Every year approximately four of every 1,000 adults has a problem with tennis elbow, or painful lateral epicondylitis. While more than 40 treatments for tennis elbow have been proposed, the two leading strategies are local injections of corticosteroids and courses of oral nonsteroidal anti-inflammatory drugs (NSAIDs). Because previous clinical studies were complicated by methodologic problems, Hay and colleagues conducted a randomized controlled trial in 23 general practices to compare treatment with local corticosteroid injections and NSAIDs for tennis elbow.

All patients 18 to 70 years of age who consulted their general practitioner because of symptoms consistent with tennis elbow were considered for the study. Exclusion criteria included structural abnormalities of the elbow and contraindications to either of the study treatments. After baseline assessment by research nurses, patients were randomly assigned to one of three treatment groups. Local injection of 20 mg of methylprednisolone plus 0.5 mL of 1 percent lidocaine was given to 51 patients. Enteric-coated naproxen in a dosage of 500 mg twice daily for two weeks was prescribed for 49 patients. The remaining 56 patients were given placebo (unmarked vitamin C tablets) and the same advice and instructions as the patients who received naproxen. All patients were given an information leaflet and provided with analgesia. Patients were reassessed four weeks, six months and one year after treatment. All of the assessments were conducted by a research nurse who was not aware of the treatment used. Patient assessment of symptoms, functional impairment and change of symptoms over time were included in the reassessments. Grip strength and physical examination findings were also documented.

Although the groups were comparable at randomization, by four weeks after treatment 92 percent of patients who received injections reported improvement or recovery compared with 57 percent of those who were treated with naproxen and 50 percent of patients who received placebo. Five patients discontinued use of naproxen because of side effects. Most injected patients reported a minor increase in pain following treatment, but no serious side effects of treatment were noted. Local skin atrophy occurred in three patients in the naproxen and placebo groups. At one year, 84 percent of injected patients, 85 percent of naproxen-treated patients and 82 percent of placebo-treated patients recorded no or low pain scores.

The authors conclude that early local injection of a corticosteroid is an effective, safe treatment for lateral epicondylitis. A two-week course of NSAIDs was found to be no more effective than placebo in this study, and the long-term outcome was good regardless of initial treatment.