Veronica Hackethal, MD
October 17, 2019
Steroid injections are frequently used to relieve pain associated with osteoarthritis of the knee and hip, but new evidence suggests the treatment may do more harm than good for some people. Experts now stress the need for better informed consent about potential risks and benefits of injections.
Data from more than 450 patients who received intra-articular corticosteroid injections for osteoarthritis at Boston University show that the treatment may speed the pace of osteoarthritis and contribute to joint destruction.
The article was published online October 15 in the journal Radiology.
"We are now seeing [that] these injections can be very harmful to the joints, with serious complications such as osteonecrosis, subchondral insufficiency fracture, and rapid progressive osteoarthritis," senior author Ali Guermazi, MD, PhD, said in a press release. Guermazi is chief of radiology at the Veterans Affairs Boston Healthcare System and professor of radiology at Boston University School of Medicine.
Some patients may be more prone than others to poor outcomes from the treatment, but it's not yet known how to identify these people. The researchers stress the importance of informed consent, and urge radiologists to take x-rays before administering steroid injections, in order to identify underlying problems that may contribute to adverse events.
"Intra-articular corticosteroid injection should be seriously discussed for pros and cons. Critical considerations about the complications should be part of the patient consent, which is currently not the case right now," Guermazi added.
Long-Term Data Has Been Lacking
The first-line treatment for osteoarthritis, which most commonly affects the hip and knee, is conservative pain control, but many patients eventually need joint replacement. Yet people with osteoarthritis are often older and have multiple medical problems that make them ineligible for surgery or long-term treatment with acetaminophen or nonsteroidal anti-inflammatory (NSAIDs) medication.
Steroid joint injections have been widely used for decades to treat patients like these, and others with inadequate pain control. While short-term complications are rare, most studies on the long-term effects are of low quality. Some evidence from animal and human laboratory studies suggests steroid joint injections may contribute to progression of osteoarthritis. Professional societies differ on whether or not to recommend steroid joint injections for osteoarthritis.
Therefore, Andrew Kompel, MD, also from Boston University School of Medicine, and colleagues reviewed the records of 459 individuals who received at least one corticosteroid injection in the hip or knee joint in 2018 at an inner city hospital in Boston.
Overall, 8% (n = 36) of patients experienced an adverse joint event after receiving a steroid joint injection. These individuals ranged in age from 37 to 79 years (mean age, 57 years) and most (72%) showed moderate osteoarthritis at baseline. They received an average of 1.4 injections and developed joint complications anywhere between 2 to 15 months after injection, with an average of 7 months.
The authors identified four main adverse joint events after steroid joint injections. The most common was accelerated progression of osteoarthritis, found in 6% of individuals (n = 26).
The second most common adverse joint event was subchondral insufficiency fracture, found in 0.9% (n = 4) of individuals. Subchondral insufficiency fracture has traditionally been thought to occur in older individuals with weak bones, but recent evidence suggests it may be more common and affect younger patients.
The condition is potentially underdiagnosed due to lack of awareness. Delayed diagnosis can lead to joint damage and eventual joint replacement. Diagnosis is important before giving steroid joint injections, which can impair healing in these kinds of fractures, according to the authors.
In addition, osteonecrosis and rapid joint destruction each affected 0.7% (n = 3) of patients, respectively.
Osteonecrosis refers to decreased blood flow to the bone that can cause breakdown of the bone, eventual fracture, and need for joint replacement. Patients with osteonecrosis but without fracture sometimes receive steroid joint injections. The authors emphasize the need to inform such patients that steroid joint injections could potentially worsen their condition.
They also note that rapid joint destruction and accelerated bone loss may occur after the first steroid injection and in patients without evidence of underlying disease on x-ray. In these patients, they suggest closely reviewing the need for injection and repeating x-rays before giving further injections.
The authors conclude: "The radiology community should actively engage in high-quality research to further understand these adverse joint findings and how they possibly relate to [intra-articular corticosteroid] injections to prevent or minimize complications."
In an accompanying editorial, Richard Kijowski, MD, of the University of Wisconsin School of Medicine and Public Health, notes several limitations of the study, including the small number of patients and lack of standardized methods.
"The report is neither a prospective clinical trial nor a retrospective observational study…The objective is to educate radiologists that the intra-articular corticosteroid injection they routinely perform with little, if any, thought about long-term safety may cause more harm than benefit," he writes.
He agreed with the authors about the importance of informed consent.
"Patients might be more than willing to take the small risk of an adverse joint event requiring eventual joint replacement for the possibility of at least some degree of pain relief after intra-articular corticosteroid injection," he concludes. "However, patients have the right to make this decision for themselves, and this requires radiologists to discuss all potential risks and benefits with the patient when obtaining written informed consent."
The study authors acknowledge that they could not determine whether these adverse joint events were already present when patients had their steroid joint injections, or if the injections caused these problems.
One or more authors owns shares in and/or has been a consultant for one or more of the following: Boston Imaging Core Lab, TissueGene, Merck Serono, Pfizer, AstraZeneca, Galapagos, and/or Roche. Kijowski has disclosed no relevant financial relationships.