November 9, 2020—It does not appear that corticosteroid injections for the management of pain associated with knee osteoarthritis contribute to worsening of the condition, according to research presented at the American College of Rheumatology Virtual Convergence 2020, which took place online from November 5 to 9.
“Cortisone injections have been used for a long time to treat pain from knee osteoarthritis,” said lead author Justin J. Bucci, MD, of Boston University School of Medicine, during an ACR press conference attended by Elsevier’s PracticeUpdate. “It provides short-term pain relief and is generally considered a safe procedure, but recent studies have questioned whether cortisone injections actually worsen knee osteoarthritis. A study from last year comparing people who were getting cortisone injections and those who did not found that the group getting injections had worse disease over time, based on their X-rays.” It is unclear whether confounding variables contributed to this association, however.
Rather than try to determine what those confounding variables might be, the authors decided instead to compare patients with osteoarthritis of the knee who received corticosteroid injections with those who received injections that have not been associated with worsening of the condition, notably hyaluronic acid injections.
The investigators looked at rates of radiographic progression and total knee replacement among patients given corticosteroid injections or hyaluronic acid injections for knee osteoarthritis in the MOST and OAI cohort studies. In both studies, patients were followed-up with radiography to the knee and were asked if they had received hyaluronic acid or corticosteroid injections in their knees in the previous 6 months. The frequency of follow-up was every 12 months for OAI and every 30 months for MOST. The studies use similar acquisition and reading protocols. Kellgren and Lawrence grades (KL) 0 to 4 and joint space narrowing (JSN) 0 to 3 were scored for both studies. In addition, in OAI, medial joint space (JWS250) was also measured.
For analysis of radiographic progression, the investigators excluded knees with baseline KL grades of 4, corticosteroid injections or hyaluronic acid injections reported at first visit, and recipients of both hyaluronic acid injections and corticosteroid injections.
In total, the investigators evaluated 792 knees, comprising 647 for which corticosteroid injections use was reported and 145 for which hyaluronic acid injections were reported. In addition, there were 124 knees for which corticosteroid injections were reported at more than one visit, and 19 knees for which hyaluronic acid injections were reported at more than one visit. The investigators compared radiographic results from the visit before the first reported injection with those of the visit after the last injection. Total knee replacement without post-injection radiography was assigned a KL grade of 4 and JSN of 3 for the visit at which total knee replacement was first present. Annualized deterioration rates were calculated for KL, JSN, and JWS250.
In an unadjusted analysis, the rate of total knee replacement was greater among knees with single exam reports of hyaluronic acid injections than of corticosteroid injections (P =.04) but not different for those reporting these injections at multiple exams, although numbers were small.
In a multivariable analysis that adjusted for age, sex, body mass index, study, and baseline KL grade, rates of radiographic progression between corticosteroid injections and hyaluronic acid injections were similar, whether patients reported receiving injections at single or multiple exams.
“We found that there was no difference in progression of knee osteoarthritis in the two groups,” said Dr. Bucci. “KL, JSN [and JSW250] … did not change at different rates between the two groups … and neither group had a significantly different rate of receiving knee replacement.”
Based on these findings, he concluded that “we can provide some reassurance to patients and physicians treating osteoarthritis that cortisone injections are not causing the disease to get worse. Patients with moderate-to-severe knee osteoarthritis that is not responding to conservative measures should still be offered cortisone injections, and patients who are getting benefit from these injections can still continue getting them without doing harm to their knees.”
Moving forward, Dr. Bucci and his team are conducting a similar analysis using MRI data.