Effect of Curcuma longa extract on serum inflammatory markers and MRI-based synovitis in knee osteoarthritis: secondary analyses from the CurKOA randomised trial

Background: Curcuma longa (CL) extract is modestly effective for relieving knee symptoms in knee osteoarthritis (OA) patients; however, its mechanism of action is unclear.

Purpose: We aimed to determine the effects of CL treatment on serum inflammatory markers over 12 weeks and to explore its potential effects on synovitis assessed by contrast-enhanced magnetic resonance imaging (CE-MRI) of the knee.

Methods: Secondary analyses were conducted on the CL for knee OA (CurKOA) trial, which compared CL (n = 36) and placebo (n = 34) over 12 weeks for the treatment of knee OA. Systemic inflammatory markers (TNFα, IL6, and hsCRP) and a cartilage extracellular matrix degradative enzyme (MMP-3) were measured. A subgroup of participants (CL, n = 7; placebo, n = 5) underwent CE-MRI at baseline and a 12-week follow-up.

Results: Over 12 weeks, there were no between-group differences in change in hsCRP, IL-6, and TNFα levels. MMP-3 levels decreased in both CL (-1.31 ng/ml [95%CI: -1.89 to -0.73]) and placebo (-2.34 ng/ml [95%CI: -2.95 to -1.73]) groups, with the placebo group having a slightly greater decrease (1.03 ng/ml [95%CI: 0.19 to 1.88]). Most (10 of 12) sub-study participants had normal synovial thickness scores at baseline. One participant had mild synovitis in each of the placebo and CL groups. Synovitis status was stable for all except two participants, one each in the CL and placebo group, whose synovitis score increased.

Conclusion: This is the first study that explored the effect of CL treatment on local and systemic inflammation using biochemical markers and CE-MRI outcomes on knee OA patients. Secondary analyses from this pilot study suggest that CL is unlikely to have clinically significant effects on systemic (inflammatory and cartilage) or local synovitis (CE-MRI) biomarkers compared to placebo. The mechanism of action for CL effect on pain remains unclear.