Solving the Enigma of Frozen Shoulder

Abstract

Frozen shoulder is a common shoulder condition affecting 2-5% of the population. It is characterised by the spontaneous onset of pain, stiffness and range of motion (ROM) loss at the shoulder. The exact pathophysiology of frozen shoulder is unclear. However it is commonly believed that a combination of capsular contracture and fibrosis of the rotator cuff interval, the subscapular recess and the coracohumeral ligament lead to global movement restriction of the glenohumeral joint. There is no gold standard clinical test for frozen shoulder. Frozen shoulder is therefore a diagnosis of exclusion and relies on the accurate assessment of active and passive ROM. The generally accepted diagnostic criteria for frozen shoulder are active as well as passive movement restriction in at least two planes of shoulder range of motion, one being external rotation. However, the accuracy of active and passive ROM assessment has not been tested in people with frozen shoulder. Further, the evidence of the effectiveness of treatments for frozen shoulder that aim to stretch the presumed tight shoulder structures has been questioned recently. The overall aim of this thesis was to analyse the effectiveness of stretch-based treatments for frozen shoulder and to investigate if capsular contracture is responsible for movement loss in frozen shoulder. Chapter 2 of this thesis contains a systematic review of stretch based treatments for frozen shoulder. The aim of the review was to analyse the current evidence regarding the effectiveness of interventions that aim to stretch the tissues of the shoulder region or release the presumed capsular fibrosis. The findings of six high quality randomised controlled clinical trials were reported and discussed. The RCTs included in the study evaluated the effectiveness of manipulation under general anaesthetic (MUA), manual therapy, distension and stretching & strengthening exercises on pain, ROM and function in frozen shoulder. Overall, it was found that mobilisation combined with stretching may result in small gains in passive ROM in the short term compared to stretching and strengthening exercises. Physiotherapy after capsular distension consisting of manual therapy and stretching and strengthening exercise provides no additional benefit in terms of pain, function, or quality of life over sham-ultrasound, but may result in improved active ROM in the short term. However, these improvements may not be clinically significant. Distension, regardless of the medium used to distend the glenohumeral capsule, had no benefit with respect to pain, disability or shoulder abduction and flexion ROM over cortisone injection alone in the short term. Distension with hyaluronic acid lead to a small increase in passive external rotation ROM compared to a glenohumeral corticosteroid injection. MUA did not confers any additional benefit over a home exercise program in terms of pain, function and ROM in people with frozen shoulder Chapter 3 contains a cohort study investigating external rotation ROM and stiffness in healthy shoulders. The effects of sex, handedness, shoulder and body position on active and passive ROM and shoulder stiffness were investigated in twenty healthy participants. The results indicate that passive external rotation ROM was significantly greater than active ROM in people with healthy shoulders. Both active and passive shoulder external rotation ROM were greater when the arm was abducted at 90 degrees compared to lower positions of abduction. There was no difference in active or passive external rotation ROM between dominant and non-dominant shoulders. Female subjects demonstrated significantly more passive external rotation ROM than males. Males had greater stiffness into external rotation range than females in supine but not in other positions. Body position only had an effect on stiffness in males. This study also found that measuring external rotation ROM with the arm by the side yields similar results to external rotation ROM measured in side-lying in 45 degrees of abduction. The latter is not commonly utilised in clinical practice but was the position required for external rotation ROM measurement for the study in Chapter 4 as dictated by the participant position in preparation for shoulder surgery. Finally, Chapter 4 contains a case series of five subjects with global restriction of active and passive shoulder movement of greater than 50% of normal ROM in external rotation and at least one other plane of movement. This study demonstrates that capsular contracture is not a major contributor to movement restriction in all patients who exhibit classical clinical features of frozen shoulder. Although all five cases presented with painful, global restriction of passive shoulder movement, four subjects demonstrated significantly greater abduction range of motion (ROM) and three demonstrated significantly greater external rotation ROM under anaesthesia. These findings highlight the need to reconsider the diagnostic process used for frozen shoulder as well as our understanding of the pathology of frozen shoulder and offers an explanation for why treatments aimed at stretching tight passive structures have not proven to be more effective.