Physical Therapy Korea

Posterior capsule tightness (PCT), which is commonly seen in overhead athletes, is a soft tissue adaptation caused by repeated tensile loads on the posterior capsule during the deceleration phase of the throwing arm [1]. PCT has also been observed in non-throwers [1,2]. The mechanism for PCT in non-throwers remains unknown; it may be related to postural and degenerative joint changes [3].

PCT is associated with alterations in the scapular and humeral kinematics. PCT causes scapular protraction by passively pulling the scapular outwards [4,5]. Additionally, it induces superior-anterior translation of the humeral head during shoulder movement [1,2,6,7], which entraps the rotator cuff tendon between the acromion and the humeral head [5,7,8]. In a cadaveric study by Muraki et al. [9], PCT increased the subacromial contact pressure primarily on the lesser tuberosity during flexion. The maximum contact pressure occurred close to the end of the flexion range.

PCT significantly limits shoulder internal rotation (IR), as demonstrated in cadaveric studies [6,10]. This decreased IR creates a significantly greater scapular anterior tilt, especially towards the end of IR, at 90° shoulder flexion or abduction position [3,5,11]. This kinematic alteration is associated with shoulder pathologies, such as subacromial impingement syndrome, adhesive capsulitis, stiff shoulders, and superior labral anterior to posterior lesions [5,11-13].

Posterior shoulder stretching exercises have been proposed to resolve PCT and IR deficit [8,14]; sleeper and cross-body stretches (CBS) are widely used. The sleeper stretch is performed in a side-lying position; the shoulder is rotated inward with the opposite hand, while the shoulder and elbow are flexed at 90°. The CBS is performed in a standing or side-lying position; the humerus is pulled horizontally over the body with the opposite hand [14]. Wilk et al. [15] modified the sleeper stretch and CBS because of improper control of both scapular and shoulder rotations, possibly leading to increased subacromial impingement. The modified sleeper stretch (MSS), is performed in a side-lying position. To reduce pain symptoms, the trunk is rotated 20°–30° posteriorly, and the humerus is rotated inward with the opposite hand [15]. However, the MSS still carries a risk of anterior translation of the humeral head. Therefore, a novel MSS (NMSS) was designed to prevent this anterior translation. In the side-lying position, the humeral head of the stretching side is pressed by the opposite hand and head to control its anterior translation. The opposite elbow is placed on the distal part of the forearm and pressed to stretch the posterior capsule and shoulder external rotators.

Several studies have investigated the effectiveness of posterior shoulder stretching exercises in restoring lost shoulder range of motion (ROM) [12-14,16-23], improving pain [12,13,21], shoulder dysfunction [12], muscle strength [20], and muscle stiffness [18], and decreasing the shear elastic modulus [24]. Different methods, such as joint mobilization [16,22,23] and scapular stabilization [17,20,24] have been combined with posterior stretching. However, to the best of our knowledge, no previous studies have investigated the effectiveness of posterior shoulder stretching exercises in decreasing scapular anterior tilting. Thus, we aimed to compare the immediate effects of two posterior shoulder stretching exercises (NMSS and CBS) on scapular anterior tilting and shoulder IR ROM.

1. Subjects

Based on the results of the pilot study (n = 12), a partial η2 (0.217) was calculated. The total sample size was set to 32 (16 NMSS and 16 CBS), considering the calculation for the two-way mixed ANOVA using G* power (ver. 3.1.9.7; Franz Faul, Heinrich Heine University D^üsseldorf, D^üsseldorf, Germany) (alpha level = 0.05, power = 0.8, ES = 0.526) [25].

Thirty-two subjects with an anterior tilted scapular and IR deficits were included in this study. Scapular anterior tilting was estimated as the distance from the posterior aspect of the acromion to the table in the supine position. A distance of ≥ 2.5 cm was considered as scapular anterior tilting (Figure 1). The inclusion criteria were as follows: (1) anterior tilted scapular and (2) a passive IR ROM of ≤ 50°, which was considered as an IR deficits. The exclusion criteria were as follows: (1) a history of shoulder injury within the last 2 years and (2) the presence of enough pain to be unable to proceed with stretching. The general characteristics of the subjects are shown in Table 1. Before being included in the study, all subjects signed a written informed consent. The study was approved by the Institutional Review Board of Yonsei University Mirae campus (IRB no. 1041849-202206-BM-109-01).

Table 1 . General characteristics of the subjects (N = 32).

Variable	NMSS group (n = 16)	CBS group (n = 16)	p-value
Sex (male/female)	8/8	7/9	0.723a
Dominant arm (right/left)	14/2	16/0	0.144a
Age (y)	24.6 ± 2.8	24.1 ± 2.5	0.640b
Height (cm)	170.9 ± 7.9	167.2 ± 7.4	0.179b
Weight (kg)	66.1 ± 12.8	60.9 ± 12.4	0.254b
BMI (kg/m2)	22.5 ± 3.0	21.7 ± 3.5	0.495b

Values are presented as number only or mean ± standard deviation. NMSS, novel modified sleeper stretch; CBS, cross-body stretch; BMI, body mass index. ^aχ²-test. ^bIndependent t-test..

Figure 1. Scapular anterior tilting measurement at rest. The length shown on display (distance from lower jaw's tip to upper jaw's tip) plus the length of the lower jaw (1,500 mm) was measured.

2. Experimental Procedure

A flowchart of the study is shown in Figure 2. Sixty-one subjects were screened to determine their eligibility for this study. Finally, thirty-two subjects with anterior tilted scapula and IR deficits were included in the study. The subjects were randomly assigned to the NMSS or CBS group using a randomization generator (www.randomizer.org), and scapular anterior tilting and shoulder IR ROM were measured. The subjects were familiarized with the stretching exercises before testing. The familiarization period ended when the subject could maintain the exercise position for 30 seconds. The subjects underwent a 10-minute wash-out period after the familiarization period. The stretching program consisted of 30 seconds of stretching and 10 seconds of rest in one set; a total of 10 sets were performed. The scapular anterior tilting and shoulder IR ROM were measured immediately after the intervention.

Figure 2. Flowchart of the study. NMSS, novel modified sleeper stretch; CBS, cross-body stretch.

3. Instrumentation

Scapular anterior tilting was measured using Vernier calipers [26]. The interclass correlation coefficient (ICC) was 0.88–0.94 [27]. Shoulder IR ROM was measured using the Clinometer and bubble level smartphone application (version 2.4; Plaincode Software Solutions, Gunzenhausen, Germany). The ICC was 0.81 (95% confidence interval: 0.70–0.88) [28]. It could measured up to 0.1° and the error range was ± 0.1°. The pressure biofeedback unit (PBU) (Stabilizer TM; Chattanooga Group Inc., Hixson, Tennessee, USA) was used to apply consistent pressure to the stationary arm to improve the reliability of shoulder ROM measurements [29].

4. Intervention

The NMSS was performed in a side-lying position with the shoulder and elbow flexed at 90°. Subjects were asked to press the humeral head of the stretching side with the opposite hand and head to prevent anterior translation of the humeral head. Additionally, they were asked to press the distal part of the forearm with the opposite elbow to provide a stretching force causing shoulder IR (Figure 3).

Figure 3. Novel modified sleeper stretch.

CBS was performed in a side-lying position with the shoulder and elbow flexed at 90°. The body was tilted 20°–30° posteriorly to restrict scapular abduction. The forearm of the side to be stretched was aligned with the opposite forearm to limit humeral ER via counter pressure. The shoulder was adducted horizontally by the opposite hand (Figure 4) [15].

Figure 4. Cross-body stretch.

5. Outcome Measurement

Scapular anterior tilting and shoulder IR ROM were measured. To avoid bias, the examiner was not allowed to read the results on the calipers or clinometer. An independent observer, blinded to the group assignment, read and recorded the results.

Scapular anterior tilting of the dominant shoulder was measured before and immediately after the intervention. Scapular anterior tilting was measured in two ways: at rest and at shoulder 60° active IR. Scapular anterior tilting at rest was measured in supine position. To assess scapular anterior tilting in the shoulder at 60° active IR, the subject was positioned supine with the shoulder at 90° abduction and elbow at 90° flexion. Thereafter, the subject actively rotated their shoulder internally to the target bar, which was set at 60° of IR. When the subject’s hand reached the target bar, the scapular anterior tilting was measured (Figure 5).

Figure 5. Scapular anterior tilting measurement at shoulder 60° active internal rotation.

Shoulder IR ROM was measured in the dominant shoulder before and immediately after the intervention. The subject was positioned supine with 90° of shoulder abduction and elbow flexion. The PBU was placed under the center of the subject’s acromion. To apply consistent pressure at the subject’s shoulder, examiners monitored the PBU gauge and regulated pressure from the initial pressure of 20 mm Hg to 30 mm Hg. The smartphone was located at the proximal ⅓ of the subject’s forearm to measure the ROM. Three ROM measurements were obtained, and the mean value was used for statistical analysis. To assess IR, the examiner pushed the clavicle, coracoid process, and humeral head to fix the humerus and scapular and passively rotated the shoulder inwardly with the opposite hand. When rotation did not occur and reached at firm end feel, it was determined as the end range. The independent observer recorded this ROM data (Figure 6) [23,29].

Figure 6. Shoulder internal rotation range of motion measurement.

6. Statistical Analysis

Statistical analysis was performed using IBM SPSS for Windows (ver. 26.0; IBM Co., Armonk, NY, USA). The Shapiro-Wilk test was used to assess normal distribution. Two-way mixed ANOVA was used to identify significant differences in scapular anterior tilting and shoulder IR ROM between two groups (NMSS vs. CBS and between factors) and within the groups (pre- vs. post-test). Paired t-test was used to identify significant differences in scapular anterior tilting and shoulder IR ROM within each group (pre- vs. post-intervention). Statistical significance was set at p < 0.05.

There were no significant intergroup differences in general characteristics (Table 1). Scapular anterior tilting (at supine and at shoulder 60° active IR) and shoulder IR ROM significantly changed over time (p < 0.05). However, there were no significant group-by-time interaction effects and no significant differences between the groups (Table 2). After stretching, the scapular anterior tilting (at supine and at shoulder 60° active IR) significantly decreased in both groups (p < 0.05). In addition, IR ROM significantly increased in both groups (p < 0.05) (Table 3, Figure 7).

Table 2 . Comparison of scapular anterior tilting and shoulder IR ROM between pre- and post-stretch (N = 32).

Variable	NMSS group			CBS group			Group	Time	Group × Time
	Pre	Post		Pre	Post		p-value
AT rest (cm)	5.80 ± 1.08	5.05 ± 1.02		5.26 ± 1.32	4.51 ± 1.33		0.191	< 0.001*	0.968
AT 60˚ (cm)	4.36 ± 1.12	3.95 ± 0.92		4.20 ± 1.29	3.77 ± 1.25		0.666	< 0.001*	0.859
IR ROM (˚)	46.69 ± 2.61	63.04 ± 8.03		45.23 ± 1.76	65.08 ± 7.24		0.844	< 0.001*	0.203

Values are presented as mean ± standard deviation. NMSS, novel modified sleeper stretch; CBS, cross-body stretch; AT rest, scapular anterior tilting at rest; AT 60°, scapular anterior tilting at shoulder active 60˚ internal rotation; IR ROM, shoulder internal rotation range of motion. *p < 0.05, by two-way mixed ANOVA..

Table 3 . Paired t-test analysis within each group (N = 32).

Variable	Group	Mean ± SD	t	p	Effect size
AT rest (cm)	NMSS	0.74 ± 0.69	4.322	< 0.001*	1.08
	CBS	0.75 ± 0.44	6.840	< 0.001*	1.71
AT 60˚ (cm)	NMSS	0.41 ± 0.43	3.019	0.002*	0.94
	CBS	0.44 ± 0.44	3.995	0.001*	1.00
IR ROM (˚)	NMSS	–16.36 ± 7.71	–8.490	< 0.001*	–2.12
	CBS	–19.85 ± 7.51	–10.568	< 0.001*	–2.64

SD, standard deviation; AT rest, scapular anterior tilting at rest; AT 60°, scapular anterior tilting at shoulder 60˚ active internal rotation; IR ROM, shoulder internal rotation range of motion; NMSS, novel modified sleeper stretch; CBS, cross-body stretch. *p < 0.05, by paired t-test..

Figure 7. Comparison of scapular anterior tilting and shoulder IR ROM in the NMSS and CBS between pre- and post-stretch. (A) Scapular anterior tilting at rest, (B) scapular anterior tilting at shoulder 60° active IR, and (C) shoulder IR ROM. NMSS, novel modified sleeper stretch; CBS, cross-body stretch; IR, internal rotation; ROM, range of motion. *p < 0.05.

This study was conducted to compare the effects of posterior shoulder stretching exercises (NMSS vs. CBS) on scapular anterior tilting and shoulder IR ROM. To the best of our knowledge, it is the first study that compares the effects of posterior shoulder stretching exercises on decreasing scapular anterior tilting.

Scapular anterior tilting (at rest and shoulder 60° active IR) was significantly decreased in both groups. At rest, scapular anterior tilting decreased by 0.74 cm (13%) and 0.75 cm (14%) in the NMSS and CBS groups, respectively. At shoulder 60° active IR, scapular anterior tilting decreased by 0.41 cm (9%) and 0.44 cm (10%) in the NMSS and CBS groups, respectively. This decrease in scapular anterior tilting after posterior shoulder stretching is supported by previous studies showing that PCT causes superior-anterior translation of the humeral head, restriction of shoulder IR ROM, scapular anterior tilting, and protraction [3,5-7,11].

The results of this study showed that scapular anterior tilting at shoulder 60° active IR was less than that at rest. This is inconsistent with the results of a previous study in which the scapula was anteriorly tilted during IR. This can be explained by differences in the measurement posture. In a study by Borich et al. [11], scapular anterior tilting was measured in a standing position, and a sling maintained shoulder 90° abduction. In this study, both measurements were performed in a supine position. At rest, scapular anterior tilting was measured with no shoulder abduction and elbow flexion; however, at shoulder 60° active IR scapular anterior tilting was measured with shoulder at 90° abduction and elbow at 90° flexion. At shoulder 60° active IR, shoulder abduction in the supine position could apply an additional gravitational load to the scapular, which could allow the scapular to be tilted posteriorly.

The shoulder IR ROM was significantly increased in both groups after the stretching exercises. The IR ROM increased by 16.36° (35%) and 19.85° (44%) in the NMSS group CBS groups, respectively. These finding is consistent with those of several previous studies. Both NMMS and CBS improved the flexibility of the posterior capsule and IR ROM. This increased IR ROM contributed to a decrease in scapular anterior tilting.

The IR ROM significantly increased in both groups after the stretching exercises; however, the increase was mildly more in the CBS group than in the NMSS group. This can be explained by the anatomical structure of collagen fibers in the posterior capsule. The radial bundle, which oriented in the medial-lateral direction, is deeper and stronger than the circular bundle, which is oriented in the superior-inferior direction [30]. CBS with horizontal adduction may stretch the radial bundle more directly than NMSS with IR, which is probably more advantageous for posterior capsule stretching. Yamauchi et al. [18] compared the stretching effects of modified CBS (MCS) and MSS and established that the teres minor stiffness reduced in the MCS group, while the infraspinatus stiffness reduced in the MSS group. Teres minor and infraspinatus muscle release can increase IR ROM by 20° [31]. This indicates that NMSS and CBS may be selectively used, depending on which structure is shortened.

Pectoralis minor tightness, thoracic kyphosis, thoracic flexed posture, and loss of lower trapezius and serratus anterior muscle activity also contribute to the scapular anterior tilting [5,26]. Pectoralis minor stretching, shoulder brace use, and scapular posterior tilting exercises were used to reduce scapular anterior tilting. The results of this study indicate that posterior shoulder stretching can be used with these interventions to reduce the scapular anterior tilting in individuals with anterior tilted scapular and shoulder IR deficits.

This study has several limitations that should be considered in future research. First, this study only established the immediate clinical effect of posterior shoulder stretching. In previous studies, a stretching program was conducted for at least 4 weeks. To observe the long-term effect of stretching, future research will require a stretching program of more than 4 weeks. Second, all the subjects were young and had no pain symptoms. Thus, it is difficult to generalize the results to all ages and symptoms. Finally, the method for measuring scapular anterior tilting is limited. In past studies, scapular anterior tilting was measured with a 3-dimentional electromagnetic tracking system or a digital inclinometer in a sitting or standing position [11,32]. However, in this study, the distance between the table and the poster admission was measured due to the limitations of the measurement posture, research environment and equipment. This method used to confirm the round shoulder posture in past studies [26]. This measurement method includes not only the scapular anterior tilting but also the scapular IR component. Therefore, in future study, scapular anterior tilting should be measured alone.

This study provides the first clinical evidence of the therapeutic effects of posterior shoulder stretching exercises on decreasing scapular anterior tilting. Both NMSS and CBS were significantly effective in decreasing the scapular anterior tilting in both positions (at rest and at shoulder 60° active IR). IR ROM significantly increased immediately after NMSS and CBS. However, there were no significant differences between the stretching methods. Thus, NMSS and CBS can be recommended to decrease the scapular anterior tilting and increase the IR ROM in individuals with anterior tilted scapular and shoulder IR deficits.

None.

None to declare.

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

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