Background
Although many guidelines now recommend against hyaluronic acid (HA) injections alone for the nonoperative management of knee osteoarthritis (OA), prior studies have disagreed about whether adding an intra-articular corticosteroid to HA results in sufficient benefit to justify the potential dose-dependent and time-dependent chondrotoxicity of corticosteroids.
Questions/purposes
(1) Do patients with knee OA treated with low-dose intra-articular corticosteroids and HA administered under ultrasound guidance have more favorable patient-reported knee OA–related outcomes from 1 to 26 weeks after treatment than those treated with ultrasound-guided HA injections alone? (2) Do patients treated with a corticosteroid co-injection exhibit more favorable physical functional performance than those treated with injections of HA alone?
Methods
We conducted a randomized, double-blind trial at a small medical center in urban Taipei, Taiwan. We considered those who fulfilled the clinical and radiographic criteria for knee OA, defined by the American College of Rheumatology as a Kellgren-Lawrence grade 2 or 3, as potentially eligible. We excluded those who had a previous knee fracture or surgery and those who had received intra-articular knee injections of HA or corticosteroids within 6 months of the trial. Based on these criteria, 78 patients were eligible, and 76 agreed to be randomized to the co-injection or HA-only groups through block randomization. The groups had 38 patients each; all participants who completed the study follow-up were analyzed in the groups to which they were randomized and treated. There was no crossover. The patients in the co-injection group were treated with a one-shot co-injection of HA with corticosteroids (Hyruan Plus: molecular weight, 3000 kDa, 10 mg/mL HA, 2 mL with triamcinolone acetonide 10 mg/dL, 1 mL) in the first week followed by two injections of HA (10 mg/mL, 2 mL) at a 1-week interval. The patients in the HA-only group received three HA injections (Hyruan Plus: molecular weight, 3000 kDa, 10 mg/mL HA, 2 mL) administered at a 1-week interval. A senior physiatrist performed ultrasound-guided knee injections using the lateral suprapatellar approach. The physiatrist, participants, and outcome assessor were all blinded to the injection contents. No differences were observed between the two groups regarding primary demographic data, baseline WOMAC and Knee Injury and Osteoarthritis Outcome Score (KOOS) scores, and physical functional performance. Three participants experienced localized swelling after the injections, and all symptoms improved after a few days of localized ice pack application. A total of 3.9% of patients (three of 76) had adverse events (mainly localized swelling of the knee, with or without accompanying pain), with no between-group differences in the frequency of adverse events. No severe side effects were reported during or after the injections. The patient-reported outcomes were the WOMAC and KOOS scores. The objective outcome was physical function performance (speed for 10-meter regular and fastest walking, and time for stair-climbing and chair-rising). The primary outcome timepoint was 6 months postinjection. A sample size of 34 participants was required in each group, based on a difference in WOMAC scores of 58 points, a significance level of 0.05, and a power of 0.8.
Results
We found no difference in improvement from baseline between the two groups regarding patient-reported outcomes at any follow-up timepoint, from 1 week after injection to 6 months. At 6 months post-treatment, compared with baseline, there were no between-group differences in improvement between the co-injection group and the HA-only group in terms of WOMAC pain scores (mean difference 1 [95% CI -2 to 3]; p = 0.62), stiffness (mean difference -11 [95% CI -13 to -9]; p < 0.001), and physical function (mean difference -4 [95% CI -5 to -3]; p < 0.001) and KOOS scores for pain, other symptoms, activities of daily living, sports and recreation, and quality of life. Except for chair-rising time at 6 months, which may have represented spurious significance because of the number of comparisons, compared with baseline, we found no differences in improvement between the two groups regarding the regular and fastest walking speed and stair ascent and descent time.
Conclusion
We found that a single co-injection of corticosteroids along with HA did not confer any additional benefit regarding patient-reported outcomes or physical function compared with a three-injection regimen of HA. Therefore, we do not recommend a one-shot co-injection of corticosteroids alongside the three-injection regimen of HA for patients with knee OA, and because prior research found HA to be ineffective, HA should not be used with or without corticosteroids for patients with knee OA.
Level of Evidence
Level I, therapeutic study.
