Effect of uric acid on lipid metabolism assessed via restricted cubic splines: A new insight

Background: Hyperuricemia can promote both blood lipids and non-alcoholic fatty liver disease (NAFLD). However, the role of the entire uric acid (UA) span, especially low concentrations below hyperuricemia, on lipid metabolism remains unclear. Methods: A cross-sectional study was designed. Data on the age, sex, UA, triglyceride (TG), total cholesterol (TC), high-density lipoprotein (HDL), and low-density lipoprotein (LDL) of 1977 participants, who underwent physical examination, were collected. NAFLD and non-alcoholic fatty pancreas disease (NAFPD) were diagnosed using abdominal ultrasound. Restricted cubic splines (RCS) linear regression model was used to evaluate the effect of the UA span on TG, TC, HDL, and LDL, respectively. RCS logistic regression model was employed to evaluate the effect of the UA span on NAFLD and NAFPD. Results: RCS linear regression model showed that TG was negatively correlated with UA at first, then exhibiting a positive correlation. Meanwhile, HDL was positively correlated with UA at first, then negatively correlated. There was a positive linear correlation between TC and UA (P for nonlinear = 0.578) and a positive nonlinear correlation between LDL and UA (P for nonlinear = 0.021). RCS logistic regression model showed that NAFLD and NAFPD were negatively correlated with UA at first and then positively correlated with UA. Conclusion: our study showed that the entire UA span has a J-shaped effect on some lipids, NAFLD, and NAFPD. Besides, TG and HDL, compared with TC or LDL, may better reflect the status of NAFLD and NAFPD.