Background: Frozen shoulder is a debilitating disease characterized by pain and progressive loss of motion in shoulder secondary to fibrotic inflammation of the joint capsule. The incidence is 2% in the general population, mostly affecting women 40 to 70 years of age. Therapeutic measures include physiotherapy, shoulder manipulation, corticosteroid injection and arthroscopic release. The purpose of this study was to compare the results of manipulation under anesthesia with or without concomitant intra-articular injection of corticosteroid.

Methods: Twenty-six patients (21 females, 5 males) with the clinical diagnosis of idiopathic frozen shoulder were evaluated in a randomized clinical trial. The average age of the patients was 55.7 years. Twelve cases had diabetes mellitus, which is considered a strong predisposing factor and has a poor prognosis. Thirteen patients received intra-articular corticosteroid injection just before manipulation. The outcome was assessed by evaluating functional score (Constant scoring system) at the six-month follow-up.

Results: The average score before manipulation was 28.7 out of 100, which significantly improved to 84.8 at six months after the procedure (p<0.05). The results were slightly better in the group who received corticosteroid injection, but the difference was not statistically significant (p=0.1). The outcome in the diabetic group was less favorable than in the non-diabetic patients (p<0.05).

Conclusions: Manipulation under anesthesia is a very effective method of treatment for idiopathic frozen shoulder. Concomitant intra-articular injection of corticosteroid does not seem to improve significantly the final outcome. Diabetes mellitus can negatively affect the results of this treatment method.