Treatment of osteoarthritis with intra-articular injections of hyaluronic acid over a 5-week period provided symptomatic relief that had noninferior efficacy and was safer than treatment with an oral NSAID in an open-label, multicenter randomized trial of older Japanese adults.
Intra-articular injections of hyaluronic acid (IA-HA) have been shown to reduce pain and improve joint function in knee osteoarthritis (OA), but studies also have reported a relatively slow speed of therapeutic onset and "considerable heterogeneity" in outcomes. NSAIDs have long been regarded as effective for symptomatic relief of knee OA, but carry the risk of serious gastrointestinal adverse events. The lack of data on the short-term effects of IA-HA compared with widely used NSAIDs spurred Dr. Muneaki Ishijima of Juntendo University, Tokyo, and colleagues to conduct the first trial to directly compare efficacy and safety of the two treatments for knee OA (Arthritis Res. Ther. 2014 Jan. 21 [doi:10.1186/ar4446]).
The investigators randomized 200 patients to 5 weeks of either weekly IA-HA or the nonselective cyclo-oxygenase inhibitor loxoprofen in three 60-mg tablets daily and evaluated patients’ percent change from baseline on the patient-reported Japanese Knee Osteoarthritis Measure (JKOM, primary endpoint) and their percent change from baseline in pain as rated on a 0- to 100-point visual analog scale (VAS, secondary endpoint). (Loxoprofen is not approved in the United States.)
After the 5 weeks of therapy, the investigators analyzed data from 98 patients who underwent IA-HA and 86 patients who received loxoprofen who had final JKOM scores available. In the IA-HA group, one patient withdrew consent before undergoing IA-HA, and one was lost to follow-up before the final JKOM analysis. In the NSAID group, six patients withdrew consent before receiving treatment, one patient was excluded because of duplicate entry into the trial before receiving treatment, and another seven were lost to follow-up before the final analysis.
The trial participants had a mean age of about 68 years, and approximately 72% were women. The patients had OA of the medial femorotibial joint with a Kellgren-Lawrence grade of 1-3.
The mean JKOM score declined 32.2% (from 32.0 to 22.0) with NSAID treatment and 34.7% (from 33.8 to 21.5) with IA-HA. This difference of –2.5% and its 95% confidence interval of –14.0% to 9.1% meant that the two treatments had similar efficacy based on the trial’s prespecified definition of noninferiority as a 10% difference or less in the upper limit of the confidence interval. The result did not change after adjustment for age, Kellgren-Lawrence grade, body mass index, and medical center.
Pain on the VAS dropped by 36.0% (from 55.5 to 31.9) with NSAID therapy and by 41.2% (from 60.1 to 31.8) with IA-HA, a nonsignificant difference.
There also were no differences in the percentage of responders in each group based on Outcome Measures in Rheumatology clinical trials and Osteoarthritis Research Society International (OMERACT-OARSI) response criteria, with 69.7% in the IA-HA group and 62.4% in the NSAID group. For this trial, the responder criteria were partly modified by use of the JKOM score. Response was defined as a decrease in joint pain or improvement in function based on at least a 50% reduction in score and a decrease of at least 20 mm on the VAS or clinical improvement on two of three measures: at least a 20% decline in joint pain and at least 10 mm on the VAS; functional improvement of at least 20% and a drop of at least 4 points on the 40-point JKOM functional subcategory scale; and a decrease of at least 20% on the patient’s global assessment score and a drop of at least 10 points from the total JKOM score.
No serious adverse events occurred in either group, but in the NSAID group, seven patients had symptoms related to a GI tract disorder, and three had allergy to loxoprofen, compared with one patient who received IA-HA who had stiffness in the treated knee after injection, a significant difference.
"The present study suggests that future randomized trials should thus be carried out with a longer duration of follow-up and larger samples, in order to identify optimal knee OA treatment alternatives," the investigators wrote.
They acknowledged the problems inherent with the potential of the trial’s open-label design to introduce bias into the results and the use of a 10% difference in the change in JKOM score to define noninferiority, although this margin was supported by a pilot study.
The authors reported no relevant financial disclosures.