Medical practice

Steroids for tennis elbow ‘not bad’

“Tennis elbow injections do ‘more harm than good’”, reports BBC News. The article was referring to steroid injections that are commonly used to treat the pain and inflammation associated with lateral epicondylitis (‘tennis elbow’).

The story is based on a recently published, well-conducted systematic review that searched the medical literature for studies assessing whether injections into the tissue surrounding a damaged tendon (including Achilles, elbow, shoulder and knee) improved pain and function. Many of these studies assessed the use of corticosteroids to treat problems associated with tennis elbow. Overall, the review found that, compared to no intervention, corticosteroids reduced pain associated with tennis elbow in the short term, but that the benefits did not last. There is no particular evidence that these injections ‘do more harm than good’. In fact, only one small study with imprecise results found a small increase in the risk of loss of skin pigmentation compared with no treatment. The BBC headline is therefore an inaccurate extrapolation of the results.

There is not enough evidence yet to determine the best injectable treatment for tennis elbow and other tendinopathies. The short-term relief provided by corticosteroids may still be appropriate for some sufferers.

Where did the story come from?

The study was carried out by researchers from the University of Queensland, Griffiths University and the Royal Brisbane and Women’s Hospital in Australia. The work received no external funding and was published in the peer-reviewed medical journal The Lancet.

BBC News has covered this review fairly well, though the headline is an unsupported extrapolation of the results and may mislead readers into thinking that the study has found harms associated with using corticosteroids to treat tennis elbow. This study has not found significant harms associated with these injections compared to placebo for this condition.

What kind of research was this?

This systematic review investigated whether injections are beneficial for tendinopathies (disorders of the tendons). A systematic review is a robust way of bringing together all the good research on a topic and combine the results to obtain a measure of effectiveness. The researchers noted that “the large number of studies about…other injection types underpins the need for a synthesis of the evidence for injection therapies”.

What did the research involve?

Tendinopathies are disorders of the tendons relating to overuse, and are common in active young and middle-aged people. A range of injection types are offered as treatment, including steroids, platelet-rich plasma, botox and other drugs. In this research, a systematic review of studies was carried out to determine which types of injections are beneficial for different types of tendinopathies (including corticosteroids). Given that there is usually no inflammation associated with these conditions, researchers are questioning whether corticosteroid injections, one of the commonly used treatments, are actually effective. They did not include studies that assessed injections into the muscles or injections directly into the joints.

The researchers were also interested in whether there were differences in the short, medium and long-term effects of the different treatments. Databases of medical literature were searched for randomised controlled trials published before March 2010. To be eligible for inclusion, studies had to have assessed whether injections into the tissue surrounding a tendon were effective for treating tendinopathies, and compared these injections with placebo or with other non-surgical interventions. The researchers also only included studies that they assessed to be of high quality.

The researchers then combined the results of the relevant trials using meta-analyses, grouping them by the type of tendinopathies, the type of injectable treatment and the term of effect (short, medium and long).

What were the basic results?

The literature searches found 41 studies, including 12 that assessed the effectiveness of steroids for tennis elbow. The other studies assessed steroids for rotator-cuff tendinopathy (tendon problems in the shoulder), for tendinopathies in other sites and other treatments for these disorders.

As described above, the research was grouped in different ways, but the results consistently showed that, compared with other interventions, corticosteroids reduced pain only in the short term. Multiple injections did not improve outcomes more than single injections.

The other treatments for tennis elbow showed that there were significant reductions in pain in the short, medium and long term following injections of sodium hyaluronate. Pain was also significantly reduced with botox in the short-term and with prolotherapy (hypertonic glucose and anaesthetic) in the medium-term.

How did the researchers interpret the results?

The researchers say that despite the effectiveness of corticosteroid injections in the short-term, non-corticosteroid injections ‘might be of benefit for the long-term treatment of’ tennis elbow.

However, they say that the response to injection should not be generalised across different sites (e.g. tendon problems in elbow, knee, shoulder, etc.) because their review shows differences depending on which joint the tendinopathy affects.


This is a well-conducted systematic review but it has some limitations, largely relating to the quality of the underlying evidence. The researchers detail some of these:

  • Firstly, the evidence available to compare steroid injections with physiotherapy and with placebo was ‘heterogenous’ (the underlying studies had relatively different methods). This affects the confidence that can be assumed in the summary results when such studies are pooled.
  • Some of the subgroups for analysis were quite small. For many, there was only one study. There may have been only limited power to detect any significant differences due to the small sample sizes

The research demonstrates a short-term benefit of corticosteroids compared with doing nothing for pain related to tennis elbow, but that this particular treatment has no effect in the longer term. There is no conclusive evidence that the steroid injections ‘do more harm than good’ as suggested by the BBC News headline. Some of the studies favoured the treatments that the corticosteroids were being compared against. However, this was often in single-study results (not pooled results) and was not the case for all comparators.

An editorial that accompanies this article suggests, “today’s review might discourage clinicians from using corticosteroids in patients who are seeking medium-term and long-term cures”. It says the clinical implication here is that one corticosteroid injection is not helpful for elbow pain at 6 to 12 months, and that multiple injections do not improve outcomes. However, given the shortage of studies for some of these analyses, especially of alternatives to corticosteroid injections for tennis elbow, the short-term benefits may still be better than nothing for many patients. The editorial says that “there is no compelling evidence that any injection for tendinopathy is a magic bullet”.

NHS Attribution