Reviewed by Dr. Jennifer Wagner
McAlindon T, et al. JAMA. 2017 May 16;317(19): 1967-1975

Symptomatic knee osteoarthritis is common and disabling. Intra-articular steroid injections are often given for pain control by suppressing inflammation. However, steroids also exhibit anti-anabolic effects on cartilage, with potential to damage joints. The objective of this 2-year double blind, placebo controlled trial was to evaluate whether steroid injections affected pain, function, or cartilage loss. Participants (n=140) were randomized to receive either 1 ml (40 mg) triamcinolone injection or 1 ml saline injection (neither mixed with local anesthetic) every three months for two years. Pain/function questionnaires and knee exams were performed every three months. MRI of the knee was done at baseline then yearly to assess cartilage thickness and damage. The study resulted with a greater rate of cartilage loss with triamcinolone vs placebo (p=0.01). Other MRI structural outcomes were no different between groups, including effusion volume. At two years, there was no significant difference in pain reduction between the groups. There were no significant differences in adverse events either. This study concluded that intra-articular triamcinolone injections resulted in greater cartilage volume loss and no greater improvement in pain as compared to saline.

Although triamcinolone knee injections did not improve pain more than saline in this trial, they resulted in cartilage volume loss. We lack treatment options for these patients, and a temporary improvement in pain and function may be worthwhile. It is also unknown whether the amount of cartilage loss seen is clinically significant, as there was no increased progression of the disease based on other MR imaging or clinical findings. Although the cartilage loss in this two-year study did not affect symptoms, it has been shown that increased cartilage loss leads to higher rates of arthroplasty over time. Limitations of the study include evaluation of pain every three months, so immediate, temporary benefits may have been missed. Second, patients continued their own pain medications, only stopping NSAIDs for 48 hours before pain evaluations. Finally, participants experienced a large placebo effect, narrowing outcome differences. In summary, this article suggests that the commonly used knee injections are no more effective than saline injections and may contribute to greater cartilage volume loss in symptomatic knee osteoarthritis.