Metabolomic Signatures of Dietary Patterns and Incident Radiographic Knee Osteoarthritis: A Case-Cohort Study From the Osteoarthritis Initiative.

Abstract

Dietary factors related to inflammation, obesity, and metabolism may contribute to osteoarthritis (OA) development. This study examines the relationship between metabolomic signatures of dietary patterns and radiographic knee OA incidence. This case-cohort study included 603 participants from the Osteoarthritis Initiative, comprising 237 incident radiographic knee OA cases and 366 noncases during a 6-year follow-up. Plasma metabolomes were analyzed using ultra-performance liquid chromatography-tandem mass spectrometry. We averaged metabolite levels at baseline and year 1 as the exposure and identified incident cases with those who had a Kellgren-Lawrence grade ≥ 2 in follow-ups. We selected 46 metabolites significantly associated with major food groups based on previous studies. Principal component analysis identified 4 metabolomic signatures related to specific food groups. Weighted logistic regression was used to examine associations between metabolomic signature scores and knee OA incidence. Among the 603 participants, after adjusting for age, sex, and other covariates, the highest quartiles of 2 metabolomic signatures associated with healthy food groups were linked to a reduced OA risk compared to the lowest quartile (odds ratio [95% CI] 0.69 [0.44-1.08] and 0.61 [0.40-0.92], respectively). However, after further adjustment for BMI, these associations weakened. The proportions of the associations mediated through BMI for these 2 signatures were 43.7% and 81.4%, respectively. The other signatures were null. Metabolomic signatures of healthy dietary patterns may be linked to a lower risk of radiographic knee OA, with BMI potentially playing an intermediary role. Further studies are needed to explore causality.