Rheumatologists and radiologists are debating the use of imaging before administration of intra-articular corticosteroid (IACS) injections for people with osteoarthritis of the knee.
The difference in opinion has arisen following an expert review panel, convened after a case series of rare but potentially serious structural outcomes were seen after IACS injections for knee OA.
The panel’s radiologists suggested that X-rays and MRI prior to repeated IACS injections might be appropriate to try and avoid these outcomes.
The panel also concluded that patients should be advised of potential long-term complications before undergoing IACS injections, including accelerated OA progression and rapid joint destruction.
Rheumatologists, however, disagree, citing the likely low prevalence of the complications.
“I wouldn’t raise this as even an amber flag for the field,” said Professor Philip Conaghan, a rheumatologist and professor of musculoskeletal medicine at the University of Leeds School of Medicine who participated in the expert panel review.
“I don’t think there is a big signal here for the field to be concerned about. I think what the field needs to be concerned about is the fact that we have so few therapies for OA,” he said in an interview with the limbic.
The panel’s review was published last week in Radiology.
Troubling complications
In one case series covering 459 injections of the knee or hip, researchers found a combined total of 8% of patients had one of four adverse joint outcomes: accelerated OA progression; subchondral insufficiency fracture; complications of osteonecrosis; and rapid joint destruction including bone loss.
The panel concluded that for informed consent to IACS injection it was no longer sufficient to provide the advice “at least it won’t harm you”.
“In addition to standard risks for short-term (or early) complications (eg, bleeding, infection, damage to intra-articular and periarticular structures along the needle path, joint pain, swelling, and stiffness) we also need to explain to the patients that possible longer-term (or delayed) events may occur that may or may not be related to the actual IACS injection,” they wrote.
But Professor Conaghan disagreed, saying there was currently no evidence that steroid injections specifically contributed to the onset of problems.
“It’s very important to sort out this business of chicken and egg,” he said.
For example, a trial published in JAMA in 2017 randomised 140 patients with symptomatic knee OA to receive either corticosteroid or saline injection every 12 weeks for two years; patients who received the IACS injections had greater cartilage loss and there was no difference in pain, but there were no instances of osteonecrosis or subchondral fracture in the trial. Prof essor Conaghan said the ongoing large Osteoarthritis Initiative, in the U.S., has also not shown any worrisome safety signals with IACS injections to this point.
Imaging stand-off
The review by the expert panel included a suggested algorithm for the use of imaging for both first-time and repeat IACS injections, noting that obtaining weight-bearing X-rays and MRI in certain circumstances may help reduce the risk of some of the negative structural outcomes.
For example, if subchondral fracture was found on imaging studies then the IACS injection could be avoided, due to the potential for glucocorticoids to inhibit the fracture healing process. However, the review noted that the rheumatologists and orthopaedic surgeons included on the expert panel did not concur with the recommended use of imaging for repeated IACS injections.
“There might be a small percentage of all the people who present with knee pain that we think is OA who’ve got a subchondral fracture,” and are treated with an IACS injection, Professor Conaghan said.
“The question is, is that a bad thing? Well, if it helps their pain, I don’t think it’s a bad thing.” Furthermore, it was unlikely that the injection itself would actually worsen the condition, he added.
Unanswered questions
The expert panel concluded that the exact cause of the negative structural outcomes in patients presenting with knee OA remains unclear, and that further study into the issue is warranted.
They also recommended that clinicians inform their patients that along with the standard short-term risks such as bleeding, swelling, stiffness, of the potential for longer-term negative outcomes that may or not be associated with the injections themselves.
In terms of future research directions, the prevalence of the negative structural outcomes remains a key question, as if they are very rare then pre-injection MRI for large numbers of patients would be prohibitively costly.
“Overall the data is reassuring,” Professor Conaghan said. “But we should be conscious that we don’t [always] know where the pain comes from … and we should be cautious about people whose symptoms either don’t respond or get worse after a steroid injection. But I think, clinically speaking, they are a very small group.”