Corticosteroid (CS) injections may worsen progression of knee osteoarthritis (OA) as seen on radiography and whole-knee MRI. Injecting hyaluronic acid (HA) instead, or managing the condition without injections, may better preserve knee structure and cartilage, according to results of two related studies presented at the Radiological Society of North America (RSNA) 2022 Annual Meeting.
The findings come nonrandomized, observational cohort studies, leading knee OA experts to call for further study in randomized trial settings. In the meantime, they advise shared decision-making between patients and clinicians on the use of these injections.
For knee OA, most patients seek a noninvasive treatment for symptomatic relief. “At least 10% of these patients undergo local treatment with injectable corticosteroids or hyaluronic acid,” the lead author of one of the studies, Upasana Upadhyay Bharadwaj, MD, research fellow in musculoskeletal radiology at the University of California, San Francisco, said in a video press release.
Researchers in both studies used data and images from the Osteoarthritis Initiative (OAI), a multicenter, longitudinal, observational study of 4796 US patients aged 45–79 years with knee OA. Participants were enrolled from February 2004 to May 2006.
The OAI maintains a natural history database of information regarding participants’ clinical evaluation data, x-rays, MRI scans, and a biospecimen repository. Data are available to researchers worldwide.
Two Studies Draw Similar Conclusions
In one study, Bharadwaj and colleagues found that HA injections appeared to show decreased knee OA progression in bone marrow lesions.
They investigated eight patients who received one CS injection, 12 who received one HA injection, and 40 control persons who received neither treatment. Participants were propensity-score matched by age, sex, body mass index (BMI) , Kellgren-Lawrence (KL) grade, Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), and Physical Activity Scale for the Elderly (PASE).
The researchers semiquantitatively graded three Tesla MRI scans that had been obtained at baseline, 2 years before the injection, and 2 years after the injection, using whole-organ MRI score (WORMS) for the meniscus, bone marrow lesions, cartilage, joint effusion, and ligaments.
They quantified OA progression using the difference in WORMS between baseline and 2-year follow-up, and they used linear regression models, adjusted for age, sex, BMI, KL grade, WOMAC, and PASE, to identify the link between type of injection and progression of WORMS.
At 2 years, the authors found the following:
There was a significant association between CS injection and postinjection progression of WORMS over 2 years for the knee overall, the lateral meniscus, lateral cartilage, and medial cartilage.
There was no significant link between HA injection and postinjection progression of WORMS.
There was no significant link between either injection type and progression of pain, as quantified by WOMAC.
There was no significant difference in progression of WORMS over the 2 years prior to injection for CS and HA injections.
“Corticosteroid injections must be administered with caution with respect to long-term effects on osteoarthritis,” Bharadwaj advised. “Hyaluronic acid injections, on the other hand, may slow down progression of knee osteoarthritis and alleviate long-term effects while offering similar symptomatic relief to corticosteroid injections. Overall, they are perhaps a safer alternative when looking at medium- and long-term disease course of knee osteoarthritis.”
In the second study, lead author Azad Darbandi, MS, a fourth-year medical student at Chicago Medical School of Rosalind Franklin University of Medicine and Science, and colleagues found that patients who received CS injections experienced significantly more medial joint space narrowing.
They identified 210 knees with imaging at baseline and at 48 months that received CS injections, and 59 that received HA injections; 6827 knees served as controls. The investigators matched 50 patients per group on the basis of confounding factors, which included age, sex, BMI, comorbidities, surgery, and semiquantitative imaging outcomes at baseline. They performed ANCOVA testing using 48-month semiquantitative imaging outcomes as dependent variables and confounding variables as covariates.
The researchers analyzed joint space narrowing, KL grade, and tibia/femur medial/lateral compartment osteophyte formation and sclerosis.
At 4 years:
Average KL grade in the CS group was 2.79; in the HA group, it was 2.11; and in controls, it was 2.37.
Intergroup comparison showed significant differences in KL grade between CS and HA groups and between CS and control groups.
Medial compartment joint space narrowing was 1.56 in the CS group, 1.11 in the HA group, and 1.18 in controls. There was a significant difference between the CS and control groups.
Other dependent variables were not significant.
“These preliminary results suggest that corticosteroid injections accelerated the radiographic progression of osteoarthritis, specifically medial joint space narrowing and Kellgren-Lawrence grading, whereas hyaluronic acid injections did not,” Darbandi told Medscape Medical News by email.
“OA radiographic progression does not always correlate with clinical progression, and further research is needed,” he added.
Proper matching of patients at baseline for confounding factors is a strength of the study, Darbandi said, while the retrospective study design is a weakness.
Experts Share Their Perspectives on the Preliminary Results
Michael M. Kheir, MD, assistant professor of orthopedic surgery at the University of Michigan Health System, who was not involved in the studies, said he would like to see further related research.
“Perhaps steroid injections are not as benign as they once seemed,” he added by email. “They should be reserved for patients who already have significant arthritis and are seeking temporary relief prior to surgical reconstruction with a joint replacement, or for patients with recalcitrant pain after having already tried HA injections.”
William A. Jiranek, MD, professor and orthopedic surgeon at Duke Health in Morrisville, North Carolina, who also was not involved in the studies, was not surprised by the findings.
“It is important to do these studies to learn that steroid injections do not come with zero cost,” he said by email.
“I am pretty sure that a percentage of these patients had no cartilage loss at all,” he added. “We need to understand which OA phenotypes are not at risk of progressive cartilage loss from steroid injections.”
Annunziato (Ned) Amendola, MD, professor and sports medicine orthopedic surgeon at Duke Health in Durham, North Carolina, who was also not involved in the studies, said he would like to know how injection effectiveness and activity level are related.
“If the injections were effective at relieving pain, and the patients were more active, that may have predisposed to more joint wear,” he said by email. “It’s like tires that last longer if you don’t abuse them.”
Shared Decision-Making and Further Research Recommended
Amanda E. Nelson, MD, associate professor of medicine in the Division of Rheumatology, Allergy, and Immunology at the University of North Carolina at Chapel Hill, said, “The lack of randomization introduces potential biases around why certain therapies (CS injection, HA injection, or neither) were selected over others (such as disease severity, preference, comorbid conditions, other contraindications, etc), thus making interpretation of the findings challenging.
“The causal relationship remains in question, and questions around the efficacy of intraarticular HA in particular, and the ideal settings for intraarticular therapy in general, persist,” Nelson, who was also not involved in the studies, noted by email. “Thus, shared decision-making between patients and their providers is essential when considering these options.”
C. Kent Kwoh, MD, professor of medicine and medical imaging at the University of Arizona College of Medicine and director of the University of Arizona Arthritis Center in Tucson, said in a phone interview that these types of studies are important because CS injections are common treatments for knee OA, they are recommended in treatment guidelines, and other good options are lacking.
But he pointed out that the results of these two studies need to be interpreted with caution and should not be used to decide the course of treatment.
“These data are hypothesis generating. They suggest association, but they do not show causation,” said Kwoh, who was also not involved in the studies. “Both studies are secondary analyses of data collected from the OAI, which was not specifically designed to answer the questions these studies are posing.
“The OAI was not a treatment study, and participants were seen only once a year or so. They may have had joint injections anytime from only days to around 1 year before their visit, and their levels of activity or pain just prior to or just after their joint injections were not reported,” Kwoh explained.
The reasons why patients did or did not receive a specific joint injection — including their socioeconomic status, race, access to insurance, and other confounding factors ― were not assessed and may have affected the results, he added.
The fact that both studies used the same data and came to the same conclusions gives the conclusions some strength, he said, but “the gold standard to understanding causation would be a randomized, controlled trial.”
Darbandi’s research received grant support from Boeing, His co-authors, as well as all experts not involved in the studies, report no relevant financial relationshiips. Bharadwaj did not provide conflict-of-interest and funding details. Kwoh reports membership on panels that have developed guidelines for the management of knee osteoarthritis.
Radiological Society of North America (RSNA) 2022 Annual Meeting.