscle	PUP	PUP with ADIM	PUP with HA	F	p-value	η2
PM/SA	0.61 ± 0.50	0.48 ± 0.35	0.32 ± 0.23	13.88	< 0.001	0.14

Values are presented as mean ± standard deviation. PUP, push-up plus; PUP with ADIM, push-up plus with abdominal drawing-in maneuver; PUP with HA, push-up plus with hip adduction; PM, pectoralis major; SA, serratus anterior..

Figure 4. Multiple comparisons of ratio among exercises. PUP, push-up plus; PUP with ADIM, push-up plus with abdominal drawing-in maneuver; PUP with HA, push-up plus with hip adduction; PM, pectoralis major; SA, serratus anterior. ***p < 0.001.

The purpose of this study was to investigate the effects of HA isometric contraction on the activation of the shoulder and trunk muscle stabilization during PUP exercises in healthy male participants. Participants performed PUP, PUP with ADIM, and PUP with HA. The differences in activation of the PM, SA, EO, and IO muscles were compared. Our hypothesis was that PUP exercises with HA would increase the activation of SA, EO, and IO compared to other PUP exercises and decrease the activation of PM and the PM/SA ratio. The study results showed that EO was more involved during the PUP with HA exercise than in the PUP exercise, and a significant difference was observed in SA when compared between the PUP and PUP with ADIM exercises [19]. In addition, the muscle activation of the PM was lower than that seen for the PUP and knee PUP with ADIM exercises, supporting our hypothesis. Therefore, if a more effective training method is needed to selectively train the SA during the early shoulder rehabilitation period, without adding burden to the shoulder, the PUP with HA exercise is sufficient [11].

Previous studies on PUP exercises that aimed at eliciting high muscle activation of the SA showed that when performing standard PUP, knee PUP, elbow PUP, and wall PUP exercises on stable and unstable surfaces, the SA was not influenced by the surface condition. Additionally, during the knee PUP exercise, the muscle activation of SA was reported to be 13.48 and 29.20 %MVIC, respectively [30,31]. During the bridge exercise, the activity of the IO due to cocontraction of the hip adductor muscles accounted for 20.39 %MVIC [24]. Our study also found that the activity of the SA (46.69 %MVIC) and IO (26.20 %MVIC) muscles exceeded a certain threshold level and appeared to have reached an appropriate level for muscle strengthening.

Muscle usage such as speed, intensity, and the nature of the movement are determined by various factors, and predicting which muscles will be additionally utilized is challenging. However, according to the muscle recruitment principle, adding one movement to a specific action would have a higher probability of recruiting additional muscles, and according to the Henneman's size principle, larger muscles would have been recruited more often and exerted greater force [32,33]. Based on this principle, Kaur et al. [34] observed that SA activity increased during forward punch plus exercise accompanied by activation of the lower extremities and trunk muscles. Also, while the hip adductor muscles are primarily responsible for HA, their cocontraction increases the tension of the thoracolumbar fascia and activates both the IO and EO, thereby stabilizing the lumbopelvic region. This connection between the hip adductor and the abdominal muscles is called longitudinal sling, the cocontraction of the hip adductor muscles promotes the cooperation between the pelvic floor and abdominal muscles, increasing intra-abdominal pressure and enhancing trunk stability [1,20,21,35,36]. For this reason, Kibler and Sciascia [37] suggested that shoulder rehabilitation, which includes scapular muscle exercise, be accompanied by activation of abdominal muscles to increase trunk stability. As such, the shoulder joint of the upper extremity and the scapulothoracic muscles are anatomically and physiologically connected to the trunk [36]. Trunk stability plays an important role in transferring greater force and energy to the upper extremities during activities of daily living, movement, or exercise, thereby allowing the body to efficiently generate and transfer energy for these activities [16,17,38]. Kim and Yoo [39] reported that the above mechanism resulted in a decrease in deviation of center of pressure and a selective increase in SA activity while performing PUP with HA. Therefore, our study predicted that the activation level of the SA would be higher when the hip adductor muscles were contracted during the PUP exercise compared to that in the previous studies, based on the aforementioned rationale.

However, this study had several limitations. First, only young males participated; thus, the results are not generalizable to older individuals or females. Second, the applicability of results to individuals with a medical condition is unknown, as only healthy participants were included. Third, long-term effects were not observed. Therefore, future studies should observe long-term effects through interventions in patient groups with shoulder disorders, as well as through comparisons based on age and sex.

In this study, when performing PUP exercise by activating the hip adductor muscles, the activation of SA and EO increased and the PM and PM/SA activity ratio decreased. This method has an advantage in increasing the activity ratio of the SA muscle, even proportionally, compared to using ADIM. If this method is used in clinical practice, it can be used to effectively lead muscle activity while reducing the burden on patients’ shoulders with an easier educational method.

None.

None to declare.

No potential conflicts of interest relevant to this article are reported.

Conceptualization: SL, JK, OK. Data curation: SL. Formal analysis: SL, JK. Investigation: SL, JK, OK. Methodology: SL, JK, OK. Project administration: OK. Resources: SL. Supervision: OK. Validation: JK, OK. Visualization: JK. Writing - original draft: SL. Writing - review & editing: JK, OK.