The aim of this study is to determine if there is any difference in terms of pain relief in patients with frozen shoulder, when given a steroid injection as a suprascapular nerve block compared to a glenohumeral joint intra-articular injection.

The investigators hypothesis is that there is a significant difference in terms of pain relief at 3 months, 6 months and 1 year between a steroid injection as a suprascapular nerve block compared to a glenohumeral joint intra-articular injection.

Participants who attend the orthopaedic clinic and are diagnosed with unilateral frozen shoulder and who are suitable to be treated with a steroid injection will be identified. If the participant consents to having a steroid injection, he/she will then be invited to participate in the trial where they will have an injection either into the glenohumeral joint (ball and socket joint of the shoulder) or as a suprascapular nerve block (injection adjacent to a nerve over the shoulder blade) under ultrasound guidance. Participants will then be followed up at 3, 6 and 12 months after their injection and asked to complete a set of questionnaires assessing pain, function and movement. The results of the groups will then be compared to see if one treatment is superior to the other.

This is a single centre, parallel, two-arm, randomized clinical trial. The trial is expected to take a total of 3 years with 12-18 months for patient recruitment, 12 months for follow up and 6-12 months for data collection, analysis and publication.

The null hypothesis is that there will be no differences in terms of pain relief at 3 months, 6 months and 1 year between a steroid injection as a suprascapular nerve block compared to a glenohumeral joint intra-articular injection.

Adult patients diagnosed with unilateral frozen shoulder who are suitable to be treated with a steroid injection will receive a verbal explanation of the study by a suitably qualified member of the research team alongside an information leaflet. Eligible, consented participants will be randomised in a 1:1 ratio by the selection of sealed, opaque envelopes with one of the two injection types documented inside. Participants will be free to withdraw from the study at any time without prejudice. The injection types include an ultrasound guided steroid injection into the glenohumeral joint (10mls of 1% lidocaine with 40mg depo-medrone) or an ultrasound guided steroid injection as a suprascapular nerve block at the spinoglenoid notch adjacent to the nerve as it traverses under the spinoglenoid ligament (10mls of 1% lidocaine with 40mg depo-medrone. Following both injections, participants are taken through a protocol of physiotherapy for their frozen shoulder over the subsequent 3 months. Participants will be seen back in the outpatient clinic 3 months after their injection to review their progress and collect follow-up data. Further follow-up data will be collected by the completion of questionnaires via post at 6 months and 12 months after the injection. The primary outcome measure of this study is the Oxford Shoulder Score (OSS) at 3 months post-injection. Secondary end points include Oxford Shoulder Score at 6 months and 1 year, pain using the Pain Numeric Rating Scale at 3 months, 6 months and 1 year, Quick Disabilities of Arm, Shoulder and Hand (QuickDASH) questionnaire at 3 months, 6 months and 1 year and EQ-5D-5L at 3 months, 6 months and 1 year.

Baseline demographic and clinical variables will be reported using summary statistics. In terms of the primary outcome, the change in total Oxford Shoulder Score from baseline to 3 months post-injection will be compared between the two groups using either the Independent Samples t-Test or Mann-Whitney U Test, with the final choice depending on an exploration of the data. The same approach will be adopted for secondary outcomes. All analysis will be performed according to the intention to treat principle.

Inclusion Criteria:

• Age greater than or equal to 18 years
• Patient diagnosed with unilateral frozen shoulder (Diagnosis of frozen shoulder will be based on clinical examination (restriction of passive external rotation in the affected shoulder > 50% compared to opposite site), in the presence of normal plain radiographs (antero-posterior and axillary projections) (other than calcific tendinopathy))

Exclusion Criteria:

• Age less than 18 years
• Lacking capacity/unable to give valid consent for participation
• Full thickness rotator cuff tear diagnosed on either Ultrasound scan or Magnetic Resonance Imaging
• Unable to complete follow up
• Unable to speak or read English
• Allergy to corticosteroid or local anaesthetic
• Simultaneous bilateral frozen shoulder.

Data will be published in peer review journal on completion of the study.