Examining the role of thoracic kyphosis in shoulder pain

Introduction: Shoulder pain is a common musculoskeletal condition that carries a substantial economic burden. Due to the challenges acknowledged with structural diagnoses, studies have labelled participants as having non-specific shoulder pain. Non-specific shoulder pain may be caused by several factors, including physical, psychological, lifestyle and cognitive factors. An increase in thoracic kyphosis is a physical factor which has been hypothesised to contribute to shoulder pain. While exercises to reduce thoracic kyphosis are a common component of the physiotherapy treatment of shoulder pain, the relationship between thoracic kyphosis and shoulder pain is unclear. The aim of this doctoral thesis was to examine the role of thoracic kyphosis in shoulder pain. Methods: This doctoral thesis presents six studies. A systematic review examines the association between thoracic kyphosis and shoulder pain, function and range of motion (Chapter 2). Three methodological studies (systematic review, reliability study, validity study) investigate the reliability and validity of methods for measuring thoracic kyphosis clinically (Chapter 3). Chapter 4 presents two clinical studies. A case series evaluates pain, disability and thoracic kyphosis in two groups of people with non-specific shoulder pain who receive two different types of shoulder exercise classes. A separate qualitative study of people with non-specific shoulder pain examines individual experiences of shoulder exercise classes. Results: A systematic review (Study I) concluded that thoracic kyphosis was not significantly different in people with and without shoulder pain, suggesting that increased static thoracic kyphosis is not a consistent postural deviation in people with shoulder pain. The cross-sectional nature of these studies prevented analysis of a causal relationship between thoracic kyphosis and shoulder pain. A second systematic review (Study II) synthesised the evidence regarding the reliability and validity of non-radiographic thoracic kyphosis measurement methods. A reliability study (Study III) found that the Flexicurve and manual inclinometer demonstrated excellent intra- and inter-rater reliability for thoracic kyphosis measurement. In a validation study (Study IV), the manual inclinometer demonstrated good concurrent validity with the gold standard radiographic Cobb angle, in contrast to the Flexicurve angle which demonstrated poor validity. In Study V, people with non-specific shoulder pain who completed a six week group exercise class demonstrated significant and clinically meaningful improvements in shoulder pain and disability at six week and six month follow-up, without a change in thoracic kyphosis beyond measurement error. The separate qualitative study (Study VI) revealed that shoulder exercise classes provided an environment conducive to peer-learning, support and motivation, facilitation towards independent exercise and highlighted beliefs regarding pain and exercise. Conclusion: Thoracic kyphosis can be measured with validity and good reliability using the manual inclinometer. However, static thoracic kyphosis, measured in relaxed standing, may not be strongly related to shoulder pain. This thesis provided preliminary quantitative and qualitative evidence to support group exercise classes for people with non-specific shoulder pain. However, a change in thoracic kyphosis was not the mechanism of clinical improvements. Future research should (i) compare the effectiveness of shoulder exercise classes, with and without thoracic extension exercises, for the treatment of non-specific shoulder pain, (ii) examine the effectiveness of group exercise classes compared to individual physiotherapy for this population and (iii) explore the mechanisms which underlie the improvements after exercise-based rehabilitation in people with non-specific shoulder pain.