Physical Therapy Korea

Shoulder pain is the most common problem in overhead athletes [1,2], and it is caused by the cumulative microtrauma brought on by high speeds and repetitive overhead motions during throwing [3]. More than 50% of adolescent baseball players are also observed to experience pain in the shoulder joint [3,4]. During the throwing motion, a high humeral angular velocity of internal rotation (IR) is produced at about 7,000 °/s, and the IR torque may be as much as 90 Nm [3,5]. It leads to an increase in the eccentric stress on the posterior rotator cuff (RC), posterior deltoid, and soft tissues [6-8]. Repetitive throwing motions thus cause imbalance in the RC muscles in overhead athletes [9-11], which can lead to various shoulder injuries [12-14].

This imbalance of the RC muscles is also closely related to scapular dyskinesis (SD) [15-22] defined as alterations in position and motion patterns of the scapular and it is believed to occur caused by changes in activity of the scapular stabilizer muscles [23,24]. Optimal scapular positioning provides a stable base of support for the RC muscles to generate maximum force [15,16]. Altered scapular kinematics can increase the demand on the RC muscles during functional activities, potentially causing weakness and imbalance in the RC muscles [19-22]. In addition, this muscle imbalance predisposes to shoulder joint dysfunction and throwing-related injuries. Previous studies have reported decreased strength in the RC muscles of overhead athletes with SD [19-22].

In overhead athletes with impingement syndrome, Tate et al. [22] found reduced muscle strength during scapular plane elevation at 90° when the scapula was in a neutral position compared to when it was repositioned. Merolla et al. [19] investigated the comparison of infraspinatus muscle strength before and after scapular repositioning and reported a greater decrease in infraspinatus muscle strength before scapular repositioning. They also demonstrated an improvement in infraspinatus strength and pain after 3 and 6 months of interventions aimed at improving the balance of the scapular stabilizer in professional volleyball players with SD [20]. Another study by Merolla et al. [21] investigated the effectiveness of strengthening exercise on the scapular stabilizing muscles to strengthen the supraspinatus and infraspinatus muscles in overhead athletes with SD who had shoulder pathology and reported increased muscle strength after 3 and 6 months.

Previous studies have confirmed that SD affects the RC muscle strength of overhead athletes. However, these studies have only focused on isometric strength, raising doubts about whether it accurately reflects the muscle strength needed for the throwing activity. Because of the high demands placed on the IR muscle groups for strong propulsion during overhead throwing [10], research has demonstrated a consistent increase in IR muscle strength in the dominant shoulder in overhead athletes across different levels, from professional institutes to college and high school [14,25-30]. Therefore, the concentric and eccentric balance of agonist and antagonist muscles is very important because the proper balance of IR and external rotation (ER) allows the humeral head to center the glenoid fossa [29] and maintain the rotation point of the glenohumeral (GH) joint for overhead activities [6,31].

The functional strength ratio (FSR), described as the strength ratio of eccentric antagonist to concentric agonist, has been used to measure the dynamic strength and balance of RC muscles related to throwing [10,32-36]. Studies have documented a decreased dominant shoulder ER-to-IR strength ratio and eccentric ER to concentric IR (ERecc/IRcon) ratio in overhead athletes [11,25,37,38]. Therefore, the evaluation of the balance of agonist and antagonist muscles should be considered. However, to our knowledge, there has been no study examining potential differences in concentric and eccentric strength and FSR of the RC muscles based on functional parameters related to throwing in individuals with or without SD. Increased demands on the IR muscle according to throwing characteristics and kinematic changes in the scapula can affect the muscle length-tension relationship. Therefore, we hypothesize that it will also affect the concentric and eccentric muscle strength of adolescent baseball players. This study aimed to: 1) determine whether there was a difference in RC muscle strength and strength ratio between the dominant shoulder with SD and the non-dominant shoulder without SD and 2) compare the RC muscle strength and strength ratio between SD types.

1. Study Participants

The sample size for this study was determined using G*power software (ver. 3.1.2; Franz Faul, Kiel University). There were no previous studies comparing concentric, eccentric strength, and FSR of RC according to SD, therefore, based on previous study that compared the isometric strength differences, the authors estimated the sample size for this study needed to detect a large effect size (d = 0.80) [19]. Power analyses indicated that at least 19 participants would be required to achieve a power of 0.85 with a large effect size of 0.8 and a significance level of 0.05 in a one-tailed test. A total of 40 adolescent baseball players (mean age: 16.93 ± 1.31 years, mean height: 176.7 ± 5.9 cm, mean weight: 74.38 ± 8.17 kg, mean experience: 6.1 ± 1.6 years) participated in this study (Table 1). The inclusion criteria were the presence of SD in the dominant shoulder, a lack of SD in the nondominant shoulder, and the absence of pain in the shoulder, neck, or upper extremities. Those without SD in the dominant shoulder or with SD in the nondominant shoulders, a history of surgery on the upper extremities, including the neck, shoulder, and elbow, or with acute upper extremity injuries were excluded. Each participant read and signed a consent form and this study was approved by the Bioethics Committee of Pukyong National University (IRB no. 1041386-202401-HR-13-02) following the Declaration of Helsinki and this study was conducted by the ethical standards of the responsible committee.

Table 1 . Demographic and clinical characteristics of participants.

Variable	Descriptive statistic	Rage
Age (y)	16.93 ± 1.31	15–19
Height (cm)	176.7 ± 5.9	161.2–193.4
Weight (kg)	74.38 ± 8.17	59.1–94.9
Experience (y)	6.1 ± 1.6	2–9
Position
First baseman	2 (4.4)
Second baseman	1 (2.2)
Third baseman	3 (6.7)
Infielder	5 (11.1)
Outfielder	9 (20.0)
Center fielder	1 (2.2)
Pitcher	17 (37.8)
Pitcher/Batter	2 (4.4)

Values are presented as number only, number (%), or mean ± standard deviation..

2. Procedures

First, the participants in this study were classified by SD type as type I to III using SD test. SD was classified by examiner’s visual observation based on scapular movements of the participants during arm elevation in the raising (concentric) and lowering (eccentric) phases. Then, after receiving an explanation of the isokinetic muscle strength test procedure, participants performed the IR and ER with maximal concentric and eccentric contractions at an isokinetic velocity of 120 °/s in supine position. The concentric and eccentric muscles for the IR and ER of the dominant shoulder with SD and the non-dominant shoulder without SD were evaluated. FSR was calculated by using the measured isokinetic strength data. The difference in isokinetic concentric and eccentric strength of IR and ER muscles and FSR between the dominant shoulder with SD and the non-dominant shoulder without SD was compared.

SD was classified based on movement pattern using the SD test, which evaluates scapular alignment and movement via visual observation during active arm elevation [39,40]. In this study, for reliable of SD classification, SD type was classified by two physical therapists with over 10 years of experience in treating musculoskeletal and sports injuries. The examiners evaluated the scapular movements of the participants during arm elevation in the raising (concentric) and lowering (eccentric) phases.

SD was classified into four types [39,41] based on scapular movement patterns, as follows: 1) Type I: the scapular inferior medial angle or the lower third of the medial scapula border is displaced posteriorly from the thorax and prominent; 2) Type II: the upper two-thirds of the medial border is displaced posteriorly from the posterior thorax and prominent; 3) Type III: insufficient scapular upward rotation or involves early or excessive scapular elevation; and 4) Type IV: no evidence of scapular inferomedial angle and medial border displacement, and early or excessive scapular elevation (normal scapular movement). The SD test has moderate intra-rater reliability (Kappa coefficient = 0.49–0.59) and moderate-to-substantial interrater reliability (Kappa coefficient = 0.49–0.64) for identifying and classifying the SD [39-41]. Only when the SD types classified by the two examiners matched were the data included in this study; any data where they did not match were excluded. In this study, as a results of SD test, SD types were identified as type I (n = 19) and type II (n = 20), respectively (Figure 1). Therefore, identified the difference in RC muscle strength and RC muscle strength ratio between type I and type II in this study.

Figure 1. Scapular dyskinesis type I (A) and type II (B).

To evaluate the concentric and eccentric strengths of the IR and ER muscles, an isokinetic dynamometer was used (Biodex dynamometer system 4, Biodex Corp.). To minimize compensatory movements of the scapula or other segments during the evaluation of the isokinetic concentric and eccentric strengths, the test was conducted in the supine position [33,42,43]. The supine position is recommended as a reliable method to evaluate the isokinetic muscle strength of IR and ER muscles because of its benefits in facilitating the range of motion of the shoulder joint and stability during evaluation [44,45]. Furthermore, adopting the supine position may help control unwanted body bounces or inconsistent repetitions during evaluation [33,42,43]. The participants were asked to lie down on an isokinetic strength test equipment chair with the backrest tilted back. Then, the participants abducted their shoulder to 90°, flexed their elbow to 90°, and rotated their shoulder to 0° while keeping the forearm pronated. The strap was wrapped around the shoulder girdle and trunk to prevent compensatory movements during the evaluation. The shoulder axis was aligned against the isokinetic dynamometer to evaluate strength across a total range of 150°, from an IR of 60° to an ER of 90° [33,42,43], at an isokinetic velocity of 120 °/s [42,43] (Figure 2). Concentric strength was evaluated first, followed by eccentric strength. Both concentric and eccentric tests consisted of five maximal effort reciprocal repetitions, and peak torque to body weight (PT/BW) data was collected in each condition. During the evaluation of isokinetic strength, the participants were not provided any feedback about torque for the IR and ER. The FSR data was obtained by calculating the strength ratios of eccentric IR to concentric ER (IRecc/ERcon) and ERecc/IRcon, respectively. The participants were given 5 minutes to rest between IR and ER strength test sessions.

Figure 2. Measurement of isokinetic concentric and eccentric torques of internal and external rotation. IR, internal rotation; ER, external rotation.

3. Statistical Analyses

The Shapiro–Wilk test for normality was conducted, and the assumption of normality was satisfied for all dependent variables. The paired t-tests were used to analyze the differences in the isokinetic concentric and eccentric PT/BW of IR, ER, and FSR between the dominant shoulder with SD and the nondominant shoulder without SD. In addition, the independent t-test was used to compare the differences in isokinetic strength and FSR according to SD types. IBM SPSS version 21.0 for Windows (IBM Corp.) software was used for all data analyses. The alpha level was set at 0.05.

In this study, the intra-rater reliability of the isokinetic concentric and eccentric strengths of the IR (intraclass correlation coefficient [ICC] = 0.968–0.990) and ER (ICC = 0.987–0.994) muscles of the dominant shoulder with SD and the non-dominant shoulder without SD was examined and found to be high (Table 2). There were significant differences in isokinetic concentric PT/BW, eccentric PT/BW, and the FSR of IR and ER muscles between the dominant shoulder with SD and the non-dominant shoulder without SD. There was a significant decrease in the concentric PT/BW of the IR (mean difference = –1.96, p = 0.009) and ER (mean difference = –1.98, p = 0.005) muscles in the dominant shoulder with SD compared to the non-dominant shoulder without SD (Table 3). Eccentric PT/BW of the IR (mean difference = –1.13, p = 0.039) and ER (mean difference = –1.99, p = 0.004) muscles were also significantly decreased in the dominant shoulder with SD compared to the non-dominant shoulder without SD (Table 3). However, no significant difference was observed in the FSR (IRecc/ERcon ratio: mean difference = 0.04, p = 0.067, ERecc/IRcon ratio: mean difference = 0.02, p = 0.47) (Table 3). Also, no significant differences were identified in any measured variables between type I and type II SD in terms of isokinetic strength and FSR (p > 0.05) (Table 4).

Table 2 . The intra-rater reliability for isokinetic concentric and eccentric PT/BW in dominant shoulder with SD and non-dominant shoulder without SD.

	Intra correlation coefficient
	Concentric IR (PT/BW)	Eccentric IR (PT/BW)	Concentric ER (PT/BW)	Eccentric ER (PT/BW)
Dominant shoulder with SD	0.980	0.983	0.994	0.987
Non-dominant shoulder without SD	0.968	0.990	0.990	0.989

IR, internal rotation; ER, external rotation; PT/BW, peak torque to body weight; SD, scapular dyskinesis..

Table 3 . Comparison of isokinetic concentric, eccentric PT/BW, and FSR of IR and ER muscles between dominant shoulder with SD and non-dominant shoulder without SD.

Variable	Dominant shoulder with SD	Non-dominant shoulder without SD	Mean difference (95% CI)	p-value
Concentric IR (PT/BW)	36.40 ± 6.13	38.36 ± 4.56	–1.96 (–3.40 to –0.53)	0.009*
Eccentric IR (PT/BW)	40.92 ± 5.15	42.06 ± 4.63	–1.13 (–2.21 to –0.06)	0.039*
Concentric ER (PT/BW)	37.43 ± 7.51	39.41 ± 6.75	–1.98 (–3.34 to –0.62)	0.005*
Eccentric ER (PT/BW)	42.00 ± 5.41	43.99 ± 5.67	–1.99 (–3.30 to –0.69)	0.004*
IRecc/ERcon ratio	1.14 ± 0.29	1.10 ± 0.22	0.04 (–0.003 to 1.88)	0.067
ERecc/IRcon ratio	1.18 ± 0.20	1.16 ± 0.17	0.02 (–0.04 to 0.08)	0.470

Values are presented as number only or mean ± standard deviation. PT/BW, peak torque to body weight; FSR, functional strength ratio; IR, internal rotation; ER, external rotation; SD, scapular dyskinesis; CI, confidence interval; IRecc/ERcon, eccentric internal rotation to concentric external rotation; ERecc/IRcon, eccentric external rotation to concentric internal rotation. *p < 0.05..

Table 4 . Comparison of isokinetic concentric, eccentric PT/BW, and FSR of IR and ER muscles between SD types (type I vs. type II) Dominant shoulder with SD and non-dominant shoulder without SD.

Variable	SD type I (n = 19)	SD type II (n = 21)	Mean difference (95% CI)	p-value
Concentric IR (PT/BW)	36.55 ± 6.19	36.27 ± 6.23	0.28 (–3.70 to 4.26)	0.89
Eccentric IR (PT/BW)	41.41 ± 4.71	40.49 ± 5.59	0.92 (–2.41 to 4.25)	0.58
Concentric ER (PT/BW)	38.29 ± 7.12	36.65 ± 7.64	1.64 (–3.21 to 6.48)	0.50
Eccentric ER (PT/BW)	43.03 ± 4.96	41.07 ± 5.75	1.96 (–1.50 to 5.41)	0.26
IRecc/ERcon ratio	1.08 ± 0.22	1.12 ± 0.21	–0.04 (–0.23 to 0.15)	0.67
ERecc/IRcon ratio	1.20 ± 0.16	1.16 ± 0.24	0.04 (–0.09 to 0.17)	0.56

Values are presented as number only or mean ± standard deviation. PT/BW, peak torque to body weight; FSR, functional strength ratio; IR, internal rotation; ER, external rotation; SD, scapular dyskinesis; CI, confidence interval; IRecc/ERcon, eccentric internal rotation to concentric external rotation; ERecc/IRcon, eccentric external rotation to concentric internal rotation..

We investigated differences in isokinetic concentric and eccentric strengths and FSR of the IR and ER muscles of the shoulder joint between the dominant shoulder with SD and the nondominant shoulder without SD. We also compared isokinetic strength and FSR based on SD types (type I and type II). The results indicated significant differences in isokinetic PT/BW of IR and ER muscles between the dominant shoulder with SD and the nondominant shoulder without SD. The PT/BW was significantly less in both concentric and eccentric strengths of the IR and ER muscles in the dominant shoulder with SD than the non-dominant shoulder without SD.

SD is characterized by altered scapular kinematics and is closely related to imbalances of the RC muscles [16,19-22]. Studies have shown that repetitive throwing motion results in altered scapular kinematics, including decreased serratus anterior (SA) and lower trapezius (LT) muscle activity, as well as decreased scapular upward rotation and posterior tilt conversely, there may be an increased scapular anterior tilt, protraction, and IR [35,46,47]. Consequently, it leads to weakness of the RC muscles. The current study demonstrated that the isokinetic concentric and eccentric PT/BW in both IR and ER muscles decreased in the dominant shoulder with SD compared to the non-dominant shoulder without SD. This result can be explained by the demand for RC muscles and the length-tension relationship caused by an imbalance of scapular stabilizer muscles. Alterations in normal scapular kinematics caused by imbalance or fatigue of the scapular stabilizer muscles cause an imbalance of the RC muscles [17-22]. The SD types observed in the participants of this study were types I and II, which are characterized by the prominence of the scapular inferior medial angle and the upper two-thirds of the medial border, respectively. SD types I and II are related to weakness of the LT and SA muscles. In this study, SD types I and II were characterized by weakness of the participants’ LT and SA muscles. Moreover, because LT and SA play an important role in upward rotation of the scapula for functional activities of the shoulder joint, weakness or fatigue of the scapular stabilizer muscle causes imbalance of these muscles, increasing the demand on the RC muscles for functional activities, which subsequently weakens the RC muscles [17-22]. Furthermore, the correct position of the scapula provides a stable base of support and an optimal length-tension relationship of the RC muscles to produce maximum force generation [15,16]. Therefore, altered scapular kinematics in the dominant shoulder are hypothesized to have increased the demands on the RC muscle for functional activity and affected the optimal length-tension relationship, resulting in a decrease in concentric and eccentric torque of the IR and ER muscles in this study. Not only RC, other shoulder rotator muscles such as pectoralis major or posterior deltoid are play role in shoulder rotational movement during overhead throwing. In addition, these muscles are attached to the clavicle and acromion therefore, it may be affected by SD. Therefore, needed to more clearly demonstrate the impact of SD on RC and other shoulder rotator muscles, including shoulder joint kinematics through further experimental research on individual muscle activity and strength.

Our result is comparable to previous studies on isometric strength. Hannah et al. [48] compared the muscle strength around the scapular and shoulder joints in individuals with and without SD. They found reduced strength of the SA and LT muscles, including the upper trapezius, along with supraspinatus and internal rotator muscles, in those with SD compared to those without SD. Merolla et al. [19-21] reported the results of an experimental study targeting overhead athletes in various sports who have an SD. In a study on volleyball and tennis players with SD and shoulder pain, the strength of the infraspinatus muscle was compared before and after scapular repositioning. The results indicated that the strength of the infraspinatus muscle was significantly improved after retraction and suggested that the scapular retraction test is a reliable measurement method to confirm whether SD causes weakness in RC muscles [19]. In addition, in a study on professional volleyball players with SD, they applied balance training of the scapular stabilization muscles for 3 and 6 months and reported significant improvement in the strength of the infraspinatus muscle [20]. A study examining changes in the strength of the supraspinatus and infraspinatus muscles following balance training of the scapular stabilizing muscles also reported an improvement in the strength of the supraspinatus and infraspinatus muscles after intervention [21]. Tate et al. [22] reported improvements in supraspinatus muscle strength and pain following scapular repositioning in a study targeting college overhead athletes with impingement syndrome. These results suggest that SD could affect RC muscle strength and shoulder symptoms. Moreover, the differences in concentric and eccentric strengths identified in this study may also affect the functional strength of IR and ER muscles related to overhead throwing. This suggests that for adolescent baseball players to improve the function of the IR and ER muscles for overhead throwing, evaluating and addressing SD is important for recovery. Although previous studies and our study have confirmed that SD may affect RC muscle strength, the specific influence of SD on the concentric and eccentric strengths of these muscles remains uncertain. Therefore, caution is advised when making generalized assumptions that SD is a key determinant of the strength of RC muscles. To clarify this, further research is needed to determine the relationship between SD and RC muscle strength.

The result of FSR showed that the IRecc/ERcon and the ERecc/IRcon ratio were not significantly different between the throwing shoulders with SD and nondominant shoulders without SD in this study. Additionally, no differences were found depending on the SD type. An optimal balance of IR and ER positions the head of the humerus in the center of the glenoid fossa [31], stabilizing the rotational axis of the GH joint for overhead activities [6,31]. For this, the eccentric control of the antagonist is very important. Regarding FSR, although no standards have been suggested, studies have reported that the ranges of ERecc/IRcon ratio and IRecc/ERcon ratio as 0.97–1.17 for the ERecc/IRcon ratio and 1.62–2.39 for the IRecc/ERcon ratio in overhead athletes [9,32,33,49]. Our results were consistent with previous studies, showing IRecc/ERcon and ERecc/IRcon ratios of 1.14 and 1.18, respectively, in the dominant shoulder with SD. For the non-dominant shoulder without SD, the IRecc/ERcon and IR/IRcon ratios were 1.10 and 1.16, respectively, in this study. Our FSR results indicate that the antagonist muscle counterbalances the concentric contraction of the agonist muscle. The FSR acts as an indicator of risk for throwing-related injury; studies have revealed that a lower ERecc/IRcon ratio in overhead athletes predisposes them to shoulder injuries [9,26,49,50]. An ERecc/IRcon ratio of < 0.66 is a risk factor for throwing-related injuries [48]. Although there was no significant difference in FSR, both concentric and eccentric PT/BW in the IR and ER muscles were reduced in the dominant shoulder with SD compared to the non-dominant shoulder without SD in this study. Management for SD must address the potential risk factors for throwing-related injuries. Future research is needed to determine not only isokinetic strength but also the relationship between SD, FSR, and the prevalence of throwing-related injuries. This will help establish a management system to address the risk factors associated with overhead throwing-related injuries.

Our study has several limitations. First, because the strength of the scapular stabilizer muscles was not measured, it was not possible to compare differences in the imbalance of the scapular stabilizer muscles according to SD. Future research is needed to quantitatively measure and compare the strength of the scapular stabilizer muscles along with changes in scapular alignment. Second, we did not measure the muscle activity of the IR and ER muscles in the shoulder joint. Although differences in torques for IR and ER were identified, because the IR and ER muscles play a role in controlling the rotation of the humeral head, further research is required to corroborate the differences in muscle activity and muscle activity ratio according to SD. Additionally, direct comparative studies with adolescent baseball players without SD are also needed in the future. Lastly, because factors not only with or without SD but also such as baseball experience and position, including training period may affect muscle strength, it is necessary to interpret the results considering these factors.

The findings indicate that the isokinetic concentric and eccentric PT/BW of the IR and ER muscles were significantly lower in the dominant shoulder with SD than in the non-dominant shoulder without SD. However, the IRecc/ERcon and ERecc/IRcon ratios showed no significant difference between the dominant shoulder with SD and the non-dominant shoulder without SD. Therefore, when establishing a strategy for addressing RC muscle weakness in adolescent baseball players with SD, it is essential to consider an approach that accounts for scapular kinematic recovery.

None.

None to declare.

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

Conceptualization: IY, TK. Data curation: IY, TK. Formal analysis: IY. Investigation: IY. Methodology: TK. Project administration: IY. Resources: IY, TK. Software: IY. Supervision: TK. Validation: TK. Visualization: IY. Writing - original draft: IY. Writing - review & editing: TK.