Serum vitamin C levels and risk of non-alcoholic fatty liver disease: results from a cross-sectional study and Mendelian randomization analysis

Abstract

Background and aims: Vitamin C, as an antioxidant, may play a role in the treatment of NAFLD. This research aimed to investigate the association of serum vitamin C levels with the risk of NAFLD and to further examine the causal relationship by Mendelian randomization (MR) method. Methods: The cross-sectional study selected 5,578 participants of the National Health and Nutrition Examination Survey (NHANES), 2005–2006 and 2017–2018. The association of serum vitamin C levels with NAFLD risk was evaluated under a multivariable logistic regression model. A two-sample MR study, using genetic data from large-scale genome-wide association studies (GWAS) of serum vitamin C levels (52,014 individuals) and NAFLD (primary analysis: 1,483 cases /17,781 controls; secondary analysis: 1,908 cases/340,591 controls), was conducted to infer causality between them. The inverse-variance-weighted (IVW) was applied as the main method of MR analysis. A series of sensitivity analyzes were used to evaluate the pleiotropy. Results: In the cross-sectional study, results showed that Tertile 3 group (Tertile 3: ≥1.06 mg/dl) had a significantly lower risk (OR = 0.59, 95% CI: 0.48 ~ 0.74, p < 0.001) of NAFLD than Tertile 1 group (Tertile 1: ≤0.69 mg/dl) after full adjustments. In regard to gender, serum vitamin C was protective against NAFLD in both women (OR = 0.63, 95% CI: 0.49 ~ 0.80, p < 0.001) and men (OR = 0.73, 95% CI: 0.55 ~ 0.97, p = 0.029) but was stronger among women. However, in the IVW of MR analyzes, no causal relationship between serum vitamin C levels and NAFLD risk was observed in the primary analysis (OR = 0.82, 95% CI: 0.47 ~ 1.45, p = 0.502) and secondary analysis (OR = 0.80, 95% CI: 0.53 ~ 1.22, p = 0.308). MR sensitivity analyzes yielded consistent results. Conclusion: Our MR study did not support a causal association between serum vitamin C levels and NAFLD risk. Further studies with greater cases are warranted to confirm our findings.