Skip navigation
Acute Pain | Chronic Pain | General

Corticosteroid injections for lateral epicondylitis (tennis elbow)

Clinical bottom line: Corticosteroid injections are effective in treating lateral epicondylitis in the short term (2-6 weeks), but there appears to be no longer term benefit. It is unclear whether this form of treatment has benefits over other treatments. Review findings are based on hospitalised patients, and are therefore not generalisable to general practice. Clear recommendations on treatment regime could not be formulated.

Corticosteroid injections are commonly used to alleviate pain in lateral epicondylitis. This condition has an incidence of 4 to 7 per 1000 per year in general practice, with the average episode lasting between 6 months and 2 years. Dutch data suggest that 10 to 30% of episodes result in an average of 12 weeks absence from work. Other treatments include analgesic medication, physical therapy and, most radical, surgery.

Systematic review

Assendelft WJJ, Hay EM, Adshead R, Bouter LM. Corticosteroid injections for lateral epicondylitis: a systematic overview. British Journal of General Practice. 1996; 46: 209-16.

Inclusion criteria were randomised controlled trials of corticosteroid injection for lateral epicondylitis. The review aimed to answer questions on optimal treatment regimen and suitability for general practice as well as efficacy.

Reviewers used conclusions on efficacy from the original reports, and also extracted data on success rates at follow-up points, with preference given to patient ratings.


Injection versus placebo

Five trials compared injection plus local anaesthetic with either local anaesthetic alone or with saline. Four of five trials reported significant benefit of injection over placebo.

Injection versus active

One high quality trial compared injection with Cyriax physiotherapy and found significant benefit with injection. Two trials compared injection with naproxen. Naproxen alone was no different to injection, and naproxen with wrist brace was better than injection (although quality of trial lower).

One trial compared injection with elbow band and wrist brace, and found significant benefit with injection. One trial compared injection with ultrasound, phonophoresis and transcutaneous electrical nerve stimulation, and found no differences. One trial compared two injection techniques and found them to be equally effective.

For all included trials, there was a tendency for higher quality trials to show a larger effect, with a pooled odds ratio of 0.15 (0.10 to 0.23) at 2 to 6 weeks, indicating a beneficial effect of corticosteroids.

Longer term efficacy

For all included trials there was a general trend that injections were effective in the short term (2-6 weeks), but not after a follow-up of greater than 6 weeks. The odds ratio for >6 weeks was 0.73 (0.37 to 1.44) indicating a lack of long term effect.

Adverse effects

Six trials reported on adverse effects. Three of six stated there were no adverse effects. Three trials reported post-injection pain lasting a few days, but this was not always attributed to corticosteroids. However, the most detailed trial reported that 50% (58/116) of patients injected with corticosteroids and local anaesthetic experienced post-injection pain compared with 31% (9/29) with local anaesthetic alone. Skin atrophy was reported in 27% (31/116) patients compared with 17% (5/29) with local anaesthetic alone.

Related topics