Intra-articular Corticosteroid Injections in the Hip and Knee: Perhaps Not as Dangerous as They Want You to Believe?

We read with concern the October 2019 report in Radiology by Dr Kompel and colleagues, whose findings have been highlighted in the media as demonstrating a high rate of serious consequences of intra-articular corticosteroid (IACS) injections into the knee and hip (1). The most frequent complication, at 6%, accelerated knee osteoarthritis, is concerning given that IACS is widely used in practice and is considered effective and safe (2,3). Because sensationalizing medication complications can itself have adverse clinical consequences, the incidence rate of adverse events following IACS must derive from solid methods. Unfortunately, as a retrospective review of clinical data, the study by Dr Kompel and colleagues is inherently susceptible to biases (eg, indication and detection) that inflate estimates of occurrence. Aspects of their study that would generate such biases include reliance on clinically available imaging, nonsystematic case detection, incomplete follow-up, and (especially) inability to be sure that the outcomes of interest were not present prior to IACS. Incomplete information regarding these patient characteristics also prevents the generalization of their results.

Another reason to suspect an inflated adverse event rate is its discordance with controlled studies and clinical practice. Among more than 10 published IACS clinical trials (representing 818 patients and 2084 injections), including two 2-year trials (∼140 patients) of repeated injections (4,5), none reported adverse joint event rates of this magnitude. One trial administered a knee IACS injection every 3 months for 2 years and performed systematic MRI surveillance for all the adverse events reported by Dr Kompel and colleagues, yet they did not detect a single one (5). Furthermore, expert panels from Osteoarthritis Research Society International and the American College of Rheumatology recently conducted separate reviews of the global literature by using systematic risk-benefit assessments, and both concluded that IACS is recommended for osteoarthritis treatment (2,3).

The study by Dr Kompel and colleagues is informative regarding the types of pathologic abnormalities observed among patients with joint pain undergoing joint injections. However, it does not allow inference of causality nor does it provide accurate measurements of the rate of occurrence of those complications. We need to be thoughtful about how such information is presented and disseminated to avoid excessive concerns about a treatment that most experts view as having a favorable balance of benefit versus harm when used appropriately.