A national cross-sectional analysis of selenium intake and risk of osteoarthritis: NHANES 2003-2016.

Abstract

The association between dietary selenium intake and arthritis, rheumatoid arthritis (RA), and osteoarthritis (OA) is inconsistent in previous studies and remain unclear. To investigate their relationship, this study was performed. Data from the National Health and Nutrition Examination Survey (2003-2016) were downloaded and further analyzed. Dietary Se intake was classified according to quartiles with quartile 1 (Q1) having the lowest intake and quartile 4 (Q4) having the highest intake. Weighted logistic regression was used to investigate the association between dietary selenium intake and arthritis, RA, and OA. Subgroup analyses were performed to verify the findings. To further examine the non-linear relationship between dietary selenium intake and OA, restricted cubic spline (RCS) was adopted. In the crude model, the highest level of dietary selenium intake was siginificantly associated with decreased risks of arthritis (OR: 0.40, 95% CI: 0.37, 0.44) and rheumatoid arthritis (OR: 0.47, 95% CI: 0.40, 0.54), respectively. In the fully adjusted model, dietary selenium intake was not associated with risk of arthritis and RA (all P > 0.05). Conversely, the risk of OA was noted for participants with higher selenium intake (odds ratio of quartile 4 = 1.33, 95% CI = 1.07-1.65, P < 0.05). In the subgroup analyses, participants with diabetes had a higher risk of OA when ingested high selenium levels than those without diabetes (P < 0.001). The results of RCS showed that significant overall trends were found between dietary selenium intake and osteoarthritis (P for overall < 0.05). However, non-linear association was not detected in this association (P for non-linear > 0.05). Using data from NHANES, this study discloses that high dietary selenium intake might be associated with risk of OA. However, the generalization of conclusion needs further examination because of the limitation of dietary questionnaire survey.