What is the evidence for corticosteroid injections in the treatment of rotator cuff pathology?
AMSTAR Rating
Summary
Patient Population:
Patients ranged in age from 50 – 69 years. Women were in the majority in 5 or 9 studies identified. Reported duration of symptoms ranged from 1 week to 3.8 years.
Intervention:
Agents injected: betamethasone, methylprednisone, triamcinolone, lidocaine, and Naproxen. Methylprednisolone and triamcinolone were used in 8/9 trials. Approaches used included lateral, anterior, posterior, posterolateral and anterolateral.
Comparison:
9 trials were identified. Two studies used specific imaging techniques to assess rotator cuff tendons. The remainder did not, however, the authors suggest that sufficient criteria were established to include individuals with rotator cuff disease, while excluding those with other conditions such as adhesive capsulitis, rheumatoid arthritis, osteoarthritis, and various other systemic illnesses. The studies identified were small, of relatively poor quality (subject to bias) and heterogeneous. No pooled analysis could be conducted given study heterogeneity.
Outcome:
- Pain: 4 studies demonstrated significant improvement in pain associated with corticosteroid injection when compared with the control condition. However, in only one study was the improvement greater than 20mm (assessed on the VAS scale).
- ROM: 3 studies demonstrated significant improvement in ROM.
- Function: Results varied. While some primary studies reported no significant between group differences in terms of function, others reported significant gains associated with sub-acromial injection.
- Overall: The authors conclude that there is relatively little evidence to support the use of sub-acromial corticosteroid injection in the treatment of rotator cuff disease.
Guideline Recommendations
| Source | Recommendation |
|---|---|
| AAOS 2010 | Inconclusive |
| ACOEM 2011 | Level B |
Outcomes Assessed
- Benefit
- Harm
- Inconclusive
Evidence supports improvement
Pain (2 point on 10 point scale)
ROM (3 degrees)
Evidence inconclusive for
Function
