Corticosteroid injections for lateral epicondylitis: a systematic review

Abstract

BACKGROUND: Lateral epicondylitis (tennis elbow) is a common complaint, for which corticosteroid injections are a frequently applied therapy. However, there were no up-to-date reviews available that systematically addressed the effectiveness and adverse effects, including questions concerning optimal timing of injections and composition of the injection fluid. AIM: The aim of the study was to assess the effectiveness of corticosteroid injections in the treatment of lateral epicondylitis (tennis elbow) by systematic review of the available randomized clinical trials. DATA SOURCES: The data sources used were randomized clinical trials identified by literature searches of the MedLine (1966-1994) and Embase (Exerpta Medica) (1980-1994) databases for the keywords epicondylitis, tendinitis and elbow, injection. References given in relevant publications were further examined. STUDY SELECTION: The criteria for selecting studies were as follows: randomized clinical trials (treatment allocation in random or alternate order); one of the treatments to include one or more corticosteroid injections (additional interventions were allowed); participants suffering from lateral epicondylitis; and publication in English, German or Dutch. Abstracts and unpublished studies were not included. DATA SYNTHESIS: Methodological quality was assessed by means of a standardized criteria list (range 1-100 points). The extracted outcomes were the general conclusion drawn by the authors of the reports on the trials, and the success rates at the various follow-up points as (re)calculated by us. The success rates were subsequently graphically displayed and statistically pooled. Separate stratified analyses were conducted according to a predetermined analysis plan. RESULTS: Twelve randomized clinical trials were identified. The median methodological score was 40 points, indicating an overall poor to moderate quality. The pooled analysis indicated short-term effectiveness (2-6 weeks): pooled odds ratio (OR) = 0.15 [95% confidence interval (CI) 0.10-0.23], chi 2 [degrees of freedom (df = 5) = 13.3], indicating statistical heterogeneity. At longer term follow-up, no difference could be detected. The studies of better methodological quality indicated more favourable results than those of lesser methodological quality. The most suitable corticosteroid to use as well as dosage, injection interval and injection volume could not be derived from the various trials. CONCLUSION: The existing evidence on corticosteroid injections for the treatment of tennis elbow is not conclusive. Many trials were conducted in a secondary care setting and clearly had serious methodological flaws, and there was statistical heterogeneity among the trials. Corticosteroid injections appear to be relatively safe and seem to be effective in the short term (2-6 weeks). Although the treatment seems to be suitable for application in general practice, further trials in this setting are needed. As yet, questions regarding the optimal timing, dosage, injection technique and injection volume remain unanswered.