Kinetic chain-based shoulder exercises: effects of lower extremity contribution on scapular muscle balance in individuals with and without scapular dyskinesis Authors: Young-Eun Choi Published: 2025 ABSTRACT AbstractThe purpose of this study was to compare the effects of kinetic chain-based forward punch plus (FPP) exercise on scapular muscle activation between individuals with scapular dyskinesis (SD) and healthy controls. A total of 30 male participants were recruited, including 15 with SD and 15 healthy individuals. Muscle activation of the upper trapezius (UT), lower trapezius (LT), serratus anterior (SA), and gluteus maximus, as well as activation ratios, were measured during FPP and FPP with contralateral leg flexion. The healthy group demonstrated decreased UT activation accompanied by increased LT and SA activation, resulting in significantly reduced UT/SA and UT/LT ratios. In contrast, the SD group showed significant increases in SA and LT; however, compensatory activation of the UT prevented improvement in muscle activation balance. These findings highlight the need for targeted neuromuscular strategies to address compensatory UT activation in individuals with SD during kinetic chain-based movements.Keywords: Scapular dyskinesis, Serratus anterior, Kinetic chain, Electromyography INTRODUCTION Scapular dyskinesis (SD) is defined as an alteration in the normal position and motion of the scapula during both static and dynamic states, with a high prevalence reported in approximately 61% of athletes and 59% of asymptomatic individuals (Burn et al., 2016; Ramiscal et al., 2022). SD may not only be a consequence but also a risk factor for injury, negatively impacting return-to-sport timelines and the recovery of daily functional activities (Hickey et al., 2018). Therefore, an appropriate intervention for SD is clinically important from both a preventive and therapeutic perspective. Shifting the paradigm from simple muscle weakness, recent studies now attribute SD primarily to impaired scapular kinetic control rather than to general weakness of the periscapular muscles (Fu et al., 2025). Electromyographic studies have characteristically observed relative overactivity of the upper trapezius (UT), underactivity and delayed activation of the serratus anterior (SA) and lower trapezius (LT) in individuals with scapular dysfunction (Sciascia and Kibler, 2022). This failure of intermuscular coordination, characterized by an imbalance between the UT and LT and SA, is considered a primary mechanism leading to abnormal scapular kinematics (Hotta et al., 2018). Furthermore, Martens et al. (2024) reported that symptomatic patients with SD exhibit decreased strength and endurance of the protractor muscles, while the function of the retractor muscles is relatively increased. This distinction underscores the need for effective rehabilitation strategies that focus on restoring intermuscular balance rather than solely on increasing absolute muscle strength. Recruitment patterns of the periscapular muscles during open and closed kinetic chain exercises have been actively studied for the functional recovery of scapular muscles. Among these, the push-up exercise has been shown to effectively activate the SA while minimizing UT activity, a finding that has been consistently supported by more recent studies examining various push-up modifications and kinetic chain (Mendez-Rebolledo et al., 2022; Wright et al., 2018). The kinetic chain is a process in which multiple joints and muscles act sequentially to generate and transfer force; thus, upper extremity function is based not on isolated shoulder joint movement but on sequential kinetic chain activity, where the energy generated from the lower extremities and trunk is transmitted to the upper limbs (Ellenbecker and Aoki, 2020; Richardson et al., 2020). Therefore, integration of the kinetic chain into shoulder rehabilitation exercises is proposed as a key clinical strategy. De Mey et al. (2013) reported that trapezius muscle activity during scapular retraction exercises varies with the posture and movement of the lower extremities, noting particularly greater activation when performed in a contralateral leg squat position. Recent studies have also shown that integration of the kinetic chain during shoulder elevation exercises affects scapular muscle activation and intermuscular ratios (Borms et al., 2020; Castelein et al., 2016; Yamauchi et al., 2015). Although the forward punch plus (FPP) has been explored with kinetic chain involvement in healthy population, such research remains limited in scope and lacks inclusion of individuals with SD (Kaur et al., 2014). Furthermore, existing research is often limited to a single group (either healthy individuals or patients) or specific exercises, leaving a scarcity of evidence directly comparing healthy and SD groups. The FPP exercise is considered a functional rehabilitation movement that resembles the forward pushing action commonly performed in sports and daily life. It simultaneously requires scapular protraction and shoulder stabilization (Kaur et al., 2020; Kim et al., 2011). Nevertheless, the muscle activation effects and rehabilitation applicability of such functional exercises have not been sufficiently verified. Therefore, addressing this gap in knowledge may provide critical insight into designing patient-specific rehabilitation strategies and developing evidence-based rehabilitation programs. Moreover, university students are a population particularly susceptible to changes in scapular movement patterns due to prolonged sitting, forward head posture, and increased use of electronic devices (Akodu et al., 2018). Therefore, a preventive approach targeting this group is also important. Therefore, the first purpose of this study was to compare the effects of a kinetic chain-based FPP on scapular muscle activation between healthy individuals and a group with SD, both consisting of young adults. The second purpose was to examine the differences in intermuscular balance, specifically the ratios of MT, LT, and SA activity to that of the UT, during the selected exercise. MATERIALS AND METHODS Participants Participants were recruited through university bulletin board announcements and voluntary applications. Those who met the inclusion criteria participated in a prescreening process and submitted a consent form prior to joining the study. A total of 30 male participants (age, 22.93±1.93 years; range, 20–28 years) were recruited for this study, comprising 15 individuals with SD and 15 healthy controls. All participants received a thorough explanation of the experimental procedures and the purpose of the study, after which they provided voluntary written informed consent. Participants capable of full active elevation of the upper extremities were included. The exclusion criteria included a history of pain or surgery in the neck or upper extremities, neurological abnormalities, being an athlete or engaging in regular strenuous exercise, lower limb length discrepancy, scoliosis or thoracic kyphosis beyond the normal range, and positive findings on impingement-related tests. The procedures of this study were approved by the Research Ethics Committee of the Kaya University (Kaya IRB-427). Clinical assessment The participants were classified into a healthy control group and a SD group based on the presence of dyskinesis confirmed through clinical assessment. The clinical assessment was performed by a physical therapist specializing in shoulder injuries, and all evaluations were conducted consistently by the same therapist. The presence of dyskinesis was evaluated using the Scapular Dyskinesis Test and the Kibler Lateral Scapular Slide Test (LSST). In the Scapular Dyskinesis Test (McClure et al., 2009; Tate et al., 2009), the participants were instructed to bilaterally elevate their shoulders through flexion and abduction to their maximum range of motion over 3 sec and then lower them over 3 sec, with their thumbs pointing upward. This movement was repeated 5 times, after which the participants performed the same movement again while holding dumbbells determined by their body weight. A 1.4-kg weight was used for those weighing <68.1 kg, and a 2.3-kg weight was used for those weighing ≥68.1 kg. All movements were recorded from a posterior view using a smartphone (Galaxy Note 20 Ultra, Samsung, Korea). The examiner reviewed the recorded videos to identify the presence of unilateral or bilateral SD and rated the scapular movement as normal, subtle, or obvious. Obvious was defined as a clear dysrhythmia or winging of >2.54 cm occurring in at least three of five repetitions during either flexion or abduction. Dysrhythmia was defined as any of the following: premature or excessive scapular protraction, excessive elevation, nonsmooth motion, or rapid downward rotation during arm lowering. Winging was defined as the posterior displacement of the medial border and/or inferior angle of the scapula away from the thoracic cage. The dominant arm was used for measurements if SD was bilateral or absent; the arm with observed dyskinesis was used if it was unilateral. The LSST measures the distance between the inferior angle of the scapula and the corresponding spinal spinous process (Curtis and Roush, 2006; Kibler, 1998; Koslow et al., 2003). The measurements were taken in three positions: with the arm at the side, with the hand on the hip, and with the arm abducted to 90° in maximal internal rotation. A positive test was determined if an asymmetry of ≥1.5 cm was observed in two or more of the three positions. Participants who met the positive criteria on both the Scapular Dyskinesis Test and LLST were classified as having SD, consistent with the diagnostic criteria (Uhl et al., 2009). To ensure that the observed dyskinesis was not confounded by pain from impingement syndrome, only participants with negative findings on all related tests were classified into the SD group. Electromyography acquisitions In this study, data on the muscle activity of the UT, LT, SA, and gluteus maximus (GMax) were collected and processed using a 2EM electromyography (EMG) system (4D-MT, Relive, Korea). Surface EMG electrodes were attached parallel to the muscle fiber orientation with an interelectrode distance of 2 cm. The EMG signals were sampled at 1,000 Hz and bandpass filtered between 0 and 500 Hz. Surface EMG electrodes were placed according to standard anatomical landmarks. The electrode for the UT was attached at the midpoint between the spinous process of the seventh cervical vertebra (C7) and the posterior tip of the acromion process, along the line of the trapezius muscle. The electrode for the SA was positioned below the axilla, in the area between the superior border of the latissimus dorsi and the inferior border of the pectoralis major (Hintermeister et al., 1998). The electrode for the LT was placed obliquely upward and laterally along the line connecting the intersection of the scapular spine and the vertebral border of the scapula with the spinous process of the seventh thoracic vertebra (T7). The electrode for the GMax was positioned midway between the greater trochanter and the third sacral vertebra (S3), at the level of the greater trochanter. All electrodes were attached at an oblique angle parallel to the orientation of the muscle fibers, and a reference electrode was placed over the sternal manubrium. The EMG signals were processed and analyzed using the root mean square (RMS) method. To normalize the activity of each muscle, maximal voluntary isometric contraction (MVIC) was measured. The MVIC of the UT was obtained in a seated position with the shoulder flexed to 90°, against manual resistance applied to the distal upper extremity. The LT MVIC was measured in a prone position with the arm abducted along the direction of the muscle fibers, against resistance to further elevation. The SA MVIC was assessed in a seated position with the arm elevated to 135°, against resistance applied to the distal upper extremity. For the GMax, the MVIC was measured in a prone position with the knee flexed at 90°, against resistance to hip extension. Each MVIC measurement was performed 3 times for 5 sec, and the average RMS value from the middle 3 sec (excluding the first and last seconds) was used for normalization, expressed as a percentage of MVIC (%MVIC). Procedure The FPP was performed with the participants standing upright, feet shoulder-width apart and parallel to each other. The movement began with the affected arm positioned alongside the body and the elbow flexed at 90°. From this position, the participants flexed the shoulder to 90°, fully extended the elbow, and internally rotated the humerus by approximately 45° while protracting the scapula. The initial position was then resumed by extending the shoulder and flexing the elbow back to the neutral. The FPP with contralateral leg flexion (FPP-CLF) was performed in the same manner as the FPP, except that participant simultaneously extended the ipsilateral leg backward and flexed the contralateral leg forward in a lunge-like motion. After the punching phase, participants returned to the initial parallel stance by extending the shoulder and flexing the elbow to neutral. The step length was standardized to 75% of each participant’s leg length and marked on the floor based on the center of the contralateral foot. All participants were provided with a practice trial before the test, followed by five repetitions of the exercise. A 5-sec rest period was given between repetitions. The movement speed was controlled using a metronome set at 60 beats per minute, with one beat allocated for each of the three phases: starting position, maximum reach, and ending position (Decker et al., 1999; Kaur et al., 2014). Data analysis All statistical analyses were performed using IBM SPSS Statistics ver. 23.0 (IBM Co., USA). Independent t-tests were used to compare participant characteristics and baseline outcome measures between the normal and SD groups. To examine the main effects and interaction effects of group (SD vs. healthy control) and intervention (FPP vs. FPP-CLF), a two-way repeated-measures analysis of variance (ANOVA) was conducted. When significant main or interaction effects were found, post hoc analyses were performed using the Bonferroni correction. The level of statistical significance was set at α=0.05. RESULTS General characteristics of the SD group and the control group This study had a total of 30 male subjects, including 15 subjects in the SD group and 15 in the healthy control group. The general characteristics of the two groups are summarized in Table 1. Table 1General characteristics of participants (n=30)VariableSD groupControl groupAge (yr)22.53±2.3123.32±1.42Height (cm)173.65±5.61173.34±6.12Weight (kg)77.49±9.8373.21±7.78BMI (kg/m2)24.71±3.6623.81±3.03Values are presented as mean±standard deviation.SD, scapular dyskinesis; BMI, body mass index. General characteristics of participants (n=30) Values are presented as mean±standard deviation. SD, scapular dyskinesis; BMI, body mass index. Muscle activation during FPP During the FPP exercise, the UT activation in the SD group was significantly higher than that in the control group (P=0.009). In contrast, the SA activation was significantly higher in the control group than in the SD group (P=0.038). No significant differences were found between the groups in the LT or GMax activations (P> 0.05). Regarding muscle activation ratios, the SD group showed significantly higher values in both UT/SA (P=0.005) and UT/LT (P=0.014) compared to the control group (Table 2). Table 2Comparison of muscle activation and activation ratios between SD and control groups during FPP (n=30)VariableSD groupControl group t P-valueUT34.97±16.6421.63±6.19−2.9090.009SA37.53±7.9443.36±6.672.1740.038LT41.93±11.0841.60±9.11−0.0870.931GMax5.59±2.747.90±2.441.5750.126UT/SA1.00±0.560.51±0.19−3.1760.005UT/LT0.86±0.410.54±0.20−2.6780.014Values are presented as mean±standard deviation.SD, scapular dyskinesis; FPP, forward punch plus; UT, upper trapezius; SA, serratus anterior; LT, lower trapezius; GMax, gluteus maximus. Comparison of muscle activation and activation ratios between SD and control groups during FPP (n=30) Values are presented as mean±standard deviation. SD, scapular dyskinesis; FPP, forward punch plus; UT, upper trapezius; SA, serratus anterior; LT, lower trapezius; GMax, gluteus maximus. Main and interaction effects Two-way repeated-measures ANOVA revealed significant main effects of time for UT, SA, LT, and UT/LT (P<0.001). Significant main effects of group were also observed for UT, SA, UT/SA, and UT/LT (P<0.05). Significant time × group interactions were found in UT (P<0.001), LT (P=0.008), UT/SA (P=0.035), and UT/LT (P<0.001), whereas no significant interactions were observed for SA (P=0.349) or GMax (P=0.102) (Table 3). Table 3Changes in outcome measure variables between FPP and FPP-CLFVariableTimeGroupTime×group F P-value F P-value F P-valueUT90.351<0.00121.515<0.00144.077<0.001 SA187.000<0.0017.6760.0100.9060.349 LT67.038<0.0010.6300.4348.3010.008 GMax429.491<0.0011.2290.2772.8610.102 UT/SA0.3270.57221.375<0.0014.9260.035 UT/LT18.056<0.00118.290<0.00137.546<0.001Values are presented as mean±standard deviation.FPP, forward punch plus; FPP-CLE, forward punch plus-contralateral leg flexion; UT, upper trapezius; SA, serratus anterior; LT, lower trapezius; GMax, gluteus maximus. Changes in outcome measure variables between FPP and FPP-CLF Values are presented as mean±standard deviation. FPP, forward punch plus; FPP-CLE, forward punch plus-contralateral leg flexion; UT, upper trapezius; SA, serratus anterior; LT, lower trapezius; GMax, gluteus maximus. Within-group changes Within-group comparisons showed that, in the SD group, the activations of UT, SA, LT, and GMax significantly increased after the intervention (P<0.001). Although the UT/SA and UT/LT ratios tended to increase, the changes were not statistically significant (P>0.05). In the control group, all muscle activations also increased significantly (P<0.05), while the UT/LT ratio significantly decreased after the intervention (P=0.027) (Table 4). Table 4Results of within-group comparisonsVariableGroupPreinterventionPostinterventionP-valueUTSD34.97±16.6454.15±16.69<0.001Control21.63±6.1925.04±8.740.020 SASD37.53±7.9449.52±7.10<0.001Control43.36±6.6757.14±6.71<0.001 LTSD41.93±11.0847.81±10.33<0.001Control41.60±9.1153.88±10.83<0.001 GMaxSD5.59±2.7449.56±12.00<0.001Control7.09±2.4444.42±6.22<0.001 UT/SASD1.00±0.561.12±0.380.131Control0.51±0.190.44±0.160.117 UT/LTSD0.86±0.411.18±0.43<0.001Control0.54±0.200.48±0.210.027Values are presented as mean±standard deviation.SD, scapular dyskinesis; UT, upper trapezius; SA, serratus anterior; LT, lower trapezius; GMax, gluteus maximus. Results of within-group comparisons Values are presented as mean±standard deviation. SD, scapular dyskinesis; UT, upper trapezius; SA, serratus anterior; LT, lower trapezius; GMax, gluteus maximus. Between-group differences in changes Between-group comparisons of the change values revealed significant differences in UT (P<0.001), LT (P=0.011), UT/SA (P= 0.035), and UT/LT (P<0.001). The change in UT activation was markedly greater in the SD group than in the control group (19.17± 7.70 vs. 3.40±5.02). The UT/SA and UT/LT ratios also increased significantly in the SD group, whereas no significant differences between groups were found for SA (P=0.349) or GMax (P=0.106) (Table 5). Table 5Results of between-group comparison in the change valueVariableSD groupControl groupP-valueUT19.17±7.703.40±5.02<0.001SA11.98±2.8313.78±6.720.349LT12.27±8.235.88±2.460.011GMax43.96±13.6137.33±6.730.106UT/SA0.11±0.28−0.07±0.160.035UT/LT0.32±0.22−0.05±0.09<0.001Values are presented as mean±standard deviation.SD, scapular dyskinesis; UT, upper trapezius; SA, serratus anterior; LT, lower trapezius; GMax, gluteus maximus. Results of between-group comparison in the change value Values are presented as mean±standard deviation. SD, scapular dyskinesis; UT, upper trapezius; SA, serratus anterior; LT, lower trapezius; GMax, gluteus maximus. DISCUSSION This study aimed to compare the effects of a kinetic chain-based FPP on scapular muscle activation between individuals with SD and healthy controls. The analysis revealed that both groups exhibited significantly greater activation of the SA and LT during the FPP-CLF compared with the FPP; however, the muscle activation ratios (UT/SA and UT/LT) showed contrasting patterns. The UT/SA and UT/LT ratios are key indicators reflecting the coordination among scapular stabilizing muscles and the ability to maintain shoulder stability. A decrease in these ratios indicates suppression of excessive stabilization and intermuscular coordination (Hori et al., 2024; Ijiri and Suzuki, 2024). In the healthy group, a decrease in the UT activity accompanied the increase in LT and SA activity, leading to a significant reduction in the UT/SA and UT/LT ratios, which can be interpreted as an improved muscle activation balance necessary for shoulder stabilization. In contrast, the SD group appeared to retain compensatory activation patterns of the UT, suggesting limited ability to control scapular upward rotation and posterior tilting. These findings imply that individuals with SD may exhibit altered neuromuscular control characteristics compared to healthy individuals. The FPP is a representative movement that elicits strong activation of the SA, and this activation is further enhanced by the integration of the lower extremities into kinetic chain (Kaur et al., 2014). This suggests the potential for a partial contribution from the serape effect (serape chain mechanism) during lower extremity engagement. Indeed, in this study, contralateral GMax activation increased significantly during FPP-CLF in both groups, supporting the possibility of myofascial chain activation from the external oblique to the SA via the thoracolumbar fascia (Shin et al., 2013; Vega Toro et al., 2016). Maenhout et al. (2010) reported that ipsilateral leg extension during a knee push-up plus increased SA activation, and Kim et al. (2011) suggested that a myofascial connection between the GMax, external oblique, and SA could contribute to SA activation. Therefore, the increased SA activation in the healthy group in our study may be partially attributable to increased GMax activation and its associated myofascial chain reaction. In the SD group, while SA activation significantly increased, the magnitude of the increase was smaller than in the healthy group, and a concomitant increase in UT activation was prominent. This finding suggests that kinetic chain engagement may not be universally effective for improving muscle activation balance. Scapular stabilization refers to the ability to maintain proper positioning and movement of the scapula relative to the thorax. This is primarily controlled by the coordination of muscles such as SA and LT. These stabilizing muscles help maintain the functional center of the shoulder joint through upward rotation, posterior tilt, and approximation of the scapula to the thoracic cage (Camargo and Neumann, 2019). However, in individuals with SD, weakness or delayed activation of the SA and LT can result in a scapular posture characterized by anterior tilt and internal rotation, which may lead to superior migration of the humeral head or shoulder impingement symptoms. Such imbalance induces compensatory overactivation of the UT, ultimately leading to an unstable scapulohumeral rhythm (Huang et al., 2018; Sciascia and Kibler, 2022). Therefore, weakness of the scapular stabilizing muscles is closely associated with the main symptoms of SD, including pain, fatigue, and restricted shoulder movement (Lopes et al., 2015). Previous studies have reported reduced neural drive to the SA and impaired recruitment of high-threshold motor units in patients with SD, indicating that the UT may be recruited compensatorily when efficient SA activation is not achieved (Kuniki et al., 2024). Furthermore, Martens et al. (2024) reported that an imbalance between shoulder protractors and retractors is closely associated with symptom exacerbation. The results of our study also support this mechanism, suggesting that kinetic chain interventions, when applied prematurely before an adequate neuromuscular control foundation is established, may paradoxically lead to inefficient muscle activation patterns in patients with SD. Clinically, kinetic chain-based exercises can enhance SA and LT activation while inhibiting UT in healthy individuals, supporting their use for improving shoulder stability. In contrast, patients with SD may not show similar benefits. Therefore, in the initial phase of rehabilitation, priority should be given to local shoulder stabilization training, particularly selective strengthening of the SA and LT and inhibition of the UT. Therefore, a stepwise approach is advisable, in which local stabilization is prioritized before progressively integrating kinetic chain-based training that includes the lower extremities and trunk, once neuromuscular control and intermuscular coordination have been restored to a certain level (Moradi et al., 2025). This study has several limitations. First, it was conducted on young adult males, which limits the generalizability of the findings to other age groups or sexes. Second, the use of surface EMG precluded the direct confirmation of SA motor unit recruitment and firing characteristics. Future research should more specifically validate the timing and efficacy of the kinetic chain approach in patients with SD through analyses that include various relative load levels and long-term clinical outcomes. Nevertheless, this study is significant in that it comparatively verified that a kinetic chain-based FPP induces different muscle activation balance responses in healthy and SD groups. This clearly demonstrates that the kinetic chain approach is not a universally positive strategy for all individuals and highlights the importance of a stepwise, individualized rehabilitation strategy. This study confirmed that a kinetic chain-based FPP induces different muscle activation responses in healthy individuals and patients with SD. The healthy group exhibited a typical response where increased SA and LT activation was accompanied by decreased UT activation, resulting in improved muscle activation balance necessary for shoulder stabilization. In contrast, the SD group did not achieve improved intermuscular balance due to compensatory UT activation, despite the increase in SA activation. These findings indicate that the kinetic chain approach does not provide the same positive effects for all individuals. Therefore, while kinetic chain exercises may be effectively utilized for performance enhancement and preventive training in healthy individuals, patients with SD may require a more tailored and progressive approach. Emphasizing local scapular control and neuromuscular coordination within kinetic chain exercises may be essential to promote balanced muscle activation and prevent compensatory UT dominance. These considerations highlight the importance of individualized rehabilitation strategies when integrating kinetic chain-based movements in clinical practice. Footnotes CONFLICT OF INTEREST No potential conflict of interest relevant to this article was reported. ACKNOWLEDGMENTS The authors received no financial support for this article. REFERENCES