To date no trials have been performed looking at whether or not intra-bursal injection of an IL-1 antagonist provides pain relief similar to that of a corticosteroid injection. The subcutaneous injection of anakinra, an IL-1 receptor antagonist, in patients with shoulder pain due to rotator cuff tendonitis and subacromial bursitis was efficacious in relieving pain but this information was presented as a case series in a letter to the editor format, so the validity of these results would require additional testing [Omoigui S, et al. 2004]. Based mainly on the data from the intra-articular administration of anakinra, there have not been any adverse trends in outcomes or safety to suggest that intra-bursal injection of rilonacept will carry an increase risk of adverse events. The purpose of this trial is to compare the improvement in pain and function of patients with clinical symptoms and signs of subacromial bursitis of rilonacept vs. corticosteroid injection (standard of care).

Background: Subacromial bursitis is an inflammatory condition which is typically triggered by altered rotator cuff mechanics. One of the mainstays of therapy is corticosteroid injection. Given the inflammatory nature of subacromial bursitis coupled with prominence of interleukin-1 (IL-1) beta on histopathologic assessment of resected subacromial bursa, targeted anti-IL-1 therapy would be an attractive alternative to corticosteroid injection.

Hypothesis: One intra-bursal injection of the IL-1 trap rilonacept (Arcalyst ®), currently FDA approved for the treatment of cryopyrin associated periodic syndrome (CAPS), is non-inferior to intra-bursal corticosteroid injection at 4 weeks post-procedure.

Methods: This study is a prospective trial lasting a total of 4 weeks. Patients with typical symptoms and signs of subacromial bursitis are randomized to either corticosteroid injection prepared in the usual fashion in clinic vs. injection of rilonacept. Prior to injection of either medication patients will complete a QuickDASH Questionnaire and provide a verbal pain score from 0 to 10. The Quick DASH Questionnaire and verbal pain score are then completed within 2 duty days of injection, 2 weeks after injection, and 4 weeks after injection by phone.

Primary Outcome: Primary outcome is improvement in QuickDASH. Secondary outcomes are improvement in the verbal pain score and monitoring for infection or other complications from rilonacept administration.

Inclusion Criteria:

• At least 18 years of age or older and at a minimum have a recent history (more than 3 days worth) of shoulder pain with moderate to severe tenderness to palpation over the subacromial bursa.

Exclusion Criteria:

• Allergies to lidocaine, marcaine, or kenalog.
• Allergies to rilonacept
• Flare of active inflammatory arthritis (such as a flare of Rheumatoid Arthritis)
• Gout or Pseudogout attack of the shoulder with subacromial tenderness
• Active infection
• Actively receiving chemotherapy, radiation therapy, or anticipating surgery for neoplasia
• Active myocardial infarction
• Clinical and/or radiographic evidence of a fracture (clavicular, humeral, or other).

Prior to starting the trial, we anticipated having to recruit 150 subjects total (75 in each arm). At our first analysis of the data, the effect size of both the rilonacept and triamcinolone were larger than expected, and a re-calculated 'n' was 29 for an alpha of 0.05 and power of 1.0000 (effect size 1.1).

1 year