The objective is to determine whether corticosteroid injection into the subacromial space was not inferior to intra-articular injection in patients with idiopathic frozen shoulder (FS), and whether combined injections had an additive effect.

Patients with idiopathic FS (n=126) were randomly assigned to receive ultrasound-guided intra-articular (IA group), subacromial (SA group), or combined IA and subacromial injections (IA+SA group). All groups received a total dose of 40mg triamcinolone acetonide. The outcome measures included the visual analog scale (VAS) for pain, American Shoulder and Elbow Surgeon (ASES) shoulder score, subjective shoulder value (SSV), and passive range of motion before and at 3, 6, and 12 weeks after treatment.

There was significant effect of time on all measurements such that all measures improved in all groups (P<0.001) during the 12 weeks after treatment. Group-by-time interactions were significant for ASES (P=0.006), VAS (P<0.001), SSV (P=0.03), and internal rotation (P=0.014). Between-group comparisons revealed a significant improvement in the IA (P<0.001) and IA+SA (P<0.001) groups as compared to the SA group. The IA+SA group demonstrated significant improvement in internal rotation as compared to the IA group (P=0.046).

The efficacy of corticosteroid injection into the SA space in idiopathic FS was inferior to IA injection up to 12 weeks; however, combination injections had an additive effect on increasing the internal rotation angle. These results indicate that although the glenohumeral joint is a major site in the pathogenesis of idiopathic FS, the SA space may be a contributing site.