Although intra-articular injection of corticosteroid is a common treatment for osteoarthritis of the knee, there is concern that long-term treatment may promote joint destruction and tissue atrophy. Arroll and Goodyear-Smith conducted a meta-analysis to determine the relative benefits and risks of intra-articlular injections of corticosteroid for osteoarthritis of the knee.

The researchers searched electronic databases and reference lists, and contacted authors to identify controlled trials and papers concerning intra-articular steroid injection for osteoarthritis of the knee. Arroll and Goodyear-Smith independently assessed the identified studies for quality and abstracted data to determine relative risk and numbers needed to treat for improvement of symptoms.

Ten studies met the inclusion criteria for the meta-analysis. The dose of prednisone that was injected varied from 6.25 mg to 80 mg. In six of these studies, significant improvement in symptoms of osteoarthritis was demonstrated up to two weeks after injections, with a relative risk (RR) of 1.66 and a number needed to treat (NNT) for one stage of improvement of 1.3 to 3.5. Neither of the two high-quality studies showed statistically significant improvement at 16 to 24 weeks, but their pooled results for improvement at 16 to 24 weeks had an RR of 2.09 and an NNT of 4.4.

The authors conclude that intra-articular corticosteroid injections improve symptoms of knee osteoarthritis at two weeks and up to 16 to 24 weeks. The NNT for benefit is low (range: 1.3 to 3.5 patients). Although the optimal dose is not clear, the authors estimate that 50 mg of prednisone may be required to show benefit for 16 to 24 weeks.