OBJECTIVES: To examine in randomized controlled trial the effectiveness and cost-effectiveness of arthroscopic acromioplasty in the treatment of stage II shoulder impingement syndrome. METHODS: We divided 140 patients into supervised exercise program (n = 70, exercise group) and arthroscopic acromioplasty, followed by a similar exercise program (n = 70, combined treatment group). The primary health outcome measure was self-reported pain on a 0-10 Visual Analogue Scale at 24 months with a two-point change defined as minimal clinically important difference (MCID). RESULTS: Results In an intention-to-treat analysis an improvement exceeding MCID took place from baseline to 24 months in both groups: self-reported pain diminished from 6.5 to 2.9 in the exercise group (N = 66) and from 6.4 to 2.5 in the combined treatment group (n = 68) (P <0.001 in both). In the combined treatment group pain relief was attained faster, but the groups did not any more differ at 24 months (P = 0.37). A similar pattern was seen in the secondary outcome measures: disability, pain at night, SDQ score, ability to work, number of painful days and proportion of pain-free patients. The mean total cost was €2961 in the combined treatment group and €1864 in the exercise group. The incremental cost-effectiveness ratio was €5852 per MCID unit, i.e., combined treatment was considerably more costly. CONCLUSION: Arthroscopic acromioplasty does not provide any significant additional value over structured and supervised exercise program alone in terms of subjective outcome or cost-effectiveness. Operative treatment should be offered judiciously.