#424 Shoulder the Evidence: Steroid injections for shoulder pain
CLINICAL QUESTION
What are the effects of steroid injections for rotator cuff related pain and is ultrasound guidance needed?
BOTTOM LINE
For rotator cuff related pain, a single steroid injection provides short-term, but not long-term pain relief (example at 6 weeks, steroids/anesthetic reduced pain from 6 at baseline to 3.2 versus 4.4 for anesthetic alone, no difference at 12 weeks). Ultrasound guidance is at least as good as landmarking for injections, and may be superior based on small, unblinded randomized, controlled trials (RCTs); however, the largest and blinded RCTs showed no difference.
Reading Tools for Practice Article can earn you MainPro+ Credits
Join NowAlready a CFPCLearn Member? Log in
EVIDENCE
- Results statistically significant unless indicated.
- Focus on largest RCTs with responder analyses. Single, subacromial injections with 40mg methylprednisolone/triamcinolone and lidocaine/bupivacaine unless indicated.
- Steroid versus no injection:
- RCT, 708 participants: “Impression of treatment” (0-10, higher=better): 7.3 (no injection) versus 8 (steroid-anesthetic) at 8 weeks, no difference 6-12 months.1
- Steroid anesthetic versus anesthetic injection:
- RCT, 108 participants, 20mg triamcinolone/lidocaine. 2
- Pain (scale 0-10, 0=no pain), baseline ~6, changed to 4.4 (anesthetic alone) versus 3.2 (steroid-anesthetic) at 6 weeks, no difference at 12 weeks.2
- Systematic reviews: Similar. 3,4
- RCT, 108 participants, 20mg triamcinolone/lidocaine. 2
- Ultrasound guidance versus landmarking for injections:
- Five RCTs. “Response” =Proportion of participants with ≥50% improvement in pain or completely recovered/much better at 6 weeks.5-9
- 256 participants, unblinded, response: 38% (ultrasound) versus 30%, not statistically different.5
- 56 participants, blinded, response: 50% (ultrasound) versus 46%, not statistically different. 6
- 60 participants, unblinded, response: 83% (ultrasound) versus 23%, number needed to treat (NNT)=2. Function, range of motion: better with ultrasound.7
- 67 participants, unblinded, betamethasone-based proprietary product, response: 64% (ultrasound) versus 38%, NNT=4. At 12 weeks: No difference.8
- Additional RCT abnormally low response rate (landmark: 5%).9
- Systematic review using later time points (example: 12 weeks): No difference.10
- Five RCTs. “Response” =Proportion of participants with ≥50% improvement in pain or completely recovered/much better at 6 weeks.5-9
- Serious adverse events: None/not reported.1,10
- Limitations: Diverse pathologies and co-interventions (examples: exercise/leaflets/nothing), suspected publication bias.4
CONTEXT
- Physiotherapy-supervised exercise (~6 sessions over ~16 weeks) likely superior to exercise leaflet,5 and similar to one physiotherapy session followed by physiotherapist-recommended exercises at home.1
- Guidelines: Single corticosteroids/anesthetic injection can be considered for short-term improvement of pain/function.11
- Risk of joint infection <0.01%.12
Latest Tools for Practice
#424 Shoulder the Evidence: Steroid injections for shoulder pain
What are the effects of steroid injections for rotator cuff related pain and is ultrasound guidance needed?
#423 Estrogen – an answer to a burning question?
Does estrogen reduce urinary tract infections (UTIs) in post-menopausal women?
#422 What’s the Scoop on Creatine? Effects on strength
How does creatine supplementation affect strength in adults?
For short term relief