Question:
In patients with the subacromial impingement syndrome, is the combination of corticosteroid injection plus exercise and manual therapy more effective than exercise and manual therapy alone?
Design:
Randomized (allocation concealed), unblinded, controlled trial with a 12-week follow-up.
Setting:
A primary care-based musculoskeletal and rehabilitation service in Leeds, UK.
Patients:
232 patients who were ≥40 years of age (mean age, 56 years; 55% women) had unilateral shoulder pain subjectively rated as moderate or severe and had a noncapsular pattern of restriction. Other inclusion criteria were ≤25% loss of lateral rotation compared with the rotation in the opposite shoulder and presence of a Neer impingement sign or positive results on the Hawkins impingement test. Exclusion criteria included blood coagulation disorders, referred pain from the cervical spine or internal organs, history of arthritic diseases, bilateral shoulder pain, neurological diagnosis, contraindication to corticosteroid injections, previous shoulder injury or surgery, and treatment to the symptomatic shoulder in the previous 6 months. 205 patients (88%) were available for 12-week follow-up.
Intervention:
Patients were allocated to a combined treatment strategy of subacromial corticosteroid injection (20 mg triamcinolone acetonide mixed with 4.5 mL of 1% lidocaine) plus exercise and manual therapy (n = 115) or exercise and manual therapy alone (n = 117). The exercise program was delivered by a physiotherapist and involved manual mobilization techniques and exercises individualized to each patient. Patients received manual therapy at least once during treatment.
Main outcome measures:
The primary outcome was the 12-week score on the Shoulder Pain and Disability Index (SPADI) (range, 0 to 100 for each of two subscales for pain and disability and a total score combining the means of each subscale; higher scores indicate greater pain and disability). The secondary outcome was a global assessment of change from baseline (5-point scale, ranging from much worse to completely recovered).
Main results:
Analysis was by intention to treat. The study had 80% power to detect a difference between groups of ≥10 points on the SPADI. Patients in the injection-plus-exercise group showed greater improvement than those in the exercise-alone group at 1 and 6 weeks (p < 0.001) (Table). At 12 weeks, there was no difference between groups in change in SPADI scores (Table). The data from the global assessment of change from baseline showed greater improvement from baseline in the injection-plus-exercise group at 1 week (p < 0.001) and 6 weeks (p = 0.006), but the groups were not different at 12 weeks (p = 0.248).
Conclusion:
In patients with the subacromial impingement syndrome, the combination of corticosteroid injection plus exercise and manual therapy was not more effective than exercise and manual therapy alone at 12 weeks.
Nuffield Department of Orthopaedics, University of Oxford, Oxford, United Kingdom
The study by Crawshaw and colleagues aims to investigate if the use of exercise therapy in the “window” of opportunity after a corticosteroid injection results in better outcomes for patients with subacromial impingement syndrome compared with exercise therapy alone.
Shoulder pain is common, particularly in individuals who are >40 years of age, and is often persistent, with half of those presenting in primary care still having symptoms 18 months later. Shoulder pain is not a diagnosis, and many previous studies have failed to discriminate between tendinopathy (impingement syndrome) and capsulitis or arthritis. This study includes clinical examination by experienced physiotherapists to pragmatically screen patients and thereby investigates a substantially more homogeneous population than those of other studies.
The finding that steroids produced significant short-term benefit in terms of pain and restoration of function is consistent with other studies. Whether or not steroid injection plus exercise therapy was any more or less effective than steroid injection alone was not addressed in this study. Nor was there a comparison with no treatment or observation. Similar to other injection studies, no short-term harmful effects or complications were reported (e.g., tendon rupture). However, we cannot determine any long-term deleterious effects of steroids with a study end point of only 12 weeks.
One-third of patients treated with exercise alone subsequently opted for steroid injection. More importantly, 14% to 19% of patients in both groups were no better or worse and an additional 72% were improved but still having problems at 24 weeks. No imaging of the rotator cuff was undertaken, and we do not know if the presence of a rotator cuff tear influences outcome.
In conclusion, surgeons should continue to use steroid injections for patients in whom a short-term improvement is desired. However, this study highlights the need for a better understanding of the disease mechanisms underlying rotator cuff tendinopathy and the discovery of more effective treatment strategies.