The serum uric acid/creatinine ratio is associated with nonalcoholic fatty liver disease in the general population

Abstract

Serum uric acid-to-creatinine ratio (sUA/CrR) may be associated with metabolic syndrome components, but limited evidence exists on a relationship between sUA/Cr and NAFLD. Here, we investigated the association between sUA/CrR and NAFLD. We performed a cross-sectional analysis in 3359 subjects who participated in the NHANES 2017–2018 survey and consumed less than 30 and 20 g alcohol (men and women, respectively), with no positive tests of viral hepatitis. Liver steatosis was defined by controlled attenuation parameter and fibrosis by stiffness measurements obtained via transient elastography. We modeled the relationship between NAFLD and relevant demographic, anthropometric, and biochemical variables. sUA/CrR was significantly higher in participants with NAFLD than those without NAFLD. LASSO logit regression showed that only logarithmized age (p = 1.2e-3), waist circumference (WC) (p = 1.8e-5), triglycerides (p = 5e-6), and sUA/CrR (p = 3e-5) were retained in the model. Multivariate logistic analysis demonstrated a significant association between sUA/CrR and NAFLD; the OR for NAFLD of one log(sUA/CrR) increase was 2.61 (95% CI: 1.86–3.68, p < 3e-8) after adjusting for relevant covariables, including aminotransaminase levels and the effect of sUA/CrR remained significant for highest WC quintiles. The model’s predictive power with vs. without sUA/CrR was slightly but significantly better (Auroc: 0.859 ± 0.006 vs. 0.855 ± 0.007, p < 1.1e-2). Mediation analysis showed that SUA/CrR modestly mediates the effect of WC and insulin resistance but not glycohemoglobin on NAFLD. In conclusion, elevated sUA/CrR was significantly associated with NAFLD in the general population. Therefore, kidney function should be closely monitored in NAFLD patients.