Objective: To investigate the effectiveness of surgical and postsurgical interventions for the subacromial impingement syndrome (SIS).
Data sources: Using terms including randomized controlled trial (RCT), shoulder impingement syndrome, rotator cuff impingement, and interventions, the Cochrane Library, PubMed, Embase, PEDro, and CINAHL were searched up to February 2009.
Study selection: Relevant systematic reviews and RCTs in any language that included patients with SIS that was not caused by systemic disease or acute trauma; that evaluated surgical or postsurgical treatment for SIS; that reported outcomes of pain, function, or recovery; and that had a follow-up of ≥ 2 weeks, were selected by 2 independent reviewers and confirmed by consensus.
Data extraction: Data on the study population, interventions, outcome measures, and length of follow-up (<3 months, 4 to 6 months, and >6 months) were extracted. Two reviewers assessed the methods of the RCTs that were included in reviews other than Cochrane reviews and in RCTs published more recently, rating RCTs that met 6 of 12 Furlan criteria as high quality. The effectiveness of the intervention was rated as moderate, limited, or no evidence, depending on the consistency and quality of the evidence.
Main results: The included studies were 11 RCTs from a Cochrane review, an additional 2 RCTs that reported on the effectiveness of surgery for SIS, and 3 further RCTs that reported on postsurgical interventions. The quality of the studies was generally poor (10/11 RCTs in the Cochrane review had a high risk of bias; and 3 of the additional RCTs were rated moderate to low quality). Arthroscopic or open subacromial decompression compared with physiotherapy or exercise was investigated in 3 trials. No evidence for differences between the treatments in pain, function, or time to recovery was found in the short, medium, or long term. Five comparisons of arthroscopic versus open subacromial decompression showed no differences in pain or shoulder function scores at any follow-up assessment. One study found no difference in short- or long-term outcomes between arthroscopic versus open subacromial decompression for the removal of calcium deposits. A comparison of the Neer versus the modified Neer surgical technique found no verifiable difference between them in the short term. In a study comparing the techniques of electrocautery and holmium laser in arthroscopic subacromial decompression, differences favoring electrocautery were found for the short and medium term on one American shoulder rating scale but not on another. A high-quality study found no differences at any follow-up in pain or shoulder function scores when arthroscopic subacromial decompression was compared with radio-frequency-based plasma microtenotomy. A high-quality RCT compared injections of platelet-leukocyte gel given postoperatively with no injection. The additional intervention was effective at 6 weeks in reducing pain and improving shoulder function and range of motion, and in return to activities of daily living, but no differences were found in shoulder stability. Three low-quality studies compared the effectiveness of various postsurgical interventions. Early progressive exercises (range of motion, and strengthening exercises) resulted in greater improvements in range of motion at 3 and 12 months than later dynamic and strengthening exercise. Reductions in pain were similar for both groups. Patients taking ketoprofen (200 mg once daily for 6 weeks) compared with those taking placebo had less pain and better shoulder scale scores, active forward flexion, and active abduction, at 6 weeks but the groups did not differ after 2 years. The use of a pain pump (0.375% ropivacaine) after arthroscopic surgery was not effective in improving shoulder function scores at the 2-year follow-up.
Conclusions: Studies of surgical interventions in subacromial impingement syndrome suggested that no technique is convincingly better than another or than conservative interventions. Evidence from the few better quality studies suggested that injections of platelet-leukocyte gel given postoperatively were effective in the short term for lessening pain and improving shoulder function, and that radio-frequency-based plasma microtenotomy was not more effective than arthroscopic subacromial decompression for improving any outcome measure.