Abstract
BACKGROUND: The association between non-alcoholic fatty liver disease (NAFLD) and osteoarthritis (OA) has not been well elucidated. The aim of the present study was to investigate the association between NAFLD and OA in the US adults. METHODS: A cross-sectional study was performed on participants in the 2017-2018 National Health and Nutrition Examination Survey (NHANES) cycle. NAFLD was defined by the vibration-controlled transient elastography. The diagnosis of OA was based on self-reported data. Weighted multiple logistic regression models and stratified analyses were performed to explore the relationship and verify the stability of the conclusions. Sensitivity analysis using multiple imputation for missing data and propensity score matching (PSM) were performed. RESULTS: In total, 2622 participants [Male: 1260 (47.8%)] were included in this study with a mean age of 48.1 years old (95% CI, 46.6-49.6 years old), containing 317 (12.8%) OA patients and 1140 NAFLD patients (41.5%). A logistic regression indicated a significant association between NAFLD and OA without adjustment [odds ratio (OR) = 2.05; 95% CI, 1.52-2.78]. The association remained stable after adjustment for covariates (OR = 1.72; 95% CI, 1.26-2.34). Sensitivity analysis of missing data with multiple interpolation and PSM found similar results. A significant and consistent association of NAFLD with OA was still observed in each subgroup stratified by age and metabolic syndrome (MetS). Stratified by sex, obesity, and sensitivity c-reactive protein (hs-CRP) category, a statistically significant association was only shown in females, those without obesity, and those without hyper hs-CRP. The results illustrated that the relationship between NAFLD and OA was stable in all subgroups and had no interaction. CONCLUSIONS: NAFLD was positively correlated with OA. Given the current pandemic of NAFLD and OA, clinicians should screen for NAFLD in arthritis patients and intervene early.