Shoulder impingement: the effect of sitting posture on shoulder pain and range of motion

Abstract

The re-education of spinal posture is an integral part of shoulder impingement management yet supporting evidence is limited. The purpose of this study was to evaluate the effect of slouched versus erect sitting posture on shoulder pain intensity and range of motion (ROM) in subjects with impingement.

A same-subject repeated-measures design was utilized. Maximum active shoulder flexion and associated pain intensity were measured in 28 subjects in slouched and erect sitting postures, using video-analysis and visual analogue scales, respectively. An intra-tester reliability study of the video-analysis system was completed and intra-class correlation coefficients calculated. Shoulder flexion differences between slouched and erect sitting posture were analysed using a repeated-measures analysis of variance (ANOVA).

The intra-tester reliability of the video-analysis method was found to be ‘excellent’ (ICC =0.99). Flexion ROM was significantly greater in the erect sitting posture (F=100.3, P<0.0001); the mean ROM difference between postures was 17.67° (±9.17°). There was no significant difference in pain intensity between postures (F=1.9, P=0.179).

An erect sitting posture appeared to increase active shoulder flexion in subjects with shoulder impingement, although there were no differences in reported pain intensity. Further research is required to investigate the long-term effects of postural re-education.

Introduction

Epidemiological studies of Western society suggest that the prevalence of shoulder disorders in the general population is high, ranging from 6% (van der Windt et al., 2000), to 14% (Miranda et al., 2001). In 1994, neck and shoulder complaints represented 18% of the total paid sick leave for musculoskeletal disorders in Sweden (Nygren et al., 1995). Studies have encountered methodological difficulties due to the poor agreement in syndrome classification and diagnosis. Arguably, the diagnosis of shoulder dysfunction has been simplified by the development of recognized clinical syndromes such as adhesive capsulitis, instability and shoulder impingement (Fu et al., 1991; Tibone and Shaffer, 1995). The term ‘shoulder impingement’ was originally introduced by Neer (1972) who described the mechanical wear of the long head of biceps and supraspinatus by the acromion during overhead activities. The understanding of the aetiology and location of shoulder impingement has been substantially improved over the last decade (Lewis et al., 2001), yet the optimal method of management remains unclear.

Physiotherapy approaches may include joint mobilization (Conroy and Hayes, 1998), deep friction massage (Cyriax, 1993), taping (Host, 1995), re-education of the rotator cuff (Thein and Greenfield, 1997) and scapular stabilizing muscles (Schmitt and Snyder-Mackler, 1999). Postural re-education of the cervical and thoracic spines is a frequently cited treatment recommendation (Keirns, 1994; Turner, 1996; Ayub, 1997). There are many theories underpinning this approach. Solem-Bertoft et al. (1993) suggested that the increased thoracic curvature accompanying a slouched posture may influence scapular kinematics and cause a reduction in the sub-acromial space. An exaggerated thoracic kyphosis has been suggested to adversely influence length–tension relationships of the shoulder girdle muscles (Grimsby and Gray, 1997) which in turn may cause mal-tracking of the humeral head within the glenoid fossa (Wilk and Arrigo, 1993). Postural correction may restore normal movement patterns and ensure that the dynamic sub-acromial space is maximized (Solem-Bertoft et al., 1993). However, evidence to support these anecdotal theories is sparse.

The correlation between slouched spinal posture and shoulder pain has been explored by comparing the posture of healthy subjects with those who have shoulder pain (Griegel-Morris et al., 1992; Greenfield et al., 1995). Greenfield et al. (1995) compared the posture of 30 subjects with ‘shoulder overuse injuries’ with matched healthy controls. No significant differences were found between groups in relation to thoracic curvature or scapular orientation. The symptomatic group was found to have a significantly greater forward head posture. However, analysis of the data revealed that the mean difference in cervical posture was only 5° whereas the measurement accuracy of photography as a tool for measuring ‘craniovertebral angle’ was found to be between 3° and 6.5° (Braun and Amondsen, 1989). Postural measurement was undertaken with the shoulder in a neutral position. Griegel-Morris et al. (1992) concluded that ‘rounded shoulders’, ‘severe kyphosis’ and ‘forward head posture’ correlated with interscapula pain but not shoulder or upper arm pain as that expected of an impingement syndrome (Conray and Hayes, 1998). Conclusions may be limited as all subjects were described as a ‘healthy’ population and the pain report of such subjects may be ambiguous.

The influence of spinal posture on shoulder range of motion (ROM) and biomechanics has been evaluated using same-subject designs (Ludewig and Cook, 1996; Kebaetse et al., 1999). Ludewig and Cook (1996) evaluated the effect of cervical position on scapula orientation on 25 healthy subjects. Results suggested that increased cervical flexion prevented upward rotation and posterior tilt. The researchers postulated that cervical flexion generated tension in levator scapulae which impeded optimal scapular kinematics. Impingement subjects have been shown to exhibit reduced scapular posterior tilt during shoulder elevation compared to the normal population (Lukasiewicz et al., 1999). Kebaetse et al. (1999) explored the effects of sitting posture on shoulder elevation in the scapular plane in 34 asymptomatic subjects. Significantly less ROM was noted in a slouched posture as compared to an erect posture. The mean difference was 23.6° and, at 90° and maximum elevation, subjects were found to have significantly less posterior scapular tilt.

There appears to be no studies that have evaluated the effect of spinal postural correction on shoulder ROM and pain in a patient population.

The current study aimed to compare the effect of slouched versus erect sitting posture on shoulder flexion ROM and pain in a population of subjects with impingement syndrome. It was hypothesized that shoulder flexion ROM would be greater and pain intensity associated with shoulder movement would be less in an erect sitting posture, than in a slouched posture.