Intra-articular corticosteroids for knee and hip OA: a systematic review of serious adverse joint outcomes

Abstract

Purpose: Intra-articular corticosteroids (IACS) are proven to provide short-term pain relief and functional improvement in individuals with knee and hip osteoarthritis (OA). However, the potential for serious adverse joint outcomes resulting from long-term use of IACS has been cause for concern. We conducted a systematic review of all RCT data comparing IACS to intra-articular saline to explore the incidence of accelerated OA progression, subchondral insufficiency fracture, complication of osteonecrosis, and/or rapid destruction of joint tissue among individuals with knee and hip OA. Methods: Medline, PubMed Central, Google Scholar, and the Cochrane Databases were systematically searched from inception to November, 2019. Reference lists of relevant reviews were searched by hand, and we sought additional data within supplements of conference proceedings that had been published up until November, 2019. Randomized controlled trials comparing IACS to intra-articular saline in participants with knee or hip OA were included. Two reviewers screened potentially relevant references and independently assessed study quality using the Cochrane Risk of Bias tool. Relevant data were independently extracted from included studies by the same two reviewers. Outcomes of interest included incidence of any one of four target serious adverse joint outcomes: accelerated OA progression, subchondral insufficiency fracture, complication of osteonecrosis, and/or rapid destruction of joint tissue. Analysis of dichotomous outcomes using the Mantel-Haenszel method was planned a priori, and effect estimates were to be reported as risk ratios (RR) with 95% confidence intervals. Results: Our search returned 516 potentially relevant references. Of these, 28 references were deemed to be potentially eligible for inclusion. Twenty two RCTs met our inclusion criteria, but 12 of them did not report safety outcomes. Ten RCTs incorporating a total of 538 patient-years of follow-up reported on safety in sufficient detail to confirm absence or incidence of the target serious adverse joint outcomes. Seven RCTs (N= 728) involved participants with knee OA, and 3 RCTs focused on hip OA (N= 162). Follow-up of the included RCTs ranged from 5 weeks to 2 years, with the most common (40% of RCTs) study duration being 6 months. 60% of RCTs involved a single IACS injection; two trials involved 8 injections total over 12 week intervals, while two other RCTs involved 3 and 5 total injections over intervals of 2 weeks and 1 week, respectively. Study quality was overall assessed to be of Low to Very Low Quality. None of the included RCTs involving participants with knee or hip OA documented incidence of accelerated OA progression, subchondral insufficiency fracture, complication of osteonecrosis, or rapid destruction of joint tissue over the course of the study follow-up. Due to a complete absence of events, statistical comparison of IACS with placebo was not possible. Conclusions: The results of our study suggest that the use of IACS is unlikely to lead to serious adverse joint outcomes in a controlled clinical setting. These findings are limited by the short follow-up and low number of total IACS injections, and may not be generalizable to patients with repeated injections over longer time periods. High-quality, long-term observational data are needed to confirm the safety of long-term use of IACS in knee or hip OA.