Purpose: Frozen shoulder or adhesive capsulitis is a painful and progressive loss of both active and passive range of motion without any known intrinsic cause. The natural history and histological stages have been described to help explain the pathogenesis. There have been conflicting reports evaluating the effects of intra-articular corticosteroid injections in the treatment to improve the natural history. We report our non-operative experience with the use of glenohumeral corticosteroid injections in patients diagnosed with adhesive capsulitis of the shoulder.

Method: 129 consecutive patients with a diagnosis of frozen shoulder were followed from 1997–2002. A detailed physical examination in both the erect and supine position documented range of motion. A VAS was used to document pain. All patients underwent a glenohumeral injection with 40 mg DepoMedrol and 9 ml 1% plain lidocaine at the time of initial presentation and at monthly follow-up with the following criteria: 1. No improvement in pain of 2 VAS levels 2. No improvement in erect abduction or forward flexion of 20° or 3. No improvement in erect or supine IR or ER of 10°. A maximum of 3 injections was used. Patients were followed until complete resolution of symptoms or if surgical intervention was needed. Successful treatment was considered if there was complete resolution of pain, full function, and patient satisfaction. Initial and follow-up ASES and HSS L’Insalata scores were recorded.

Results: Thirty-one patients were lost to follow-up leaving 98 patients to be evaluated. There were 69 females with average age of 40.7 years and 29 males with average age of 53.2 years. Overall success was 71.4% (71% females, 72.4% males). Successful treatment occurred at 4.15 months in females and 4.5 months in males. 85.7% of both female and male patients recovered with 1 or 2 injections. Poor prognostic indicators were Diabetes Mellitus, absent physiotherapy, workman’s compensation, post-operative stiffness cases, dominant arm, and stage 3 cases. Average ASES scores were 41.8 at presentation and 92.7 at resolution and HSS L’Insalata scores were 52.5 at presentation and 91.0 at resolution. There were no complications with our technique.

Conclusion: Glenohumeral corticosteroid injections for the patient with adhesive capsulitis are considered to be safe and an effective method of treatment for resolution of pain and improvement in functional range of motion. We recommend glenohumeral corticosteroid injections at the time of presentation and with close follow-up for frozen shoulder as part of the initial treatment regime. We have suggested an algorithm for the timing of intra-articular injections based on pain and objective range of motion.