Introduction

Shoulder impingement syndrome (SIS) is a common musculoskeletal disorder observed in both athletic and non-athletic population [11, 15, 16, 19]. In fact, SIS has been reportedly diagnosed to account for 44–65% of all shoulder complaints resulted in loss of function, and motor control deficits [30]. This syndrome has been described by a narrowing of the subacromial space that causes encroachment of the subacromial tissues, including the rotator cuff and the subacromial bursa [19]. It could be usually caused by repeated, tedious actions or movements with arm raised above the head in swimming, throwing, installing devices, playing golf, shoveling snow, etc. In elderly individuals, the pain-provoking factor is a contracture of the anterior shoulder complex structures caused by a rounded shoulder posture with protracted shoulders, and injury or degenerations in the joint area [12].

However, conflicting, altered scapular positioning and movement patterns may be critical risk factors for SIS [6, 16, 24]. In this regard, kinematic changes may occur in patients with shoulder impingement resulting in a further decrease in supraspinatus muscle activation [1, 10, 13, 25]. Excessive superior or anterior translation of the humeral head on the glenoid fossa, inadequate external rotations of the humerus, a decrease in normal scapular upward rotation, and posterior tipping on the thorax during shoulder elevation could cause shoulder impingement. Additionally, the hypothesized kinematic alterations in scapular motions have been linked to decrease in serratus anterior muscle activity, to increase in upper trapezius muscle activity, or to unbalanced forces between the upper and lower trapezius muscle [18].

Therefore, it seems that to treat SIS, the focus should be on stretching, strengthening, and neuromuscular control exercises for scapula muscles [7, 8]. It has been reported that a scapular-focused treatment (SFT) improves scapular upward rotation (as one of the main SIS incidence causes) [1] and movement patterns [16].

As monitoring scapular movements during exercise is impossible, shoulder exercises need feedbacks with verbal, visual, and kinesthetic stimuli to help the patients with shoulder pain to control abnormal movements, and maintain posture [28]. In this regard, Weon et al. [31] reported that real-time visual feedback using a video can provide an immediate and continuous feedback to correct scapular movements during independent shoulder flexion exercises, and this results in better activation of the upper trapezius and serratus anterior muscles in those patients with scapular winging.

No randomized controlled trial (RCT) has so far conducted to combine SFT and feedback (SFTF) to treat patients with SIS. Therefore, this study investigated the effects of an 8-week SFT with and without feedback on pain, function and scapular kinematics in participants with SIS. It was hypothesized that adding feedback to a shoulder exercise treatment in clinical setting may enhance the outcomes in participants with SIS.

Materials and methods

The study was performed in accordance with the ethical standards in the World Medical Association Declaration of Helsinki (2002). Ethical approval was obtained from the Ethical Committee of the Kharazmi University and Faculty of the Sport Sciences and the ethical standards in sport and exercise science research were respected (IRB ID number = Kharazmi.REC 20190601/34).

This RCT (UMIN000036399) with a 1:1 ratio allocation was conducted in and approved by the department of Sport Sciences. All the participants were fully informed about the nature of the study, and they signed an informed consent agreement. The examiners were blinded to group assignment.

Participants

The participants were recruited by orthopedic physicians via flyers displayed at hospitals and the university in May and June 2019. Based on the criteria, 75 males and females were randomly assigned into the scapular-focused treatment (SFT) (n = 25), SFT with Feedback (SFTF) (n = 25), and control (n = 25) groups (Fig. 1). The group allocation was performed using a computer-generated, randomized list, created by a person not involved in recruiting procedure. After pre-intervention measurements, the first author opened the sealed envelopes and assigned the individuals into one of the groups. There was no significant difference between participants’ demographic data at baseline (Table 1). The sessions which the participants attended in each group are as follows: SFT group, 22 ± 1 sessions, SFTF group, 21 ± 1 sessions, and control group 22 ± 1 sessions. No adverse events were reported.

Fig. 1
figure 1

Flow chart

Full size image
Table1 Participants demographics and characteristics
Full size table

Inclusion and exclusion criteria

Participants with shoulder pains, lasted > 6 weeks, were included in the study. A consulting orthopedic surgeon diagnosed the participants with SIS if they exhibited at least two of the following items: painful arc during flexion or abduction; a positive Neer or Hawkins–Kennedy test; and painful resisted external rotation, abduction, or Jobe test [21, 29].

Participants were excluded if they had a history of surgery, fracture, or dislocation and traumatic onset of shoulder pains; existence of type 3 acromion; massive rotator cuff tear; a long head of bicep tendon tear; or degenerative joint disorder at the shoulder complex, any rheumatologic, systemic, or neurologic disorders; any neuro-musculoskeletal disorder (including cervical radiculopathy) [21, 23, 29], a body mass index > 30 kg/m2; being pregnant, and, receiving steroid injections and physical therapy during the previous 6 months preceding the study [21, 23, 29].

Procedures

The participants were assessed before (pre-intervention) and after the 8-week interventions (post-intervention). Pain, function and scapular kinematics were meatured as dependent variales. Only the involved side of each participant was assessed.

Primary outcome measure (pain)

Current pain severity at rest, during shoulder movements, during the prior week, and least pain during the prior week were assessed with VAS. Its minimal clinical difference (MCD) is reported more than 2.5 points (25 mm), scores ranged from 0 (no pain) to 10 (maximum pain) [2, 22].

Secondary outcomes (DASH, and 3D scapular kinematics)

Disability of the arm, shoulder, and hand (DASH) questionnaire was used to assess the upper limb functions. The DASH questionnaire has been identified and used as an upper extremity specific orthopedic outcome assessment tool [9, 20]. This version of the DASH questionnaire has been proved to be valid and highly reliable, with intra-observer and interobserver intraclass correlation coefficients (ICCs ˃ 0.90) [9, 20]. A recent study showed that a 10-point decrease on the DASH questionnaire can be considered a clinically important improvement [9].

Also, the scapular motion was recorded with nine infrared cameras (Vicons MX13, Oxford Metrics Ltd., Oxford, UK). Markers were fixed on the seventh cervical vertebra, eighth thoracic vertebra, on the humerus lateral and medial epicondyle, the angle of the acromion, the angle of the acromion jugular notch of the sternum, xiphoid process of the sternum, base of the scapular spine, inferior angle of the scapula, and midpoint between the most anterosuperior aspect of the acromioclavicular joint (Fig. 2). An experienced physiotherapist positioned the markers according to the recommendations of the international society of biomechanics (ISB) [32].

Fig. 2
figure 2

Reflective markers on the posterior of the scapula; a spinous process of the seventh cervical vertebra, b base of the scapular spine, c inferior angle of the scapula, d spinous process of the eighth thoracic vertebra, e midpoint between the most anterosuperior aspect of the acromioclavicular joint and the angle of the acromion, f the angle of the acromion, g on the humerus lateral and medial epicondyle

Full size image

All data were collected at 200 Hz. Skin-based method, which inherently has some errors, was used to measure the kinematics [33].

The shoulder was guided to maintain in scapular plane at 90° of shoulder flexion, by two plastic guides while performing the tests. The test was performed in sequence of 0° maximum 120°–0° elevation every 2 s guided by a metronome (30 beats per min). Each subject performed five trials to familiarize with the test. Scapular positions at 90° and 120° of elevation were extracted during arm raising/lowering, the arc at which impingement might have occurred. Scapular angles were calculated based on the local coordinate system of the scapular relative to the local coordinate system of the thorax and protraction/retraction. The upward/downward rotation and anterior/posterior tilt were extracted using Euler sequencing [5]. These methods are aligned with the recommendations of ISB [4]. Scapular variable IR, PT MCD (1.8–1.6) and MCD value across arm angle, ranged from 2.9 to 7.7.

A difference of 4° or more in scapular kinematics was considered clinically relevant [11].

Kinematic data were analysed in sagittal and scapular planes during arm elevations. To ensure the test is performed correctly, the subject placed the arm on a flat surface while the thumb pointing to the ceiling. Three tests were performed, each in 3 s. Three complete cycles of movements were thereby completed. This procedure has been reported to be reliable in asymptomatic individuals and individuals with SIS [4, 27].

Scapular focused-exercise routine

The SFT protocol used in this research was based on previous studies. This protocol consisted of an 8-week SFT intervention (including isometric stretching, intrinsic and eccentric isotonic exercises). This training protocol includes a shoulder position training (in the first week), rotator cuff muscle strength (second, third, fifth, sixth and seventh weeks), and flexibility (fourth and eighth weeks) [7, 22]. The patients were instructed on how to perform the related exercises focusing on range of motion, strength, and function. At the week 8, they attended a one-on-one check review to ensure compliance with the exercise sheet.

Feedback intervention

The feedback intervention was conducted according to the previous studies [14, 31]. For a 15-min familiarization process, participants performed each exercise four times. During the towel Slide shoulder extension, maximum forward flexion, scapular movement in direction of the 9 o'clock, and the scapula retraction, focusing on the posterior scapula tilt was guided by the therapist using tactile and verbal feedbacks in the scapula and pre-scapular musculature. Each exercise was performed four times with a 1-min rest between each repetition while participants were looking at a monitor located 1 m away from them and at the height of 1 m from the ground. Modified Borg scale was used to assess subjective perception of the volunteer's exertion at the end of the test [3].

Statistical analysis

Statistical power calculations were conducted based on a type I error probability level of alpha = 0.05, type II error probability level of beta = 0.20, and an anticipated dropout rate of 20%, indicating 62 participants for the study [1, 14, 29].

Kolmogorov–Smirnov and Levene tests were used to assess the distribution and homogeneity of variance before performing analysis of covariance, respectively (ANCOVA).

Between-group comparisons were analyzed with a 3 × 2 analysis of variance test, including group interventions (SFT, SFTF, control) and time (baseline, or 8 wk). Furthermore, within-group comparisons were performed using 1-way analysis of variance, and Bonferroni correction was used to adjust the significance levels.

For DASH and VAS scores, intergroup comparisons were calculated using the nonparametric Kruskal–Wallis ANOVA test with multiple comparisons, and intragroup comparisons were calculated using the Friedman ANOVA test with multiple comparisons.

In addition, the effect size (Cohen´s d) was calculated to determine the difference magnitude between groups, and it was interpreted as small if d ≤ 0.5, moderate d = 0.5–0.8 and large d ˃ 0.8.

Results

The primary outcome

For pain, both the SFTF and SFT groups had significant within-group changes at 8 weeks. Also, there were significant differences between SFTF vs. SFT (mean difference = 1 ± 0.6; ES (95% CI) = − 0.3(− 0.4 to − 0.1); P = 0.04) SFTF vs. control (mean difference = − 4.5 ± 0.7; ES (95% CI) = 1.3(0.7–1.9); P = 0.01), and also SFT vs. control (mean difference = 3.5 ± 0.1; ES (95% CI) = 1.0(0.4–1.6); P = 0.01) (Table 2).

Table 2 Primary outcome measure (pain) from baseline to 8-weeks
Full size table

Secondary outcomes

The DASH scores significantly improved from pre- to post-intervention in both experimental groups at 8 weeks. Also, there were significant differences between SFTF vs SFT (mean difference = − 6.1 ± 0.7; ES (95% CI) = 1.8(1.1–2.4); P = 0.04), SFTF vs control (mean difference = 15.1 ± 0.9; ES (95% CI) = 4.4(3.3–5.4); P = 0.01), and also SFT vs control (mean difference = 9 ± 0.2; ES (95% CI) = 2.6(1.8–3.4); P = 0.01) (Table 3).

Table 3 Secondary outcome measure (DASH) from baseline to 8-weeks
Full size table

Table 4 shows the results of kinematics data. Comparisons between baseline and 8-week follow-up at scapular internal rotations indicated that there were no within-group differences at the sagittal and scapular plane in both experimental groups (Table 4).

Table 4 Secondary outcome measure (scapular kinematics during elevation in the sagittal and scapular planes) from baseline to 8-weeks for
Full size table

Comparisons between baseline and 8-week follow-up at scapular upward rotation and tilt showed that the scapula was more upwardly rotated and posteriorly tilted in SFTF group, whereas there were no differences found in the SFT and control group. Also, there were significant differences between SFTF vs. SFT (sagittal plane, mean difference = 3.4 ± 2.6; ES (95% CI) = 1.0 (0.4–1.5); P = 0.03; or scapular plane, mean difference = 4 ± 0.5; ES (95% CI) = 1.1 (0.5–1.8); P = 0.01), SFTF vs control (sagittal plane, mean difference = 8.1 ± 3.5; ES (95% CI) = 2.3 (1.6–3.1); P = 0.01; scapular plane, mean difference = 7.5 ± 0.3; ES (95% CI) = 2.2 (1.4–2.9); P = 0.01) (Table 4).

Considering the scapular tilt, there was a significant difference between SFTF vs. control (sagittal plane, mean difference = 3.2 ± 0.5; ES (95% CI) = 0.92(0.3–1.5); P = 0.01) (Table 4).

Discussion

The most important findings of the study were the significant differences with a larger effect size between SFTF and SFT in pain, function and scapular upward rotation and tilt.

Consistent with the present research, studies reported adding visual feedback to the scapula-stabilization exercises could decrease pain and disability, and improve scapula range of motion, strength [17] and control of the movement [26] in patients with winging scapula.

It seems improving scapula kinematics (upward rotation and tilt) during SFTF (which emphasized on rotator cuff muscle strength and tight shoulder muscle flexibility) could decrease sub-acromion bursa inflammation and soft tissue impingement. Also, it is well supported that visual feedback could improve the eccentric contraction, control ability and electromyography (EMG) activity of the serratus anterior muscle [17, 31]. Although the present study did not evaluate EMG data, it may be inferred that improving kinematics (scapula upward rotation and tilt) after SFTF could result in improvement scapular muscle activities and reduce sub-acromion space.

Moreover, studies reported adding manual therapy [5] or scapular stabilization exercises [29] to a shoulder stretching and strengthening program could not enhance scapula kinematics in patients with shoulder pain. The ESs in kinematics outcomes of the present study were larger than two mentioned studies outcomes. This can be related to the SFT, emphasizing on isometric stretching, intrinsic and eccentric isotonic exercises, the kinds of the feedbacks and the time of the treatments. Finally, the research results indicated a large ESs in some kinematic outcomes (upward rotation and tilt), especially in the SFTF group. Especially, this difference between two studies in terms of clinical significance of the interventions might be associated with using feedback in this study. Using visual, patients have time to observe the exercises and movements they are performing. On the other hand, the auditory feedback provides patients with an opportunity to correct their movement by the feedback they received from an experienced supervisor.

The findings of this study indicate adding feedback to SFT protocol definitely provides more advantages to relieve pain and improve and shoulder kinematics in individuals with SIS. Adding feedbacks to SFT could clinically enhance the shoulder kinematics resulting in relieving pain and improvement in function.

Conclusion

The study confirms the effectiveness of exercise treatment to control pain and improve function in patients with SIS. Also, the results suggest a rehabilitation program integrated with feedback can be effective to reduce pain, and improve function and kinematic in patients with SIS.